Characterization of undecaprenol kinase from Lactobacillus plantarum

112

Biochimica et Biophysics Acta, 574 (1979) @ Elsevier/North-Holland Biomedical Press

112-122

BBA 57399

CHARACTERIZATION OF UNDECAPRENOL LACTOBACILLUS PLANTARUM

JACK R. KALIN and CHARLES Department

(Received

Key

of Biochemistry,

January

words:

M. ALLEN,

University

KINASE FROM

Jr.

of Florida,

Gainesville,

FL 32610

(U.S.A.)

2nd, 1979)

Undecaprenol

kinase;

Isoprene

metabolism;

(Laclobacillusplantarum)

Summary A membrane-bound undecaprenol kinase from Lactobacillus has been identified by observing the ATP-dependent phosphorylation of [ 14C]undecaprenol. The product of this reaction was shown to be [L4C]undecaprenyl monophosphate by comparison of its chromatographic mobilities with authentic undecaprenyl monophosphate. It was also shown that 32P from [T-~~P]ATP was incorporated into undecaprenyl monophosphate. The kinase was partially solubilized by a variety of methods utilizing Triton X-100. Both the membraneassociated and solubilized enzymes required Mg2+, Triton X-100 and dimcthylsulfoxide for activity. The enzyme preferentially phosphorylated the C4s, CsO and Cs5 polyprenols. Geranylgeraniol (C,,) and dolichol (CloO), however, were utilized only 6% and 13% as well as undecaprenol, respectively. Despite the &fold difference in apparent V values, the apparent K, values for dolichol and undecaprenol were both 14 PM. The apparent K, for the nucleotide cosubstrate, ATP, was 2 mM. No other nucleoside triphosphate could substitute for ATP.

Introduction The role of long chain polyprenols as glycosyl carriers in bacterial cell wall biosynthesis has been well established [ 11. The eleven unit isoprene compound, undecaprenyl monophosphate, functions to accept sugars from their activated nucleoside diphosphate forms and transfer them to the growing polysaccharide. In fact, the availability of this monophosphate has been suggested to be rate controlling in peptidoglycan synthesis in Staphylococcus aureus [2,3]. Its availability appears to be dependent on the enzymatic dephosphorylation of undecaprenyl pyrophosphate [ 41 and the rephosphorylation of the free alcohol

113

[2]. The enzyme involved in this latter step, isoprenoid alcohol phosphokinase *, has been characterized in S. aureus as an extremely hydrophobic, membrane-bound protein [ 51. The only other report to date of a similar kinase has been in Klebsiella aerogenes [6,7]. This enzyme was found to be butanol extractable, however, unstable once solubilized. We present here evidence for the presence of this enzyme activity in Lactobacillus plantarum. Furthermore, we describe an alternative method for extraction of the kinase from membranes, which may be useful for isolating other polyprenol kinases which are unstable in butanol. Materials and Methods Sorbents for chromatography were silicic acid (BIO-SIL A, Bio-Rad Laboratories), silica gel G on plastic sheets (Sil-G, Macherey-Nagel), and Kieselguhr G (E. Merck). A”-[ 1-‘4C]Isopentenyl pyrophosphate (0.45 Ci/mol) and farnesyl pyrophosphate of mixed stereochemistries were prepared as described earlier [ 81. trans, trans-Famesyl pyrophosphate, trans, trans, trans-geranylgeraniol, and cis, trans, trails-geranylgeraniol were generously provided by Dr. Tsuneo Baba. [ 1-3H]Dolichol (10.1 Ci/mol) was obtained from New England Nuclear. Undecaprenol was purified from L. plantarum and dolichol from pig liver as previously described [8,9]. Solanesol was generously provided by Aromatics International, Atlanta, GA. All nucleotides and the enzymes, DNAase I and lysozyme, were obtained from Sigma Chemical Co. [y-32P]ATP was generously provided by Dr. Judy Thompson. All other chemicals were of reagent grade. Thin-layer

