Oestradiol induced changes of soluble protein synthesis in human endometrium

J. steroid Biochem. Vol. 18, No. 5, pp. 535-540, Printed in Great Britain. All rights reserved

0022-473 l/83/050535-06$03.00/0 Copyright 0 1983 Pergamon Press Ltd

1983

OESTRADIOL INDUCED CHANGES OF SOLUBLE PROTEIN SYNTHESIS IN HUMAN ENDOMETRIUM J. K~RNYEI, J. A. SZBKELY*and MARIETTAVBRTES Institute of Physiology and *Department of Pbstetrics and Gynecology, University Medical School, PECS, Hungary (Received 22 Jul.v 1982) SUMMARY The effect of an in vitro application of oestradiol-17-b on the soluble protein synthesis in human endometrium during the menstrual cycle was investigated. Results obtained by double-label technique and SDS-polyacrylamide gel or Cellogel electrophoresis show that synthesis of soluble proteins in human endometrium is affected by oestradiol in both follicular and luteal phases of the cycle. One of the proteins obtained after oestradiol treatment seems to be similar to IP of rat uterus as judged from its molecular weight (41,000 daltons) and electrophoretic mobility (RF: 1.18-1.24 relative to BSA on Cellogel). The magnitude of this protein synthesis correlates significantly with cytoplasmic oestradiol receptor concentration in endometrium throughout the cycle. The highest induction of IP like protein(s) synthesis at follicular phase and a decreased induction during the luteal phase was detected. No quantitative relationship between the changes of progesterone receptor concentration and IP like protein synthesis was found. INTRODUCTION Steroid receptor levels in human endometria fluctuated in response to the cyclic variation in ovarian steroid secretion [l-3]. Little information is available concerning further steps in the mechanism of hormone action in human oestrogen sensitive tissues. The

induction of the synthesis of a specific protein, called induced protein (IP) in rat uterus is the earliest known biosynthetic tissue response after oestrogen binding. The molecular weight of IP is about 40,000 daltons and its electrophoretic mobility relative to BSA on Cellogel electrophoresis is 1.09-1.19 (for review see: [4,5]). In human endometrium showing secretory histological appearance an increased incorporation of [35S]-methionine into a specific protein which migrated on SDS-polyacrylamide gels at a molecular weight of about 55,000 daltons was induced by oestradiol[6]. As far as progesterone effects are concerned, a specific soluble protein with molecular weight of 50,000 daltons was found in tissue culture of human endometria [7, S] and the presence of 48,000 dalton cytosolic protein in luteal phase endometria was reported by Joshi et aI.[9]. The synthesis of specific proteins migrating on SDS gels at molecular weight of 24,000 and 36,000 daltons [lo] or 46,000 daltons [ 1l] were induced by oestradiol in human breast cancer cell lines (MCF-7). On the other hand no IP response after oestradiol treatment were observed in Evsa T breast cancer cells [ll]. Preliminary results from our laboratory have shown that soluble protein synthesis in human endometria is affected by oestradiol[12].

In the present study the effect of an in oitro application of oestradiol on the soluble protein synthesis in human endometrium during the menstrual cycle was investigated. METHODS Tissues

Endometria (n: 56) were obtained from women with regular menstrual cycles aged between 22-47 years undergoing either curettage for sterility in the second half of the cycle for histological evaluation or hysterectomy for uterine leiomyomas. In the second cases immediately after abdominal or vagina1 exstirpation of the uterus the endometrium was scraped from the uterine cavity. All samples of endometrium tissues came directly from the operating theatre to the laboratory, where after washing in ice-cold 0.15 M NaCl solution to remove mucus and blood, they were minced and immediately worked up. Part of the endometrial tissue was kept for histological evaluation. The phase of menstrual cycle was determined by endometrial histology, the day of ovulation by basal body temperature. Samples showing different types of endometrial pathology or showing no signs of secretoric transformation arising from the second part of the cycle were excluded from the study. For comparison, the effect of in tritro application of oestradiol on soluble protein synthesis of rat uterus were also studied. Uteri from adult cycling rats ovariectomized 10 days before killing were used. Determination qf oe.stradio/and progesterone binding

