Estradiol-17β and 2-hydroxyestradiol-17β-induced differential production of prostaglandins by cells dispersed from human intrauterine tissues at parturition

PROSTAGLANDINS

ESTRADIOL-176 AND 2-HYDROXYESTRADIOL-17B_INDUCED DIFFERENTIAL PRODUCTION OF PROSTAGLANDINS BY CELLS DISPERSED FROM HUMAN INTRAUTERINE TISSUES AT PARTURITION

David M. Olson, Karolina Skinner and John R.G. Challis MRC Group in Reproductive Biology Department of Obstetrics and Gynaecology University of Western Ontario London Ontario Canada N6A 5A5 ABSTRACT Prostaglandin (PGE, 6-keto PGFl,) output by cells dispersed from human amnion and decidua in the presence of increasing levels (o-5000 ng/ml) of estradiol-17B (E2) or 2-hydroxyestradiol-178 (2-OH E2) was studied in relaTissues were obtained from women at tion to parturition. term either before (CS) or after (SL) spontaneous labor and vaginal delivery. In the absence of estrogens, the output of both PGs from amnion increased significantly with labor. No significant increase in decidua PG output occurred with labor. Neither estrogen influenced CS amnion PG output. However, both E2 and 2-OH E2 stimulated SL amnion PGE output (2-OH E2>E2) while having no affect on 6-keto PGFlu output. Only the highest dose of 2-OH E2 stimulated PGE output in CS decidua, but both estrogens significantly inhibited 6-keto PGFle output in this tissue. In SL decidua only 2-OH E2 significantly stimulated PGE, and neither estrogen affected 6-keto PGFlcl output. These results might suggest that estrogens modulate PG biosynthesis at the level of endoperoxide to primary PG conversion.

MAY

1983 VOL. 25 NO. 5

639

PROSTAGLANDINS

INTRODUCTION We have recently shown that dispersed ceils from human amnion, chorion, decidua and placenta synthesize and metabolize prostaglandins (PGs) E, F and 6-keto PGF when incubated in short-term cultures (1). Furthermoretcl large increases in PG output occurred in the amnion and chorion in association with labor whereas only small or negligible increases occurred in the decidua and placenta. Identification of the factors which are responsible for this increased PG output from the fetal membranes at parturition would be useful for further understanding the initiation of human parturition. Among the several hormones or agents which have been shown to modulate PG synthesis and metabolism, both (2-4). Recently, in vivo and in vitro, are the estrogens -7 estradiol-17S (E2) and one of its major metabolites, 2hydroxyestradiol-178 (2-OH E2) (5), have been shown to stimulate simultaneously PGE and PGF synthesis while inhibiting 6-keto PGFle production (6,7). The catechol estrogens, such as 2-OH E2, are present during human At term, the placenta contains 2-hydroxylase pregnancy. is excreted in the activity (8). Two hydroxyestradiol urine (9,lO) of pregnant women. Plasma levels of 2-OH E2 are higher during the third trimester of pregnancy than in non-pregnant women (11). Since these estrogens (E2 and 2-OH E2) may have divergent effects upon the biosynthesis of the myometrial contractile stimulant PGE2, and the myometrial inhibitor, PG12 (12-14), we investigated their roles in PG production from the fetal amnion and maternal decidua in relation to parturition. MATERIALS

AND METHODS

Subjects Seven women at term pregnancy but not in labor, who were delivered by elective cesarean section (CS), and four women who delivered vaginally following spontaneous labor (SL) at term were used in this study. The outcomes of pregnancy and weights of the fetuses were all within the normal range. Cell

Dispersal

and Incubation

The prodecure used has been described previously (1). Briefly, placentas with attached fetal membranes The amnion were obtained immediately following birth. was manually separated from the chorion, and the decidua

MAY

1983 VOL. 25 NO. 5

PROSTAGLANDINS

was scraped from the chorion using a steel blade The tissues were washed glass microscope slide. times in 0.9% saline.

or a several

A portion of the tissue was then incubated at 370C in a shaking water bath, under air, for 20-30 min in a and 0.05% trypsin solution containing 0.1 % collagenase inhibitor in Ca2+, Mg2+-free Dulbecco's phosphate bufTissues were removed and, after discarding fered saline. the first suspension, were incubated again for l-l.5 h in fresh solution. Dispersed cells were filtered and washed twice in Eagle's Minimum Essential Medium (MEM) The final viable cell conwith gentle centrifugation. Estrogens prepared in ethanolic centration was 1x105/ml. stock solutions were added directly to the MEM prior to The final concentrations of ethanol addition of cells. in MEM was 0.5% (v/v) or less. Incubations were carried out for 6 h at 37OC and Each experiment interminated by freezing at -7OOC. cluded cells without additions frozen at time zero and blank tubes containing additions but no cells. The values obtained in these tubes were subtracted from the experimental values to obtain the net PG output. Prostaglandins E and PG12 (as measured by its inactive hydrolysis product 6-keto PGFl,) were assayed directly from the medium by sensitive and specific radioimmunoassays previously described (15). Materials Estradiol-176 (1,3,5,(10)-estratrien-3, 17B-dial) was obtained from the Sigma Chemical Co. (St. Louis, MO), and 2-hydroxyestradiol-l78(1,3,5,(lO)-es~ratrien-2,3,17~triol) was purchased from Steraloids, Inc. (Wilton, NH). Estrogens were used without further purification. Statistical

