THEJOURNAL OF BIOLOGICAL
CHEMISTRY
The Purification and Characterizationof CTP:Phosphorylcholine
Cytidylyltransferase from Rat Liver*
Paul A. Weinhold, MaryEllen Rounsifer, and Douglas A. Feldman
From the Veterans Administration Medical Center and the Department of Biochemstry, University of Michigun,
Ann Arbor, Michigan 48105
We have purified CTP:phosphorylcholine cytidylyl-
transferase from rat liver cytosol 2180-fold to a spe-
cific activity of 12,250 nmol/min/mg of protein. The
purified enzyme was stable at -70 “C in the presence
of Triton X-100 and 0.2 M phosphate. The purified
enzyme gave a single protein and activity bandon
nondenaturing polyacrylamide electrophoresis. Sepa-
ration by sodium dodecyl sulfate-polyacrylamide elec-
trophoresis indicated that the purified enzyme con-
tained subunits with M , of 39,000 and 48,000. Gel
filtration analysis indicated that the native enzyme
was a tetramer containing two 39,000 and two 48,000
subunits. The purified enzyme appeared to bind to
Triton X-100 micelles, one molecule of tetramerlmi-
celle. Maximal activity wasobtained with 108 PM phos-
phatidylcholine-oleic acid vesicles (8-10-fold stimu-
lation). Phosphatidylglycerol produced a 4-5-fold in-
crease in activity at10 p ~ The
. pH optimum and true
K , values for CTP and phosphorylcholinewere similar
to those reported previously for crude preparations of
cytidylyltransferase. The overall behavior of cyti-
dylyltransferase during purification and subsequent
analysis suggested that it has hydrophobic properties
similar to those exhibited by membrane proteins.
The reaction catalyzed by CTP: choline-phosphate cyti-
dylyltransferase (EC 2.7.7.15) is a principal rate-limiting step
inthe biosynthesis of phosphatidylcholine. For example,
pulse-chase experiments with hepatocytes ( l ) , isolated Type
I1 pneumonocytes from adult (2) and fetal (3) lung, a myoblast
cell line (L,) (4),and HeLa cells (5) indicated thatthis
reaction was the rate-limiting step in the incorporation of
radioactive choline into phosphatidylcholine. Furthermore,
variations in the rate of phosphatidylcholine synthesis, pro-
duced in cell cultures by a variety of agents and conditions,
have been correlated with alterations in the rate of cytidylyl-
transferase-catalyzed reaction (5-12).
The mechanisms which regulate cytidylyltransferase activ-
ity are not understood. However, the results from a variety of
studies have produced some insight into the regulatory prop-
erties of the enzyme. The activity of soluble forms of cyti-
dylyltransferase is stimulated by anionic phospholipids and
by phosphatidylcholine fatty acid vesicles (13-15). Binding of
the enzyme to the vesicles appears to be associated with the
* This work was supported by the Veterans Administration and by
grants from the National Institute of Child Health and Human
Development and from the Biomedical Research Council, University
of Michigan. The costs of publication of this article were defrayed in
part by the payment of page charges. This article must therefore be
hereby marked “advertisement” in accordance with 18U.S.C. Section
1734 solelyto indicate this fact.
5104
vel. 261, No. 11, Issue of April 15,pp. 5104-5110,1966
Printed in U.S.A.
(Received for publication, August 26, 1985)
stimulation (15). Translocation of enzyme activity from cy-
tosol to microsomal membranes is correlated with an increase
in the intracellular conversion of phosphorylcholine into
phosphatidylcholine (16-22) and suggests that membrane-
bound enzyme may be the active form (23). Translocation is
promoted by freefatty acids (20-22), byapparent modification
in membrane composition (16-19), and by 12-0-tetradecanoyl
phorbol 13-acetate (24). Furthermore, evidence suggests that
phosphorylation of cytidylyltransferase also may alter the
activity and properties of cytidylyltransferase (25).
Insightinto the relative importance of these enzymatic
properties and their relevance to different physiological and
metabolic processes would be greatly facilitated by the avail-
ability of purified enzyme. Although attempts to purify cyti-
dylyltransferase have been reported (26, 27), a reproducible
procedure which gives highly purified stable enzyme has not
been obtained. In thispaper we describe the purification and
properties of cytidylyltransferase. The procedure is reproduc-
ible and yields highly purified enzyme which is stable for at
least 2 months when stored at -70 “C in the presence of
Triton X-100.
EXPERIMENTALPROCEDURES
Materials-Female rats (200-250 g) were purchased from Holtz-
man Company, Madison, WI. Phosphatidylcholine (type XI-E from
egg yolk), phosphatidylglycerol (from egg yolk phosphatidylcholine),
N-octyl P-D-glucopyranoside (octyl glucoside),Triton X-100, Nonidet
P-40, CTP, oleic acid, phenylmethylsulfonyl fluoride, and activated
charcoal (acid washed with phosphoric and sulfuric acids) were pur-
chased from Sigma. Phospho[methyE-’4C]choline was obtained from
New England Nuclear. DEAE-Sepharose CL-GB was from Pharmacia
Fine Chemicals. Hydroxyapatite and Bio-Gel A-1.5m were obtained
from Bio-Rad.
