~ et BiavhysieaActa, 1136(1992)68-74 © 1992ElsevierSciencePabl/shersB.V.All fightsresewed0167-4889/92/$05.[]0

68

BBAMCR 13192

Platelet-activating factor-mediated synthesis of prostaglandins in rat Kupffer cells C h a n d r a s h e k h a r R. G a n d h i , M i c h a e l S. D e B u y s e r e a n d M e r l e S. O l s o n Department of Biorhem~o; The Unirer~y of Tea~sHealth ScienceCenter, San Antonio, TX (USA)

(Received It G~t~oer1991) (Revised manuscriptreceK'ed3 February1992~

Keywords: Prostaglandin;Kupffercell; Liver.P'mtelet-aczivatingfactor. Protein Kina~eC S~althesls of prostaglandins was stimulated in rat Kupffer cells upon challenge with platelet-~,ctivating factor (PAl:). PAF-mediated synthesis of pmstaglandins w~s inh~ited by the Ca2+ ion chelator (EGTAL ~ Ca2* channel antagonist (nifedipine) and U66985, a structural analogue and antagonist of the biological effects of PAl: in other cellular systems, lnh~itors of protein kinase C, smurosporioe and polymiJdn B, did not affect PAl-induced prostaglandia synthesis. Phorbol 12-myri~ate 13-acetate (PMA), an activator of protein kinase C~ stimulated synthesis of prostaglandins in K,~pffercells; PAl and PMA exvficd addhive actiors on this process. Both PAl- and PMA-stimulated prostaglandin production, was inh~ited by TMB-8. PAl-stimulated synthe~'~ of ptostaglandios also was inhibited upon treatment of Kupffer ceils with pertnssls toxin. Cholera toxin, in contrast, stimulated the production of prostaglandins in a concentration-dependem manne~ cholera toxin and Po~J: together had an ad~C,ve effect. These results suggest that PAl-induced synthesis of prostaglandip.sis stimulated via a specific receptor coupled to a pertussls toxin-sensitive G-protein, is dependent upon extracellular Ca2+ and is not influenced by l~rotein Kinase C activation, Since PAY and prostaglandins are produced in the liver under conditions such as endutoxemia, PAF-mediated synthesis of these lipid antacoids may he of importance in the regulation of hepatic function during pathophysiologicalepisodes.

Introduction Eicos,~oids are synthesized by various types of macrophages in response to a number of stimuli [1--4] Arachidonic acid metabolites are associated with several biological processes such as inflammation, platelet aggregation, vasodilation, vasoconstriction, etc. [5]~ Prostaglandin-stimulated DNA synthesis and mitotic activities in cultured hepatocytes [6,7] and the sTnthesis of eicosanolds by liver [8] and by isolated liver cells [9] following partial hepatectomy have ~ugges:cd that these antacoids may be involved in liver regeneration. In addition, certain prostaglandins such as PGI 2 at:d

Correspondence to: C.R. Gandhi, Departmentof Biochemistry,The Universityof TexasHealth ScienceCenter. San Antonio,TX 78284~60, USA. Abbreviations: EGTA, ethylene glycol bis(~-aminocthy| ether)N,N,N',N'-'etraacetic acid; IBMX, isobu~lmethylxanthine;NDGA. nordih,ydroguaiareficacid; PAl:, platelet-activating favor (l-O-al~l2-ace~lglycerophosphoclmline); PGD2, prostaglandin D2; PGE2, pmstaglandin E2; TMl:bS,3,4,5-trimetho~henzoic acid 8-(dinthylamino)octylester.; PM~,, phorbol 12-myristate 13-acetate.

