[CANCER RESEARCH 58. 2260-2264. May 15. I998|

ll-O-Tetradecanoylphorbol-lS-acetate-induced Apoptosis in LNCaP Cells IsMediated through Ceramide Synthase1

Mark Garzotto,2 Margaret White-Jones,2 Youwei Jiang, DésiréeEhleiter, Wen-Chieh Liao,Adriana Haimovitz-Friedman, Zvi Fuks, and Richard Kolesnick3

Laboratory of Signal Transduclion ¡M.C., M. W. J., Y. J., K. K.¡and the Departments of Radiation Oncology [D. E., W. C. L. A. H. F., Z. F.I and Surgery [M. C.], MemorialSlnan-Kettering Cancer Center. New York, New York 10021

ABSTRACT

Protein kinase C i I'M i activation is often antiapoptotic, although in a

few cell types PKC initiates apoptosis by an unknown mechanism. Recentinvestigations showed that activation of PKCa by 12-0-tetradecanoyl-phorbol 13-acetate (TPA) induced apoptosis in LNCaP prostate cancer

cells. The present studies examine the mechanism of this effect and showthat (/<•novo ceramide generation through the enzyme ceramide synthase

is required. TPA induced rapid ceramide generation, which was detectable by l h and increased linearly for 12 h. TPA-induced apoptosis was

measurable by 12 h and was progressive for 48 h. Investigations into themechanism of TPA-induced ceramide generation revealed that acid and

neutral sphingomyelinase activities were not enhanced. However, TPAinduced an increase in ceramide synthase activity that persisted for atleast 16 h. Treatment with (unionism B,, a specific natural inhibitor ofceramide synthase, abrogated both ceramide production and TPA-in

duced apoptosis. Ceramide analogues bypassed fumonisin B, inhibition toinitiate apoptosis directly. Thus, ceramide appears to be a necessary signalfor TPA-induced apoptosis in LNCaP cells. This represents the first

description of a pathway by which PKC may signal apoptosis.

INTRODUCTION

The PKC4 family of serine/threonine kinases is involved in signal

transduction of a myriad of biological effects. Responses demonstrated to occur secondary to phorbol ester-mediated PKC activation

include cellular transformation, proliferation, differentiation, and resistance to apoptosis (1,2). Phorbol ester treatment protects cells fromundergoing apoptosis caused by TNF-a (3, 4), chemotherapy (5),

growth factor withdrawal (6), and irradiation (7), whereas inactivationof PKC with specific inhibitors can induce cell death directly (8-10).

Alternatively, phorbol esters have been shown to induce apoptosis ina few cell types via PKC activation (11-14).

One cell line in which phorbol ester activates PKC and signalsapoptosis is the human prostate cancer cell LNCaP. Studies indicatethat the phorbol ester, TPA, causes selective membrane translocationof PKCa, and pretreatment with PKC inhibitors blocks apoptosis (11,15). Because exposure of LNCaP cells to analogues of the lipidsecond messenger ceramide also induced apoptosis (16), we hypothesized that ceramide might mediate apoptosis in response to TPA.

Ceramide has emerged as a signal for induction of apoptosis by avariety of stresses in numerous cellular types (17, 18). Stimulatedelevation in cellular ceramide has been shown to occur by two distinctpathways. In most instances, cytokines, hormones, and environmentalstresses induce rapid catabolism of membrane-bound sphingomyelin

Received 12/29/97; accepted .1/18/98.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 with18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported by CaPCURE foundation award to A. H. F. and a grant to

R. N. K. and Z. F. from the Pepsico Foundation.2 These authors made significant and independent contributions.' To whom requests for reprints should be addressed, at Laboratory of Signal Trans

duclion. Memorial Sloan-Kcltering Cancer Center. 1275 York Avenue. New York. NY10021. Phone: (212)639-8573; Fax: (212)639-2767.

4 The abbreviations used are: PKC, protein kinase C; TNF. tumor necrosis factor; TPA.

12-O-tetradecanoylphorbol-13-acetate; FB,. fumonisin B,; OPA. o-phthaldehyde; DAG.diacylglycerol; HPLC. high-performance liquid chromatography.

to ceramide by the action of neutral or acidic sphingomyelinases (17,18). Alternatively, ceramide may be synthesized de novo via thecondensation of the sphingoid base sphinganine and fatty acyl-CoA

catalyzed by the enzyme ceramide synthase (also termed sphingosinejV-acyltransferase; EC2.3.1.24) to form dihydroceramide, which is

subsequently oxidized to ceramide.Bose et al. (19) reported that prolonged activation of ceramide

synthase can be induced by daunorubicin treatment of P388 and U937cells (19). The Vmax of ceramide synthase was increased by 70%,resulting in sustained elevations of ceramide (and sphingomyelin)levels for at least 8-12h. Furthermore, FB, a fungal toxin that specifically inhibits ceramide synthase, blocked daunorubicin-inducedceramide generation and apoptosis. Because FB, did not block TNF-

induced ceramide generation that occurred by sphingomyelinase activation nor TNF-initiated apoptotic death, it was concluded that the

mechanism of cell death involving ceramide synthase activation wasspecific for daunorubicin. Recently, FB, has been shown to abrogatedaunorubicin-induced apoptosis in hen granulosa cells (20) and CPT-11-induced apoptosis in L929 cells (21).