chromatography

TLC was carried out on silica gel sheets using diisobutylketone/acetic acid/ water (8 : 5 : 1, v/v/v; solvent I), CHC13/CH30H/H20 (65 : 25 : 4, v/v/v; solvent II), and benzene/ethyl acetate (9 : 1, v/v; solvent III). Reverse phase TLC was carried out on Kieselguhr G plates impregnated with paraffin oil [S] using acetone/H,0 (23 : 2, v/v; solvent IV). Polyprenols were visualized by staining with an anisaldehyde reagent [lo]. Substrate

preparation

Polyprenyl pyrophosphate was synthesized from farnesyl pyrophosphate and A3-[ l-‘4C]isopentenyl pyrophosphate using a hydroxyapatite-purified undecaprenyl pyrophosphate synthetase according to the methods described by Allen et al. [ 111. [ “C]PoIyprenol was then prepared by enzymatic hydrolysis of the polyprenyl pyrophosphate by sequential phosphatase treatments yielding first undecaprenyl monophosphate, then the free alcohol [12]. The [14C]polyprenol, suspended in 3.0 ml of CHC13, was applied to a 1 cm X 5 cm column of silicic acid which had been previously equilibrated with 80 ml of CHC13. The column was then washed with CHC13 until no more [14C]polyprenol was eluted (approximately 15-20 ml). This eluate contained 60--80% of the radioactivity originally added as A3-isopentenyl pyrophosphate. prefer to use the name this paper.

* We

undecaprenol

kinase to describe

this enzyme

and will refer to it as kinase in

114

The [ i4C]polyprenol chromatographed as a single radioactive component by TLC in three different solvent systems with RF values characteristic of undecaprenol (Table I) [ 81. Reverse phase TLC in solvent IV was also employed to determine the chain length of the polyprenol. The C& polyprenol, undecaprenol, was shown to be the only product derived from preparations using truns, truns-farnesyl pyrophosphate as substrate by comparing its chromatographic mobility with cochromatographing L. pluntarum undecaprenol or ficaprenol. The polyprenols derived from preparations using famesyl pyrophosphate of mixed stereochemistries, however, contained major and minor components, corresponding to the Cs5 and CsO polyprenols, respectively [ 121. The [ 14C]undecaprenol prepared using trans, truns-farnesyl pyrophosphate had a calculated specific activity of 3.64 Ci/mol. For routine assays, however, the mixture of [14C]polyprenols, prepared using farnesyl pyrophosphate of mixed stereochemistries, was used and had a calculated specific activity of 3.47 Ci/mol (assuming a Cs5 : CsO ratio of 5 : 3 as determined from peak areas on reverse phase TLC). Enzyme assay [‘4C]Polyprenol, (1400 dpm, 0.18 nmol) in 20 ,ul of ethanol, was added to 0.25 ml of a solution containing 200 mM Tris buffer pH 8.0, 40 mM MgC&, 5 mM NaF, and 40% (v/v) Me,SO. Sufficient Triton X-100 was added to bring the final concentration to 0.15% (v/v). Water and the enzyme preparation were then added to make a final volume of 1 ml. The reaction was initiated by the addition of 20 ,ul of 0.15 M ATP and incubated for 15 min at 33°C. These assay mixtures were equilibrated at 33°C for 5 min prior to addition of the enzyme and ATP. The reaction was terminated by extraction with 2.5 ml of CHC13/ CH30H (1 : 1) and the lower phase aspirated and saved. The aqueous phase was then reextracted with 1 ml CHC13. The organic phases were pooled, and CH30H added to give a final concentration of CHC1JCH30H of approximately 2 : 1. Polyprenol and polyprenyl monophosphate contained in the extracts were analyzed by a column chromatographic technique. The CHC1&H30H extracts were applied to 0.5 cm X 2 cm columns of DEAE-Sephadex LH20 prepared as described elsewhere [ 131. The unconverted polyprenol washed through with 5 ml TABLE