Endometrial tissue samples (n: 31) were homogenized in 5 mmol sodium phosphate pH 7.4, 1 mmol 535

thioglycerol. 30”,, (v/v) glycerol (Phosphate-thioglycerol buffer) with Braun homogenizer and centrifuged at 800 9 for 30 min. For determination of oestradiol binding. aliquots of supernatants were incubated at 4 C for 16 h with IO-’ M [3H]-oestradiol (SA lO(J Cii mmol, Radiochemical Centre. Amersham) with or without IOOO-fold excess of diethylstilboestrol (KochLight Labs Ltd, Colnbrook, Bucks, England). After incubation the supernatants were treated with 025?j0 charcoal. 0.0025”,, Dextran 20 in Phosphate-thioglycerol buffer pH 7.4 for 10 nun with occasional shaking and centrifuged at 2000 g for 20 min. Aliquots of the supernatant were added to the scintillation solution, cooled and counted. For determination of progesterone binding the supernatants were with lo-*M incubated [3H]-RS020 (SA 82 Ci/mmol. New England Nuclear Corporation) with or without lOOO-fold excess of norethisterone (Sigma Chemical Company, U.S.A.) at 4’C for 16 h and then the binding was measured by the above mentioned charcoal absorption technique. All determination were done in triplicates. Receptor bound tritiated steroids were determined as the difference between values obtained with or without lOOO-fold excess of unlabelled hormones. Protein determinations were performed by the method of Lowry et al.[13]. Protein synthesis Hormone treatment was performed iti vitro. Minced tissue samples with equal weights were incubated in KrebssRinger bicarbonate buffer pH 7.4 at 37’C under 95”” O,-5”,, CO, with or without (controls) lo-* M oestradiol for 1 h. 20 @X/ml [3H]-leutine (SA 50 Ci/mmol, Radiochemical Centre, Amersham) for hormone treated samples, and 5 pCi/ml [i4C]-leucine (SA 330 mCi/mmol, Radiochemical Centre, Amersham) for control samples were added to the incubation medium [ 141 and the incubation flasks were agitated for a further 1 h in a Dubnoff shaker at about one stroke per second. In some control experiments, reverse labellinp was used. The tissue samples were then washed thoroughly with ice cold Krebs Ringer bicarbonate buffer. the 3H and i4C labelled samples were mixed and cohomogenized in 1.5 mm01 Na, EDTA and centrifuged at IO5 g for 1 h. Cytosol proteins were analyzed by electrophoresis. IP synthesis was quantitated as described by Katzenellenbogen and Gorski[ 141. Electrophoresis Cellogel (Chemetron, Milan) electrophoresis on 250 [cm thick strips in 0.04M sodium barbitone pH 8.6 was carried out at 20 V/cm toward the positive pole for 90 min. Gels were fixed in 5”,, trichloracetic acid then cut into 2 mm sections for counting of radioactivity. Bovine serum albumin marker (BSA, Schwartz-Mann, Orangeburg. NY) was run simultaneously on a separate gel.

In some experiments. proteins from the --lP” region of Cellogel electrophorcgram (mobility : 1.I-t 1.iO relatively to BSA) were excised and applied to j”,, SDS-polyacrylamide gels (3 x 75 mm), [I I] and &ctrophoresed at 6 mA/gel for 5 hr. Molecular weight determinations were made as described by Weber and Osborne[l5] using crystalline BSA (mol. wt 67.000). ovalbumin (3 times crystallized, mol. wt 45.000). chymotripsinogen (6 times crystallized. mol. wt 25.000). myoglobin (mol. wt 17.8OOJ.

Cellogel slices (2 mm) were placed in scintillation vials containing 0.5 ml Soluene 350 (Packard Instrument Co.). The polyacrylamide gels were fixed in 5”,, TCA for 30 min before slicing. The gels were cut into 1 mm sections and two slices were placed in each vial containing 0.5 ml Soluene. The vials were capped and after at least 12 h at room temperature the scintillant was added (5 g PPO, 0.3 g POPOP per litre toluene). Counting was performed in a Packard spectrometer Model 3380 with correction for crossover of 14C into 3H channel. The 3H/‘4C ratio in each slice was determined. RESULTS

The synthesis of soluble proteins in the endometrium was markedly affected by oestradiol administration and the electrophoretic pattern of newly synthetized proteins varied with the menstrual cycle (Figs 1 and 2). However, the major 3H/‘4C peak observed at al1 stages of the cycle had a mobility similar to that of IP from rat uterus. The electrophoretic mobility of IP from both rat uterus and human endometria relative to BSA on Cellogel was 1.1X-1.24. In the endometrium the most pronounced IP-like response was found during follicular phase. it was at least 556-fold higher than that observed in the luteal phase of the cycle. Besides the IP-like protein(s), the synthesis of several other protein fractions, especially in late follicular phase. were also affected by oestradiol. No induction of IP-like protein(s) were observed in samples exposed irt vitro to testosterone, dexamethasone or progesterone. IP-like protein(s) from Cellogel strips were subjected to electrophoresis on SDS-polyacrylamide gel, which separate proteins according to size rather than net charge (Fig. 3). As the results indicate, the fraction was not pure and contained at least three components with molecular weights of 62.000. 41.000 and 36.000, respectively. The 41,000 dalton peak showed the highest stimulation. The concentration of oestrogen and progesterone receptors in unstimulated endometria varied with the menstrual cycle (Table 1.). The maximal concentrations of cytoplasmic oestrogen receptors occurred in early follicular phase, followed by a continuous decrease reaching the lowest level in late luteal phase. The concentration of progesterone receptor in the cytoplasm was highest in the late follicular phase and

