Mechanisms of short-term (second range) regulation of the activities of enzymes of lipid and phospholipid metabolism in secretory cells

MECHANISMS OF SHORT-TERM (SECOND RANGE) REGULATION THE ACTIVITIES OF ENZYMES LIPID AND PHOSPHOLIPID METABOLISM IN SECRETORY CELLS

OF OF

HANS-DIETER SOLING, WERNER FEST, KATRIN MACHOCZEK, TYEDE SCHMIDT, HERMANN ESSELMANN and MANFRED FISCHER Abteilung Klinische Biochemie, Universit~itG6ttingen, D-3400 G6ttingen, Federal Republic of Germany

INTRODUCTION

In many eukaryotic cell types, signal transmission involves activation of the phosphoinositide pathway (for a review see Ref. 2). This activation results from the receptor-mediated increase in the activity of phosphoinositidespecific phospholipase(s) C. However, activation of phospholipase C is not the only step in the metabolism of phospholipids exhibiting altered activity during the early phase of signal transduction. We have recently shown in experiments with guinea pig parotid acinar cells (1) that during the first 10-20 sec following stimulation of either 13-adrenergic or muscarinic receptors a significant rise in the turnover of diacylglycerols occurs simultaneously with an increased transfer of fatty acids from acyl-CoA to neutral lipids. An analysis of 1,2-sn- and 2,3-sn-diacylglycerols revealed further that stimulation via 13-adrenergic agonists led to the activation of a triglyceride lipase (1). In order to get an indication of the mechanism(s) leading to the rapid changes of diacylglycerol metabolism, we have analyzed the activities of the enzymes designated (1) to (7) in Figure i in isolated guinea pig parotid gland lobules 30 sec after stimulation with isoproterenol (13-adrenergic agonist) or carbachol (muscarinic agonist). In addition, we have analyzed whether calcium/phospholipid binding proteins (calpactins) might be able to affect the activities of various phosphoinositide-specific phospholipases of the C-type. M A T E R I A L S AND METHODS

Isolated guinea pig parotid lobules were prepared and incubated as described previously (3). They were stimulated with either 2 x 10-5 M isoproterenol or 10-5 M carbachol. After 30 sec of stimulation, the lobules 35

H.-D. SOLING, et al.

36

Glycerol

'I

clpGlycero-P 2 1~r~-~'-------- Acyl-CoA

- co, °AI,27:

Phollne_ .ho~ph at idy I-

5~

ACyI-CoA

Triglycerides

!

t~FFA

CoA-SH

2,3 - Diacylglycerides

r l 3- Monoglycerides

Glycerol

FIG. 1. Main reactions involved in the production and consumption of phosphatidate, 1,2-sn- and 2,3-sn-diacylglycerols.The reactionsmarked (1) through (7) represent the enzymes analyzed in this work.

were chilled and homogenized in 5 ml of ice-cold TKS-buffer (Tris-Cl 50 mM, KCI 25 mM, DTE 1 mM, sucrose 0.25 M; pH 7.4). The 1,000 g supernatant was spun for 60 min at 100,000 g and the resulting pellet ("microsomes") resuspended in TKS-buffer to a protein concentration of 4--6 mg/ml. The 100,000 g supernatant was designated "cytosol". Enzyme activities were measured as follows: glycerol kinase (final concentrations) Tris-Cl 0.1 M, pH 7.4, MgCI2 2 mM, ATP 1.35 mM, [3H]glycerol 0.5 mM (specific activity 104 dpm/nmol); glycerolphosphate acyltransferase according to (4); lyso-phosphatidate acyltransferase and diacylglycerol acyltransferase according to (5). Phosphatidate phosphohydrolase was measured according to (6) plus and minus Mg 2+ in the presence and absence of 10 mM fluoride (7) in order to eliminate the contribution of phospholipase(s) C to the cleavage of phosphatidate; CDP-diacylglycerol synthetase according to (8), and D A G kinase under the following conditions (final concentrations) Tris-Cl 0.1 M, NaF 20 mM, deoxycholate 1 mM, DTE 0.5 mM, MgCI2 10 mM, diolein 20/xM, (~/-32p)-ATP 20/zM (specific activity4 X 104 dpm/nmol).