I

TLC MOBILITIES

OF THE KINASE

SUBSTRATE

AND PRODUCT

Mobilities expressed as RF values were determined using Sil-G sheets. Undecaprenyl monophosphate standard was prepared from undecaprenyl pyrophosphate using the crude phosphatase from M. lysodeihticus. Data in brackets indicate the reference number. Solvent I Substrate Undecaprenol Product Undecaprenyl Undecaprenyl Undecaprenyl

monophosphate monophosphate pyrophosphate

* For ficaprenyl

monophosphate.

standard

0.95 0.94 0.60 0.58 0.56 0.33

181

system II

III

0.95

0.68 0.74 0.00 0.00

0.53 0.52 0.51 0.09

[231

*

C&t]

115

of CHC13/CH30H (1 : 1). The charged, phosphorylated product was eluted with 3 ml of CHC1&H30H (1 : 1) containing 0.25 M ammonium formate. Ammonium formate was removed from the eluted fraction by washing with 1.5 ml of water, and the CHC13 phase was transferred to a scintillation vial. The solvent was evaporated over low heat and radioactivity was quantitated by counting in toluene scintillation fluid containing 0.4% Omnifluor (New England Nuclear). Enzyme preparation L. plantarum (ATCC 8014) was grown at 30°C in stationary culture to late log phase, harvested, and washed as previously described [ 81. 30 g of cells were then suspended in 250 ml of a solution containing 1 M sucrose, 30 mM Tris buffer, pH 8.0, 1 mM glutathione, and 62 mg lysozyme, then incubated at 37°C for 30 min. All subsequent steps, unless otherwise noted, were performed at 0---5°C. The cells were collected by centrifugation at 30 000 X g for 30 min, shocked by resuspension in 270 ml of 10 mM Tris/maleate buffer, pH 6.6, and subsequently passed twice through a French pressure cell at 18 000 lb/ inch*. This slurry was then treated with several milligrams of deoxyribonuclease I at 37°C for 3-5 min or until the viscosity decreased. The resulting suspension was centrifuged at 30 000 Xg for 30 min and the supernatant discarded. The white, fluffy upper layer of the pellet was gently removed and saved. A suspension of the remaining brown layer (unbroken cells) in 40 ml of 10 mM Tris buffer (pH 7.5) was passed through the pressure cell, and centrifuged as before. The fluffy upper layer of the pellet was pooled with the other membraneous fraction, resuspended by homogenization in 40 ml of 50 mM Tris buffer, pH 7.5, and centrifuged once again. The large membraneous layer was carefully removed and rehomogenized in an equal volume of 50 mM Tris buffer, pH 7.5. Typical yields were approximately 250 mg of membrane protein/30 g wet weight of bacteria. This material represents the membrane-associated enzyme and was stable for several months when stored at -20°C at a protein concentration of 8-10 mg/ml. Enzyme extraction Attempts to solubilize the kinase activity from the membrane fraction using butanol, butanol/acetic acid, or butanol/pyridinium acetate (pH 4.2) [ 141 were unsuccessful. The detergents, deoxycholate, cetylpyridinium chloride, and sodium dodecyl sulfate, all at 1% (w/v) were also ineffective in extracting activity. Sonication (2 min) of the membranes, however, with 1% deoxycholate supplemented with 0.5-2s Triton X-100 [15] followed by centrifugation at 120 000 Xg resulted in the extraction of 24-26s of the kinase activity. Similarly, sonication (2 min) in 0.5% Triton X-100/6 M urea [16] solubilized up to 45% of the activity. An alternative procedure giving similar yields of solubilized activity, without using urea, involved extraction of the membrane preparation with 0.5% Triton X-100 (v/v) containing 0.5 M NH&l at a final protein concentration of 4-5 mg/ml. NH&l proved to be a more effective salt than either NaCl or KCl. This suspension was adjusted to pH 8 with NH,OH, then frozen and thawed [17] three times using a solid CO,/acetone bath and centrifuged at 120 000 X g