Oestradiol induced protein in human endometrium

“C dpm

.. .... !im

537

3H dpm

..dpm

-t

4000 -3000 -2000 2500 -1000

’ -0 0

0

60 mm

30

30

60 mm

Fig. 1. In vitro effect of oestradiol-17/? on soluble protein synthesis in human endometria from A: early follicular phase, B: late follicular phase. Equal weights of human endometria were incubated in the presence or absence of lo-* M oestradiol for 1 h at 37°C and then for further 1 h with C3H]-leucine (hormone treated) or [‘%I-leucine (controls). Soluble fractions were electrophoresed on Cellogel strips as described in Methods. 3H (solid line) and l“C radioactivity (broken line) were quantitated (bottom). As an indicator of IP synthesis 3H/‘4C ratios were calculated (top). Ordinates: 3H and 14C radioactivity or 3H/‘4C ratios. Abscissa: migration in mm towards positive pole (from left to right). Arrows mark the position of BSA, peaks marked by stars show electrophoretic mobility similar to rat uterine IP.

B

A

1 7 6 5 4I

7 6 5I

0

60

30

mm ‘k

0 dpm

3H dpm 5000

600 4000

:.,?........ ‘.., \

4OOC

,., Y 30

60mm

“.

‘..

5000

4000

‘...,

. .. .

....

‘.....

:.

:..,,

2000 :.

‘..

,

‘...L

3000 2000 1000

‘.. 0 0 30 0 20 60 60 mm Fig. 2. In vitro effect of oestradiol-17p on soluble protein synthesis in human endometria from A: early luteal phase, B: late luteal phase. For further details and abbreviations see legend of Fig. 1. -3

0

tradiol

stimulation

16t

0

I--

8Omm

40

3H dpm

“C dpm ... . .

jOI3

250

0 80 mm

to

0 I

.

.

.

I

6

70 60 50 Go 30 20 x ddaltons

Fig. 3. SDS-acrylamide gel electrophoresis pattern of Cellogel-IP region from endometrial cytosol of late follicular phase. Labelled cytosol proteins from endometria (see Legend of Fig. I.) were subjected to Cellogel electrophoresis. The region corresponding to IP position for rat uterine cytosol (mobiiit~ 1.14 1.30 relatively to BSA) was excised. applied to Y,, SDS-po~~~icr~~larnidegel. The gels were cut into 1 mm sections and the& 3H (solid line) and ‘-‘C radioactivity (broken line) were determined (bottom). As an indicator of IP bynthesis 3H,“JC ratios were calculated (top). Ordinates: ‘H and “C radioactivity or .3H,‘“JC ratios. Abscissa: rni~r~~~l~lll in mm towards positive pole (from left to right dircclion). Arrow indicates the position of hromphenol blue marker. its level was also declined during luteal phase, Our results correspond well to those found by others [l-3, 16, 171. In Fig. 4. levels of cytoplasmic oestradiol or progesterone receptors present in endometria before oesTable

I. Cytoplasmic

oestrogen

and

--_ Phase of the menstrual cycle * Early follicular (days -IO to -5) Late follicular (days -3-U) Early luteal (days 1-S) Late lute31 (days h-13)