REGULATION OF ENZYMES OF LIPID METABOLISM

37

The catalytic subunit of cAMP-dependent protein kinase was purified from beef heart according to (9), Ca2+/calmodulin-dependent protein kinase II from rabbit skeletal muscle according to (10). Protein phosphatases 1 and 2A were purified from rabbit skeletal muscle by a modification of the m e t h o d given in (11), protein phosphatase 2B from bovine brain according to (12) by Dr. G. Mieskes in our department. Lipocortin 1 (p36), protein I (p82) and lipocortin 2 (p35) were purified from beef lung according to (13). Phosphatidylinositide-specific phospholipase C was enriched from rat liver cytosol, human platelet membranes and rat brain membranes according to (14), (15), and (16), respectively. Phospholipase C activity with phosphatidylinositol as substrate was measured for 20 min at 37°C under the following conditions (final concentrations): Tris-maleate 50 mM, p H 6.0, CaCI 2 2 mM, [3H]phosphatidylinositol 60/zM (specific activity 10,000 dpm/nmol). Phospholipase C activity with phosphatidylinositol-4,5bisphosphate was performed for 20 min at 37°C under the following conditions (final concentrations): Hepes 18 mM, pH 7.0, LiCI 8.5 mM, MgCI 2 2 mM, CaCI 2 0.1 mM, [3H]PIP2 0.6/.tM (specific activity 1,000 dpm/pmol). U n d e r these conditions the tests were linear with time. RESULTS

Effects of Stimulation with Isoproterenol or Carbachol on Enzyme Activities The activities of glycerol kinase, glycerolphosphate acyltransferase, and CDP-diacylglycerol synthetase did not significantly change during stimulation (results not shown here). Mg2+-dependent phosphatidate TABLE 1. EFFECTS OF STIMULATION FOR 30 SEC OF ISOLATED GUINEA PIG PAROTID GLAND LOBULES WITH CARBACHOL (10-5 M) AND ISOPROTERENOL (2 x 10-5 M) ON THE ACTIVITIES OF LYSO-PHOSPHATIDATE ACYLTRANSFERASE, DIACYLGLYCEROL KINASE AND DIACYLGLYCEROL ACYLTRANSFERASE Enzyme activity as percent of unstimulated controls (%) Lyso-PA-acyltransferase DAG-kinase DAG-acyltransferase Carbachol Isoproterenol Carbachol Isoproterenol Carbachol Isoproterenol 392 ± 149 (174-669)

221 774 ± 138 ± 270 (115--473) (324-1147)

253 ± 96 (252-390)

750 ± 330 (377-1229)

239 ± 79 (136--356)

The results are related to the activities measured in microsomes from unstimulated cells which were set to 100%. Under the experimental conditions, the mean activities in microsomesfrom unstimulated cells for lyso-phosphatidateacyltransferase, diacylglycerol kinase and diacylglycerolacyltransferasewere 164 (range 109 - 274), 21 (range 11 - 31), and 153 (range 109 - 282) pmol/mg proteirdmin, respectively. The values given represent mean values + S.D. from 8 separate experiments. The lowest line represents the range of values obtained for each experimental condition.

H.-D. SOLING, et al.

38

phosphohydrolase decreased by about 23% in the membrane fraction and by about 35% in the cytosolic fraction following stimulation with isoproterenol, and by 14% in the membrane fraction and by 7% in the cytosolic fraction following stimulation with carbachol. When the activity determinations were performed in the presence of the activator oleate (2.5 mM), the observed relative changes were almost the same. In contrast, the activities of lyso-phosphatidate acyltransferase, diacylglycerol acyltransferase, and diacylglycerol kinase exhibited significant increases following stimulation with both types of agonists (Table 1). The activity increases were clearly higher with carbachol than with isoproterenol. The time course of activity changes was comparable to that observed earlier for the transfer of labelled fatty acids from the acyl-CoA pool to neutral lipids (1) in that the maximum of enzyme activation occurred between 10 and 60 sec following initiation of stimulation, followed thereafter by a rapid decrease (Table 2). TABLE 2. TIME-DEPENDENCY OF THE EFFECT OF CARBACHOL (10-5 M) ON ACTIVITY CHANGES OF LYSO-PHOSPHATIDATE ACYLTRANSFERASE, DIACYLGLYCEROL ACYLTRANSFERASE AND DIACYLGLYCEROL KINASE IN GUINEA PIG PAROTID MICROSOMES 0 Enzyme activity (dpm/mg protein) Lyso-PA-acyltransferase 1,2-DAG-acyltransferase 1,2-DAG-kinase

1850 1293 2728

Time after beginning of stimulation (min) 0.5 1 2 5 10

20

8807 50676 12200 5208 3442 2513 20235 42166 31343 38823 20470 19920 60397 19981 7307 9165 9312 7120

Parotid gland lobules were prepared and incubated with carbachol as in Table 1. At the time intervals after addition of carbachol given in the Table, lobules were removed, chilled and microsomeswere prepared and used for enzyme activity determinations as given in the Methods section.

In Vitro Activation and Inactivation of Enzymes in the Presence of Protein

Kinases and Protein Phosphatases The rapid activation of the three enzymes, lyso-phosphatidate acyltransferase, diacylglycerol acyltransferase and diacylglycerol kinase, followed by a rapid inactivation pointed to the possibility that the activity changes might have resulted from protein phosphorylation and dephosphorylation. Therefore, microsomes from guinea pig parotid gland lobules were incubated in the presence of the catalytic subunit of cAMP-dependent protein kinase or of Ca2+/calmodulin-dependent protein kinase II under the conditions presented in the legend to Table 3. Both kinases led to a significant increase of the activities of all three enzymes.

