[CANCER RESEARCH 50, 1879-1885. March 15, 1990]

Conjugation of 9-Deoxy-A9,A12(E)-prostaglandin D2 with Intracellular Glutathione

and Enhancement of Its Antiproliferati ve Activity by Glutathione Depletion1

Jacob Atsmon, Michael L. Freeman, Michael J. Meredith,2 Brian J. Sweetman, and L. Jackson Roberts II'

Department of Pharmacology and Medicine fJ. A., B. J. S., L. J. R.j, I'anderbilt Center for Radiation Oncology [M. L. F.], Department of Biochemistry and the Center

of Molecular Toxicology [M. J. M.¡,Vanderbilt University School of Medicine, Nashville, Tennessee 37232

ABSTRACT

The major dehydration product of prostaglandin D2, 9-deoxy-A9,A'2(E)-prostaglandin D2, is a potent cytotoxic compound. Like other

cytotoxic prostaglandins, this compound possesses an a, /3-unsaturatedketone group to which cytotoxic activity has been attributed. This prostaglandin was found to readily conjugate with glutathione (GSH) in vitro.When 9-deoxy-A',A'2(E)-prostaglandin D2 was incubated with Chinese

hamster ovary or hepatoma tissue culture cells, it was rapidly taken upand was recovered in the cell lysate primarily as a GSH conjugate inwhich the keto group at C-l 1 and the A12double bond had been reduced.

Identification of the GSH conjugate was accomplished by analysis byfast atom bombardment mass spectrometry following purification by highperformance liquid chromatography. This GSH conjugate and its cystei-nylglycinyl and cysteinyl metabolites were also identified in the cellculture medium. 9-Deoxy-A',A12(E)-prostaglandin D3 inhibited cell pro

liferation of these two cell lines in a concentration dependent manner.Depletion of intracellular glutathione by treatment with diethyl maléateand buthionine sulfoximine decreased the amount of intracellular conjugated prostaglandin recovered, and significantly enhanced the antiprolif-erative effect of 9-deoxy-A'-A'2(E)-prostaglandin D2 on the growth of

these cell lines in a concentration dependent fashion. We conclude thatintracellular GSH may modulate the antiproliferative activity of 9-deoxy-A9,A'2(E)-prostaglandin D2 and, possibly, of other cytotoxic prostaglan

dins.

INTRODUCTION

In recent years, there have been numerous reports describingthe cytotoxic activity of prostaglandin D2 in a large variety oftumor cell lines (1-8). It has been observed that PGD24 dehy

dration products are formed in the medium during incubationin cell culture. This was established by identifying these compounds in the medium, and by showing that when PGD2 isincubated in the presence of serum or vertebrate albumin,dehydration products are formed, probably due to exposure toan alkaline microenvironment in the albumin binding sites (9).These compounds exert much greater cytotoxic activity thanPGD2 and most likely are responsible for the cytotoxic activityoriginally ascribed to PGD2 (10-13). PGD2-derived compoundsare members of a larger group of eicosanoids which displaycytotoxic activity. Among them are PGA compounds which aredehydration products of prostaglandin E (14-17) and the marine ecosanoids, clavulones and punaglandins (18, 19). Theexact mechanism by which these compounds exert their cyto-

Received 6/14/89; revised 11/27/89.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 inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

' This work was supported by Grants GM15431, CA 38079. ES 03272. and

ES 00267 from the NIH.2 Present address: Department of Biochemistry, School of Dentistry, Oregon

Health Sciences University. Portland OR 97201.5To whom requests for reprints should be addressed at the Department of

Pharmacology and Medicine. Vanderbilt University School of Medicine, Nashville TN 37232-6602.

' The abbreviations used are: PGD¡.prostaglandin D¡;GSH. glutathione:PGA, prostaglandin A: PGF¡,prostaglandin F¡;PBS. phosphate-buffered saline:HPLC, high pressure liquid chromatography; CHO, Chinese hamster ovary;HTC, hepatoma tissue culture: FAB, fast atom bombardment; MS. mass spectrometry; DMSO, dimethyl sulfoxide: BSO, buthionine sulfoximine: DEM. diethyl maléate.

toxic effect is not yet fully understood but has been a subject ofmany investigations (14, 20-26). However, a common featurein all these compounds is the presence of a reactive «,0-unsat-urated ketone in the cyclopentenone ring. Prostaglandins thatlack an a,0-unsaturated ketone are not cytotoxic, and it has,therefore, been assumed that the «,#-unsaturated ketone isessential for exerting cytotoxic activity (27-29).

«,0-Unsaturated carbonyls are very susceptible to nucleo-philic addition reactions with thiols such as glutathione; themost abundant nonprotein thiol in vivo (30). GSH has beenpreviously shown to conjugate with some cytotoxic prostaglandins of the PGA group (31, 32). It was noted that the inhibitoryeffect of PGA on the proliferation of tumor cells was markedlydiminished when GSH was added to cell culture medium containing PGA (29). The interaction of a cytotoxic prostaglandinwith intracellular GSH, however, has not been reported.

We have recently shown that 9-deoxy-A9,A12(E)-PGD2, the

major dehydration product of PGD2, also conjugates in vitrowith GSH (33). This reaction is very rapid reaching completionwithin about 10 min in the absence of GSH-5-transferase andGSH-5-transferase further enhances the rate of conjugation byapproximately 2-fold. As 9-deoxy-A9,A'2(E)-PGD2 is known to

be a potent inhibitor of cell proliferation, we examined whetherthis compound may interact with intracellular GSH and howsuch an interaction may effect its antiproliferative activity. Twocell lines were studied: rat hepatoma tissue culture and Chinesehamster ovary cells.