Cytidylyltransferase Assay-Cytidylyltransferase activity was de-
termined as follows. Each reaction contained 1.6 mM phospho-
ryl[’4CH3]choline (1200dpm/nmol), 3.0 mM CTP, 10.0 mM ME?+,2.0
mM EDTA, 45 mM NaCl, 0.2 mM phosphatidylcholine-oleic acid (1:l
molar ratio), and 65 mM Tris, pH 7.5. The phosphatidylcholine-oleic
acid vesicles wereprepared by sonication as described previously (15).
The reaction was stopped by the addition of 0.5 mlof charcoal
suspension (6% charcoal in 20 mM phosphorylcholine). Water (7 ml)
was added and the mixture vortexed briefly. The charcoal was sedi-
mented by centrifugation and the water decanted. The charcoal pellet
was suspended in 7 ml of water, vortexed, and the contents poured
onto aMillipore filter (0.45 ,A,25-mm diameter) which waspositioned
in a vacuum-filtering manifold (AmiconCorp.). The charcoal was
collected by vacuum and washed on the filter with 3 volumes of 7 ml
of H20. Scintillation vials were placed under the filter andthe
charcoal extracted four times with 1.0 ml of extraction solvent
(ethanol/ammonium hydroxide/water, 19O:ll:llS). Each round of
extraction involved a 5min incubation before the vacuum wasapplied.
This method resulted in 64-68%recoveryof CDP-choline anda
background of 60-100 dpm. The recovery of CDP-choline was deter-
mined in each set of assays by adding a known amount of radioactive
CDP-choline to a complete assay mixture. All assays were corrected
for recovery of CDP-choline.
 CTP:Phosphorylcholine
Cytidylyltransferase
Purification 5105
Electrophoresis-Polyacrylarnide gel electrophoresis was per- tant was discarded, and the precipitate was resuspended in
formed on 150-mm thick gels using the buffer system described bybuffer A (two-thirds the volume of the original cytosol) by
Laemmli (28). Electrophoresis under nondenaturing conditions washomogenization with a motor-driven Teflon pestle glass ho-
performed in 7% gel.which contained 0.1% Nonidet P-40.
mogenizer. The pH of the mixture was 7.2. The mixture was
Electrophoresis in the presence of sodium dodecyl sulfate (SDSl)
was performed on 12.6% gels with a 4% stacking gel. Samples werecentrifuged at 20,000 X g for 20 min, and the supernatantwas
discarded.
prepared by dialysis against 10 mM Tris buffer, pH 6.8, and dried in
a Savant Speed Vac concentrator. The dried samples were dissolved The pH 5.0 precipitate was suspended by homogenization
in 60 mM Tris, pH 6.8,containing 2% SDS, 5% P-mercaptoethanol, in 20 mM octyl glucoside dissolved in buffer A (7 ml of octyl
and 10% glycerol. After electrophoresis, gels were fixed for 1 h in
glucoside solution/100 mg of protein in the original cytosol).
20% trichloroacetic acid, stained overnight in 0.1% Coomassie R-250
Phenylmethylsulfonyl fluoride was added to obtain a 0.5 mM
in 10% acetic acid, 50% ethanol, and destained by diffusion in ethanol,
5% acetic acid mixtures. concentration.
After stirring at room temperature for 20 min the mixture
In order to determine cytidylyltransferase activity after nondena-
was centrifuged at 20,000 X g for 20 min. The supernatant
turing electrophoresis, 3-mm slices of gel were extracted in 100 pl of
50 mM Tris, 0.15 M NaCl, 1.0 mM EDTA, pH 7.5,overnight at 4 "C. was removed for subsequent fractionation on DEAE-Sepha-
The liquid was separated from the gel, and 100 p l of reaction ingre-
rose. The octyl glucoside extract contained 0.64 k 0.13 ( N =
dients were added to theextract. The mixture was incubated at 37 "C
5) nmol of phospholipid/mg of protein or 59.8 k 10 pmol of
for 2 h. The reaction was stopped and processed by the usual charcoal
method. total phospholipid. Since 225-250 pmol of phosphatidylcho-
Analytical Methods-Protein was determined by the Bradford line had been added to thecytosol before the pH 5 precipita-
method (29) using reagent purchased from Pierce Chemical Co. tion, about 25-30% of the phospholipid was extracted by octyl
Bovine serum albumin was used as standard. Standards were prepared
glucoside.
with the same concentrations of enzyme buffer ingredients that wereA DEAE-Sepharose column (1.6 x 12 cm) was equilibrated
in the sample.