39

REGULATION OF ENZYMES OF LIPID METABOLISM TABLE 3. STIMULATION OF ENZYME ACTIVITIES IN ISOLATED G U I N E A PIG PAROTID MICROSOMES DURING IN VITRO PHOSPHORYLATION FOR 10 MIN WITH THE CATALYTIC SUBUNIT OF cAMP-DEPENDENT PROTEIN KINASE OR CALMODULIN-DEPENDENT PROTEIN KINASE II Enzyme activities as percent of controls without kinase (%) Lyso-PA-acyltransferase DAG-kinase DAG-acyltransferase Calm.-PK PK-A Calm.-PK PK-A Calm.-PK PK-A 1183 ± 770 (330-2084)

502 ± 126 (333--636)

979 ± 320 (478--1430)

752 ± 210 (458-926)

1137 ± 490 (660-2057)

514 ± 210 (218-695)

The microsomes were isolated as given in the Methods section and incubated with either the catalytic subunit of cAMP-dependent protein kinase or Ca2+/calmodulin-dependent protein kinase II for 10 min as given in (21). Incubations under identical conditions but in the absence of ATP served as controls. The activities are expressed as percent of controls which were set to 100%. Mean values + S.D. from 7 separate experiments are presented. The lowest line gives the range of relative activities for each experimental condition.

As indicated in Figure 2, protein phosphatase 2A when added after prior activation with Ca2+/calmodulin-dependent protein kinase II induced a decrease of enzyme activities. As indicated in Figure 3, protein phosphatase 1 seemed to be the most active enzyme under the conditions tested.

CAL-KIN. 1400

o

12oo CAL-KIN

CAL-KIN.

+

1ooo

.......

PHOSPHATASE

.=

~!ii:!i ':::::-::::::;: ":':':':':" ::::::::::::::

CAL~KIN '

800

CAL-KIN +

>

>"

400

uJ

CONTR

co.r.,

2O0

co.T., p.os

LYSO-PA ACYLTRANSFERASE

co.T..

÷

~.os

DAG-KINASE

co.T. +

co.T., p.os

i~!!!!! ,'.:.:.:.:.:.: 1.:.:.:,,.-,-.

:~:~:!:!:!:i:~ ~.":':':'.".:~:i:"

DAG~ACYL TRANSFERASE

FIG. 2. Microsomes which had been incubated in the presence and absence of Ca2+/calmOdulindependent protein kinase II under the conditions given in the legend to Table 3 were subsequently treated with 35 p.g/ml of purified catalytic subunit of protein phosphatase 2A for 20 min and the activity of the three enzymes measured as given in the Methods section. Means from 3 separate experiments.

40

H.-D. SOLING, et al. CONTI~DL(NO pFIOTEINPHOSPHATA,SE]

o~°~-----O~oJ/° 8( NO ImOTEINPHOGPItATAIR > k> m.-

E

(5(

Is

I'tC$1~ATASE 2O 2A r=

.<

~

4O

+0-

i

0

~

Pt~SI~ATASE 24

/

pItlOSPHATAS£ I

s-

2'Of ~L

i

0

I

15

I

~10

I

45

I

60

0

UffSTIMULATEOCONTROL

15 '

310 TIME (MIN)

:~

20

TIME (rain)

FIG. 3. Inactivation of lyso-phosphatidate acyltransferase (left panel) and diacylglycerol

kinase (right panel) by the catalytic subunits of protein phosphatase 1 or 2A or by protein phosphatase 2B. Microsomes from guinea pig parotid gland lobules were first incubated with or without CaZ+/calmodulin-dependent protein kinase as given in the legend to Table 3. After removal of protein kinase by centrifugation and washing of the microsomes, the microsomes were incubated with 12/zg/ml, 35/~g/ml, and 20/~g/ml of protein phosphatase, 1, 2A and 2B respectively for the indicated times.