MATERIALS AND METHODS

Unlabeled PGD2 was the generous gift of John Pike of the UpjohnCo., Kalamazoo, MI. [5,6,8,9,12,14,15-'H]PGD2 (100 Ci/mmol) (50

¿iCi)was purchased from New England Nuclear (Boston MA). Humanserum albumin (prepared from fraction V albumin) and DMSO werefrom Sigma Chemical Co. (St. Louis, MO). DL-Buthionine-5,/?-sulfox-¡minewas purchased from Chemlog (NJ), DEM was obtained fromAlfa Products (Danvers, MA). Labeled or unlabeled 9-deoxy-A',A12(E)-

PGD2 was generated by incubating PGD2 with human serum albuminas formerly described (3). Unlabeled 9-deoxy-A',A12(E)-PGD: was also

kindly supplied by Ono Pharmaceutical Co., Osaka. Japan.CHO cells, growing exponentially in monolayer culture, were main

tained at 37°Cand at pH 7.4 in McCoy's Medium 5A (GIBCO),

supplemented with 10% fetal bovine serum (GIBCO). sodium bicarbonate (2.2 g/liter), penicillin G sodium (100 units/ml), and streptomycinsulfate (100 mg/ml) (GIBCO). HTC cells were cultured under identicalconditions by using Dulbecco's modified Eagle's medium (Sigma) sup

plemented with 20% calf bovine serum (HyClone) and antibiotics asabove. Experiments with HTC cells were performed in medium containing 10% calf serum.

To determine the effect of intracellular GSH depletion on the inhibition of cell proliferation by 9-deoxy-A',A'2(E)-PGD2, 8 x 10" HTCcells or 3 x IO5CHO cells were seeded in 25-cttr tissue culture flasks

containing 4 ml of medium and were allowed to incubate for 24 h priorto the commencement of the study.

The cells were then exposed for l h to 100 ^M DEM (dissolved in0.5 ml DMSO) plus 50 ^M BSO. This was followed by rinsing twicewith Dulbecco's PBS (GIBCO) and further incubation with medium

containing 50 n\i BSO. Cells not exposed to DEM/BSO treatments1879

on August 6, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GSH DEPLETION AND 9-DEOXY-A9,YI(E)-PGD2

were incubated for l h with medium containing 0.5 HIM DMSO,followed by rinsing and incubation with BSO-free medium. 9-Deoxy-A9,A12(E)-PGD2, dissolved in 95% ethanol, was diluted serially with

medium to various concentrations and was then added to the flasks.The final concentration of ethanol was 0.1%. After 48 h of exposure to9-deoxy-A9,A12(E)-PGD2 at 37°C,the flasks were rinsed with Dulbec-co's PBS, and the cells which remained anchored to the surface were

trypsinized and counted. These experiments were repeated at least 3times with a single seed stock. Intracellular GSH levels were measuredfrom cells rinsed twice with Dulbecco's PBS and lysed in 10% perchloric

acid. GSH was assayed by high performance liquid chromatography(34).

To assess the uptake of 9-deoxy-A',A12(E)-PGD2, unlabeled and[3H)-9-deoxy-A9,A'2(E)-PGD2 were dissolved in Duibecco's PBS to a

concentration of 10 jig/ml. The growth medium was aspirated from75-cm2 flasks which contained exponentially growing cells, and thecells were rinsed twice with PBS. Prostaglandin-containing PBS wasthen applied (4 ml/flask) and the cells were maintained at 37°Cin 5%

Co2, 95% air. After incubating for 10 or 120 min, the PBS was aspiratedand acidified immediately to pH 3 by addition of l N HC1 to inhibitfurther conjugation with GSH. The cells were washed 4 times with ice-cold PBS and were then scraped into 4 ml of 1% ammonium acetatesolution (pH 3), sonicated on ice (3 times for 15 s) and centrifuged for20 min at 18,000 x g. The pellet was washed twice with the ammoniumacetate solution, dissolved in l N NaOH, and its radioactivity wasmeasured. The supernatant, as well as the incubation medium (PBS),were either extracted with ethyl acetate or applied on a Ci8 SEP-PAK(Waters Associates, Milford, MA), which was preconditioned by rinsingwith acetonitrile and 50 mM aqueous ammonium acetate (pH 3.4).After loading the sample, the SEP-PAK was rinsed with 10 ml each ofammonium acetate, heptane, and finally, with 95% ethanol. The ethanoleluate was dried under nitrogen and then subjected to reverse phasehigh pressure liquid chromatography (CI8 Alltech column, 4.6 x 250mm) with acetonitrile/50 mM aqueous ammonium acetate (pH 3.4)(25:75, v/v) run isocratically at 1 ml/min. Fractions (1 ml) werecollected and the compound was detected by either absorbance at 245nm or by radioactivity. The isolated peaks were further analyzed by fastatom bombardment mass spectrometry. FAB mass spectra were obtained by using a VG 70-250 HF gas chromatography/MS instrument,equipped with a standard unheated VG FAB ion source and a standardsaddle field gun (Ion-Tech Model BUN) producing a beam of xenonatoms at 8 keV and 1 mA. The mass spectrometer was adjusted to aresolving power of 2500, and spectra were obtained by using an accelerated voltage of 6 kV. Scans were obtained in the negative ion mode.

To measure the activity of HTC cells to metabolize PGD2 to 9a,l lß-PGF2, cells were sonicated on ice for 15s. The lysate was then incubatedfor 2 h at 37°Cin the presence of 0.5 mg/ml of PGD2 and in the

absence or presence of 1 mg/ml of NADPH. After incubation, themixture was acidified to pH 3 by addition of l N HC1 and centrifugedat 13,000 x g. The 9t*,ll/3-PGF2 was then extracted, purified, andanalyzed by gas chromatography/MS as described (35).

Láclatedehydrogenase activity was measured with Sigma kit 226-B.7-Glutamyl transpeptidase activity in CHO cells was carried out asdescribed by Meister et al. (36).