Analysis

Data were analyzed using non-parametric methods since transformation of the data did not remove heterogeneous variance. In certain cases (Fig. l-3) the data were normalized to percent change from the control (zero) value prior to analysis. The Kruskal-Wallis one-way analysis of variance (16) was used-to analyze the effects of estrogen treatment upon the output of each PG for each tissue studied. When appropriate, comparison of two individual treatments was made using the Wilcoxon paired sample test (16) (within subjects or the Mann Whitney U test (16) (between sub-

MAY 1983 VOL. 25 NO. 5

641

PROSTAGLANDINS

jects). Throughout the text, statistical is obtained at the ~'0.05% level.

significance

RESULTS The absolute values for PGE and 6-keto PGFla output (pq/105 cells/6h) for amnion and decidua from CS and SL women are given in Table 1. As we reported previously (l), there was a large increase in the output of each PG by amnion cells obtained from SL women when compared to amnion cells obtained from CS women. There was no siqnificant increase with labor in output for either PG from the decidua. However, for the seven CS women, the ratio of decidua PGE to 6-keto PGF output was 0.9OrtO.23 (g-+SEM) while in the four SLlaomen this ratio difbecame 3.53111.89. These ratios were significantly ferent (p=O.O4) by the Mann-Whitney U test. There was no change in the amnion PGE/G-keto PGFlc, ratio with labor. Table

Prostaqlandin output (pq/105 cells/6h) by amnion and decidua cells x?SEM (n)

1.

cs

Amnion --

Decidua

161?61

PGE 6-keto

PGFlc,

(7)

129-t63 (7)

270-+60 (7) 349+76

(7)b

SL Amnion -~ PGE 6-keto

PGFlcl

Decidua -

998+416

(4)a

546?159

(4)

626+140

(4)a

3552233

(4)

a - Significantly different from CS amnion value (pcO.05, Mann-Whitney U test) b-

642

Significantly different value (pcO.05, Wilcoxon

from CS amnion paired sample test).

MAY

1983 VOL. 25 NO. 5

PROSTAGLANDINS There was no significant effect of E2 or 2-OH E2 ;; PGE or 6-keto PGFle output by CS amnion cells (Table . There were, however, significant effects of estrogen coincubation upon PG output by the SL amnion (Fig. 1). In the SL amnion, as both E2 and 2-OH E2 concentrations increased, a significant increase in PGE output occurred (Kruskal-Wallis ANOVA). While both estrogens stimulated an increase in PGE output, the effect was more evident with the catechol estrogen (Fig. 1). Neither estrogen had any effect upon 6-keto PGFlc, output in this tissue. Table 2.

Effects of estradiol-17P and 2-OH estradiol on prostaglandin output by CS amnion Z+SEM (n=7)

Estradiol-178 0

1

(ng/ml)

100

5000

161?61

130+43

117+45

123?47

6-keto PGFlc, 129+63

163264

131?68

94*40

PGE

2-OH Estradiol -178 (ng/ml) 0

100

5000

PGE

161?61

99+38

86+34

6-keto PGFlc,

129+63

154+70

89538

Unlike cells dispersed from the CS amnion, the cells from the CS decidua did respond to increasing levels of estrogens in the incubation medium (Fig. 2). Estradiol-17B had no significant effect upon PGE output but significantly inhibited 6-keto PGFle output. A slight increase in PGE output at the highest E2 dose was evident, but this was not significant. In the presence of 2-OH E2, however, there was a significant increase in PGE output at the highest concentration. Likewise, an inhibition in 6-keto PGFlc output was also observed with the catechol estrogen.

MAY 1983VOL. 25 NO. 5

643

PROSTAGLANDINS

50

E2

SL

-25 t -50

1 -t 0

a

I

I

IO’

IO2

IO3

Id

ESTROGEN (ng/ml) Fig.

1.

The effect of increasing doses of E2 (top) and 2-OH E2 (bottom) on 6-keto PGFlc, (open circles) and PGE (closed circles) output The by cells from SL amnion is shown. results are expressed as the percent change The means+ from the control (zero) value. SEM of measurements from 3 (E2) and 4 (2-OH E2) women are presented for each PG.