Since both protein kinases catalyzed an activation of the enzymes, the question was whether they would phosphorylate the same or different sites on the enzymes. As shown in Table 4, addition of either kinase after phosphorylation by the other kinase did not lead to a further increase in enzyme activity. This indicates that both protein kinases activate the enzymes by phosphorylating the same site(s). The activities of glycerol kinase, glycerolphosphate acyltransferase, phosphatidate phosphohydrolase and CDP-diacylglycerol synthetase remained unaffected by incubation with either of the two protein kinases (results not shown here). These results indicated that the flux changes observed earlier (1) had resulted from the activation of lyso-phosphatidate acyltransferase, diacylglycerol acyltransferase and diacylglycerol kinase through phosphorylation by either a cAMP-dependent protein kinase (stimulation with 13-adrenergic agonists) or a Ca2+/calmodulin-dependent

REGULATION OF ENZYMES OF LIPID METABOLISM

41

TABLE 4. EFFECTS OF SEQUENTIAL PHOSPHORYLATION OF GUINEA PIG PAROTID MICROSOMES WITH THE CATALYTIC SUBUNIT OF cAMP DEPENDENT (PK-A) OR CaE+/CALMODULIN-DEPENDENT PROTEIN KINASE (Calmod-Kin) Enzyme activity (percent of activity determined in unphosphorylated microsomes) (Calmod-Kin) (PK-A) 10 min 20 rain 30 min 40 min Lyso-phosphatidate acyltransferase Diacylglycerol kinase Diacyiglycerol acyltransferase

Lyso-phosphatidate acyitransferase Diacylglycerol kinase Diacylglycerol acyltransferase

357 + 98 358 + 100 529 + 123

408 + 91 411 + 12 632 + 157

366 + 98 399 + 51 469 + 87

373 + 108 377 +5 597 + 140

(PK-A) 10 min 20 rain

(Calmod-Kin) 30 min 40 min

249 + 83 316 + 79 509 + 118

278 -+ 88 374 + 88 588 +_ 164

266 + 113 350 + 109 527 + 146

337 + 109 367 + 65 541 + 134

The microsomes were first incubated with either Ca2+/calmodulin-dependent protein kinase or the catalytic subunit of cAMP-dependent protein kinase for 10 rain and 20 min. After 20 min, the second protein kinase was added and the incubation continued for another 10 or 20 min. For each time-point and condition the enzyme activities were determined and expressed relative to the activities measured in unstimulated microsomes which were set to 100%. The values are means + S.D. from 3 separate experiments for each condition.

p r o t e i n k i n a s e ( m u s c a r i n i c agonists). If this were t r u e , o n e w o u l d expect that the e n z y m e s in m i c r o s o m e s isolated f r o m s t i m u l a t e d intact p a r o t i d g l a n d l o b u l e s s h o u l d be less activatable in vitro by p r o t e i n k i n a s e s t h a n m i c r o s o m a l e n z y m e s f r o m u n s t i m u l a t e d cells. This was i n d e e d the case. I n the e x p e r i m e n t s d e p i c t e d in T a b l e 5, i n c u b a t i o n of m i c r o s o m e s from s t i m u l a t e d cells with p r o t e i n k i n a s e s led only to a negligible a c t i v a t i o n w h e r e a s the activities in m i c r o s o m e s from u n s t i m u l a t e d cells were significantly e n h a n c e d u n d e r the same c o n d i t i o n s .

Effects of Lipocortins on Phosphoinositide-specific Phospholipase(s) C I n the c o n t e x t of r e g u l a t i o n of signal t r a n s m i s s i o n , c o n t r o l of p h o s p h o l i p a s e ( s ) A2 by l i p o m o d u l i n ( = l i p o c o r t i n 1 = calpactin 2) has b e e n c o n s i d e r e d for a l o n g t i m e as a physiological m e c h a n i s m p a r t i c u l a r l y

320 (245--446) 238 (196--285) 415 (213--732)

341 (257-474) 250 (241-269) 444 (198--730)

114 (94--132) 107 (94-121) 96 (66-116)

106 (91-119) 96 (80-113) 100 (71-118)

123 (97-147) 115 (112-119) 110 (87-130)

122 (106-135) 114 (106--130) 128 (99--161)

Enzymatic activity (% of activity in microsomes incubated without protein kinases) Microsomes from Carbachol-stimulated Isoproterenol-stimulated unstimulated controls microsomes microsomes PK-A Calm.-PK PK-A Calm.-PK PK-A Calm.-PK

Stimulation of Iobules for 30 sec, isolation of microsomes and incubation with protein kinases were performed as given in the Methods section and in the legend to Table 3. Mean values from 3 separate experiments. The numbers in parentheses give the range of results in each experimental group.

Diacylglycerol acyltransferase

Diacylglycerol kinase

Lyso-phosphatidate acyltransferase

Enzyme examined

TABLE 5. EFFECTS OF STIMULATION OF INTACT GUINEA PIG PAROTID GLAND LOBULES WITH EITHER CARBACHOL OR ISOPROTERENOL ON THE STIMULATION OF ENZYMATIC ACTIVITIES OF LYSO-PHOSPHATIDATE ACYLTRANSFERASE, DIACYLGLYCEROL KINASE AND DIACYLGLYCEROL ACYLTRANSFERASE IN ISOLATED MICROSOMES BY THE CATALYTIC SUBUNIT OF cAMP-DEPENDENT (PK-A) OR BY Ca2+/CALMODULIN-DEPENDENT PROTEIN KINASE II (Calm-PK)

Z

O,

.:z

REGULATION OF ENZYMES OF LIPID METABOLISM

43

o 3

<

2

< _=

0

0 ----------O

0. 3

1.67

2~50

~

3133

0

4~16

I ~ (1~ • mr1)

A

10090.