Phospholipid was quantified by total phosphorus as described with buffer A (300ml of buffer at 40 ml/h). All elution
previously (30).Phospholipid was extracted from enzyme prepara- solutions were prepared in buffer A. The octyl glucoside
tions and column fractions using the Bligh and Dyer method (31).extract was applied at a flow rate of 25-30 ml/h followed by
100 ml of buffer A. The column was sequentially eluted with
RESULTS AND DISCUSSION equilibration buffer, a 200-ml gradient of0.15-0.3 M NaC1,
150 ml of 0.3 M NaC1, and 50 ml of 0.4 M NaCl containing 50
Purification of Cytidylyltrurzsferase-Female rats (200-250 mM octyl glucoside.
g) were anesthetized with chloral hydrate and the liver per- Cytidylyltransferase activity was eluted by 0.4 M NaC1, 50
fused thoroughly with 0.9% NaC1. All procedures were per- mM octyl glucoside. The activity was eluted in two overlapping
formed at 4 "C. Liver (90-100 g) was homogenized at 5 ml/g peaks (Fig. 1).A large amount of phospholipid was eluted
tissue in buffer A (150 mM NaC1,50 mM Tris, 1.0 mM EDTA, with the first peak. Other experiments indicated that elution
2 mM DTT, 0.025% sodium azide, pH 7.4). Phenylmethylsul- of the enzyme using a 0-50 mM octyl glucoside gradient in 0.4
fonyl fluoride, dissolved in isopropyl alcohol, was added im- M NaCl separated the two peaks. Subsequent purification of
mediately to the homogenates to achieve a 1.0 mM final each peak separately gave purified cytidylyltransferase with
concentration. the same specific activity and electrophoretic profile. Thus,
Cytosol was isolated by sequential centrifugation at 10,000 the two peaks apparently are not due to separate enzyme
X g for 10 min and at 125,000 x g for 60 min. A phosphati-
species but probably result from co-elution of enzyme with
dylcholine-oleic acid suspension (5 mM phosphatidylcholine, different amounts of phospholipid. We routinely combined
10 mM oleic acid) was prepared by mixing the appropriate both peaks for further chromatography on hydroxyapatite.
volumes of stock solutions of phosphatidylcholine and oleic The pooled enzyme contained 1.22 & 0.2 (n = 5) pmol of
acid in a round-bottom flask. The solvent was evaporated phospholipid/mg of protein or a total phospholipid of 13 k 4
under vacuum in a rotaryevaporator. Buffer A was added and pmol. This was about 22% of the phospholipid in the octyl
the mixture was sonicated with a probe sonicator for 60 min glucoside extract.
at 4-10 "C. This mixture was added to thecytosol to produce A column of hydroxyapatite (1.6 x 7.5 cm) was equilibrated
a phosphatidylcholine to protein ratioof 80 nmol/mg protein. with 150 ml of buffer A at a flow rate of 20 ml/h. The pooled
The cytosol-lipid mixture was stirred at room temperature for
20 min. The addition of phosphatidylcholine-oleic acid sig-
nificantly increased the yield of enzyme through the subse- I r n B C D I
quent acetic acid precipitation and octyl glucoside extraction.
Preliminary experiments indicated that maximal yields were
obtained with 1:2 molar mixture of phosphatidylcholine to
oleic acid at thedesignated lipid to protein ratio.
-
0
-1
E
The pH of the cytosol-lipid mixture was reduced to 5.0 by 2.52-
the addition of 1.0 M acetic acid. The mixture was placed on GO8- 2.00a
ice for 10 min. The mixture was centrifuged at 20,000 X g for f 0.6 - 1.5 %
z
20 min. Cytidylyltransferase activity wasrecovered in the 204- 1.0 E
b
precipitate. The supernatant contained no detectable cyti- &02- 0 5 1:
01 0 2
dylyltransferase activity. Approximately 53% of the cytosol 0 20 40 60 80 100 120 140 180
160 200
activity and 14%of the protein were recovered in the precip- FRACTION NUMBER
IC"----------3.0 ml .:- l.Ornl4
itate. The activity measurements with the precipitate may
not be reliable since the activity recovered in the octyl glu- FIG. 1. Chromatography of octyl glucoside extract on
coside extract was 72% of the cytosol activity. The superna- DEAE-Sepharose. The column was equilibrated with 0.15 M NaCl,
50 mM Tris, pH 7.5,2 mM EDTA, 2 mM dithiothreitol and eluted
-~~ ~~ ~

sequentially with equilibration buffer ( A ) ,a linear gradient of 0.15-

The abbreviations used are: SDS, sodium dodecyl sulfate; DTNB, 0.3 M NaCl ( B ) ,0.3 M NaCl (C), and 50 mM octyl glucoside, 0.4 M
5,5'-dithiobis-(2-nitrobenzoic acid);NEM, N-ethylmaleimide; NaCl (Dl.All elution solutions were prepared in the equilibration
pCMB, p-chloromercuribenzoate; DTT, dithiothreitol. buffer. CT, cytidylyltransferase.