116

TABLE

II

UNDECAPRENOL Membranes 25°C

were

and

120

000

sent __..

the

KINASE suspended

allowed X g for

number

Solubilization

SOLUBILIZATION

to 60 of

in 0.02

stand

min

was

for

assayed

experiments

M Tris

3040

FROM buffer

min. for

(PH

The

enzyme

CRUDE 8.0)

containing

supernatant activity

MEMBRANES the

fraction

and

protein.

indicated

components

obtained The

after

numbers

at O’C

centrifugation

in parentheses

or at

repre-

performed.

technique

% solubilized

from

membranes _____

Enzyme 1%

Triton

x-100

at o”c

1%

Briton

X-100

at 25’C

(4)

1%

Triton

X-100

+ 10%

glycerol

at O°C

1%

Triton

X-100

+ 10%

glycerol

at 25’C

1 mM

EDTA

1 mM

EDTA

glycerol * Final

at O’C wash at 25”

suspension

(2)

(3)

(2) (7)

16-19

4040

15-25

39-54

16

3* followed

by

1%

Triton

X-100

+ 10%

Protein

4147

39-71

*

at O°C

activity

40-80

23-32 5

25-35 8-14

C (6) for

solubilization

contained

5 mM

Tris,

PH

8.0.

for 1 h. The resulting supernatant was dialyzed overnight with one buffer change against 50 ~01s. of 10 mM Tris buffer (pH 8.0) containing 0.5% Triton X-100. Once the activity was extracted with either urea- or salt-supplemented Triton X-100, the enzyme remained active and soluble following dialysis to remove the urea or salt. This was important because urea concentrations greater than 0.5 M or salt concentrations greater than 0.1 M in the assay inhibited the kinase activity. The dialysate contained approximately 30-50% of the membrane-associated enzyme activity at a 1.5-fold purification. Solubilized enzyme activity was stable for several weeks when stored at -20°C at a protein concentration greater than 2 mg/ml. The NH4C1/Triton-solubilized enzyme was the preparation used in most of the experiments described here, unless otherwise noted. The use of higher detergent concentrations at 25°C [ 181, however, proved to be more effective in extracting enzyme activity. The addition of 10% glycerol as a stabilizing agent [19] also improved enzyme solubilization (Table II). Enzyme extractability under these conditions depended on the protein/ detergent ratio. At least 2.2-2.7 mg Triton X-lOO/mg protein was required for maximum solubilization. The pH of the extractions was also important since below pH 8.0, an opaqueness was often observed in the supernatants. It was also observed that some protein was removed without any enzymatic activity by washing the crude membranes with dilute EDTA [20] (Table II). This afforded a higher purification upon subsequent extraction with Triton/glycerol and gave a purification of 5-6-fold over the untreated membrane preparation. Protein determinations were made according to the method of Lowry et al. [21] using bovine serum albumin as the standard. For samples containing Triton X-100, sodium dodecyl sulfate was included in the assay [22]. Results Product

characterization

The product

of the kinase activity

cochromatographed

with or had RF values

117

characteristic of undecaprenyl monophosphate in three different TLC solvent systems (Table I). Undecaprenyl pyrophosphate chromatographed quite differently in solvents I and II. Experiments were performed with [y-32P]ATP to verify the presence of a monophosphorylated product. One tube was incubated for 90 min with [‘4C]undecaprenol (0.62 PM, 4750 dpm) and unlabelled ATP (3 mM), while a second tube was incubated for 90 min with unlabelled undecaprenol (50 fl) and [y-32P]ATP (0.45 mM, 342 000 dpm). After incubation, the mixtures were pooled, extracted with CHC13/CH30H and analyzed by TLC in solvent I. A radiochromatographic scan (Fig. 1A) shows the presence of 14C-labelled substrate at an RF of 0.95 and both 14C- and 32P-labelled products at an RF of 0.62. An alternate scanning method revealed a single component at an RF of 0.62 corresponding to 32P-labelled undecaprenyl monophosphate (Fig. 1B). Hence, the same component was observed monitoring either 14C or 32P incorporation into the product. The product was also characterized, for comparative purposes, by paper chromatography followings biosynthesis using unlabelled undecaprenol and [y-32P]ATP, according to the method of Sandermann and Strominger [ 141. To verify the identity of the polyprenol moiety of the kinase product, the monophosphate was prepared with the membrane-associated kinase as described in Fig. 1 (scaled up five-fold) and purified using DEAE-Sephadex LH20. This material was then, in turn, hydrolyzed using the yeast phosphatase and analyzed by reverse phase TLC in solvent IV. The resulting alcohols migrated as Cs5 and CsO polyprenols in proportions similar to those of the substrate. These results suggest that the kinase was indeed producing polyprenyl monophosphate and that both Cs5 and C& were equally phosphorylated.