° 80" ;

la o

70.

~ 50'

30i 20. <

~ 10I

A T

2,5

I

5,0

i

7.5

PHOSPHOLIPID

•

!

1

10,0

15,0

(~11 (3H)-E.coli)

FIG. 4. Inhibition of pancreatic phospholipase A2 by lipocortin 2 as a function of its concentration (upper panel) and relief of the inhibition at higher substrate concentrations (lower panel). The substrate ([3H]oleate labelled E. coil membranes) was prepared and the test performed according to (18).

44

H.-D. SOLING, et al.

when it had been reported that phosphorylation of lipomodulin would decrease or abolish its inhibitory action on phospholipase A 2 (17). However, recently it has been reported that lipomodulin inhibits phospholipase A2 by binding to its substrate rather than by directly interacting with the enzyme (18). A study examining the potential effects of the lipocortins on phosphoinositide-specific phospholipase(s) C has not been published so far. As depicted in Figure 4, lipocortin 2 strongly inhibits pancreatic phospholipase A2 and this inhibition is overcome by increasing the substrate concentration. A similar concentration of lipocortin 2 also inhibits the activities of phospholipase C from rat liver cytosol (Table 6). Similar results were obtained for the human platelet enzyme (results not shown here). As indicated in Figure 5, for the rat brain membrane enzyme, the inhibition occurred irrespective of whether the substrate was phosphatidylinositol (PI) or phosphatidylinositol-4,5-bisphosphate (PIPe). An increase in the concentration of phosphatidylinositol tended to overcome the inhibition (Fig. 5). In the case of phosphatidylinositol-4,5-bisphosphate the concentration necessary to overcome the inhibition was not reached due to lack of sufficient amounts of this substrate. Similar effects were obtained using the same concentrations of lipomodulin (lipocortin 1) or protein pI (results not shown here). In order to clearly demonstrate that the inhibition of phospholipase C by lipocortins occurred by binding to the substrates, a centrifugation-binding assay was performed (Fig. 6). In the left panel of this figure, phosphatidylinositol and lipocortin 2 were spun under identical conditions in separate tubes. When lipocortin 2 was first incubated with

TABLE 6. INHIBITION OF PHOSPHOINOSITIDE-SPECIFIC PHOSPHOLIPASE C FROM RAT LIVER CYTOSOL BY INCREASING CONCENTRATIONS OF LIPOCORTIN 2 Concentration of lipocortin 2 (/~g/ml)

Phospholipase C activity (dpm)

0 20 40 60

1,222 356 43 5

Lipocortin 2 and phosphoinositide-specific phospholipase C were prepared and the activity of the enzyme determined as given in the Methods section. The enzyme was tested with [3H]phosphatidylinositol as substrate. Enzyme activity is expressed as radioactivity of [3H]inositol recovered in the aqueous phase.

45

REGULATION OF ENZYMES OF LIPID METABOLISM NO INHIITOR

//

~owoR

o

~

|3 i o

2'

m

10.

p

o

WITH pN (~4 J~O,'m*)

~ 8b

D

NO ENZYME

60

~i0

PI~PHATilDYMNOGITOL (tiM)

o:62 PP2

1~3

~:~s

~JM)

FIG. 5. Inhibition of phospholipase C from human platelet membranes by lipocortin 2 (24/~g/ml) as a function of the concentration of phosphatidylinositol (left panel). The inhibition by the same concentration of lipocortin 2 of phospholipase C from rat brain membranes is shown on the right panel. The concentrations of PIP2 used in the experiment were not sufficient to overcome the inhibitory effect of the lipocortin.

phosphatidylinositol in the presence of 10/,LM free calcium, lipocortin 2 and phosphatidylinositol banded in the same position (middle panel in Fig. 6). The lipocortin peak had moved up, the phosphatidylinositol peak down. Omission of calcium almost completely abolished the interaction between lipocortin and phosphatidylinositol (right panel in Fig. 6). These findings indicate that the inhibition of phosphoinositide-specific phospholipase(s) C by lipocortins does indeed result from their binding to the substrates and not from a direct interaction with the enzyme. Therefore, we do not believe that the lipocortins play a physiological role in the regulation of phospholipases during signal transduction. DISCUSSION

An analysis of 1,2-sn- and 2,3-sn-diacylglycerols following stimulation of guinea pig parotid gland cells with isoproterenol or carbachol had indicated that isoproterenol leads to an increase of 2,3-sn-diacylglycerols, carbachol to an increase of 1,2-sn-diacylglycerols (1). This is compatible with the view that B-adrenergic agonists activate a triglyceride lipase via phosphorylation by a cAMP-dependent protein kinase whereas carbachol leads to an increase of 1,2-sn-diacylglycerols mainly by stimulation of phosphoinositide-specific phospholipase(s) C. In spite of these different effects of the two types of

46

H.-D. SOLING, et al. SEPAPiATE WJNS Of p3e AND (3H)-PHOSPHA T IOYLINOSITOL

90, 80.