5106 CTP:Phosphorylcholim Cytidylyltransferase Purification
enzyme from the DEAE-Sepharose column was diluted with the protein in the cytosol was removed by the combined pH
buffer A to 20 mM octyl glucoside and applied to thecolumn 5 precipitation and octyl glucoside extraction. The use of
at 15 ml/h. Preliminary experiments indicated that the elu- detergents to solubilize and elute cytidylyltransferase from
tion of cytidylyltransferase activity depended upon both the the columns was critical in thepurification process.
phosphate concentration and the type and concentration of The purified enzyme has been stored at -70 "C in 0.03%
detergent. Purification was achieved by washing the column Triton X-100, 0.2 M potassium phosphate-buffer A for 2
with potassium phosphate and detergentconcentrations months without loss of activity. Approximately 30% of the
whichremoved contaminatingproteinsbut did not elute activity was lost after 2 weeks at 4 "C. A decrease in NaCl
cytidylyltransferase activity. We washed the column with 150 below 75 mM led to rapid loss of activity. The presence of 0.2
ml of 0.15 M potassium phosphate in buffer A followed by 75 M phosphate and Triton X-100 was important in order to
ml of 0.2 M potassium phosphate. All phosphate-buffer A retain enzyme activity during storage. In fact, dilution of the
mixtures were adjusted to pH7.5. An octyl glucoside gradient, phosphate and Triton X-100 concentrations 10-fold resulted
150 ml of 0-50 mM octyl glucoside in 0.2 M potassium phos- in 50% loss of activity within 1 h at 4 "C.
phate-buffer A, was applied. This was followed by sequential Electrophoretic Characterization-Purified cytidylyltrans-
washes with 75 ml of 50 mM octyl glucoside, 0.2 M phosphate- ferase gave a single band on nondenaturing polyacrylamide
buffer A, and 60 mlof 100 mM octyl glucoside, 0.2 M potassium electrophoresis (Fig. 3). Enzyme activity coincided with the
phosphate-buffer A. Cytidylyltransferase activity was eluted single protein band. When electrophoresis was attempted
with 0.03% Triton X-100, 0.2 M potassium phosphate-buffer without Nonidet P-40 in the gels, the protein did not enter
A (Fig. 2). the gel.
Several peaks of protein andphospholipid were eluted prior Comparison of the migration of cytidylyltransferase to pro-
to the Triton X-100 elution of cytidxlyltransferase. More than tein standardsgave an estimated molecular weight of 240,000.
98% of the phospholipid applied to the column was eluted Since the electrophoretic migration of native proteins depends
prior to theenzyme. We could not detect phospholipid in the on both the charge and size of the protein, this molecular
final enzyme preparation. However, the limit of detection was weight is an approximation.
such that phospholipid content below 300 nmol/mg protein Gel electrophoresis in the presence of sodium dodecyl sul-
would not have been measurable. fate indicated that thepurified cytidylyltransferase contained
A summary of the purification is shown in Table I. This two subunits which appeared to be present in equal amounts
data is the average of five preparations. More than 96% of (Fig. 4). The molecular weights of the two subunits were

1
determined by RF versus logM , plots to be 38,900 and 47,800
(referred to as 39,000 and 48,000 subunits)..Lane B contains
fi approximately 5 times more protein than lane A . The smaller
subunit does not band as sharply when large amounts of
P I protein are applied. However, the larger sample enables a
1 300 better estimate of the purification. Two faint protein bands
can be seen in lune B in addition to thetwo major bands. We
estimated that the faint bands contributed no more than 1-
2% of the total protein.
Although we included phenylmethylsulfonyl fluoride in the
3..
f 150 0.6 preparation buffers to inhibit protease activity, the possibility
E still exists that the subunits mayhavebeen produced by
-2 E0.2
0.1
0.4 - protease activity. Therefore, a separate experiment was con-
t 100
V Ea
._
- ducted in which cytidylyltransferase was purified in the pres-
50 0.2 g ence of a mixture of protease inhibitors. The mixture was
VI
2 added immediately following homogenization and again dur-
0 o n ing the octyl glucoside extraction. Thus, thehomogenate and
0 20 40 60 80 100 120 140 160
FRACTION NUMBER the octyl glucoside extraction buffer contained 0.25 mM phen-
FIG. 2. Chromatography of the DEAE poolof cytidylyl- ylmethylsulfonyl fluoride, 1.0 mM diisopropyl fluorophos-
transferase (CT) on hydroxyapatite. The column was equili- phate, 5 pg/ml pepstatin A, 0.1 pg/ml p-tosyl-L-lysine chlo-
brated in the same buffer as described in the legend to Fig. 1. The romethyl ketone, 5 pg/ml soybean trypsin inhibitor, and5 pg/
column was eluted sequentially at a flow rate of 15 ml/h with 0.15 M ml antipain. The final purified enzyme had the same specific
phosphate (A),0.2 M phosphate ( B ) ,a linear gradient of 0-50 mM activity and yield as the routine preparation. More impor-
octyl glucoside in 0.2 M phosphate (c),50 mM octyl glucoside, 0.2 M
phosphate (D), 100 mM octyl glucoside, 0.2 M phosphate ( E ) ,and 0.5 tantly, SDS-polyacrylamide gel electrophoresis analysis
mM Triton X-100, 0.2 M phosphate ( F ) . All solutions were made in showed the same two subunit bands. Furthermore, enzyme
equilibration buffer. prepared in the absence of any protease inhibitor also con-

TABLEI
Summary ofcytidylyltransferase purification
The results are the average f S.D. of five separate preparations using 95-100 g of rat liver for each.