In the present studies, we explored the mechanism by which phorbol ester signals the apoptotic response in LNCaP cells. We havedefined an obligatory role for ceramide generation via activation ofceramide synthase. These studies represent the first description of amechanism by which activation of the PKC pathway can signalapoptosis.

MATERIALS AND METHODS

Materials

Fatty acid-free BSA, palmitoyl-CoA, bis-benzimide (Hoechst 33258), cer

amide Type III. FB,. DMSO. OPA, and sphinganine were purchased fromSigma Chemical Co. D-?rv'/im-C,4-Sphingosine was from Matreya, Inc. TPAwas purchased from LC Laboratories. C2-ceramide, C2-dihydroceramide, and

Escherichia coli diacylglycerol kinase were from Biomol. Cardiolipin waspurchased from Avanti Polar Lipids. [l-14ClPalmitoyl-CoA (57.1 mCi/mmol)and [-y-"P]ATP (3000 Ci/mmol) were obtained from DuPont (NEN). Octyl-

ß-D-glucopyranoside was from Calbiochem. C2-ceramide was stored at-20°C and dissolved in DMSO (final concentration, 0.4%) just prior to each

experiment. FB, was diluted in sterile saline (0.9%) and stored at 4°Cuntil use.TPA was dissolved in DMSO, aliquoted, and stored at -20°C until use.

Cell Culture

LNCaP cells were cultured in RPMI 1640 supplemented with 10% fetalbovine serum. For apoptosis and lipid studies, cells were plated at a density of1.5-2.5 x IO5 cells/ml onto either 6- or 12-well plates and treated 36 h later.

For ceramide synthase assays, cells were plated at a similar density onto100-mm dishes.

PKC Studies

Cells (20 X 106/point) were homogenized in lysis buffer [5 mM NaCl, 4 mM

Tris-HCl (pH 7.4), 2 mM EDTA. 5 mM EGTA. 1 mM MgCU. 2 mM DTT, 10

fig/ml each leupeptin, aprotinin, and soybean trypsin inhibitor, and 1 mMphenylmethylsulfonyl fluoride] at 4°C,and cytoplasm (supernatant) and mem

brane (pellet) fractions were separated by centrifugation at 100,000 X g for

2260

Research. on October 22, 2015. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

CERAMIDE SYNTHASE SIGNALS LNCaP APOPTOSIS

1 h. The membrane was resuspended in lysis buffer containing 1% TritonX-100, incubated on ice with frequent mixing for 30 min, and then centrifuged

for 30 min at 100,000 X g to remove insoluble material. Cytoplasmic andmembrane-bound PKCs were partially purified by passage through a DEAE-

cellulose column, and PKC activity was measured using the BIOTRAK ProteinKinase C Enzyme Assay System according to the manufacturer's instructions

(Amersham International, Buckinghamshire, United Kingdom).

Lipid Studies

DAG Kinase Assay. Following incubation with TPA, FB,, or both. LNCaPcells, which are typically loosely adherent, were detached by gentle pipetting.The cells were pelleted by centrifugation (500 X g for 5 min), washed withice-cold PBS, and extracted with 2 ml of chloroform:methanol:l N HCI

(100:100:1 v/v/v), 540 /il of buffered saline solution (135 mM NaCl. 1.5 mm

CaCl2, 0.5 mM MgCl2, 5.6 mM glucose, and 10 mM HEPES, pH 7.2). and 60/il of 100 mM EDTA. Lipids in the organic phase extract were dried under N2and subjected to mild alkaline hydrolysis (0.1 N methanolic KOH for l h at37°C)to remove glycerophospholipids. Samples were re-extracted, and the

organic phase was dried under N2. Ceramide was quantified by the DAGkinase assay as described (22) .

HPLC Assay. Cell ceramide content was also determined by a non-enzy

matic method, which is a modification of a technique used for amino acidanalyses (23). Lipids were extracted as described for the DAG kinase assayabove. Ceramide in the organic phase was measured after deacylation tosphingoid base and derivitization with OPA as described by Merrill et al. (23).Briefly, aliquots of the organic phase were mixed with D-<?ryr/iro-C,4-sphin-

gosine, which was used as an internal standard. The organic phase was driedunder N2, resuspended in 0.5 ml of l N KOH in 90% methanol, and incubatedfor 1.5 h at 100°Cto deacylate ceramide to C18-sphingosine (24). Incubations

were neutralized with 0.5 ml of l N HC1 in methanol and extracted by adding1 ml of CHC13 and 750 /¿Iof l N NaCl. The extracted lipids were dried underN2, dissolved in 100 /il methanol. and mixed with 100 fil of OPA reagent,which was prepared fresh daily by mixing 99 ml of 3% (w/v) boric acid inwater (pH adjusted to 10.5 with KOH) and 1 ml of ethanol containing 50 mgof OPA and 50 /tl of 2-mercaptoethanol. After incubation for 5 min at 22°C,

500 fu of methanol:5 mM potassium phosphate (pH 7.0; 90:10. v/v) wereadded, and the samples were clarified by brief centrifugation. Aliquots (20 /il)were quantified by reverse-phase HPLC using a Nova Pak C18 column (60 A,