PGE 2 seem to protect hepatocytes against chemical damage and lvjpoxia [10]. Several reports have attempted to characterize a role for prostaglandins h~ i'.q~ regulation of hepatic glycogenolysis [11-14]. However, t'~e precise mechanism by which prostaglandins exhibit glycogenolytic effects in the liver is somewhat controversial [15]. Hepatic sinusoidal Kupffer cells and endothelial cells synthesize eicosanolds upon appropriate stimulation [16,17]. In contrast, hepatocytes do not synthesize but catabolize arachidonic acid metabolites

[18.1 Platelet-activating factor, a potent lipid autacoid, has been shown to mediate allergic and inflammatory responses in many cell and tissue .Wpes [19,20]. This !abo~tory has demonstrated that infusion of endotoxin into the peffused rat liver causes stimulation of the synthesis of PAF and eicosanoids [21]. Since the liver is involved in the removal of endotoxins from the circulation, under conditions such as endomxemia, PAF is likeb" to play a major role in hepatic pathophysiology. In order to better understand the role of PAl= in liver pathophysiology, we undertook this iovestiga'-.ion to characterize PAF-mediated synthesis of prostagiandius by rat Kupffer ce~ls.

69 Materials and Methods

Reagents. Collagenase (Type IV from Clastridium hi~olyticum), proteinase (Type XIV, from Streptomyces gr/seus), bovine serum albumin (fraction V), TMB-8, nifedipine, ibuprofen, nordihydrognaiaretic acid, EGTA, phorbol 12-myristate 13-acetate and heat-inactivated fetal bovine serum were purchased from Sigma, SL Louis, MO. Willian,,'s medium E (GIBCO Labs, New York, NY), A23187 (Calbiochem, San Diego, CA), methyl formate (Eastman Kodak, Rochester, NY), [5,6,8,9,11,12,14,15-3H(N)] arachidon~c acid (100 Ci/mmoD (New England Nuclear, Boston, MA), and [3H]PGE2, [3H]PGD2 and the radioimmunoassay kit for cyclic AMP (Amerslmm, Arlington Heights, IL) were also purchasc'd. All other chemicals and reagents purchased locally w:re of highest purity available. Isolation and culture of Kupffer celL~. Kupffer cells were isolated from livers of male Spragae-Dawlcy rats (200-250 g), using a modification of the centrifugal elutriation procedure of Knook and Sleyster [22] as described previously [23]. Kupffer cells wexe suspended at 2. l06 cells/ml in WTJI.;am's medium E supplemented with 2 m M L-glutam, ine, 10% heat-,:uactivated fetal calf serum, 5000 U / m l penicillin and 5 ~ 3 0 / t g / m l sh-eptomycin. ,Miquots (2 ml) of this cell suspension were placed in 35-mm tissue culture plates and after an overnight incubation in an atmosphere of 95% 0 2 / 5 % CO 2 at 37°C, the attached cells were washed twice and placed in fresh medium containing 2% fetal calf serum (instead of 10%) and 0.5/LCi/ml [3H]arachidonic acid. The cells were incubated in this medium for 24 h, washed twice with the nonradioactive medium and incubated in Wiiliam's medium E containing 2% felal calf serum. The relative distribution of [3H]arachidordc acid in various phospholipids from TLC separathm was: phnsphatidylcholine, 43%; phosphatidylethanedamine and plasmalogens, 31%; phnsphatidyiinositoL 19%; polyphospholinnsitides, 2%; and the remainder, 9%. After 30 rain in the incubator, c e l l ~ere stimulated with PAl= and the reaction was terminated 15 min later with the addition of ethanol [24]. Extraction ofprostaglandins. Prostaglandins were extracted essentially as described by Powell [24]. Cells were scraped from the plates and ~he medium and cells were transferred quantitatively m tubes. The plates were washed with an additional 1 ml ethanol and water was added to the combhled extracts to bring the final concentration of ed~anol to 15%. The extract was centrifuged and the pellet was washed with 3 - 4 ml 15% ethanol. Combined superuatants were acidified to pH 3.5 with 1 M HCi, applied to SEP-PAK columns (CIs cartridges from Waters Associates, Milford, MA), washed with 15% ethanol and finally prostaglandins were el,,ted with methyl formate. Analysis ofprostaglandins. Methyl formate was evap-