Specific Purification Yield
Step Protein Activity activity
mg min" nmol nmol min" mgprotein-I -fold %
Cytosol 3,129 f 168 17,510 f 1,677 5.7 f 0.07 1 100
Octyl glucoside extract 87 f 6 11,920 f 822 139 -C 14 25
69 k 2 f 2.5
DEAE-Sepharose 117,080
f 1.3 f 560 639 f 53 115 k 9 42 f 5
Hydroxyapatite 0.09 f 0.03 1,043 f 260 12,250 f 1,390 2,180 k 160 6.4 f 2
 CTP:Phosphorylcholine
Cytidylyltransferase
Purification 5107
CytaseActivity
(dpm CDPcholine)

3000.
2000.
IO00 -
0
3
2.0
FIG. 3. Polyacrylamide electrophoresis of purified cyti-
dylyltransferase under nondenaturing conditions. A 7% poly-
acrylamide gel with a 3% stacking gel, both containing 0.1% Nonidet I.0
P-40, was used. Ten micrograms of protein was applied.
I
0 0.I

FIG.5. Gel filtration chromatography of cytidylyltransfer-

ase. Purified cytidylyltransferase (U cytidylyltransferase
), at
the DEAE-Sepharose stageof purification (M), and liver cytosol
in the presence of Triton X-100 (W-- 4)and liver cytosol without
Triton X-100 (E- - - 0 ) were analyzed on a Bio-Gel A-1.5m column.
The column was equilibrated and eluted with 0.15 M NaCl, 50 mM
Tris, 2 mM EDTA, 2 mM DTT, 0.5 mM Triton X-100, 0.2 M phos-
phate, pH7.5, for standards andsamples exceptin thecase of cytosol
inthe absence of Triton X-100, in which case the column was
equilibrated in the same buffer without Triton X-100. The column
was calibrated with a Bio-Rad standard protein mixture containing
thyroglobulin, 570,000; aldolase, 153,000;ovalbumin, 44,000; and my-
oglobin, 17,000. The void volume (76 ml) and internal volume (187
ml) were determinedwith 0.06-pm polystyrenebeads and3H20,
respectively. Four separate calibration profiles were obtained and the
Kd values averaged. The K d values for standards were the same with
or without Triton X-100. The plot of log M,uewus K d is shown in
the inset with the symbol (X) marking the position of cytidylyltrans-
ferase. The Kd values from each run varied by no more than +5%.

5)).This is the same apparent M , as that estimated by

A B C
FIG.4. Separation of purified cytidylyltransferase by pol-
yacrylamide gel electrophoresis in the presence of SDS. Lane
A contained 3 pgof enzyme. Lane B contained 15 pgof enzyme.
Proteinstandards, from the top:phosphorylase b, 92,500; bovine
serumalbumin, 66,200; ovalbumin 45,000; carbonicanhydrase,
21,500; lysozyme, 14,400.
tained the same subunit composition. These results strongly
suggest that the subunits are not produced by proteolysis
during the preparation.
Gel Filtration Analysis-Purified cytidylyltransferase
eluted from a Bio-Gel A-1.5m gel filtration column as a single
uniform peak with a K d of 0.41 (Fig. 5). This corresponded to
an apparent M , of 240,000 when compared to standard pro-
teins eluted underthe same conditions(log M , versus K d (Fig.
electrophoresis. The equilibration and elution of the column
with 0.5 mM Triton X-100 and 0.2 M potassium phosphate
were essential for recovery of enzyme activity when either
purified enzyme or enzyme at theDEAE stage of purification
was applied. The elution K d for cytidylyltransferase at the
DEAE-Sepharose stage of purification differed from that of
the highly purified enzyme. The apparent M , of the DEAE
enzyme activity was 330,000. In order to gain more informa-
tion about the elution differences, we applied liver cytosol
onto the same column in the presence and absence of Triton
x-100.
Cytidylyltransferase activity in cytosol eluted with an ap-
parent M , of 355,000 when analyzed in the presence of the
Triton X-100 whereas the activity eluted with an apparent
M , of 174,000 when the chromatography was performed with-
out TritonX-100 in the elution buffer.
We interpreted these results to mean that purified cyti-
dylyltransferase eluted as a tetramer of two 39,000 and two
48,000 subunits bound to one micelle of Triton X-100. The
concentration of Triton X-100 required for recovery was
above the critical micelle concentration, which has been re-
ported to be about 0.3 mM (31). The number of monomers of
Triton X-lOO/micelle has been reported to range from 111to
140 depending upon the conditions underwhich the measure-
ments were made (32, 33). The number of monomers/micelle
decreases with decreasing temperature (32). Since we per-
formed the gel filtration chromatographyat 4 "C, it is reason-
able to use the value of 111 monomers/micelle. This gives a
micelle weight of 66,700. Since tetramers of cytidylyltransfer-
ase subunits would have a M , of 174,000,a tetramer bound to
one micelle of Triton X-100 would have an apparent M , of
5108 CTP:Phosphorylcholine
Cytidylyltransferase
Purification
240,700 which is essentially the same as the observed M , of
240,000 obtained with purified cytidylyltransferase. This
interpretation agrees with the generalization that theamount
of detergent bound to proteins is often that of one micelle of
Triton/moIecule of native protein (34). Furthermore,the fact
that cytidylyltransferase activity in cytosol in the absence of
Triton X-100 eluted from the gel filtration column with a M,
of 174,000 supports the conclusion that the native enzyme
consists of a tetramer of two 39,000 and two 48,000 subunits.