4 jim, 3.9 X 150 mm; Waters). Fluorescent lipids were eluted isocratically

with methanol:5 mM potassium phosphate, pH 7.0 (90:10), at a flow rate of 0.6ml/min, detected by spectrofluorometer (excitation wavelength, 340 nm; emission wavelength, 455 nm), and corrected for recovery using the D-erythro-Cl4-

sphingosine internal standard. Ceramide levels were determined by comparison with a concomitantly run standard curve of known amounts of ceramide(Sigma type III). The retentions time for C14-sphingosine and D-erythro-C,g-

sphingosine were 4 and 10 min, respectively.Direct Comparison of Ceramide Levels by the DAG Kinase and HPLC

Assays. A recent study has questioned the validity of the DAG kinase assayfor measuring ceramide levels (25). To address this issue, we compared thequantity of ceramide measured by the DAG kinase assay with the HPLCmethod using lipid extracts from LNCaP cells treated with TPA (10 ng/ml) ordiluent for various times between 0 and 12 h. Fig. 1 shows that the ceramidelevels determined by these methods correlated directly as determined by linearregression analysis (r = 0.943). Using the t test for correlation, this r value was

found to be highly significant (P < 0.005), indicating that both methods

register the same answer.Ceramide Synthase Assay. Microsomal membranes were prepared as

described previously (19). Briefly, cells were washed with ice-cold PBS,

scraped off the plate, and resuspended in 300 /il homogenization buffer [25mM HEPES (pH 7.4), 5 mM EGTA, 50 HIM NaF, and 10 /ig/ml each ofleupeptin and soybean trypsin inhibitor]. Cells were disrupted by sonication onice, and lysates were pelleted at 800 X g for 5 min. The postnuclear supernatant was centrifuged at 250,000 X g for 35 min. The microsomal membranepellet was resuspended in 0.5 ml of homogenization buffer. Assays of ceramide synthase activity were performed as described previously (19). Briefly,microsomal membrane protein (75 /ig) was incubated with 2 mM MgCl2, 20/iM fatty acid-free BSA, indicated concentrations of sphinganine, 70 /IMunlabeled palmitoyl-CoA, and 0.2 /iCi of [l-'4C]palmitoyl-CoA at 37°Cfor

Uo.X *

5-8•¿�°«,0) O!5£iIIa °-0

750-

500-

250-

250 500 750

Ceramide by DAG Kinasepmol/106 cells

Fig. 1. Ceramide determinations by DAG kinase and HPLC correlate linearly. Lipidswere extracted from 0.5 X IO6 cells treated with TPA (10 ng/ml) or diluent lor various

times between 0 and 12 h. Identically treated samples were then assayed for ceramidelevels using either the DAG kinase or HPLC methods. The ceramide levels determined hythese methods correlated closely, as determined by linear regression analysis (r = 0.943).Using the I test for correlation, this r value was found to be highly significant (P < 0.005 ).indicating that both methods register the same answer.

1 h. Following extraction of the lipids with 2 ml of chlorofornrmethanol ( 1:2),radiolabeled dihydroceramide was resolved by TLC using a solvent system ofchloroform:methanol:3.5 N ammonium hydroxide (85:15:1, v/v/v), detected byiodine vapor staining based on comigration with commercially available ceramide (Sigma type III), and quantified hy liquid scintillation counting. Underthe conditions used, the substrate was not rate limiting, and the reaction waslinear for time and enzyme concentration.

Quantification of Apoptotic Cells. Morphological changes in the nuclearchromatin of cells undergoing apoptosis were detected by staining with theDNA-binding fluorochrome bis-benzimide trihydrochloride (Hoechst-33258),

as described previously (26). Briefly, LNCaP monolayers were detached with0.25% trypsin and 0.02% EDTA in HBSS and combined with the floatingpopulation. The cell pellet was then washed in PBS, resuspended in 50 n\ of3% paraformaldehyde. and incubated for 10 min at 22°C.The fixative was

removed, and the cells were resuspended in 20 /il of PBS containing 8 /ig/mlof Hoechst-33258. After a 15-min incubation at 22°C,an 8-/il aliquot was

placed on a glass slide, and 400 cells were scored for the incidence of apoptoticchromatin changes under an Olympus BH2 fluorescence microscope using aBH2-DMU2UV Dich Mirror Cube filter. Cells displaying at least three apop

totic bodies were scored as apoptotic.

Statistical Analysis

Statistical analysis was performed by Student's t test and I test for

correlation coefficient. Linear regression analysis was by the method ofleast squares.

RESULTS

TPA Activates PKCa and Induces Apoptosis in LNCaP. Initialstudies compared the effect of TPA on PKC activation and apoptosis.These experiments extended the investigations of Powell et al. (15),which showed, by Western analysis, that TPA induced selective andprolonged translocation of PKCa from cytoplasm to membrane. Similarly, we demonstrated translocation of greater than 90% of PKCactivity from cytoplasm to membrane by 15 min of TPA stimulation(data not shown). This resulted in a 6-7-fold increase in membrane-

bound activity. Increased activity in the membrane persisted for up to24 h; the ED50 for this event was 3-4 ng/ml TPA (data not shown).