RESULTS

Uptake, Conjugation, and Metabolism of 9-deoxy-A9,A'2(E)-

PGD2 by HTC and CHO cells. Experiments examining cellularuptake of initiated 9-deoxy-A9,A'2(E)-PGD2 revealed the pres

ence of tritiated prostaglandin in the lysate of both HTC andCHO cells within 10 min. About 1.7% of the total counts weretaken up by HTC cells and 0.3% by CHO cells. As shown inTable 1, most of the intracellular counts were detected in thecell supernatant after lysis and centrifugation. Only a smallfraction remained bound to the cell pellet. When 9-deoxy-A9,A'2(E)-PGD2 is conjugated with a polar moiety such as GSH,

it cannot be extracted with ethyl acetate at pH 3 and remainsin the aqueous fraction. This is in contrast to unconjugated

Table 1 Effect of GSH depletion on recovery of intracellular

Recovery of [3H]-9-deoxy-A',A'2(E)-PGD2 in the supernatant and cell pelletfollowing lysis of cells that had been incubated with 9-deoxy-A',A12(E)-PGD2.

9-deoxy-A',A'2(E)-PGD2

CellsHTCCHOTreatmentControl

BSO/DEMControl

BSO/DEM|GSH]°52

±5C

22±87.6

±1.20.8 ±0.3Supernatant*1

746 ±191 (SS)'

672 ±39(94)276

±52 (80)97 ±32 (85)Pellet"137

±14170±2980

±9113 ±16

" nmol/106 cells (l h after treatment).* pmol/10" cells recovered after 10 min of incubation.c Mean ±SEM.d Numbers in parentheses, percentage conjugated.

1600

1200-

Q. BOOO

400

IO 30 70

ELUTION VOLUME (ML)Fig. 1. Reverse phase HPLC of HTC cell supernatant. Following a 10-min

incubation with ['HI^-deoxy-A'.A'^EJ-PGD;, HTC cells were washed, sonicated,and centrifuged. The cell supernatant was extracted by using a C[8 SEP-PAKcartridge and the extract was chromatographed onto a d«Alltech column solventsystem: acetonitrile/50 mM aqueous ammonium acetate (pH 3.4) (25:75, v/v),run isocratically, 1 ml/min, 1-ml fractions.

PGD2 which readily extracts into ethyl acetate at pH 3. Whenwater and ethyl acetate are mixed, a small amount of the water(approximately 10%) is taken up by the organic phase. Therefore, when a conjugate is extracted with ethyl acetate, about10% will also be recovered in the ethyl acetate layer. Of thecounts, 88 and 80% were recovered in the aqueous fractionfollowing extraction of the intracellular radioactivity of theHTC and CHO cell supernatant, respectively, indicating thatessentially all of the tritiated compound was in the form of apolar conjugate.

Reverse phase HPLC analysis of the intracellular radioactivematerial following incubation of tritiated 9-deoxy-A9,A12(E)-

PGD2 with HTC cells is shown in Fig. 1. A major peak waspresent eluting at a retention volume of 21 ml. Analysis of thiscompound by negative ion FAB-MS gave the mass spectrumshown in Fig. 2. This revealed a prominent ion at m/z 644which was interpreted as the M—H~ ion. This ion is 4 atomicmass units higher than the M—H~ ion in the negative ion FABmass spectrum of the GSH conjugate of 9-deoxy-A9,A12(E)-

PGD2 (33).We have previously found that 9-deoxy-A9,A12(E)-PGD2 is

efficiently transformed by 11-ketoreductase and a double bondreductase(s) in GSH depleted 100,000 x g supernatant of rabbitliver to a compound in which the carbonyl at C-ll had beenreduced to a hydroxyl and in which one of the double bondshad been reduced.5 We have previously also shown that GSHpreferentially conjugates across the A9 double bond instead ofacross the A12double bond (33). In view of these data, we have

interpreted the mass spectrum shown in Fig. 2 to be consistentwith the structure shown in which GSH is attached at C-9 andin which the carbonyl and the A12 double bond have been

reduced. An observation which is consistent with the proposed

5D. F. Wendelborn and L. J. Roberts II. unpublished data.

1880

on August 6, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GSH DEPLETION AND 9-DEOXY-A'.^'¡(E)-PGD¡

?75

5 <

NH2

NHCOCH2CH2CHCOOM

CHCONHCH2COOH

coon

III .L.I ! ill (1 liti« il

[M-HJ(SULFOXIDE)

sn

1 I II«SI SII rii

m/zFig. 2. Negative ion FAB mass spectrum of the major compound that eluted at 21 ml in Fig.

structure is that when analyzed by UV spectroscopy, this compound no longer exhibited a UVmax absorption at 245 nm,indicating the presence of an «,/3-unsaturated carbonyl characteristic of 9-deoxy-A9,A12(E)-PGD2 (13), but exhibited a UVmax

absorption at 226 nm. There is also a prominent ion at m/z660 in the mass spectrum in Fig. 2, 16 a.m.u. higher than theM—H ion at m/z 640. This was considered to be derived fromoxidation of part of the GSH conjugate to the sulfoxide duringisolation of the compound prior to analysis. As indicated inFig. 2, the stereochemical configuration of the GSH conjugateat C-9 is not indicated, as this information is not provided byFAB-MS analysis. If the conjugation is largely catalyzed byGSH-5-transferase, the conjugation at C-9 may be expected tobe stereoselective. However, because GSH conjugates with 9-deoxy-A9,A'2(E)-PGD2 rapidly nonenzymatically, the intracel-lular conjugate may be racemic at C-9.

When HTC cells were incubated with PBS containing [JH]-9-deoxy-A9,A'2(E)-PGD2 (10 /¿g/ml)for 10 min, only 10% of

the tritiated compound which remained outside the cells was ina conjugated (unextractable) form. However. 63% of the compound recovered in the PBS was conjugated after 120 min ofincubation. When the extracellular radioactive material wasanalyzed by reverse phase HPLC following a 120-min incubation with HTC cells, 4 radioactive peaks were detected: compound I at 10 ml, compound II at 21 ml, compound III at 36ml, and compound IV at 60 ml elution volume (Fig. 3).