Presumably, the higher apparent M, of cytidylyltransferase
in the DEAE preparation and in cytosol when eluted in the
presence of Triton X-100 reflects the binding of other hydro-
phobic proteins totheTriton X-100 micelles along with
cytidylyltransferase. Whether these proteins have any meta-
bolic relationship to cytidylyltransferase remains to be deter-
mined.
The binding of cytidylyltransferase to micelles of Triton X-
100 is consistent with our previous observations that cyti-
dylyltransferase binds to phospholipid vesicles (15) and with
previous observations that lipids converted crude prepara-
tions of cytidylyltransferase from an apparent M, of 180,000-
200,000 to large aggregate forms (13, 35).
Lipid Requirementsfor EnzymeActiuity-Phosphatidylcho-
line-oleic acid vesicles (1:l molar ratio) gave the greatest
stimulation of purified cytidylyltransferase activity (Fig. 6).
Although phosphatidylglycerol was active at lower concentra-
tions than phosphatidylcholine-oleic acid, it did not stimulate
the activity to the same extent as the higher concentrations
of phosphatidylcholine-oleic acid. Oleic acid alone gave min-
imal stimulation andphosphatidylcholine alone was inactive.
Phosphatidylethanolamine and lysophospkatidylethanola-
mine were also inactive (data not shown). Both phosphati-
dylcholine-oleic acid and phosphatidylglycerol increased the
apparent V, without changing the apparent K,,, values (data
not shown).
Although the pattern of lipid stimulation for the purified
enzyme was the same as that obtained previously for crude
enzyme preparations ( E ) , maximal stimulation of purified
enzyme was achieved at lower concentrations of lipids. This
was particularly striking withphosphatidylglycerol which pro-
duced a 4-fold increase in activity at 2.5 pM.
All enzyme preparations had some activity without added
lipid. This was between 10-20% of the maximal activity. The
preparationsallcontainedTriton X-100. Typical enzyme
assays contained 0.002-0.003% Triton. Additional Triton X-
100 inhibited the lipid-independent activity but did not inhibit
the activity in the presence of optimalconcentrations of
phosphatidylcholine-oleic acid. We do not know whether the
p
LM Lipid
FIG.6. Stimulation of cytidylyltransferase by lipids. cyt-
dylyltransferase activity was determined in thepresence of phospha-
tidylcholine-oleic acid vesicles (1:l molar ratio) ( 0 )phosphatidyl-
glycerol (0),oleic acid (O), or phosphatidylcholine (=).
lipid-independent activity is an inherent property of the en-
zyme or whether it reflects a small amount of lipid associated
with purified enzyme.
Kinetic Properties-The true K, values for CTP andphos-
phorylcholine were determined. Initially, velocity versusCTP
concentrations at different phosphorylcholine concentrations
and velocity versus phosphorylcholine concentration at dif-
ferent CTP concentrations were determined. The apparent
V,,, values were obtained by reciprocal plots. A secondary
plot of apparent V,,, values versus phosphorylcholine or CTP
concentrations were made to obtain true K,,, values. These
plots were linear. The K , for CTP was 0.29 m ~and , the K,
for phosphorylcholine was 0.14 mM. Both secondary plots
intersected at the same V,,, value. The maximal velocity
obtained for the enzyme preparation used in the experiments
was 9048 nmol/min/mg of protein. This specific activity was
similar to the specific activity of 9200 nmol/min/mg of protein
obtained by the routine assay procedure. This indicates that
the routine assay is optimal and that substrate inhibitiondoes
not occur at the concentrations used in the assay. The K,
values are essentially the same as previously reported by Choy
et al. (26) for partially purified cytidylyltransferase from rat
liver.
The activity of the enzyme was maximal at pH 7.0 wit.h a
20 and 40% decrease from maximal at pH 6 and 8, respec-
tively. The activity at pH6.5 and 7.5 was nearly the same as
that at pH 7.0. Although the maximal activities obtained with
phosphatidylcholine-oleic acid or with phosphatidylglycerol
were different, the shapes of the pHcurves were identical.
Inactivation by Sulfhydryl Reagents-Purified cytidylyl-
transferase was inactivated by 5,5'-dithiobis-(2-nitrobenzoic
acid) (DTNB), N-ethylmaleimide (NEM), and p-chloromer-
curibenzoate (pCMB), butnot by iodoacetamide (Fig. 7).
pCMB was the most potent, producing 80% inactivation at
0.6 p ~ NEM. and DTNB apparently reacted with the same
type of sulfhydryl since the slope of the log % control versus
time plots was the same, reflecting the same first order rate
constants (Fig. 7). pCMB, on the other hand, appeared to
react with the same type of sulfhydryl after short preincuba-
tions but after longer preincubations reacted with a second
type of sulfhydryl.