We next examined the apoptotic response of LNCaP to TPAtreatment. For these studies, cells were treated with 1-100 ng/ml TPAfor the indicated times, followed by bis-benzimide staining to assess

apoptosis. Apoptosis was evident after 12 h of stimulation with TPA

2261

Research. on October 22, 2015. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

CEKAMIDK SYNTHASK SIGNALS LNCaP APOPTOSIS

60-

40

20-

S—a—»—H—aj. •¿�*••¿�*•J. i

B40

-30

-20

-10

- 0

0 8 12 16 24 48 0 2 4 6 8 10

1500-

S 1250-^O

•¿�S1000-«

Ets 750-

1200

- 1000

800

600

3 6 9 12 15

Time (h)

0 2 4 6 8 10

TPA (ng/ml)

Fig. 2. Effect of TPA on ceramide levels and apoptosis in LNCaP cells. A. apoptosistime course. Intact LNCaP cells were exposed to H) ng/ml TPA or diluent DMSO (finalconcentration. 0.4%). Quantification of morphological changes of nuclear apoptosis wasperformed hy staining the cells with the DNA-specific fluorochrome Hoechst-33258.

Segmentation of the nucleus into three or more chromatin fragments was consideredapoplotic. Bach value represents the mean of duplicale determinations from three independent experiments; bars, SE. A minimum of 400 cells was scored for the incidence ofapoptosis in each experiment, fi, apoptosis dose response. LNCaP cells were treated as inA for 24 h with l-IO ng/ml TPA. Cells were scored for apoptosis as in A. Each value

represents the mean ±range of duplicate determinations from two independent experiments; bars, SE. C, time course of ceramide generation. LNCaP cells were treated as inA for the indicated times. Lipids were extracted, and ceramide levels were quantified bythe DAG kinase reaction as described in "Materials and Methods." Each value represents

Ihe mean of duplicate determinations from three experiments; bars, SE. D, dose dependence of ceramide generation. LNCaP cells were exposed to I-K) ng/ml TPA for 6 h, andceramide was quantified as in C. Each value represents the mean of triplicate determinations from one of three similar experiments; bars, SE.

and was progressive for 48 h (Fig. 2a). TPA-induced apoptosis wasconcentration dependent (Fig. 2b), with 3-10 ng/ml TPA elicitingmaximal cell death; the EDM) for this event was ~2 ng of TPA/ml

(Fig. 1b). Thus, the ED50 for apoptosis correlated closely with theEDM) for translocation of PKCa activity to membrane upon TPAtreatment. Removal of TPA after 20 min did not affect the level ofapoptosis at 24 h (data not shown). Morphological changes of apoptosis were accompanied by DNA ladder formation (data not shown).The delay in induction of LNCaP apoptosis after TPA suggests that anintermediary event might be involved.

TPA Induces Ceramide Elevation in LNCaP Cells. To determine whether ceramide elevation occurred prior to TPA-induced

apoptosis, we measured ceramide levels after treatment with thisagent. As shown in Fig. 2c, exposure of LNCaP cells to TPA (10ng/ml) resulted in a 40% increase in ceramide levels within 3 h. Asmaller increase in cellular ceramide levels was routinely detected by1 h of TPA stimulation (data not shown). Ceramide generation waslinear for 12 h, at which point the level of ceramide had increased100% over baseline. Increased ceramide levels were sustained for atleast 24 h (data not shown). The dose-dependence of TPA-induced

ceramide generation at 6 h was similar to that for TPA-induced

apoptosis. Three to 10 ng/ml produced maximal elevation in ceramide; an ED50 for this event was ~ 1 ng/ml (Fig. Id). These studies

indicate that ceramide generation occurs early after TPA treatmentand precedes the onset of apoptosis by many hours.

Mechanism of Ceramide Generation in LNCaP Cells. Stimulated elevation in cellular ceramide levels has been shown to occur byhydrolysis of sphingomyelin to ceramide via the action of neutral oracidic sphingomyelinases or by activation of the synthetic enzyme,ceramide synthase (18). Initial studies examined the effect of a maximal dose of TPA for induction of apoptosis on neutral and acidicsphingomyelinase activities. LNCaP cells display detectable levels ofneutral and acidic sphingomyelinase, with maximal velocities of 100pmol/mg protein/min and 1200 pmol/mg protein/min, respectively. Atno time between 1 min and 24 h was either sphingomyelinase activityelevated (n = 5). We therefore investigated whether the TPA effect

represented de novo synthesis of ceramide via activation of ceramidesynthase. Cells were treated with TPA for 45 min, and microsomalmembranes were prepared. The kinetics of ceramide synthase activitywith increasing sphinganine concentrations are depicted in Fig. 3a.Eadie-Hofstee transformation of these data revealed that LNCaP cells

manifest a basal level of ceramide synthase activity with a maximalvelocity of 100 pmol/min/mg protein and a Km of 1.4 JLIM.A similarmaximal velocity was obtained using palmitoyl-CoA as substrate(data not shown); the Km for palmitoyl-CoA was 1.0 /UM.A 60%

increase in the Vmaxwith no change in Km occurred in response to 100ng/ml TPA (Fig. 3b). Similar results were obtained at 16 h of TPAtreatment and with a dose of 10 ng/ml TPA (data not shown). A TPA