Compound II had a retention volume the same as that of theGSH conjugate shown in Fig. 2, and analysis of this compoundby negative ion FAB mass spectrometry gave a mass spectrumessentially identical to that shown in Fig. 2. On this basis,compound II was identified as the GSH conjugate of 9-deoxy-A9,AI2(E)-PGD2 depicted in Fig. 2, in which the GSH is attached at C-9 and in which the carbonyl at C-l 1 and the A12

double bond have been reduced. The negative ion FAB massspectrum of compound III gave a base ion at m/z 458 (Fig. 4).This is the expected M—H ion for the cysteine conjugatederivative of compound II. The negative ion FAB mass spectrum of compound IV was characterized by a base ion at m/z515 which is consistent with the M—H ion of the cysteinylgly-

3000 -

2000

1000

IV

0 10 20 30 40 50 60 70

ELUTION VOLUME IMLI

Fig. 3. Reverse phase HPLC of HTC cell medium. Following incubation ofHTC cells for 120 min in PBS containing |'H]-9-deoxy-A*,A'2(E)-PGDj. the

medium was collected and analyzed by reverse phase HPLC as described in Fig.

cinyl conjugate derivative of compound II (Fig. 5). The natureof these results suggested that 9-deoxy-A9,A'2(E)-PGD2 initiallyundergoes conjugation with GSH across the A9 double bond,

followed by further metabolism by 11-ketoreductase, a doublebond reducíase, 7-glutamyl transpeptidase, and dipeptidase.

Conjugation with GSH almost certainly occurs prior to reduction of the carbonyl at C-l 1 since the carbonyl renders the A9

double bond susceptible to nucleophilic addition. To furthersupport the suggestion that these cells contain 11-ketoreductase

activity, the ability of these cells to transform PGD2 to 9«,11/3-PGF2, the product of 11-ketoreductase metabolism of PGD2

(37), was examined. PGD2 was incubated with lysed HTC cellsfor 2 h in the presence of NADPH. Analysis by GC/MSrevealed that the HTC cell lysate indeed catalyzed the transformation of PGD2 to 9a,l l/a-PGF2 (data not shown).

When CHO cells were similarly treated with PBS containing9-deoxy-A9,A'2(E)-PGD2, 8% of the compound recovered in the

medium was conjugated after 10 min and 38% was conjugatedafter 120 min of incubation. No conjugation was observed ifthe PBS was preincubated in the presence of cells for 120 min,removed, and then exposed for 2 h to the prostaglandin. Also,no free (reduced or oxidized) GSH was detected in the PBSafter 120 min of incubation with either cell line when 9-deoxy-

1881

on August 6, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GSH DEPLETION AND 9-DEOXY-.}'.V:<E)-PGD;

Fig. 4. Negative ion FAB mass spectrum of peak III inFig. 3.

NH2

CH-COOH1

mMN»68SI41Xniir-z<SOH

EOHÃŽ1TT

1 *«i <«M[M-H]mlìJljliLii.iti JÉ

4H 5M

m/z

CM

Fig. 5. Negative ion FAB mass spectrum ofpeak IV in Fig. 3.

A9,AI2(E)-PGD2 was not present. These results indicate thatconjugation of 9-deoxy-A',Al2(E)-PGD2 with GSH occurs ex

clusively inside the cell and the conjugates then subsequentlyexit the cell and appear in the culture medium.

Effects of GSH Depletion on Cellular Uptake and Antiprolif-erative Activity of 9-Deoxy-A9,A12(E)-PGD2. A 1-h treatment

m/zwith DEM/BSO decreased the intracellularGSH concentrationin HTC and CHO cells (Table 1). In HTC cells, the GSH levelsfell from a control value of 52.0 ±5.0 to 22.0 ±8.0 nmol/106

cells. In CHO cells, the 1-h DEM/BSO treatment lowered

GSH levels from a control value of 7.6 ±1.2 to 0.8 ±0.3nmol/IO6cells. These data show that the DEM/BSO exposure pro-

1882

on August 6, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GSH DEPLETION AND 9-DEOXY-A',Ai:(E)-PGDz

duced a 55 and 89% reduction in the GSH concentration,respectively. These are statistically significant as determined bya paired Student's t test (P < 0.001 ). There was also a significantreduction in the amount of 9-deoxy-A9,A'2(E)-PGD2 recovered

from both HTC and CHO cells following incubation of 9-deoxy-A9,A12(E)-PGD: with GSH depleted cells (Table 1).

These uptake experiments showed a 52% decrease (P < 0.002)in the compound recovered in HTC cells and a 41% decrease(P < 0.02) in the compound recovered in CHO cells as determined by a paired Student's t test. The reduction was observed

in the cell supernatant after lysis. No significant change wasmeasured for either cell line in the pellet from the lysed cells.