orated under N 2, the residue was dissolved in 140 ttl 36% acetic acid (0.1% v/v) in acetonitrile and filtered (0.45 p. Miilipore Type HV filter). 9 0 / t l of the f'fltrate was used for the analysis of prostaglandins by high pressure liquid chromatography. The analysis of prostaglandins was performed essentially as described by Peters et al. [25], using a Varian HPLC apparatus, an Ultrasphere ODS colmml (4.6 mm × 25 cm) and 0.1% (v/v) aqueous acetic acid (pH 3.7) and acetonitrile as the solvent system at a flow rate of 1.5 ml/min. Fractions (30 s) were collected and quantified by liquid scintillation spectroscopy. Various prostaglandins were identified by. spiking the cell extracts with a mixture of [3H]prostaglandin standards. Determination of cAMP. Kupffer cells were washed three times and were placed in William's medium E containing 2% fetal calf serum. The cells were challenged with various agents for the indicated times in the presence or absence of 0.5 mM isobutylmethylxanthine. The reaction was terminated by the addition of 0.5 ml 5% ice-cold TCi,, after aspirating the medium. The plates were scraped and washed o~ce with 0.5 ml TCA. After centrifugation, the T C A superuatants were washed six times with water-saturated ether, lyopbylized and analyzed for cAMP content using the radioimmunoassay technique as described by Amersham. Statistical analyses. Significance of results in the various experiments was derived from the Student's t-test. The data presented represent values _+ S.E. front a single experiment performed in duplicate or triplicate. There was some variation in the results obtained from similar experiments performed using separate batches off cells; however, the patterns of results were similar. Results

Effect of P/tF on synthesis of pr(~staglandins A.~. depicted in Fig. 1, a signif,.caat stim~!atien of prostaglandin synthesis in cultured Kupffer cells occurred at 30 s following the addition of P e l f with maximal effect apparent after 5 rain of incubation. Fig. 2 illustrates the dose-response relationship between PAF and the synthesis of prostaglandins; an increase in the synthesis of prostaglaudins was apparent at 1 nM PAl:. Maximal stimulation of prostaglandin synthesis was observed at 10 nM PAt: and the magnitude of stimulated PGD 2 synthesis was much greater than that observed for PGE 2 at all concentrations of PAF tested (Fig. 2). A similar dose-response relationship between the P A F concentration and phosphoinositide metabolism in Kupffer cells has been observed [23].

Calcium requirement The experiments shown in 'Fable I were designed to ascertain whether an increase in cytosolic Ca 2+ via its

TABLE H Effect of U66¢)85o~. PAF-induced p~staglandin 5yluhesis in the Kupfferce~ Kepffer c e ~ were labelled with 0.5 ~C'i/ml [3H]arachidonic acid. ~ b e d and challenged ~ t h ~ nM PAl: in the presence or absence of I ;tM U ~ . Cells were also challenged with 200 nM lyso-PAF and the ~ ~ s terminated at 20 rain. Results are expressed as the average of duplicate determinations _4-S.E. from a representative of fi~e separate experiments. . •

. s

.

.

.

10

. Is

PGE 2

20

Tins (ran) Fig. I. "[-rme cuu~se of the PAF-i~luced prostag~ndin synthesis in

cultured Kupffer ceils. Kupffer cells in prima~ culture were labelled with [ 3 H ] a r a c ~ acid for 24 I1. washed and stimulated with 25 nM PAl: for ;.n~,cated time intervals. For other details, ~ Materials and Methods. Each value b an average of dupilea:e determinations from a representative of foor separate exp'~.,iments ~S.F.. (~), PGE~; (e). PGD2.