120-

1 90

I

oa

Control

TPA (100 ng/ml)

160

100

50

0 2 4 6 8 10 0 1 10

Sphinganine (\iM)

B

Vmax (pmol/min/mg)

Diluent 100

100

TPA 160

TPA (ng/ml)

Km (uM)

1.4

1.4

Fig. 3. Measurement of ceramide synthase activity in response to TPA. A, ceramidesynthase activity. TPA ( lOOng/ml I was added to LNCaP cells for 45 min, and microsomalmembranes were prepared as in "Materials and Methods." The ceramide synthase reaction

mixture containing microsomal membrane protein (75 fig), 2 mm MgCU, 20 /¿Mfattyacid-free BSA. 0.5-IO JIM sphinganine, 70 >¿Munlabeled palmitoyl-CoA, and 0.2 /¿Ciof[' '4C]palmiloyl-CoA was incubated at 37°Cfor I h. Lipids were extracted, and radio-

labeled dihydroceramide was resolved by TLC using a solvent system of chloroform:methanol;3.5 N ammonium hydroxide (85:15:1), detected by iodine vapor staining basedon comigration with commercially available ceramide (Sigma type III), and quantified hyliquid scintillation counting. Each value represents the mean of four separate experiments;bars, SE. B. determination of Km and Vmzfi from A using Eadie-Hofstee analysis. C doseresponse. LNCaP cells were exposed to I-IOO ng/ml TPA and assayed for ceramidesymhase activity as described in A using 5 H.Msphinganine. Each value represents themean of triplicate determinations from one of three separate studies; bars, SE.

2262

Research. on October 22, 2015. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

CERAMIDE SYNTHASE SIGNALS LNCaP APOPTOSIS

1500

S&

1000

750-

500

•¿�30

oI-10

0 4 8 12 0 10 20 30 40 50

Time (h)

Fig. 4. Effect of FB, on TPA-induced ceramide generation and apoptosis. A. TPA (10ng/ml) and/or FB, (lOO ¿uu)were added to LNCaP cells prepared as in Fig. 2, and cellswere collected 6-12 h later. Ceramide was quantified by the DAG kinase reaction as in

Fig. 2. Each value represents the mean ±range of duplicate determinations from one ofthree separate experiments; bars. SE. B, LNCaP cells were treated as in A with FB, (100/¿M)and/or TPA (3 ng/ml). and the incidence of apoptosis was determined as in Fig. 2.Each value represents the mean ±range of duplicate determinations from one of fiveseparate studies; bars, SE.

dose of 1 ng/ml was sufficient to increase the activity of ceramidesynthase (Fig. 3c), and a near-maximal effect occurred with 10 ng/ml

TPA (Fig. 3c). Higher concentrations of TPA (30 and 100 ng/ml) didnot elicit statistically greater increases in enzymatic activity. Thesestudies demonstrate that TPA induced an increase in ceramide synthase activity preceding the elevation in ceramide levels.

In previous studies, ceramide, synthesized via ceramide synthase inresponse to daunorubicin treatment of P388 murine leukemia cells,was subsequently converted to sphingomyelin (19). Ceramide generation in LNCaP cells was also accompanied by sphingomyelin synthesis (data not shown). Resting LNCaP cells contain —¿�3000pmol ofsphingomyelin/106 cells, as measured by phosphorus assay (data not

shown). TPA treatment resulted in an increase in sphingomyelincontent after 3 h, which peaked at 138 ±4% of control at 7 h of TPAstimulation (P < 0.001 versus control; data not shown).

FBj Blocks TPA-induced Ceramide Generation and Apoptosis.To further examine the role of ceramide synthase in TPA-induced

ceramide elevation and apoptosis, experiments were performed usingFB,, a specific competitive inhibitor of this enzyme. Treatment ofLNCaP cells with 100 JAMFB, immediately prior to phorbol estermarkedly reduced TPA-induced ceramide elevation (Fig. 4a). In sep

arate studies, 50 ¡J.MFB, was about half as effective (data not shown).FB, had no effect on basal ceramide levels under the conditions ofthese studies (data not shown). FB, treatment also blocked TPA-

induced apoptosis (Fig. 4b). Pretreatment with FB, for up to 8 h priorto TPA did not result in greater inhibition of apoptosis compared withFB, treatment immediately before the addition of TPA (data notshown).

To provide further evidence that TPA-induced apoptosis of LNCaP

cells was mediated via ceramide generation, cells were treated withC2-ceramide, a cell-permeable ceramide analogue. C2-ceramide in

duced apoptosis to a similar extent as TPA. However, ceramideanalogue-induced apoptosis was not inhibited by the addition of FB,

(Fig. 5), indicating that ceramide exerts its effect downstream of FB,.TPA treatment with up to 250 ng/ml (a dose 80 times that which givesmaximal apoptosis) did not overcome the inhibition of cell death byFB, (data not shown), suggesting that FB, was not acting as acompetitive inhibitor of PKC activation by TPA.