~ 70

i

60

(3H)- PHOSpflATIOYUNOSITOL

(3H) - ~ H A T IDYUNO61T Or. ÷ p34S; NO GALCIUM

+ tm~ + e ~ + 41o - s ll)

,600

p36

p36 7

p36

tl f! II I

~t

PI

II I t

400

! P!

!~ 30,

200 !

A

20.

z

100 10. I/ •

2345678

12345678 Froction number

. . . .

,.

% .

. ."-.

12345678

FIG. 6. Interaction between lipocortin 2 (= caipactin 1 = p36) and phosphatidylinositol. Radioactive [3H]phosphatidylinositol-containing lipid vesicles were prepared by sonicating 2 mg [3H]phosphatidylinositol (specific activity 1/zCi/mg) and 2 mg phosphatidyiethanolamine in 1 ml of 10 mM Tris/Cl, pH 8.0 containing 50 mM NaCI. Lipocortin 2 (80 ttg) was dissolved in 180/~1 of 10 mM imidazole, pH 6.8, containing 40 mM KCI, 2 mM MgCI2, 10 mM EGTA and either no added calcium or sufficient CaCI2 to reach a free calcium concentration in the final system of 10/zM. The solution was added to 50 FI of the [3H]phosphatidylinositol liposome suspension and sonicated. Sufficient 80% (w/v) sucrose (in the imidazole buffer) was added to bring the mixture to 55% sucrose. 760/~1of this solution was filled into a centrifuge tube and overlaid with 900 ill each of 45%, 35%, and 25% sucrose (w/v) and finally 1.6 ml of the imidazole buffer mentioned above. The tubes were spun in the SW 65 swing-out rotor (Spinco) for 2 hr et 300,000 g(max)and fractionated beginning with the most dense fraction. The position and amount oflipocortin 2 was determined by SDS-polyacrylamide gel electrophoresis and staining with Coomassie Blue, that of phosphatidylinositol by liquid scintillation counting of [3H]radioactivity.

agonists on lipid metabolism early during signal transduction, b o t h led to an increased transfer of fatty acids f r o m the a c y l - C o A pool to diacylglycerols and triglycerides (1). T h e results described here explain this observation. T h e increased transfer o f fatty acids f r o m a c y l - C o A to the neutral lipids results f r o m an activation o f two acyltransferases and diacylglycerol kinase. This activation is achieved by covalent modification o f these e n z y m e s by protein phosphorylation. Since this p h o s p h o r y l a t i o n can be catalyzed by a c A M P - d e p e n d e n t as well as by Ca2+/calmodulin d e p e n d e n t protein kinases it b e c o m e s clear w h y 13-adrenergic as well as muscarinic agonists lead to an increased acyl-transfer in intact cells. A l t h o u g h the direct i n c o r p o r a t i o n of

REGULATION OF ENZYMES OF LIPID METABOLISM

47

32p-phosphate into the enzyme proteins has not yet been demonstrated, the following observations lend strong support for this interpretation. (1) The changes of enzyme activities in microsomes from stimulated intact cells persist after extensive washing of the microsomes which makes it unlikely that aUosteric factors are responsible for the activity changes. (2) The time course of activity changes parallels the time course of flux changes observed previously (1). (3) The activity changes can be produced in vitro with purified protein kinases and protein phosphatases. (4) The nature of the protein kinases which activate the enzymes in vitro is such that they should become activated in intact cells under conditions where the flux changes are observed. (5) In vitro activation by protein kinases is almost abolished when microsomes from cells are used which had been stimulated before by either carbachol or isoproterenol which indicates that stimulation of intact cells leads to the same changes as in vitro treatment of the isolated microsomes with protein kinases. While lyso-phosphatidate acyltransferase and diacylglycerol acyltransferase are clearly intrinsic membrane proteins, diacylglycerol kinase is distributed between cytosol and microsomes. We have analyzed in the present study mainly activity changes of the microsomal enzyme. However, stimulation of intact cells with either isoproterenol or carbachol led to similar increases of activities of the cytosolic and the microsomal enzyme (results not shown here). The same holds for in vitro activation with purified protein kinases (Table 7). It seems of interest in this context that phosphatidate phosphohydrolase which also shows a bimodal distribution between membranes and cytosol exhibited only minor activity changes following stimulation of intact cells and no activity changes after