Contt~ PAl: Uff~85/PAF L~o-PAF

(dl~/plate)

PGD, (dpm/plate)

172+_24 449-4.33 256 + 14 185-4.13

128+-12 823+-15 432 -4.29 II I _.- 4

mobilization f r o m intracellular s t o r e s [29] w a s sufficient t o stimulate t h e synthesis o f p r o s t a g l a n d i n s in t h e K u p f f e r cells, it is c l e a r f r o m these results t h a t b o t h E G T A , a n d t h e p i n n a m e m b r a n e C a 2+ c h a n n e l a n tagonisL nifedipine, strongly inhibited P A F - i n d u c e d synthesis o f p r o s t a g l a n d i n s indicating the d e p e n d e n c e o f .t.;,~,,.g p r o c ¢ ~ o n e ~ r a c e l l u l a r C a 2+. Inhibition o f A 2 3 1 8 7 - s t i m u l a t e d p m s t a g l a n d i n synthesis b y E G T A ( d a t a n o t s h o w n ) f u r t h e r s t r e n g t h e n s this assertion. Interestingly, T M B - 8 , a n intracellular calcium a n t a g o nist [ 2 ~ 2 9 ] (Table i) also inhibited P A l - i n d u c e d p r o s t a g l a n d i n synthesis in K u p f f e r cells. P~

I~Jl

Fig. 2. Effect of PAl= omcentratinn on the synthesis of pmstaglandins in cultured Kupffer cells. Kupffer cells in primat~ culture were washed and placed in a medium containing 0.5 ~Ci/ml [3H]arachidonle acid. Following a 24 h incubation, cells were washed three times and placed in nonradioactive medium. After 15 min at 3TC in an incubator, cells were challenged with the indicated concentrations of PAl: for 20 min. For other details, see Materials and Methods. Each ,,-dlae represents an average of duplicate determinations + $.E. from a representative of flee separate experiments. ( o ) PGE,; (e) PGD2.

TABLE 1 Efffct~, of TMB.8, EGTA and nifedipine on PAF-induced prostaglmMin synthesis in the primary cultures of mt lO~pffer cells Kupffer celi~ were labelled with I ~tCi/ml [3Hhrachidonic acid, washed and placed in a medium containing 100 ;tM TMB-8. 2 mM EGTA or 100/~M nifedipine. Cells were then stimulated with 25 nM rAF for 20 rain. Each value is an average of duplicate determinations +_S.E. from a representative of three separate experiments.

Control PAF TMB-8/PAF EGTA/PAF Nifedipine/PAF

PGE2 (dpm/plate) 115+_10 325 -4=_70 98+ 9 IP 5 ~33_~ 13

PGD 2 (dpm/plate) 79_+ 8 594 + 81 1164-_11 147+_12 139-4. 4

Spec///c/~ of PAF P A l - i n d u c e d stimulation o f p r o s t a g l a n d i n synthesis in K~pffer cells w a s f o u n d t o b e inhibited by U66985, a s t r u c t u r a l a n a l o g u e ant" a n t a g o n i s t o f m a n y P A l - m e d i a t e d processes (Table !1). F u r t h e r m o r e , bTsa-PAF w a s u n a b l e to stimulate p r o s t a g l a n d i n synthesis (Table !i). T h e s e results indicate : h a t P A F - s t i m u l a t e d synthesis o f p r o s t a g l a n d i n s in t h e ~(mpffer cells o c c u r s via specific receptors. Effect o f bzhibitors Fig. 3 illustrates the effect o f i n h ~ i t o r s o f cyclooxyg e n a s e ( i b u p r o f e n ) a n d lipoxygenase ( t ~ D G A ) o n P A F - i n d u c e d p r o s t a g l a n d i n synthesis. A s e x p e c t e d ~ u p r o f e n c a u s e d a d o s e - d e p e n d e n t inhibition o f P A F i n d u c e d p r o s t a g l a n d i n synthesis in K u p f f e r cells. Significant i n h ~ i t i o n o f P A F - i n d u c e d p r o s t a g l a n d i n p r o d u c tion r ~ c u r r e d a t 1 / z M ~ u p r o f e n (Fig. 3A). NI~.~A has b e e n implicated as a n i n h ~ i t o r o f the lipoxygenase p a t h w a y [30,31]. However, in the p r e s e n t study N D G A inhibited p r o s t a g l a n d i n synthesis involving t h e cyclooxy~enase p a t h w a y f o r a r a c h i d o n i c acid w e t a b o l i s m in a c o n c e n t r a t i o n * d e p e n d e n t m a n n e r [Fig. 313]. A signific a n t inhibition with N D G A o c c u r r e d b e t w e e n 0 . 1 - 1 / t M a n d c o m p l e t e inhibition o c c u r r e d a t 5 0 / t M , Also, b o t h ¢ f t h e s e a g e n t s inhibited A23187-stimulated

prnstaglandin synthesis in Kupffer cells (data not shown).