MAY

1983 VOL. 25 NO. 5

PROSTAGLANDINS

CL +

CS 2-OHE2 25 .

2 -+

o" CL

J

-25 t

Fig.

2.

The effect of increasing doses of F2 (top) and 2-OH E2 (bottom) on 6-keto PGFla (open circles) and PGE (closed circles) output by cells from CS decidua is shown. The results are expressed as the percent change from the control (zero) value. The means? SEM of measurements for PGE from 5 (E2) and 5 (2-OH E2, except 1 ng/ml where n=2) women and for 6-keto PGF from 7 (E ) and 7 (2-OH E2, except 1 rig/i??!? where n=2 ? women are presented.

MAY 1983VOL. 25 NO. 5

645

PROSTAGLANDINS

ESTROGEN (r~~hnl, Fig. 3.

646

The effect of increasing doses of E2 (top) and 2-OH E2 (bottom) on 6-keto PGFlcU.(open circles) and PGE (closed circles) output by cells from SL decidua is shown. The results are expressed as the percent change from the control (zero) value. The means? SEM of measurements from 3 (E2) and 4 (2-OH E2) women are presented for each PG.

MAY

1983 VOL. 25 NO. 5

PROSTAGLANDINS

The response of cells from the SL decidua was similar to that of cells from the SL amnion (Fig. 3). Estradiol-176 had no effect upon the output of either PG. increased PGE2 in On the other hand, 2-OH E2 selectively a dose-related fashion while having no effect upon the 6-keto PGFlcl output. DISCUSSION The two estrogens used in this study, E2 and 2-OH E have been shown to have differential effects upon Psi and 6-keto PGFla output by amnion and decidua cells from CS and SL women. These different PG responses to the estrogens might suggest that their sites of action are at the level of endoperoxide conversion to primary prostaglandin and not at either the arachidonate deacylation or cyclooxygenase steps of the arachidonic acid cascade. Additionally, the stimulatory effect of 2-OH E2 upon PGE output was greater than that of E2. This raises the possibility that the metabolism of E2 to 2-OH E2 might act as a regulatory step in the biosynthesis of PGs. Our results closely parallel the observations reported by Kelly and Abel (6,7) using homogenates of whole rat uterus or human uterine endometrium and exogenously added arachidonic acid. At concentrations of 1OOuM (our highest concentration was 17.4~M), 2-OH E2 stimulated large increases in PGE and PGF synthesis in both rat and human tissue and, in the rat, inhibited 6-keto PGFla production. The magnitude of the PGE and PGF responses in the presence of 100~M E2 was less, and the synthesis of 6-keto PGFl"was slightly stimulated rather than inhibited. Our data do not resolve the discrepancy between the lack of estrogen effect in CS amnion cells and the divergent action of estrogen on PG output in CS decidua cells. We speculate that this discrepancy may be related to the relative activities of the PG synthesizing enzymes in these two tissues and the level of arachidonic acid metabolism at which the estrogens may be acting. If the estrogens are influencing endoperoxide metabolism, then their effect upon PG output is dependent upon the rate at which endoperoxide is formed. In CS amnion, endoperoxide formation may be relatively low. The deacylation of arachidonic acid from phosphatidylinositol, one of the primary sources of arachidonate (17,18), is dependent upon the enzymes phospholipase C, diacylglycerol lipase and monoacylglycerol lipase (19). Diacylglycerol lipase activity is relatively low until labor is initiated Following its liberation, arachidonic acid is (20,211.

MAY 1983VOL. 25 NO. 5

647

PROSTAGLANDINS cyclized by the incorporation of molecular oxygen, a reaction catalyzed by the enzyme cyclooxygenase. In the CS amnion, cyclooxygenase (prostaglandin synthetase) activity is low when compared to the activity of the enzyme in tissues during labor (22). The situation may differ in the CS decidua. Diacylglycerol lipase activity is much greater than in the CS amnion (20). Moreover, cyclooxygenase activity is as high before labor as during labor (22). Consequently, endoperoxide formation may be greater in decidua than amnion prior to labor. The relative levels of PG output in the CS decidua are higher than in CS amnion and may reflect these possible differences in endoperoxide formation (Table 1). Studying the regulation of PG biosynthesis in decidua may be valuable even though in our in vitro system there is little or no increase in PG
648

MAY

1983VOL.25NO. 5

PROSTAGLANDINS

REFERENCES 1.

Olson, D.M., K. Skinner and J.R.G. Challis. Prostaglandin output in relation to parturition by cells dispersed from human intrauterine tissues. Submitted for publication, 1982.

2.

Thorburn, G.D. of parturition.

3.

Caldwell, B-V., S.A. Tillson, W.A. Brock and L. Speroff. The effects of exogenous progesterone and estradiol on prostaglandin F levels in ovariectomized ewes. Prostaglandins 1:217, 1972. -

4.