TABLE 7. ACTIVATION OF SOLUBLE DIACYLGLYCEROL KINASE FROM GUINEA PIG PAROTID GLAND LOBULES BY IN VITRO INCUBATION WITH Ca2+/CALMODULIN-DEPENDENT PROTEIN KINASE (calmod-PK) OR THE CATALYTIC SUBUNIT OF cAMP-DEPENDENT PROTEIN KINASE (PK-A) Controls 100 + 4.1

Activity of diacylglycerol kinase (% of control) After treatment After treatment with calmod-PK with PK-A 290 + 40

245 + 26

The "cytosol" was prepared as given in the Methods section. Incubation with protein kinases was performed under conditions similar to those given in the legend to Table 3. Before determining enzyme activity, the ATP from the protein kinase reactions was removed by centrifugation through Sephadex G-25. The enzyme activities following treatment with protein kinases were related to the activity in the untreated cytosol which was set to 100%. Mean values _+ S.D, from 5 experiments. The activity under control conditions was 137 + 6 pmol/mg protein/min.

48

H.-D. SOLING, et al.

incubation with cAMP-dependent or Ca2+/calmodulin-dependent protein kinases. It has recently been reported (19) that a GTP-mediated stimulation of CDP-diacylglycerol synthetase might favor membrane fusion processes. This could be of particular importance in secretory cells. However, as reported here, conditions leading to increase~exocytosis (stimulation of parotid gland acinar cells by isoproterenol or carbachol) did not affect the activity of CDP-diacylglycerol synthetase. Our results show that the activation of phospholipases during the early phase of signal transduction is accompanied by the activation of additional steps of lipid and phospholipid metabolism. We assume that this finding is not restricted to exocrine secretory cells. Although clear-cut evidence for a physiological role of lipocortins in the regulation of phospholipase(s) A2 is lacking, such a function is discussed even in the more recent literature (e.g., Ref. 20). We have shown here that the inhibitory properties of lipocortins are not restricted to phospholipases of the A2-type but apply also to phosphoinositide-specific phospholipases of the C-type. However, this effect results from an interaction of the lipocortins with the substrate phospholipids and not from a direct interaction with the phospholipases itself. This is in line with the observations reported by Glenney's group (18) who showed that the inhibition of phospholipase A2 resulted from an interaction of the lipocortins with the substrate rather than with the enzyme itself. The interesting properties of the lipocortins and of the other members of the calpactin family (Ca2+-dependent interaction with (membrane?)phospholipids; substrates for serine/threonine- and tyrosinespecific protein kinases) point to an important though yet unidentified physiological role of these proteins, but it seems highly unlikely that they are directly involved in cellular signal transduction by regulating the activities of phospholipases. SUMMARY

In isolated guinea pig parotid gland lobules the activities of the following enzymes were measured 30 sec after stimulation with either 2 x 10-5 M isoproterenol or 10-5 M carbachol: glycerol kinase (EC2.7.1.30), glycerolphosphate acyltransferase (EC2.3.1.15), lysophosphatidate acyltransferase (EC2.3.1.51), phosphatidate phosphohydrolase (EC3.1.3.4), diacylglycerol acyltransferase (EC2.3.1.20), diacylglycerol kinase (EC 2.7.1.107), and CDP-diacylglycerol synthetase (EC 2.7.7.41). Lyso-phosphatidate acyltransferase, diacylglycerol kinase, and diacylglycerol acyltransferase exhibited significant increases following stimulation by both types of agonists. Stimulation of the activities of these three enzymes occurred also following in vitro incubation with the catalytic subunit of

REGULATION OF ENZYMES OF LIPID METABOLISM

49

cAMP-dependent protein kinase or a Ca2+/calmodulin-dependent protein kinase II. These effects could be reversed by incubation with various protein phosphatases. When cells were first stimulated with either type of agonist, subsequent incubation with protein kinases was almost ineffective. Activation by the two types of protein kinases was not additive, indicating that they activate by phosphorylating identical sites on the enzyme proteins. The other enzymes examined showed no or only minor changes and their activities could not be affected by in vitro incubation with the two types of protein kinases. The results explain the rapid changes in acyl-group transfer from acyl-CoA to neutral lipids observed previously during the first seconds after stimulation of guinea pig parotid gland lobules with isoproterenol or carbachol (1). An analysis of a potential role of lipocortins for the regulation of phosphoinositide-specific phospholipases C reveals that these proteins do indeed inhibit these enzymes, but that this inhibition results from a calcium-dependent interaction of the lipocortins with the phospholipid substrate. A physiological role of lipocortins for the regulation of phospholipases is doubtful.