Effect of bacterial towns Both phospholipase C- and phospho!ipase A2-mediated hydrolysis of phosphoinositid(~ in Kupffer cells were i n h ~ i t e d upon treatment with pertussis toxin; cholera toxin inhibited the former process only [23]. When ~imilar experiments were performed to e'- dluate the involvement of G-proteins in P A l - s t i m u l a t e d prostaglandin synthesis, treatment of Kupffer cells with pertussis toxin resulted in the inhibition of PAF-induced synthesis of prostaglandins (Fig. 4A). Half maximal inhibition occurred at 1.25 n g / m l pertussis toxin and complete inhibition at 125 n g / m l . A23187-stimulated synthesis of prostaglandins was not affected by pertnssis toxin (data not shown), indicating the specificity of the effect of pertussis toxin on P A l - m e d i a t e d processes. Cholera toxin, o n the other hand, produced a concentration-dependent stimulation of prostaglandin synthesis with half maximal effect at approx. 1 / t g / m l of this toxin. Interestingly, cholera to~dn at all concentrations tested and P A l had additive effects on the synthesis o f prostaglandins in Kupffer cells (Fig.

413).

o

.

U

.

1.25

.

.

1:[.§

.

Sfl

.

12[[

500

P s m m s s l s TOXIn (r~lml)

|Q ~ t •

.o.-"°"

.~ . . . . . . . . . . . . . . . . . . . . .

-*"°1 r--o .......~

0

2

4

e

8

lo

Fig. 4. Effect el pertussis and cholera toxin on PAF-induced synthesis of e;ostaglandins in Kupffer cells. Kupffer cells in primary culture labelled with 0.5 /zCi/ml [3H]arachidonic acid were washed three times and then incubated in the pre.°.cnceof indicated concentrations of (A) pertussis toxin or (B) cholera toxin. After 3 h, 25 nM PAF or carrier (0.9% NaCI/0.2% albumin) was added to the plates and the reaction was terminated 20 tu;n later. Values are averages of duplicate determinations frem 3 or 4 separate experiments. (A} t[:]), PGE2; (Is). PGD2. Circles represent levels of PGE a (o) and PGD2 (e) in the nonstimulated cells. (B) (z~, a), PGE2; (o, e), PGD2. Solid symbols represent incubations in the presence of carrier and open symbolsincubations in the presence of PAF.

Effects of cAMP •

,Jo

5,o

ao

4o

so

o

t i

•

o.I

1 NOQA ( ~ J )

lO

1o0

Fig. 3. Effect of ibuprofen and NDGA on PAF-stimulated synthesis of prostaglandins in Kupffer cells. Kupffer cells labelled with 0_5 p.Ci/ml [3H] aracbidonic acid for 24 h were washed three times and challenged with 25 nM PAF for 20 min in the presence of various concentrations of ibupmfen or NDGA. lbuprofen and NDGA were added :5 rain prior to the addition of PAF. Results are expressed as average of duplicate values +_S.E. flora a representative of four experiments. (D). PGE2; (11). l~3D2. Circles represent levels of PGE 2 (o) and PGD2 (t) in the nonstimalated cells.