The inactivation produced by the reagents was not reversed
by dithiothreitol(20 mM) or by mixtures of dithiothreitol and
100
80
-
P 60
E
$ 40
8 20
0
0.5 I .o I .5
Time fmin)
FIG.7. The effect of sulfhydryl reagents on cytidylyltrans-
ferase activity. In the inactivation versus reagent concentration
experiments, cytidylyltransferase (40 ng) was preincubated for 15 min
at 37 "C in 50 r l of 0.15 M NaCl, 50 mM Tris, 0.2 mM DTT which
contained varying concentrations of iodoacetamide (01,DTNB (O),
NEM (O),or pCMB (m). After the preincubation, 50 pl of substrate
plus phosphatidylcholine-oleic acid was added, and a 10-min reaction
was performed. In the timed experiments, cytidylyltransferase (70
ng) was preincubated for the length of time indicated on the plot in
the presence of either 2 mM DTNB, 1 mM NEM, or 0.7 pM pCMB.
At the end of the preincubations, an assay for enzyme activity was
performed as described above. In both types of experiments, the %
control was calculated from the appropriate preincubated controls.
 CTP:Phosphorylcholine
Cytidylyltransferase
Purification 5109
TABLEI1 high concentrations of phosphate were required for inhibition,
The effect of CTP (6 mM) and phosphorylcholine(3 mM) on the we were interested in the effect of phosphate because of its
inactivation by NEM (3 mM) pronounced stabilizing effect on the enzyme.
Cytidylyltransferase (36 nmol/min/ml) was preincubated at 4 "C Cytidylyltransferase was not inhibited by ATP, CDP, or
for 10 min in 0.15 M NaCl, 50 mM Tris, 2 mM EDTA, 0.2 mM DTT CMP at concentrations up to 3.0 mM at 0.4 mM CTP.
or in the same buffer plus the additions indicated. After the 10 min, Divalent cations inhibited cytidylyltransferase activity.
50 ~1 was added to substrate mixtures that were adjusted so that the
final concentration of all substrates was the same. Enzyme activity Calcium and Mn2+at 3 mM produced 20 and 42% inhibition,
was determined a t 37 "C. The results are theaverage k S.D. for three respectively, while 3 mM Znz+completely inhibited the activ-
separate experiments. ity. Theseinhibitions were observed in the routine assay
Additions nmol/min/ml 76 Inactivation system containing 10 mM Mg2+.
31 +- 2.0
In summary, we have obtained for the first time highly
None
NEM 8.2 f 1.1" 73 purified cytidylyltransferase. The purified enzyme is stable
when stored at -70 "C in the presence of Triton X-100 and
CTP 37 f 2.1b 0.2 M potassium phosphate. Its pHoptimum, K,,, values, and
CTP + NEM 28 t 2.6"2*4'.' lipid requirements are similar to those observed previously
with impure enzyme. Although definite proof of the subunit
Phosphorylcholine 35 f 2.3
composition of the purified enzyme requires detailed amino
Phosphorylcholine + NEM 17 f 1.521"*'
acid analysis of the subunits, the data in this paper suggests
Significantly different values, p < 0.01. that native cytidylyltransferase contains two nonidentical
Significantly different values, p < 0.05. subunits which are associated as a tetramer. This tetramer
Significantly different values, p < 0.02.
possesses strong hydrophobic properties which is consistent
with the previous observations that cytidylyltransferase can
be induced to bind to biological membranes. Although the
mechanisms which modulate the properties of cytidylyltrans-
ferase to cause translocation of the enzyme are not under-
stood, it is tempting tospeculate that one of the subunitsmay
function in the catalytic process and theother may be involved
in the interaction with membranes. Thus, agents which pro-
mote translocation, such as fatty acids, may do so by inter-
acting with one of the subunits. On the other hand, results by
Kent and co-workers (16-19, 36) on the effects of treatment
20 of cells with phospholipase C suggest that modification of
10
biological membrane surfaces is an important component of
the translocation phenomena. The availability of purified
0 cytidylyltransferase will enable a more detailed analysis of
0 IO 20 30 40 50 60 70 80 90 100 110 120
mM Phosphate this potentially important regulatory process.
FIG. 8. The inhibition of cytidylyltransferase activity by
phosphate. In the double reciprocal plots, the 0 refers to activities REFERENCES
in the presence of 50 mM phosphate. 1. Pritchard, P. H., and Vance, D. E. (1981)Biochem. J . 196,261-
267
0-mercaptoethanol (20 mM plus 10 mM). However, the addi- 2. Post, M., Batenburg, J. J., Schuurmans, E., and van Golde, L. M.
tion of dithiothreitol prevented the inactivations and termi- G. (1982)Biochim. Biophys. Acta712,390-394
3. Post, M., Batenburg, J. J., van Golde, L. M. G., and Smith, B. T.
nated the inactivation when added after varying periods of (1984)Biochim. Biophys. Acta795,558-563
preincubation with NEM or DTNB. 4. Cornell, R. B., and Goldfine, H. (1983)Biochim. Biophys. Acta
Both CTP andphosphorylcholine provided protection from 750,504-520
inactivation by NEM (Table 11).CTP was significantly more 5. Vance, D. E., Trip, E. M., and Paddon, H. B. (1980)J . Bid.