To further define the structural determinants for ceramide signaling,cells were treated with exogenous C2-dihydroceramide, a ceramideanalogue that lacks the trans C4-5 double bond on the sphingoid

backbone. Exogenous dihydroceramide is poorly converted to ceramide in mammalian cells (26), and at equimolar concentrations (up to50 fjLM),C2-dihydroceramide failed to induce apoptosis (data not

shown).

DISCUSSION

The present studies define a mechanism by which TPA signalsapoptosis in LNCaP prostate cancer cells. TPA induced ceramideelevation within 1 h of treatment, and the level of cellular ceramideincreased linearly for 12 h. Thus, ceramide elevation preceded apoptosis, which was detected at 12 h. There was a close correlationbetween the dose dependence of PKCa activation, ceramide elevation, and apoptosis, all of which manifested ED5()s between 1 and 4 ngof TPA/ml. The enzymatic mechanism of TPA-induced ceramideelevation was determined and involved an increase in the activity ofceramide synthase, the regulatory enzyme for ceramide synthesis.TPA signaled ceramide synthase activation within 45 min, and thedose dependence for this event correlated closely to ceramide generation. Ceramide synthesis appeared obligatory for induction of apoptosis by TPA because FB, blocked TPA-induced ceramide generation

and cell death, and ceramide analogues bypassed the blockade andrestored the apoptotic response. These studies comprise the firstevidence for activation of ceramide synthase via a signal transductionpathway.

Most often, phorbol esters have been shown to promote an anti-apoptotic state (27), perhaps via activation of the mitogen-activated

protein kinase cascade (28). However, in a few instances, in additionto LNCaP, phorbol esters have been shown to be proapoptotic. In thisregard, EBV-infected Burkitt Lymphoma cells display substantial

apoptosis in response to phorbol esters, whereas uninfected Burkittcells are resistant (29). Furthermore, immature thymocytes appear tobe susceptible to phorbol ester-induced apoptosis (13). Perhaps most

relevant to the present studies are the findings of de Vente et al. (14).These investigators overexpressed PKCa in MCF-7 breast cancer

cells and compared the effects of TPA on the parental and modifiedlines. The MCF-7 parental line was growth arrested in response totreatment, whereas the MCF-7-PKCa line exhibited a cytotoxic re

sponse. These studies indicate that the status of PKCa may regulatethe type of response elicited by phorbol ester treatment.

Our studies on the susceptibility of LNCaP cells to TPA-induced

Ooo

Ia 20

10-

o-y10 20 30 40 50

Time (h)

Fig. 5. C2-ceramide-induced apoptosis is not inhibited by FB,. LNCaP cells weretreated with 50 /AMC2-ceramide in the presence or absence of 100 fiM FB,, andapoptosis was quantified by bis-benzimide staining as described in Fig. 2. Each valuerepresents the mean ±range of duplicate determinations from one of three separateexperiments; bars. SE.

2263

Research. on October 22, 2015. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

CBRAMIDE SYNTHASE SIGNALS LNCaP APOPTOSIS

apoptosis are consistent with and extend the findings of other labo-

ratories(ll, 12, 15). Powell et al. (15) demonstrated that TPA inducedselective and prolonged translocation of PKCa to the membrane priorto apoptosis. Zhao et al. (30) showed that the action of TPA involvedRb dephosphorylation and G()-G, arrest, events also reported to occur

in response to ceramide treatment in other cell types (31, 32). WhetherG()-G, arrest, like apoptosis, is signaled through ceramide in LNCaP

cells requires further investigation.The present study expands the potential role of ceramide synthase

activation as a mechanism for generation of ceramide during induction of apoptosis. In previous studies from our laboratory, evidencewas presented that daunorubicin-induced apoptosis, presumably sec

ondary to DNA damage, was mediated by ceramide synthase activation. Witty et al. (20) corroborated this finding, showing that daunorubicin-induced, but not UV-induced, apoptosis was abolished by FB,

in hen granulosa cells. Recently, Suzuki et al. (21) showed that FB,could inhibit apoptosis secondary to treatment of murine fibroblastswith the chemotherapeutic agent CPT-11 and molecularly ordered

ceramide generation upstream of caspases. Whether ceramide generated via the synthase interacts with the same targets as ceramidegenerated through sphingomyelinases is presently uncertain. However, the recognition that ceramide synthase is preferentially concentrated in mitochondria and that ceramide analogues induce generationof reactive oxygen species in isolated mitochondria by inhibition ofcomplex III and permeability transition strongly suggest a role forceramide synthase as a regulator of apoptosis at the mitochondria!level (33-35).

Prostate cancer is the most common cancer in men in the UnitedStates and accounts for 42.000 deaths annually (36). Although surgeryand radiation are effective for treatment of localized disease, there ispresently no effective chemotherapeutic regimen for metastatic prostate cancer. The present studies constitute the first evidence of asignaling pathway for the death response in prostate cancer. Furtherunderstanding of pathways involved in the death of prostate cancercells might allow for the development of new systemic therapies forthis disease. Whether ceramide is more generally involved in the deathresponse of prostate cells to other stresses, including androgen-with-

drawal and ionizing radiation, is presently unknown.

REFERENCES

1. Blobe, G. C., Obeid, L. M.. and Hannun. Y. A. Regulation of protein kinase C androle in cancer biology. Cancer Metastasis Rev. 13: 411-431. 1994.