Both pertussis toxin and cholera toxin are known to elevate c A M P levels in certain cells by m.odulating adenylate cyclase activity [36]. Therefore, it was important to assess whether these toxins stimulate the production of c A M P and also whether elevated c A M P influences PAF-~nduced synthesis of prostaglandins. Upon incubation of Kupffer cells with cholera toxin, pe~ussis toxin and PAl:, only cholera toxin was found to produce an elevation in c A M P and this effect could be observed only in the presence of an inhibitor o f the phosphodiesterase, IBMX (Table III) or theophylline (data not shown). Preincubation of Kupffer cells with dibutyryl c A M P caused an inhibition of PAF-mediated synthesis of prostaglandins (Table IV). Interestingly, the presence of IBMX or theophylline did not affect cholera toxin-stimulated prostaglandin synthesis (data not shown).

Effect of protein kinase C actication As P A F generates diacylglycerol via phospholipase C-mediated hydrolysis of phosphoinositides [23] and

72 TABLE Ill

TABLE V

S~lw.sis of cAMP in the Kupffer cells in response to PAF and bacterial toz~ns.

Effect of 77ffB-8 on PAF- and ~£4-induced prostaglandin synthesis in cultured lOJIJffer cells

Kupffer cells in primary culture were washed and placed in a medium ±0.5 mM isobutylmethylxanthine. Various agents were added to the plates at iDdicated concentrations and incubated for 3 h (cholera toxin and permssis t~xin) or 30 rain (PAF). The reaction was terminated and cAMP levels were estimated as descn3~ed in Materials and Methods. Values expressed are the means of duplic~.tc determinations ± S.E. from a representative of four separate experiments. Similar effects of these agents oa cAMP s3"nthesis v,'ere observed when theopl~line (5 raM) was added instead of IBMX.

Kupffer cells in p~h'nary c-alturc labelled with 0.5 ,uCi/ml [3H]arachidonic acid for 24 h were washed three times and placed in nonradioaclb.e medium. After 15 mio in the inc,ahetor, 10 #1 DMSO (A and 13)or 10OnM PMA ;.n DMSO (C and D) were added to the plates. Folk~ing a 3 rain incubatiork cells were challenged with ~.he carrier (0.2% alb'amin in 3.9% NaCI) (A and C) or 25 nM PAF (B and D). Pro:,taglandins were extracted after 20 rain and analyzed as described in Materials arid Methods. TMB-8 (100 p-M) was added 5 ~.hl prk~ to the addition of PAF or PMA. Each value is a mean of duplicate detenninalioes+ S.E. from a representative of three separate experiments.

Condition

cAMP (Innoles/plate) + IBMX

Control PAF (20nM) Cholera toxin (2.5/t g/ml) Pcrtussis toxin (50 ng/ml)

2.25.+_0.13 2.13+9.08 19.2 +2.63 2.54_+0.04

- IBMX !.98_+0.22 2.08±0.15 4.39±0.87 231 _+0.41

Cond~ion

( PGE2 + FGD2 )

(A) Control

-TMB-8 (dpm/plate) 325+ 18 2100_+ 312 755+ 31 35~+- 391

(B) PAF

since activation o f p r o t e i n kinase C results in the stimulation o f p r o s t a g l a n d i n synthesis in K u p f f e r cells [33], t h e influence o f this signal t r a n s d u c t i o n process o n P A F - m e d i a t e d p r o s t a g l a n d i n synthesis w a s evalua t e d . Inl~'bitors o f p r o t e i n kinase C (e.g., 50 n M s t a u r o s p o r i n e o r 5 0 / z M polymyxin B) did not affect P A l s t i m u l a t e d p r o s t a g l a n d i n synthesis ( d a t a not shown). A n ~ t i v a t o r o f p r o t e i n kinase C, P M A , in a g r e e m e n t with previous observations [33], s t i m u l a t e d p r o s t a glandin synthesis (Table V). int:r~,t'.'~g~y, P M A a n d P A F t o g e t h e r exhibited a n additive effect o n the p r o d u c t i o n o f p r o s t a g l a n d i n s a n d the stimulation b o t h by P A l a n d P M A w a s a t t e n u a t e d by T M B - 8 (Table V). Discussion K u p f f e r cells, the r e s i d e n t m a c r o p h a g e s o f t h e h e p atic sinnsoids influence overall h e p a t i c m e t a b o l i s m t h r o u g h t h e i r capability to synthesize various a n t a c o i d m e d i a t o r s [34] iaclud'.'ng eicosanoids [17,18,35-37]. T h e glycogenolytie effect o f P A l : in the p e r f u s e d r a t liver in TABLE IV Effect of dibutyryl cAMP on PAF-stimulated synthesis of prostaglondins in Kupffer cells