To assess the effect of GSH depletion on the antiproliferativeactivity of 9-deoxy-A9,A'2(E)-PGD2, cells were treated with

increasing concentrations of the prostaglandin for 48 h in thepresence or absence of BSO as described in "Materials andMethods." Cells which remained on the surface of the flasks

were trypsinized and counted. Trypan blue dye exclusion testsand colony forming assays showed these cells to be viable,whereas the cells floating in the medium did not exclude trypanblue (data not shown). At the end of the 48-h BSO exposure,the GSH concentration in HTC cells was 3% of control valuesand in CHO cells it was 8% of control values. Fig. 6 representsthe mean of 3 experiments with each cell line. In both cell linesthere was a significantly enhanced inhibition of cell proliferation by 9-deoxy-A9,A'2(E)-PGD2 when the cells were exposedto DEM/BSO treatment. The dose-response curve, characterizing HTC cells, was biphasic (Fig. 6A). There was an initialthreshold or shoulder followed by an exponential curve. Addition of DEM/BSO reduced the size of the shoulder and reducedthe slope of the exponential curve. In CHO cells, the dose-response curve was best fitted by a simple exponential (Fig.6B). Exposure to DEM/BSO reduced the slope. To quantitatethe data, inhibition of cell proliferation, expressed as relativecell number, was plotted against the prostaglandin concentration (Fig. 6, C and D). Relative cell number represents thequotient obtained after the cell number measured in the presence of 9-deoxy-A9,A'2(E)-PGD2 was divided by the cell number

measured in its absence. Cytotoxicity produced by exposure to4, 8, or 10 Mg/ml of 9-deoxy-A9,A12(E)-PGD2 was significantlyincreased in both cell lines as determined by a paired Student's

t test (P < 0.05; Fig. 6, C and D). The concentrations of 9-deoxy-A9,A'2(E)-PGD2 required to reduce proliferation by 50%

were derived from these data (Fig. 6, C and D) and decreasedfrom 9.1 ±0.9 to 4.8 ±0.1 Mg/ml in HTC cells and from 9.5±1.9 to 4.0 ±0.9 //g/ml in CHO cells when cells were depletedof glutathione by DEM/BSO treatment.

DISCUSSION

In the present study, we have shown that 9-deoxy-A9,A12(E)-

PGD2 is rapidly taken up by HTC and CHO cells and essentiallyquantitatively converted intracellularly to a polar conjugate.Narumiya et al. (2l, 22) have also observed a rapid uptake ofthe same compound in the LI210 murine leukemia cell line,most of which was recovered in the nucleus in a nonextractableform. They, however, did not identify the nature of the nonextractable material. In contrast to their findings, most of theradioactivity in our experiments was recovered in the supernatant after lysis of the cells, not in the pellet. This could beattributed, at least in part, to technical differences in cellprocessing, such as sonication and centrifugation. As we haveshown in the HPLC and FAB-MS analyses, the compound is

HTC CHO

10s BSO

. A

BSO

. B

8 10 8 10

|iq / ml9 - Deoxy A', A11(E) - PGD2

S 10°

- D

6 8 10

iig / ml9 - Deoxy A«,A'2(E) - PGD,

8 10

Fig. 6. Effect of glutathione depletion on proliferation of cell exposed to 9-deoxy-A'.A'2(E)-PGD2. Plated cells were treated with BSO/DEM for 1 h, fol

lowed by rinsing and applying fresh medium containing BSO and increasingconcentrations of 9-deoxy-A9,A':(E)-PGD; (with BSO. •¿�).Other cells were

treated with DMSO for 1 h. rinsed and supplied with fresh medium containingonly 9-deoxy-A*,A'2(E)-PGD2 (without BSO, O). Cells were allowed to incubate

for 48 h. Attached cells were then counted. Points, average of at least Ìexperiments; bars, SEM. . additive effect of GSH depletion and exposure to 9-deoxy-A°.Al;-PGD2. A, HTC cells; B. CHO cells. C and D. the data shown in A

and B are expressed as the relative cell number which represents the quotientobtained after the cell number measured in the presence of 9-deoxy-A*.A'J(E)-

PGD2 was divided by the cell number measured in its absence. The intercept ofthe dashed line with the A' axis represents the estimated concentration of 9-deoxy-A',A'!(E)-PGDj required to decrease proliferation by 50%. Error bars are shown

when they exceed the diameter of the symbol. C. HTC cells; D, CHO cells.

largely conjugated with intracellular GSH in CHO and HTCcells. The GSH conjugate then undergoes further metabolism,resulting in reduction of the keto group at C-l 1 by 11-ketore-ductase, reduction of the A12double bond, and further metabolicdegradation by -y-glutamyl transpeptidase and dipeptidase.

In both cell lines, 9-deoxy-A9,A'2(E)-PGD2 is not only rapidly

taken up by the cells, but also appears in the medium within 2h as a reduced conjugate of GSH or its metabolites. When cellswere incubated for 2 h without the prostaglandin, there was noevidence of effluxed free GSH (or its metabolites). Moreover,9-deoxy-A9,A'2(E)-PGD2 remained unconjugated when added

to PBS which had been preincubated with cells. This suggeststhat the compound is first taken up by the cell, subsequentlyconjugated with GSH, metabolized, and then effluxed. Following incubation of tumor cells for 12 h with PGA,, Vulliez-LeNorman et al. (38) have also identified cysteine and cysteinyl-glycine adducts of PGA, in the medium.

9-Deoxy-A9,A12(E)-PGD2 inhibits proliferation of HTC and

CHO cells, as has been reported in many other cell lines (12,13, 21, 22). In our study, we have shown that depletion ofintracellular GSH in both CHO and HTC cells by DEM/BSOtreatment enhanced their sensitivity to the antiproliferativeeffect of 9-deoxy-A9,A'2(E)-PGD2. Simply depleting intracellu-