ORlOGlN

5

10

FRONT 15

Alcohol

Concentration

co/,)

-

Fig. 1. Radiochromatogram scans of the kinase product analyzed by TLC using solvent I. (A) An upper detector scan revealed both 14C and 32P. (B) A lower detector scan which failed to detect 14C through the plastic backing of the TLC sheet, revealed only 32P. Incubation conditions were those described in the text except that the final concentration of Triton X-100 and NaF were 0.3% and 5 mM. respectively. Fig. 2. Dependence of membrane-associated enzyme activity on alcohol concentration. The effect of methanol (m). ethanol (0). propan (A) and butanoi (+) were investigated in the presence of 10% Me2SO using incubation conditions described in Materials and Methods. The effects of ethanol in the absence of MeZSO (0) were also investigated. Protein concentrations were all 0.41 mg/ml.

118

Reaction conditions Temperature and pH optima were determined in experiments with membrane-associated enzyme. The temperature optimum was observed at 30 35”C, hence 33°C was chosen as the routine assay temperature. The pH vs. activity profile was quite broad showing optimal activity from pH 7.5 to 9.0, hence, pH 8.0 was routinely used. The kinase activity in unextracted membranes was activated by Me,SO with activity increasing linearly up to a final concentration of 10% (v/v). We attempted to substitute ethanol for Me,SO in the assay buffer, but found it was less effective (Fig. 2). However, the level of enzyme activity was greater if, in addition to Me,SO, low concentrations of ethanol were included. Even greater stimulation was observed with methanol, while less activation was observed with propanol and butanol. Kinase activity decreased as all alcohol concentrations were increased. Consequently, undecaprenol was added to the assay tubes in 20 4 of ethanol giving a final alcohol concentration of 2% (v/v). Other reaction components of the buffer were varied individually in order to optimize conditions for the 0.5 M NH4C1/Triton X-100-solubilized enzyme. Optimal Me,SO concentration were observed at 20% (v/v) for this enzyme preparation, however, 10% was routinely used for enzyme assays. The Triton X-100 concentration giving optimal activity changed slightly with protein concentration (Fig. 3), when using either the membrane-associated or extracted enzyme. We chose 0.15% detergent to use routinely. Higher detergent concentrations markedly inhibited activity. Sandermann and Strominger [14] have suggested that the activity of the kinase from S. aureus may also be dependent on an optimal protein/detergent ratio. It is apparent that great care must be taken to assay the enzyme under near optimal conditions in order to make meaningful comparisons of the activity of different protein preparations. In an attempt to inhibit possible phosphatases, NaF was included in the incu-

0: Trlton

02

x IOC

07 0.4 Concentratlir

0.5

SC

vr

Fig. 3. Dependence of enzymic activity on Triton X-100 concentration. The effects of Triton X-100 were determined using the membrane-associated enzyme (0) or NHaCl/Triton X-lOO-solubiJized enzyme (0) under the incubation conditions described in Materials and Methods at the protein concentrations indicated. The solubilized enzyme was prepared as described in Materials and Methods except that the final Triton X-100 concentration for extraction was 0.2%.