effective than phosphorylcholine. These results suggest that Chem. 255,1064-1069
6. Paddon, H. B., and Vance, D. E. (1980)Biochim. Biophys. Acta
sulfhydryl groups are necessary for optimal catalytic function. 620,636-640
Inhibition of Cytidylyltransferase Actiuity-cytidylyltrans- 7. Pelech, S.L., Pritchard, P. H., and Vance, D. E. (1981)J. Biol.
ferase activity was inhibited by potassium phosphate (Fig. 8). Chem. 256,8283-8286
Maximal inhibition was approached at 120 mM phosphate, 8. Vigo, C., and Vance, D. E. (1981)Biochem. J . 200,321-326
although only about 60% inhibition was obtained. Inhibition 9. Pritchard, P. H., Chiang, P. K., Cantoni, G. L., and Vance, D. E.
of enzyme activity produced by 240 mM phosphate could be (1982)J. Biol. Chem. 257, 6362-6367
10. Pelech, S. L., Pritchard, P. H., and Vance, D. E. (1982)Biochim.
reversed by simple dilution of the enzyme sample prior to Biophys. Acta 713, 260-269
assay. Thus, the inhibition was reversible. Dialysis of the 11. Lim, P. H., Pritchard, P. H., Paddon, H. B., and Vance, D. E.
enzyme to remove phosphate resulted in considerable loss of (1983)Biochim. Biophys. Acta 753, 74-82
activity and, therefore, was not a satisfactory way to demon- 12. Haagsman, H. P., van den Heuvel, J. M., van Golde, L. M. G.,
strate reversibility of the inhibition. and Geelen, M. J. H. (1984)J. Biol. Chem. 259, 11273-11278
13. Feldman, D. A., Kovac, C. R., Dranginis, P. L., and Weinhold, P.
The inhibition was noncompetitive toward both phosphor- A. (1978)J. Bid. Chem. 253,4980-4986
ylcholine and CTP. The addition o f NaCl to achieve the same 14. Feldman, D. A., Dietrich, J. W., and Weinhold, P. A. (1980)
ionic strength as 120 mM KZHPO, produced 20% inhibition. Biochim. Biophys. Acta620, 603-611
Thus,the phosphateinhibitionisnot due justto ionic 15. Feldman, D. A,, Rounsifer, W. E., and Weinhold, P. A. (1985)
strength. The fact that maximal inhibition by phosphate is Biochim. Biophys.Acta 429,429-437
about 60% and that the inhibition by phosphate is noncom- 17. Sleight, R.,
16. Sleight, and Kent,C. (1980)J. Bid. Chem. 255,10644-10650
R., and Kent, C. (1983)J. Biol. Chem. 258, 824-830
petitive with substrates suggests that phosphate may act by 18. Sleight, R., and Kent, C. (1983)J. Bioi. Chem. 258,831-835
changing the conformation of the enzyme. Although relatively 19. Sleight, R., and Kent, C. (1983)J. Biol. Chem. 258;836-839
5110 CTP:Phosphorylcholine
Cytidylyltrunsferase
Purification
20. Pelech, S. L.,Pritchard, P.. H., Brindley, D. N., and Vance, D. E. Comrnun. 72,714-719
(1983) J.Biol.
Chem. 258,6782-6788 28. Laemmli, U. K. (1970) Nature 227, 680-685
21. Pelech, S. L., Cook, H. W., Paddon, H. B., and Vance, D. E. 29. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254
(1984) Biochim.Biophys. Acta 795,433-440 30. Weinhold, P. A., and Villee, C. A. (1965) Biochim.
Biophys. Acta
22. Weinhold, P. A., Rounsifer, M. E., WiIliams, S. E., Brubaker, P. 106,540-550
G., and Feldman, D. A. (1984) J. Biot.Chem. 259, 10315- 31. Bligh, E. G., and Dyer, W. J. (1959) Can. J. Biochem. Physiol.
10321 37,911-917
23. Weinhold, P. A., Feldman, D. A., Quade, M. M., Miller, J. C., 32. Helenius, A., McCaslin, D. R., Fries, E., and Tanford, C. (1979)
and Brooks, R. L. (1981) Biochim.Biophys. Acta 665,134-144 Methods Enzymol. 56, 734-749
24. Pelech, S. L., Paddon, H. B., and Vance, D. E. (1984) Biochim. 33. Kagawa, Y.(1972) Biochim. Biophys. Acta 265, 297-378
Biophys. Acta 795,447-451 34. Tanford, C., and Reynolds, J. A. (1976) Biochim.
Biophys.
Acta
25. Pelech, S. L., and Vance, D. E. (1982) J. Biol.Chem. 257,14198-457,133-170
14202 35. Choy, P. C., Farren, S. B., and Vance, D. E. (1979) Can. J.
26. Choy, P. C., Lim, P. H., and Vance, D. E. (1977) J. Biol. Chem. Biochern. 57,605-612
252, 7673-7677 36. Wright, P. S., Morand, J. N., and Kent, C. (1985) J. Biol. Chem.
27. Choy, P. C., and Vance, D. E. (1976) Biochem. Biophys.
Res. 260,7919-7926