1883

on August 6, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GSH DEPLETION AND 9-DEOXY-A9,A"(E)-PGD¡

lar GSH in both CHO and HTC cells without exposing themto 9-deoxy-A9,A'2(E)-PGD2 was associated with approximately

a 33% reduction in proliferation. GSH depletion by itself hasbeen known to inhibit cell proliferation (39-42). However, this

effect is not uniform and depends on the cell line and the degreeto which GSH concentrations are reduced (43-46). The mechanism by which 9-deoxy-A9,A12(E)-PGD2 exerts its antiprolif-

erative action probably is not simply by conjugating with GSHand thus lowering intracellular GSH concentrations. Although9-deoxy-A9,A12(E)-PGD2 inhibits proliferation of HTC cells, we

could not detect an appreciable reduction in intracellular GSHconcentration following exposure of the cells to the prostaglan-

din for 4 h (data not shown). This is not an unexpected findingin view of the fact that GSH concentrations are very high andcomparatively only small quantities of 9-deoxy-A9,A'2(E)-PGD2

are taken up by the cells.The question whether any of the 9-deoxy-A9,A'2(E)-PGD2

adducts (GSH, cysteinyl-glycine, and cysteine), formed in thesecells may in part be responsible for the cytotoxic activity of 9-deoxy-A9,A12(E)-PGD2 has not been resolved in the present

work. GSH conjugates of some xenobiotics, or their metabolites, may lead to bioactivation of an enzymatic system whichresults in toxic metabolites (47). Gouin et al. (48) identified acysteinyl-glycine conjugate of PGA, in which the C-9 keto grouphad been reduced in the medium following incubation of PGAwith tumor cells, and reported that this conjugate possessedcytotoxic activity. Formation of a cysteinyl-glycine derivativefrom a GSH conjugate requires an enzymatic reaction involving7-glutamyl transpeptidase (49). Although the HTC cells have7-glutamyl transpeptidase activity which resulted in the formation of a cysteinyl-glycine conjugate of 9-deoxy-A9,A12(E)-

PGD2, no 7-glutamyl transpeptidase activity was detected inthe CHO cells (data not shown). Moreover, the fact that GSHdepletion enhanced the inhibitory effect of 9-deoxy-A9,A'2(E)-

PGD2 would not seem consistent with the possibility that theGSH conjugate of this prostaglandin or one of its derivatives isactually responsible for the antiproliferative activity of thisprostaglandin in CHO and HTC cells.

These results, therefore, suggest that the availability of un-conjugated 9-deoxy-A9,A'2(E)-PGD2 within the intracellular

matrix may be required for this compound to exert its cytotoxiceffect, at least in CHO and HTC cells. It has been suggestedthat 9-deoxy-A9,A12(E)-PGD2 expresses this activity via an in

teraction in the nucleus (23). When the prostaglandin entersthe cell, some of it may be transported to the target organelle(e.g., nucleus) where it may bind to sulfhydryl residues onproteins, etc. The remainder may rapidly conjugate with cyto-solic GSH or efflux out of the cell. We have found that although9-deoxy-A9,A'2(E) PGD2 quantitatively conjugates nonenzy-

matically with GSH in vitro in the presence of a molar excessof GSH, the rate of conjugation is dependent on the concentration of GSH even though there is a molar excess present. Athigher concentrations of GSH, the rate of the conjugationreaction is more rapid. Therefore, it is possible that whenintracellular concentrations of GSH are reduced, the rate ofconjugation with GSH is slower. This may then allow more ofthe unconjugated compound to interact with critical factorsinvolved in cell proliferation.

The role of GSH levels in modifying the cellular response toanticancer drugs has been extensively studied in recent years,and the effect of GSH depletion on potentiating the responseis being currently investigated (50-54). In the present study, wehave obtained direct evidence that 9-deoxy-A9,A'2(E)-PGD2

conjugates with GSH intracellularly, and that this interaction

can modulate the cytotoxic activity of the compound. 9-Deoxy-A9,A12(E)-PGD2 represents a group of prostaglandins which are

potential antineoplastic drugs. The fact that its action is enhanced by GSH depletion in cell culture merits further investigation regarding the possibility that in vivo GSH depletion mayalso enhance the cytotoxicity of the compound.

REFERENCES

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Fukushima, M., Kato, T., Ueda, R., Ota, K., Narumiya. S., and Hayaishi, O.Prostaglandin D2, potential antineoplastic agent. Biochem. Biophys. Res.Commun., IOS: 956-964, 1982.Kikuchi, Y., Miyauchi. M.. Oomori. K., Rita, K., Kizawa, I., and Kalo, K.Inhibition of human ovarian cancer cell growth in vitro and in nude mice byprostaglandin D2. Cancer Res., 46: 3364-3366, 1986.Higashida, H., Kano-Tanaka, K., Natsuume-Sakai, S., Sudo, K., Fukami, H.,Nakagawa, Y., and Miki, N. Cytotoxic action of prostaglandin I>•¿�on mouseneuroblastoma cells. Int. J. Cancer, 31: 797-802, 1983.Sakao. T., Yamaguchi, N., Kawai, K., Nishino, H., and Iwashima. A. Prostaglandin I>..inhibits the proliferation of human neuroblastoma cells. CancerLett.. 77:289-294. 1983.Keyaki, A., Manda. H., Yamashita, J.. Tokuriki, Y., Otsuka, S. I., Yamasaki,T., and Gi, H. Growth inhibitory effect of prostaglandin D2 on mouse gliomacells. J. Neurosurg., 61: 912-917, 1984.Simmet. T., and Jaffe, B. Inhibition of B-16 melanoma growth in vitro byprostaglandin D2. Prostaglandins, 25: 47-54, 1983.Kawamura, M.. and Koshihara, Y. Prostaglandin I) strongly inhibits growthof murine mastocytoma cells. Prostaglandins Leukotrienes Med., 12:85-93,1983.Sakai. T., Yamaguchi, N., Shiroko, Y., Sekiguchi, M., Fujii, G.. and Nishino,H. Prostaglandin I* inhibits the proliferation of human malignant tumorcells. Prostaglandins, 27: 17-26, 1984.Fitzpatrick, F. A., Ligget, W. F., and Wynalds, M. A. Albumin-ecosanoidinteractions. J. Biol. Chem.. 259: 2722-2727, 1984.Fukushima. M., Kato, T., Ota, K., Arai, Y., Narumiya, S., and Hayaishi, O.9-Deoxy-A9-prostaglandin D2, a prostaglandin D2 derivative with potentantineoplastic and weak smooth muscle-contracting activities. Biochem. Biophys. Res. Commun., 109:626-633. 1982.Tanaka, H., Yamamuro, T., Matsumoto, M., Kotoura, Y., and Tanaka, C.The effects of PGD2 and 9-deoxy-A9-PGD2 on colony formation of murineosteosarcoma cells. Prostaglandins, 30: 167-174, 1985.Narumiya, S.. and Fukushima, M. AI2-Prostaglandin J2. an ultimate metabolite of prostaglandin I>, exerting cell growth inhibition. Biochem. Biophys.Res. Commun., 127: 739-745. 1985.Kikawa. Y., Narumiya, S., Fukushima, M., Wakatsuka, H.. and Hayaishi, O.9-Deoxy-A',Al!-13.14-dihydroprostaglandin D2 formed in human plasma.Proc. Nati. Acad. Sci. USA, 81: 1317-1321, 1984.Santoro, M. G., Crisari, A.. Benedetto, A., and Amici, C. Modulation of thegrowth of a human erythroleukemic cell line (K562) by prostaglandins:antiproliferative action of prostaglandin A. Cancer Res., 46: 6073-6077,1986.Turner, W. A., Taylor, J. D.. and Honn, K. V. Effect of prostaglandin "A"