119

bation buffer. Kinase activity was stimulated at 1.25 mM, however, it was markedly inhibited at concentrations above 5 mM. In fact, enzyme activity appeared to be quite sensitive to most salts. We observed a differential inhibition with several anions above 10 mM concentration with the following order of decreasing sensitivity: NaF > Na2HP04, K,HPO, > NaCl, KCl, NH4C1 > TrisHCl. The enzyme itself, however, was stable to high salt since addition of salt to, and then removal from an enzyme preparation did not affect activity. The concentration of Mg2+ required for optimal enzyme activity was 10 mM with inhibition occurring at higher concentrations. Mn2’ and Co2+ were only at their optimal concentrations of about 50% and 25% as effective as Mg2’ 2.5 mM and 10 mM, respectively. Other divalent catipns tested, namely Ca”, Cu’+, and Zn2’ could not effectively substitute for Mg2+. The solubilized enzyme showed a linear dependence with time of incubation up to 30 mm. Routine assays were, therefore, performed using 15-min incubations. A phosphatase control was performed under these standard assay conditions using [ ‘%]undecaprenyl monophosphate as substrate rather than [ 14C]undecaprenol. The extent of hydrolysis of the monophosphate to free alcohol was less than 20% of the extent of phosphorylation of undecaprenol. Substrate specificity

ATP alone

served as the nucleoside

triphosphate

cosubstrate

for both the

Fig. 4. Double-reciprocal plots of activity versus polyprenol concentration. (A) [ 1 4 Cl Undecaprenol prepared from trans,tmns-famesyl pyrophosphate (2800 dpm) was supplemented with increasing amounts of unlabelled undecaprenol, purified from L. plantarum, to achieve the indicated substrate concentrations. (B) [l-3HlDobchol (30 000 dpm) was supplemented with increasing amounts of unlabelled dolichol purified from pig liver to achieve the indicated substrate concentrations. The amount of product formed was calculated from the radioactivity found in the product and the specific activity of the polyprenol being used as substrate in each experiment. All experiments were carried out under the incubation conditions described in Materials and Methods. Protein concentrations were 0.23 mg/ml.

120 TABLE III POLYPRENOL

SUBSTRATE

SPECIFICITY

Unlabelled substrates and 0.83 PCi y[ 32P] ATP (50 MM and 2.9 mM, respectively) were incubated with the salt Triton-solubilized enzyme under the conditions described in Materials and Methods. These mixtures were then extracted twice with 2.5 ml CHC13/CH3OH (2 : 1) and the pooled CHCl3 phases washed with 1 ml HzO. The organic phases were then evaporated in vials and counted using a toluene-based scintillation fluid as described in Materials and Methods. TLC analysis in solvent I of duplicate extracts showed that no radioactivity was present in these organic phases as ATP or inorganic phosphate. Unlabelled substrate concentrations were determined by the method of Amenta [241 using squalene as the standard. Each value has been corrected for 32~ incorporated in the absence of added polyprenol. Substrate