2. Dekker. L.. and Parker. P. Protein kinase C-a question of specificity. Trends Bio-chem. Sci.. 19: 73-77. 1994.

3. Obeid. L. M.. Linardic, C. M., Karolak, L. A., and Hannun, Y. A. Programmed celldeath induced by ceramide. Science (Washington DC). 259: 1769-1771. 1993.

4. Jarvis. W. D.. Kolesnick. R. N.. Pomari. F. A.. Traylor. R. S., Gewirtz, D. A., andGrant. S. Induction of apoptotic DNA damage and cell death by activation of thesphingomyclin pathway. Proc. Nati. Acad. Sci. USA, 91: 73-77. 1994.

5. Grant, S., Jarvis, W., Swerdlow, P., Turner, A.. Traylor, R., Wallace. H.. Lin, P.,Pcttit, G., and Gewirtz. D. Polenliation of the activity of l -ß-n-arabinofuranosylcy-tosine by the protein kinase C activator bryostatin 1 in HL-60 cells: association withenhanced fragmentation of mature DNA. Cancer Res.. 52: 6270-6278. 1992.

6. Lotem, J., Cragoe, E. J.. and Sachs. L. Rescue from programmed cell death inIcukemic and normal myeloid cells. Blood. 78: 953-960. 1991.

7. Haimoviti-Friedman. A.. Balaban. N.. McLoughlin. M., Ehleiter, D.. Micheali. J..Vlodavsky. I., and Fuks. '¿.Protein kinase C mediates basic fibroblast growth factor

protection of endothclial cells against radiation-induced apoptosis. Cancer Res., 54:2591-2597. 1994.

8. Jarvis, W. D.. Turner. A. J.. Porvirk. L. F., Traylor, R. S.. and Grant. S. Induction ofapoptotic DNA fragmentation and cell death in HL-60 human promyelocytic leukemia cells by pharmacological inhibitors of protein kinase C. Cancer Res., 54:1707-1714, 1994.

9. McConkey. D. J.. Hartzell, P.. Jondal. M.. and Orrenius. S. Inhibition of DNAfragmentation in thymocytes and isolated thymocytc nuclei by agents that stimulateprotein kinase C. J. Biol. Chem.. 264: 13399-13402. 1989.

10. Munin.i S. J., Nod/enski. E.. Weichselbaum. R. R.. and Quintans. J. Protein KinaseC inhibition induces apoptosis and ceramide production through activation of aneutral sphingomyelinase. Cancer Res., 56: 2711-2714. 1996.

11. Young, C. Y., Murtha, P. E.. and Zhang. J. Tumor-promoting phorbol ester-induced

cell death and gene expression in a human prostate adenocarcinoma cell line. Oncol.Res.. 6: 203-210, 1994.

12. Day, M. L.. Zhao, X., Wu, S., Swanson. P. E.. and Humphrey. P A. Phorbolester-induced apoptosis is accompanied by NGFI-A and c-fos activation in androgen-sensitive prostate cancer cells. Cell Growth Differ., 5: 735-741, 1994.

13. Kizaki, H.. Tadakuma, T.. Odaka. C.. Muramalsu. J., and Ishimura. Y. Activation ofa suicide process of thymocytes through DNA fragmentation by calcium ionophoresand phorbol esters. J. Immunol.. 143: 1790-1794, 1989.

14. de Vente, J. E., Kukoly, C. A.. Bryant. W. O., Posekany, K. J., Chen. J., Fletcher.D. J., Parker, P. J., Pettit, G. J., Lozano, G., Cook. P. P., and Ways, D. K. Phorbolesters induce death in MCF-7 breast cancer cells with altered expression of proteinkinase C isoforms. Role for p53-independent induction of gadd-45 in initiating death.J. Clin. Invest., 96: 1874-1886, 1995.

15. Powell, C. T.. Brittis. N. J.. Stec. D., Hug, H., Heston, W. D., and Fair, W. R.Persistent membrane translocation of protein kinase Ca during 12-O-tetradecanoyl-phorbol-13-acetate-induced apoptosis of LNCaP human prostate cancer cells. CellGrowth Differ., 7: 419-428, 1996.

16. Herrmann. J. L.. Menter. D. G.. Beham. A., von Eschenhach. A., and McDonnell.T. J. Regulation of lipid signaling pathways for cell survival and apoptosis by bcl-2in prostate carcinoma cells. Exp. Cell Res.. 234: 442-451, 1997.

17. Hannun. Y. A. Functions of ceramide in coordinating cellular responses to stress.Science (Washington DC), 274: 1855-1859. 1996.

18. Spiegel, S.. Foster, D. A., and Kolesnick. R. N. Signal transduction through lipidsecond messengers. Curr. Opin. Cell Biol., 8: 159-167, 19%.

19. Böse,R.. Verheij. M.. Haimovitz-Friedman. A., Scotto. K., Fuks, Z.. and Kolesnick.R. Ceramide synthase mediates daunorubicin-induced apoptosis: an alternative mechanism for generating death signals. Cell, 82: 405-414, 1995.