series on tumor cells in vitro. In: T. V. Powels, R. S. Bockman, K. V. Honn,and P. Ramwell (eds.), Prostaglandins and Cancer: First International Conference, pp. 369-373. New York: Alan R. Liss, Inc., 1982.Suzuki. K., Kobayashi. N., Moriya, Y., Abiko, Y., and Suzuki, H. Inhibitionof human gingival carcinoma cell growth by prostaglandins. Gen. Pharmacol.,/ 9: 273-276, 1988.Ikai, K., Ujihara, M., Kashihara, M., and Fukushima, M. Inhibition of theproliferation of transformed epidermal cells in culture by various prostaglandins. J. Invest. Dermatol., 89: 69-72, 1987.Fukushima. M.. and Kato. T. Antitumor marine ecosanoids: clavulones andpunaglandins. Adv. Prostaglandin Thromboxane Leukotriene Res., 75: 415-418, 1985.Honda, A., Mori, Y., Iguchi, K., and Yamada, Y. Antiproliferative andcytotoxic effects of newly discovered halogenated coral prostanoids from theJapanese stolonifer Clavuria viridis on human leukemia cells in culture. Mol.Pharmacol., 32: 530-535, 1987.Bhuyan, B. K., Adams, G. E., Badiner. G. J., Li. L. H.. and Barden, K. Cellcycle effects of prostaglandins A,. A . and I>..in human and melanoma cellsin culture. Cancer Res.. 46: 1688-1693. 1986.Narumiya, S., and Fukushima. M. Site and mechanism of growth inhibitionby prostaglandins. I. Active transport and intracellular accumulations ofcytopentone prostaglandins, a reaction leading to growth inhibition. J. Pharmacol. Exp. Ther.. 239: 506-511. 1986.Narumiya, S., Ohno, K., Fujiwara, M., and Fukushima, M. Site and mechanism of growth inhibition by prostaglandins. II. Temperature-dependenttransfer of a cyclopentoenone prostaglandin to nuclei. J. Pharmacol. Exp.Ther., 259:506-511, 1987.Narumiya, S., Ohno, K., Fukushima, M., and Fujiwara, M. Site and mechanism of growth inhibition by prostaglandins. III. Distribution and binding ofprostaglandin A2 and A12-prostaglandin J2 in nuclei. J. Pharmacol. Exp.Ther., 2«: 306-311. 1988.

1884

on August 6, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

GSH DEPLETION AND 9-DEOXY-A',A1!(E)-PGD¡

24. Ohno, K., Sakai, T., Fukushima, M., Narumiya, S., and Fujiwara, M. Siteand mechanism of growth inhibition by prostaglandins. IV. Effect of cyclo- 39.pentenone prostaglandins on cell cycle progression of d-enriched HeLa S3cells. J. Pharmacol. Exp. Ther., 245: 294-298. 1988.

25. Koshihara. V. and Kawamura. M. Inhibitory effect of prostaglandin l> onDNA synthesis in nuclei. Prostaglandins Leukotrienes Med., 23: 253-265, 40.1986.

26. Todo, S.. Hasida, T.. Shimizu. Y., Imashuku. S.. Takamatsu. T.. and Fujita.S. Cell kinetic studies of PGD2 cytotoxicity on the in vitro growth of human 41.neuroblastoma. Prostaglandins Leukotrienes Med.. 23: 55-65, 1986.

27. Kato, T.. Fukushima, M., Kurozumi. S., and Noyori. T. Antitumor activityof prostaglandin A, and A"-prostaglandin J2 in vitro and in vivo. Cancer 42.Res., 46: 3538-35642, 1986.

28. Bregman, M. D., Funk. C., and Fukushima. M. Inhibition of human mela- 43.noma growth by prostaglandin A. D. and J analogues. Cancer Res.. 46:2740-2744, 1986.

29. Honn, K. V., and Marnelt, L. J. Requirement of a reactive «,^-unsaturated 44.carbonyl for inhibition of tumor growth and induction of differentiation by"A" series prostaglandins. Biochem. Biophys. Res. Commun.. 129: 34-40,

1985. 45.30. Boyland, E., and Chasseaud, L. F. Enzymes catalysing conjugations of

glutathione with «,/J-unsaturated carbonyl compounds. Biochem. J.. 109:651-661. 1968. 46.

31. Cagen, L. M.. Pisano, J. J.. Ketley. J. N., Habig. W. H., and Jakoby, W. B.The conjugation of prostaglandin \, and glutathione catalyzed by homogenous glutathione-5-transferascs from human and rat liver. Biochim. Biophys.Acta, 398: 205-208, 1975. 47.

32. Ham, E. A., Oien, H. G., Ulm, E. H., and Kuehl, F. A. The reaction of PGA,with sulfhydryl groups; a component in the binding of A-typc prostaglandinsto proteins. Prostaglandins, 10: 217-229, 1975. 48.