Kupffer cells in primary culture labelled with 0.5 /tCi/ml [3H]arachidonic acid for 24 h were washed and incubated in the presence of 2 mM d~utyuI cAMP for 2 h. Cells were then challenged with 25 nM PAF for 20 rain. Each value represents an average of duplicate determinations ± S.E. from a representa~:ve of three separate experiments.

Control PAF DBcAMP/PAF

PGE2

PGD 2

(dpm/plate)

(dpm/plate)

12.5+-1! 420 4-__34 250 _+21

99+ 5 598 +_29 329 ± 18

(C) PMA (D) PAF+ PMLA

+TMB-8 (dpm/plate) 318+_27 455 +_35 298+- 15 499 _+23

situ [38,39] w a s s u g g e s t e d t o o c c u r via the p r o d u c t i o n o f p r o s t a g l a n d i n s [12-14]. W h e r e a s h e p a t i c glycogenolysis c a n b e i n d u c e d a t P A F c o n c e n t r a t i o n s as low as 0.I n M , significant stimulation o f p r o s t a g l a n d i n synthesis in K u p f f e r cells d o e s n o t o c c u r at c o n c e n t r a t i o n s o f P A l : u p t o 1 n M (Fig. 2). F u r t h e r m o r e , P A l : c a u s e s a n i m m e d i a t e a n d t r a n s i e n t increase in glucose p r o d u c t i o n in the p e r f u s e d r a t liver, w h e r e a s n o significant inc r e a s e in p r o s t a g l a n d i n synthesis w a s o b s e r v e d until 30 s a f t e r stimulation o f K u p f f e r cells with P A F (Fig. 1). T h e s e observations, t o g e t h e r with t h e inability o f t h e cyclooxygenase inhibitor, i b u p r o f c n , to a t t e n u a t e significantly P A F - i n d u c e d h e p a t i c glycogcnob'si~ [15], indic a t e t h a t prostaglandL't~ a r e n o t o b l i g a t o r y m e d i a t o r s o f P A l - i n d u c e d h e p a t i c glucose p r o d u c t i o n . A surprising finding in this investigation w a s inhibition o f P A l - m e d i a t e d p r o s t a g l a n d i n synthesis by N D G A , which is u s e d as a specific inhibitor o f t h e lipoxygenase p a t h w a y o f a r a c h i d o n i e acid metabolism. NDGA at concentrations between 5 and 20/tM and eicosatetraynoic acid ( a n inhibitor o f b o t h l i p o ~ g e n a s e a n d c y c l o o ~ g e n a s e ) w e r e f o u n d to inhibit a r a c h i d o n i c acid s t i m u l a t e d K + c h a n n e l s in c a r d i a c cells [30,31]. I n d o m e t h a c i n , in contl~LSt, d o e s n o t affect a r a c h i d o n i c acid s t i m u l a t e d K + c h a n n e l s in these cells [30]. T h e s e results suggest t h a t N D G A specifically i n h ~ i t s the lipox~ygenase p a t h w a y in c a r d i a c cells not affecting t h e m e t a b o l i s m o f a r a c h i d o n i c acid via the cycloox-ygenase pathway. O u r results, however, clearly indicate t h a t N D G A a t c o n c e n t r a t i o n s u s e d in this investigation [30,31] ( 5 - 2 0 / z M ) inhibits t h e cyclooxygenase p a t h w a y in K u p f f e r cells (Fig. 313). T h u s , N D G A c a n n o t b e c o n s i d e r e d a specific inhibitor o f the lipoxygenase activity.