REFERENCES 1. H. D. SOLING, E. MACHADO-DeDOMENECH, J. KLEINEKE and W. FEST, Early effects of 13-adrenergic and muscarinic secretagogues on lipid and phospholipid metabolism in guinea pig parotid acinar cells, J. Biol. Chem. 262, 16786-16792 (1987). 2. M . J . BERR1DGE, Inositol trisphosphate and diacylglycerol: Two interacting second _ messengers, Annu. Rev. Biochem. 56, 159-194 (1987). 3. E. MACHADO-DeDOMENECH and H. D. SOLING, Effects of stimulation of muscarinic and 13-catecholamine receptors on the intracellular distribution of protein kinase C in guinea pig exocrine glands, Biochem. J. 242, 749-754 (1987). 4. S. YAMASHITA and S. NUMA, Glycerophosphate acyltransferase from rat liver, Methods Enzymol. 71,550-554 (1981). 5. S. YAMASHITA, K. HOSAKA, Y. MIKI and S. NUMA, Glycerolipid acyltransferases from rat liver; 1-acylglycerophosphate acyltransferase, 1-acylglycerophosphorylcholine acyltransferase, diacylglycerol acyltransferase, Methods Enzymol. 71,528-550 (1981). 6. A. MARTIN, P. HALES and D. N. BRINDLEY, A rapid assay for measuring the activity and the Mg 2+ and Ca 2+ requirements of phosphatidate phosphohydrolase in cytosolic and microsomal fractions of rat liver, Biochem. J. 245, 347-355 (1987). 7. R. G. LAMB, K. FOSTER and M. McGUFFIN, A distinction between rat liver phosphatidate phosphatase and phospholipase C, Biochim. Biophys. Acta 921, 67-74 (1987). 8. J. R. CARTER and E. P. KENNEDY, Enzymatic synthesis of cytidine diphospbate diglyceride, J. Lipid Res. 7, 678-683 (1966). 9. K. A. PETERS, J. G. DEMAILLE and E. H. FISCHER, Adenosine 3':5'monophosphate-dependent protein kinase from bovine heart: Characterization of the catalytic subunit, Biochemistry 16, 5691-5697 (1977). I0. J. R. WOODGET, M. T. DAVISON and P. COHEN, The calmodulin-dependent glycogen synthase kinase from rabbit muscle - - purification, subunit structure and substrate specificity, Eur. J. Biochem. 136, 481--487 (1983). JAER 2 ~

50

H.-D. SOLING, et al.

11. H. Y. L. TUNG, T. J. RESINK, B. A. HEMMINGS, S. SHENOLIKAR and P. COHEN, The catalytic subunits of protein phosphatase-1 and protein phosphatase-2A are distinct gene products, Eur. J. Biochem. 138, 635-641 (1984). 12. A. TALLANT, R. W. WALLACE and W. Y. CHEUNG, Purification and radioimmunoassay of calmodulin-dependent protein phosphatase from bovine brain, Methods Enzymol. 102, 244-256 (1983). 13. N.C. KHANNA, M. TOKUDA and D. M. WAISMAN, Purification of three forms of lipocortin from bovine lung, Cell Calcium 8, 217-228 (1987). 14. T. TAKENAWA and Y. NAGAI, Purification of phosphatidylinositol-specific phospholipase C from rat liver, J. Biol. Chem. 256, 6769-6775 (1981). 15. V. MANNE and H. F. KUNG, Characterization of phosphoinositide-specific phospholipase C from human platelets, Biochem. J. 243, 763-771 (1987). 16. I. LITOSCH, Guanine nucleotide and NaF stimulation of phospholipase C activity in rat cerebral-cortical membranes, Biochem. J. 244, 35-40 (1987). 17. F. HIRATA, The regulation of lipomodulin, a phospholipase inhibitory protein, in rabbit neutrophils by phosphorylation, J. Biol. Chem. 256, 7730--7733 (1981). 18. F.F. DAVIDSON, E. A. DENNIS, M. POWELL and J. R. GLENNEY, JR., Inhibition of phospholipase A2 by "lipocortins" and calpactins - - an effect of binding to substrate phospholipids J. BioL Chem. 262, 1698-1705 (1987). 19. M. JOLICOEUR, A. GUENETI'E and J. PAIEMENT, Relationship between GTP-stimulated incorporation of [3H]CTP and membrane fusion in rat liver microsomes, Abstr. 4th. Int. Congr. of Cell Biol., Montreal Abstr. Nr. P 6.5.2., 1988. 20. M. WAITE, The phospholipases, pp. 262-264 in Handbook of Lipid Research, Vol. 5,(1987). 21. C. DOMENECH, E. MACHADO-DeDOMENECH and H. D. SOLING, Regulation of acetyl-CoA:l-alkyl-sn-glycero-3-phosphocholine O2-acetyltransferase (lyso-PAFacetyltransferase) in exocrine glands - - evidence for an activation via phosphorylation by calciurrdcalmodulin-dependent protein kinase, J. BioL Chem. 2262, 5761-5676 (1987).