nm~l 32~ incorporated

Undecaprenol Dolichol Solanesol Cis, trans-geranylgeraniol

2.51 0.26 2.13 0.12

into product

membrane-associated and solubilized enzyme activities. When the nucleotides AMP, CTP, UTP and GTP were each used at 3 mM concentrations, no activity was seen. A small amount of activity was supported by ADP which could be attributed to the presence of an adenylate kinase activity catalyzing the formation of ATP and AMP from ADP. An apparent K, for ATP of 2 mM was observed using 29 ~_IMundecaprenol and the solubilized enzyme. using the solubilized The apparent K, and V values for undecaprenol enzyme at 3 mM ATP were 14 /IM and 15 nmol undecaprenyl phosphate/mg protein per 15 min, respectively (Fig. 4A). Dolichol, the eukaryotic CIOOpolyprenol, also had an apparent K, of 14 m (Fig. 4B). However, the V (1.9 nmol dolichyl phosphate/mg protein per 15 min) was only 13% of that of undecaprenol. Solanesol, the all tr~n.s-C,~ polyprenol from tobacco leaves, served as an excellent substrate, exhibiting approximately 85% of the activity of undecaprenol at equimolar concentrations (Table III). The shorter chain, C20 polyprenols, trans, tram, trans-geranylgeraniol and cis, trans, trans-geranylgeraniol, however, were very poor substrates. These compounds supported only 5% of the activity of undecaprenol at equimolar concentrations. Discussion We have developed a column assay technique for undecaprenol kinase which quantitates the conversion of radiolabelled undecaprenol into its phosphorylated derivative. The previously described method [14] uses paper chromatography to monitor the incorporation of 32P from [T-~~P]ATP into a polyprenyl phosphate. The column method provides a more rapid and convenient means of separating the phosphorylated polyprenols for subsequent analysis and characterization. Its application in a manner similar to that described here might be of more general use in assaying enzymes such as the polyprenyl phosphate-dependent glycosyltransferases. The radiolabelled polyprenol substrate was prepared by sequential phosphatase treatments of undecaprenyl pyrophosphate and identified as undecaprenol by TLC. The kinase products prepared using either [ “C]undecaprenol or

121 [y.32P]ATP cochromatographed with or had RF values characteristic of undecaprenyl monophosphate. The kinase appeared to preferentially phosphorylate C&-X5, polyprenols. Shorter (C,,) and longer (Coo) chain substrates were not utilized as well. The specificity for the polyprenol substrates tested was similar to those described for the S. aureus [ 21 and K. aerogenes [25] enzymes with the exception that solanesol (C,,) was as good a substrate as undecaprenol. The apparent V values for undecaprenol and dolichol differed greatly whereas the apparent K, values~were the same, i.e. 14 a. These K, values are considerably less than the value determined for the partially purified kinase from S. aureus (57 @I) [ 141. The apparent K, of 2 mM for ATP is considerably greater than the value determined for the S. aureus kinase (57 PM) [ 141. The two bacterial kinases are similar in their behavior with respect to divalent cations, pH vs. activity profiles, inhibition by butanol, and the Triton X-100 concentration where optimal activity occurs [ 2,141. These kinase activities differ, however, in several respects. Undecaprenol kinase, first described by Higashi et al. [2], was solubilized from membranes by extraction into acidic butanol. We were unable to extract any kinase activity from L. pluntarum membrane preparations using this technique. For this reason we pursued more conventional approaches to membrane solubilization. A combination of chemical and physical treatments resulted in the solubilization of up to 80% of the membrane-associated kinase activity. This is in contrast to the 6% solubilization obtained by butanol extraction of the S. aureus enzyme. The use of Triton X-100 in solubilizing the kinase offers an alternative method to butanol extraction which has been unsuccessful or yielded an unstable enzyme in both L. pluntarum and K. aerogenes [7]. The L. plunturum enzyme was quite sensitive to salt, being inhibited over 50% by 0.2 M NaCl. The S. uureus kinase, however, was stimulated by salt concentrations up to 0.5 M [5]. In addition, the L. planturum enzyme was almost completely inhibited by 50 mM phosphate buffer, the concentration used to determine the pH optimum of the S. uureus enzyme. While these two enzymes catalyze the same metabolic process, they differ in several enzymologically important parameters. The purification of the S. aureus kinase has made available an excellent system with which to study lipid-protein interactions and enzyme activity [ 26-291. We chose to isolate and characterize undecaprenol kinase from L. plan tarum in preparation for analogous studies on lipid activation. Preliminary results have indicated that both L. plantarum and synthetic phospholipids are capable of reactivating a solubilized enzyme preparation depleted of phospholipid by gel filtration chromatography (Kalin, J.R., unpublished observations). The metabolically related enzyme, undecaprenyl pyrophosphate synthetase, has already been partially purified from this organism and some of its lipid requirements described [ 301. The identification of both the synthetase and kinase as membrane-associated enzymes, involved in making available undecaprenyl phosphates, affords us an opportunity to compare these enzymes in the same organism with respect to their lipid requirements. The biosynthesis of dolichyl monophosphate utilizing the L. pluntarum enzyme has made it possible to develop an assay system for identifying an