20. Witty. J. P., Bridgham, J. T., and Johnson, A. L. Induction of apoptotic cell death inhen granulosa cells by ceramide. Endocrinology, 137: 5269-5277, 1996.

21. Su/uki, A., Iwasaki, M., Kalo, M., and Wagai, N. Sequential operation of ceramidesynthesis and ICE cascade in CPT-11-initiated apoptotic death signaling. Exp. CellRes., 233: 41-47, 1997.

22. Dressier. K. A., and Kolesnick. R. N. Ceramide 1-phosphate, a novel phospholipid inhuman leukemia (HL-60) cell synthesis via ceramide from sphingomyelin. J. Biol.Chem., 265: 14917-14921. 1990.

23. Merrill, J-A. H.. Wang, E.. Mullins. R. E.. Jamison. W. C. L., Nimkar, S.. and Liotta,D. Quantitation of free sphingosine in liver by high performance liquid chromatog-raphy. Anal. Biochem., 171: 373-381, 1988.

24. Veldhoven, P. P. V.. Bishop. W. R.. and Bell, R. M. Enzymatic quantification ofsphingosine in the picomole range in cultured cells. Anal. Biochem.. 183: 177-189,

1989.25. Watts, J., Gu, M.. Polverino. A., Patterson. S., and Aebcrsold, R. Fas-induced

apoptosis of T cells occurs independently of ceramide generation. Proc. Nati. Acad.Sci. USA, 94: 7292-7296. 1997.

26. Kok. J. W.. Nikolova-Karakashian. M.. Klappe. K.. Alexander. C.. and Merrill. A. H.,Jr. Dihydroceramide biology. J. Biol. Chem., 272: 21128-21136, 1997.

27. Haimovitz-Friedman. A., Kan. C. C.. Ehleiter. D.. Persaud. R. S., McLoughlin, M..

Fuks, Z.. and Kolesnick. R. N. Ionizing radiation acts on cellular membranes togenerate ceramide and initiate apoptosis. J. Exp. Med., 180: 525-535, 1994.

28. Xia. Z.. Dickens. M.. Raingeaud. J.. Davis. R.. and Greenberg. M. Opposing effectsof ERK and JNK-p38 MAP kinases on apoptosis. Science (Washington DC), 270:1326-1331, 1995.

29. Ishii, H., and Gobe. G. Epstein-Barr virus infection is associated with increasedapoptosis in untreated and phorbol ester-treated human Burkitt's lymphoma (AW-

Ramos) cells. Biochem. Biophys. Res. Commun., /92: 1415-1423. 1993.

30. Zhao. X., Gschwend, J., Powell. C.. Foster. R.. Day, K., and Day. M. Retinoblastomaprotein-dependent growth signal conflict and caspase activity are required for proteinkinase C-signaled apoptosis of prostate epithelial cells. J. Biol. Chem.. 272: 22751-

22757, 1997.31. Dbaibo. G. S.. Pushkareva. M. Y.. Jayadev. S.. Schwarz. J. K., Horowitz. J. M..

Obeid, L. M.. and Hannun. Y. A. Relinoblasloma gene product as a downstreamtarget for a ceramide-dependent pathway of growth arrest. Proc. Nati. Acad. Sci.USA. 92: 1347-1351. 1995.

32. Kuroki. J.. Hirokawa. M., Kitabayashi, A., Lee, M., Horiuchi. T.. Kawabata, Y.. andMiura, A. B. Cell-permeable ceramide inhibits the growth of B lymphoma Raji cellslacking TNF-a-receptors by inducing GO/G1 arrest but not apoptosis-a new model fordissecting cell-cycle arrest and apoptosis. Leukemia (Baltimore), W: 1950-1958,

1996.33. Garcia-Ruiz, C.. Colell. A.. Mari, M., Morales. A., and Femandez-Checa, J. C. Direct

effect of ceramide on the mitochondrial electron transpon chain leads to generationof reactive oxygen species. Role of mitochondrial glutathione. J. Biol. Chem., 272:11369-11377, 1997.

34. Arora. A. S.. Jones, B. J., Patel. T. C., Bronk, S. F.. and Gores, G. J. Ceramide induceshepatocyte cell death through disruption of mitochondrial function in the rat. Hepa-tology. 25: 958-963. 1997.

35. Pastorino, J., Simbula. G., Yamamoto. K., Glascott, P. J., Rothman. R., and Farber,J. The cytotoxicity of tumor necrosis factor depends on induction of the mitochondrialpermeability transition. J. Biol. Chem.. 27/: 29792-29798. 19%.

36. Parker, S., Tong, T.. Bolden. S.. and Wingo. P. Cancer statistics, 1997. CA CancerJ. Clin., 47: 5-27. 1997.

2264

Research. on October 22, 2015. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

1998;58:2260-2264. Cancer Res   Mark Garzotto, Margaret White-Jones, Youwei Jiang, et al.   LNCaP Cells Is Mediated through Ceramide Synthase

-Tetradecanoylphorbol-13-acetate-induced Apoptosis inO12-

  Updated version

  http://cancerres.aacrjournals.org/content/58/10/2260

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  .permissions@aacr.orgDepartment at

To request permission to re-use all or part of this article, contact the AACR Publications

Research. on October 22, 2015. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from