Abel, M. and D.T. Baird. The effect of 178-estradiol and progesterone on prostaglandin production by human endometrium maintained in organ culture. Endocrinology 106:1599, 1980.

5.

Fishman, estrone: Biochem.

6.

Kelly, R.W. and M.H. Abel. Catechol estrogens stimulate and direct prostaglandin synthesis. Prostaglandins -20:613, 1980.

7.

Kelly, R.W. and M.H. Abel. A comparison of the effects of 4-catechol oestrogens and 2-pyrogallol oestrogens on prostaglandin synthesis by the rat and human uterus. J. Steroid Biochem. -14:787, 1981.

8.

Fishman, J. and D. Dixon. 2-hydroxylation estradiol by human placental microsomes. istry 6:1683, 1967.

9.

Gelbke, H.P., H. Hoogen and R. Knuppen. Identification of 2-hydroxyestradiol and the pattern of catechol oestrogens in human pregnancy urine. J. Steroid Biochem. 6:1187, 1975.

10.

Cohen, S.L., P. Ito, Y. Suzuki and F.E. Alspector. The preparation of pregnancy urine for an estrogen profile. Steroids -32:279, 1978.

11.

Kono, S., G.R. Merriam! D.B. Brandon, D.L. Loriaux and M.B. Lipsett. Radioimmunoassay and metabolism of the catechol estrogen 2-hydroxyestradiol. Metab. 54:150, 1982. J. Clin. Endocrinol.

MAY

1983 VOL.25N0.5

and J.R.G. Physiol.

Challis. Endocrine Rev. -59:863, 1979.

control

J., R.I. Cox and J.F. Gallagher. 2-hydroxyA new metabolite of estradiol in man. Arch. Biophys. 90:318, 1960.

of Biochem-

649

PROSTAGLANDINS 12.

Inhibition by PG12 of J.R.G. Challis. Lye, S.J., myometrial activity in non-pregnant ovariectomized sheep. J. Reprod. Fert. 66:311, 1982.

13.

G.C. Folco, M. Fano and Omini, C., R. Pasargiklian, Pharmacological activity of PG12 and its F. Berti. rdgolite 6-0x0-PGFl, on human uterus and fallopian Prostaglandins . -15:1045, 1978.

14.

M. Fano and Omini, C., G.C. Folco, R. Pasargiklian, F. Berti. Prostacyclin (PG12) in pregnant human uterus. Prostaglandins -17:113, 1979.

15.

Evans C.A., T.G. Kennedy, J.E. Patrick and J.R.G. Challis. Uterine prostaglandin concentrations in sheep during late pregnancy and adrenocorticotropininduced labor. Endocrinology 109:1533, 1981.

16.

Zar, J.H. Biostatistical Analysis. Inc., Englewood Cliffs, N-J., 1974,

17.

Schwarz, B-E., F.M. Schultz, P.C. MacDonald and J.M. Johnston. Initiation of human parturition. III. Fetal membrane content of prostaglandin E2 and F recursor. Obstet. Gynecol. 46:564, 1975. 2a p

18.

Curbelo, V., R. Bejar, K. Benirschke and L. Gluck. Premature labor. I. Prostaglandin precursors in human placental membranes. Obstet. Gynecol. 57~473, 1981. -

19.

DiRenzo, G.C., J.M. Johnston, T. Okazaki, J.R. Okita, P.C. MacDonald and J.E. Bleasdale. Phosphatidylinositol-specific phospholipase C in fetal membranes and uterine decidua. J. Clin. Invest. 67:847, 1981.

20.

Okazaki, T., N. Sagawa, J.E. Bleasdale, J.R. Okita, P.C. MacDonald and J.M. Johnston. Initiation of human parturition. XIII. Phospholipase C, phospholipase A2 and diacylglycerol lipase activities in fetal membranes and decidua vera tissues from early and late gestation. Biol. Reprod. -25:103, 1981.

21.

Okita, J.R., P.C. MacDonald and J.M. Johnston. Initiation of human parturition. XIV. Increase in the diacylglycerol content of amnion during parturition. Am. J. Obstet. Gynecol. 142:432,1982.

650

Prentice-Hall p. 124.

MAY 1983VOL. 25 NO. 5

PROSTAGLANDINS

22.

Okazaki, T., M.L. Casey, J.R. Okita, P.C. MacDonald Initiation of human parturition. and J.M. Johnston. Biosynthesis and metabolism of prostaglandins XII. in human fetal membranes and uterine decidua. Am. J. Obstet. Gynecol. 139:373, 1981.

23.

Abel, M.H. and R.W. Kelly. Differential production of prostaglandins within the human uterus. Prostaglandins -18:821, 1979.

Editor: Marc Bygdeman

MAY

Received: 12-9-82

1983 VOL. 25 NO. 5

Accepted: 3-15-83

651