33. Atsmon, J., Sweetman, B. J., Baertschi, S. W., Harris, T. M., and Roberts,L. J., II. Formation of thiol conjugates of 9-deoxy-A*.^'2(E)-prostaglandinD2 and Al!(E)-prostaglandin D2. Biochemistry, in press, 1990.

34. Fariss, M. C., and Reed. D. J. Measurement of glutathione and glutathione 49.disulfide efflux from isolated rat hepatocytes. In: R. A. Harris and N. W.Cornell (eds.). Isolation. Characterization and Use of Hepatocytes. pp. 349- 50.355. Amsterdam: Elsevier/North-Holland. 1983.

35. Wendelborn. D. F., Seibert, K.. and Roberts, L. J., II. Isomerie prostaglandin 51.F3compounds arising from prostaglandin I> : a family of icosanoids producedi/i vitro in humans. Proc. Nati. Acad. Sci. USA, SS: 304-308, 1988.

36. Meister, A., Tate, S. S., and Grifith, O. W. -y-Glutamyl Transpeptides.Methods Enzymol., 77: 237-253, 1981. 52.

37. Liston. T. E.. and Roberts. L. J„II. Transformation of prostaglandin D2 to9n,l l/3-(155)trihydroxyprosta-(5Z,13£')-dien-l-oic acid (9«.l1rf-prostaglan-

din F2): a unique biologically active prostaglandin produced enzymatically in 53.vivo in humans. Proc. Nati. Acad. Sci. USA, 82: 6030-6034, 1985.

38. Vulliez-Le Normand, B., Gouyette, A.. Heidet, V., Nagel. N., and Dray. F. 54.Structural identification of prostaglandin A, biotransformation products

from tumor cells. Prostaglandins, 35: 515-521, 1988.Dehtlefsen, L. A., Biaglow, J. E., Peck, V. M., and Ridinger, D. N. Toxiceffects of extended glutathione depletion by buthionine sulfoximine on murine mammary carcinoma cells. Int. J. Radial. Oncol. Biol. Phys.. 12: 1157-1160, 1986.Fidelius, R. K., Ginouves, P., Lawrence, D., and Tsan, M. F. Modulation ofintraccllulur glutaihione concentration alters lymphocyte activation and proliferation. Exp. Cell Res.. 170: 269-275. 1986.Hennighausen. G. Effects of diethyl maléateon non-protein sulfhydryl content and cellular functions of mouse thymocytes in vitro. Arch. Toxicol.,(Suppl.) 9: 297-301. 1986.Lewko, W. M. Glutathione levels in cultured heart cells. Biochem. Pharmacol., 36: 219-233. 1987.Duthie, S. J., Colcman, C. S., and Grant, M. A. Status of reduced glutathionein the human hepatoma cell line. HEP G2. Biochem. Pharmacol., 17: 3365-3368. 1988.Miller, A. C., and Henderson, B. W. Effect of m -buthionine-.Ç.A-sulfoximineon the growth of EMT6 and RIF mouse tumor. J. Nail. Cancer Inst., 77:505-510. 1986.Dorr, R. T., Liddil. J. D., and Soble, M. J. Cytotoxic effects of glutathionesynthesis inhibition by L-buthionine-(S'A)-sulfoximinc on human and murine

tumor cells. Invest. New Drugs. 4: 305-313. 1986.Clark, E. P., Epp, E. R.. Biaglow. J. E.. Morse-Gaudio. M., and Zachgo, E.Glutathione depiction, radiosensitization and misonidazolc potentiation inhypoxic Chinese hamster ovary cells by buthionine sulfoximine. Radiât.Res..98: 370-380. 1984.Anders, M. W., Lash, L.. Dekant. W., Elfarra. A. A., and John, D. R.Biosynthesis and biotransformation of GHS-S conjugates to toxic metabolites. CRC Crii. Rev. Toxicol.. IS: 311-341, 1988.Gouin, E.. Vulliez-Le Normand. B.. Gouyette. A., Heidet. V.. Nagel. N., andDray, F. Tumor cell biotransformation products of prostaglandin A, withgrowth inhibitory activity. Biochem. Biophys. Res. Commun., 141: 1254-1260, 1986.Meister. A., and Anderson, M. E. Glutathione. Annu. Rev. Biochem., 52:711-760. 1983.Mitchell, J. B. Glutathione modulation and cancer treatment. ISI Atlas Sci.Pharmacol., 2: 155-160, 1987.Wolf. C. R., Lewis, A. D.. Carmichael. J.. Adams. D. J., Allen, S. G.. andAnsell. D. J. The role of glutathione in determining the response of normaland tumor cells to anticancer drugs. Biochem. Soc. Trans., 15: 728-730,1987.Lee, F. Y. F., Vessey, A. R.. and Siemann, D. W. Glutathione as a determinant of cellular response to doxorubicin. Nati. Cancer Inst. Monogr..6: 211-215, 1988.Fogo, A., Hamilton, T. C., Young. R. C.. and Ozols, R. F. Multidrugresistance in ovarian cancer. Cancer (Phila.), 60: 2075-2080, 1987.Clark, E. P. Thiol-induced biochemical modification of chemo- and radiores-ponses. Int. J. Radial. Oncol. Biol. Phys., 12: 1121-1126, 1986.

1885

on August 6, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1990;50:1879-1885. Cancer Res   Jacob Atsmon, Michael L. Freeman, Michael J. Meredith, et al.   Antiproliferative Activity by Glutathione DepletionIntracellular Glutathione and Enhancement of Its

with2(E)-prostaglandin D12∆,9∆Conjugation of 9-Deoxy-

  Updated version

  http://cancerres.aacrjournals.org/content/50/6/1879

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

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/50/6/1879To request permission to re-use all or part of this article, use this link

on August 6, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from