122

enzymatic activity responsible for dolichol phosphorylation in mammalian cells. This activity has the unusual requirement for CTP as the nucleoside triphosphate substrate instead of ATP [ 131. Acknowledgements We would like to acknowledge the preliminary work on the description of the phosphokinase and development of the column assay method by Jonathan Sack. This work was supported by a grant from the NIH (GM 23193). References 1 Hemming, F.W. (1974) in Biochemistry of Lipids (Goodwin, T.W., ed.). PP. 39-97, Butterworths London 2 Higashi, Y., Siewert. G. and Strominger, J.L. (1970) J. Biol. Chem. 245, 3683-3690 3 Willoughby. E.. Higashi, Y. and Strominger. J.L. (1972) J. Biol. Chem. 247. 5113-5115 4 Goldman, R. and Strominger. J.L. (1972) J. Biol. Chem. 247, 5116-5122 5 Sandermann. H. and Strominger, J.L. (1971) Proc. Natl. Acad. Sci. U.S. 68, 2441-2443 6 Poxton, I.R., Lomax, J.A. and Sutherland, I.W. (1974) J. Gem Microbial. 84. 231-233 7 Poxton, I.R. and Sutherland, I.W. (1976) Microbios. 15.93-103 8 Keenan, M.V. and Allen, C.M. (1974) Arch. Biochem. Biophys. 161, 375-383 9 Burgos, J., Hemming, F.W., Pennock, J.F. and Morton, R.A. (1963) Biochem. J. 88, 470482 10 McSweeney, G.P. (1965) J. Chromatogr. 17.183-185 11 Allen, C.M., Keenan, M.V. and Sack, J. (1976) Arch. Biochem. Biophys. 175, 236-248 12 Baba. T. and Allen, C.M. (1978) Biochemistry 17, 5598-5604 13 Allen. C.M.. Kalin, J.R.. Sack, J. and Verizzo, D. (1978) Biochemistry 17, 5020-5026 14 Sandermann, H. and Strominger, J.L. (1972) d. Biol. Chem. 247, 5123-5131 15 Reimer. B.L. and Widnell, C.C. (1975) Arch. Biochem. Biophys. 171. 343-347 16 Garewsl, H.S. and Wasserman, A.R. (1974) Biochemistry 13,4063-4071 17 Schwartz, N.B. and Roden, L. (1975) J. Biol. Chem. 250,5200-5207 18 Janson. J.-C. (1975) J. Gen. Microbial. 88,209-217 19 Dean. W.L. and Tanford, C. (1978) Biochemistry 17, 1683-1690 20 Steck, T.L. and Yu, J. (1973) J. Supramol. Struct. 1. 220-232 21 Lowry, O.H. Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 22 Wang. C.S. and Smith, R.L. (1975) Anal. Biochem. 63,414-417 23 Warren, C.D. and Jeanloz, R.W. (1973) Biochemistry 12. 5038-5045 24 Amenta. J.S. (1964) J. Lipid Res. 5, 270-272 25 Sutherland, I.W. (1977) In Surface Carbohydrates of the Prokaryotic Cell (Sutherland, I.W.. ed.), PP. 27-96, Academic Press. London 26 Sandermann. H. (1974) Eur. J. Biochem. 43,415-422 27 Gennis, R.B. and Strominger, J.L. (1976) J. Biol. Chem. 251, 1264-1269 28 Gennis, R.B. and Strominger, J.L. (1976) J. Biol. Chem. 251,1277-1282 29 Genti, R.B., Sinensky, M. and Strominger, J.L. (1976) J. Biol. Chem. 251, 1270-1276 30 Allen, C.M. and Muth, J.D. (1977) Biochemistry 16,2908-2915