Since PAF-induced prostaglandin synthesis was inhibited by a specific antagonist U66985 (Table ll), it was of interest to ascertain whether the FAF receptorcoupled arachidonic acid cascade occurred via activation of G-proteins. While pertussis toxin inhibited PAF-stimulated prostaglandin synthesis, cholera toxin stimulated prostaglandin synthesis and together cholera toxin and PAF had additive effects. These results imply that the stimulatory effects of cholera toxin and PAF on prostaglandin synthesis in Kupffer cells are independent of each other. Cholera toxin and pertussis toxin elevate intracellular cAMP levels by r-:odulating adenylate cyclasc activity in certain cell ~pcs [32]. Stimulation of prostaglandin synthesis in the RAW264.7 murine macrophage cell line both with cholera toxin and pertussis ~oxin was suggested to occur via cAMP [40]. PAFomediated synthesis of prostaglandins was inhibited in Kupffer cells incubated with dibutyryl cAMP (Table IV). Although cholera toxin produced an increase in cAMP in the presence of IBMX (Table l i d and thcophylline (not shown),the presence of these agents did not affect cholera toxinmediated prostaglandin synthesis (not shown), suggesting that cholera toxin may override the inhibitory effect of cAMP. At the present time the precise mechanism of the action of cholera toxin on PAF-rnediated prostaglandin synthesis is not clear. However, the possibility of a distinct cholera toxin-sensitive pathway cannot be ruled out. Interestingly, the effects of cholera toxin and pertussis toxin on prostaglandin synthesis were similar to those observed for the deacylation of phosphatidylinositol [23], suggesting a similar mechanism of action of the toxins on these two processes. Receptor-mediatod phospboinositide metabolism leads to the release of diacylglycerol and mobilization of intracellular Ca 2+ and thus provides a common transductic3 mechanism for diacylglycerol- and Ca 2*mediated cellular processes. Pho:~ol e.,.ters, which mimic the actions of diacylglycecol by activation of protein idna~.,~ C, were shown to enhance prostaglaudin syhtilesis in cultured rat Kupffer cells by stimulating N a + / H + exchange [17]. We have extended these ob~.rvatlons in experiments designed to illustrate protein kinase C effects on PAF-stimulated prostaglandin synthesis. The lack of any effect of s~aurosparine and polymixin B on PAF-stimulated prostaglandin sT~.thesis suggests that protein Idnase C is not involved direoly in influencing this process. This assertion is strengthened by the additive effect PMA and PAF on prostaglandin synthesis (Table V). The intraceilular Ca 2+ antagonist, TMB-8, inhibited lns(1,4,5)P3 generation in Kupffer cells which occurs within seconds following PAl: addition, but did not affect the extracellular Ca2+-dependent deacylation and phosphodiesteric hydrolysis of phosphoinositides [23|. Inhibition of PAFmediated prostaglandin synthesis in Kupffer cells by

EGTA and nifedipine (Table I) suggests the extracellular Ca2+-dependence of this process. TMB-8 has been reported to be an intracellular Ca 2+ antagonist [26-29] as well as an inhibitor of Ca 2+ influx [41,42] and protein kinas¢ C [43]. The results presented in this investigation (inhibition of PAl::- and PMA-mediated prostaglandin synthesis) and in our previous p a ~ r (inhibition of PAF-induced Ins (1,4,5)P 3 formation) [23] imply that the effects of TMB-8 on PAl=- and PMA-mediated processes are nonspecific and careful evaluation must be made before asserting the role of Ca 2+ and protein kinase C in cellular processes based on the effects of TMB-8. Infusion of several agents such as zymosan (Fisher, g A., Robertson, S.M., Steinhelper, M.E., Revtyak, G., Kumar, R., Hanahan, D J . and Olsor M.S., u~published data), lgG [l 1] and lipopolysacct~aride [21] stimulate synthesis of PAF as well as prost~.glandins in the liver. Although a definitive role for prostaglandins in hepatic metabolism has not been pr