and

the nx~pnitudc synthesis

The results presented here show that synthesis of soluble proteins in human endometrium is affected by oestradiol stimulation in both follicular and luteal phases of the cycle. Responses elicited by the hormone are different in the oestrogen rich and progesterone rich phases. One protein is similar to IP of rat uterus as judged from its molecular weight and electrophoretic mobility. The magnitude of this protein synthesis correlates well with cytoplasmic oestradiol receptor concentratj(~i~ in the endometrium throughout the menstrual cycle. These data could be explained by hormonally induced changes in either cellular protein synthesis or protein degradation. To exclude the role of hormonally induced changes in protease activity [7] some experiments were conducted in the presence of a specific protease inhibitor. diisopropy~ fluorop~~ospl?at~ (DFP, [18] ) without affecting the results. The synthesis of uteroglobin in rabbit as well as in human uterus during luteal phase of the cycle hzas been published [7. 191, Since. in our experiments, the changes were more prolloLlnced in follicular phase, it seems to be unlikely that our results are related to uteroglobin synthesis. Apparently the present results are at variance with the recently published data by Iacobelli rt a1.[6]. After oestrogen treatment they found an increased synthesis of a 55,000 daltons protein only during the secretory phase in 5 of 72 endometria studied. At the present no clear cut exp~an~ltion can be offered; how-

progesterone endometrium

receptor

Estradiol receptor cc. (frnoi~rn~ protein)

concentmtions

in human

Progesterone receptor cc. (fmof/m~ protein)

.Y,n,,> p..‘.U,

.\

6%

-132 1003

157

I(

372

17oMlx1

7Xh

SJ7--IOOX

7

157

3GY9

.?7?

247-h I7

IO

57

49

lh-193

II ’

.Y

of ocstrogcn

(RF: I.18 1.24 relative to RSA on Ccllogell are compared. It is apparent that the magnitude of IP synthesis and the number of cytoplasmic ocstradiol binding sites is signi~cant~y correlated IF := 0.83). while no correlation with progesterone receptor levels (r = 0.19) were found. Further qliantitat~v~ relationship between the levels of either oestradiol or progesterone receptors and the magnitude of different peaks in 3Hj1JC ratios has not been observed. of IP like protein

induction

IO-148

.~,,,,&.‘n>.,\ X-185

* The day of ovulation is taken as reference (day 0). u = number of subjects. .? = mean of estimations.

Oestradiol

induced

protein

in human

endometrium

539

B ;;

$

a -2

A

3o .

. . Z zo- . . E . . . .P P, 9. . f lo. > : ’ . Z _ . .* . 0 . . % o?: ::. q:. :,* 0 500 1000

cytoplasmic

oestradiol

receptor

cytoptasmic

cc,

progesteron

receptor

cc

Fig. 4. Correlation between cytoplasmic oestrogen (A) or progesterone receptor (B) concentration (fmol/mg protein, x) and the relative height of “IP” peak (y) calculated by determining the increase in the 3H/14C ratio peak at “IP” region (RF: 1.18-1.24 relative to BSA on Cellogel) over the baseline, i.e. ratio of 3H/‘4C throughout the gel exclusive of the peak regions. A: II = 31. regression equation: y= 2.19~ + 215.9, r = 0.83. P < 0.001. B: II = 31, r = 0.19, P $ 0.1.

ever our feeling is that the considerable differences in the methods used in the two studies are responsible for this discrepancy. On the basis of our results it is suggested that in human endometrium a specific protein(s) with similar properties to rat uterine IP is induced by oestradiol. The pattern of IP synthesis in rat uterus during oestrus cycle and by oestradiol administration was dependent on the blood level of ovarian steroids, and among other factors, the presence of translocatable cytoplasmic oestrogen receptors is essential for the induction of IP [20,21]. In human endometria, as our results show, the magnitude of IP synthesis and the cytoplasmic oestradiol receptor levels during menstrual cycle was significantly correlated. Maximal induction of IP synthesis during follicular phase and a decreased rate of induction in the luteal phase was detected. Since progesterone has multiple antioestrogenic effects on the endometrium [2,3,22-241, the lowered rate of IP synthesis in the luteal phase may be related to these antioestrogenie effects. The major component of IP in rat uterus was identified recently by Reiss and Kaye[25] as the brain type BB isozyme of creatine kinase. The presence of BB isozyme of creatine kinase in human endometrium, breast and breast tumors has also been reported [26]. Though the functional role of IP is still not clear. these data are promising and further investigations of early changes induced by oestradiol simultaneously with receptor assays could be useful in elucidating both clinical and basic biological features of oestrogen action. Acknowledgement-The authors wish to thank Dr R. J. B. King, Hormone Biochemistry Department Imperial Cancer Research Foundation, London, England, for much helpful discussion. REFERENCES

1. Gem T., Eiletz J.. Kreuzer G.. Pollow K. and SchmidtGollwitzer M.: Untersuchungen zur endokrinen regu-

2.

3.

4.

5.

6.

7.

8.

9.

IO.

11.

12.

13

14 15.

lation des endometriums wahrend mensuellen zyklus. Geburtsh. Frauenheilk. 40 (1980) 990-999. Tseng L. and Gurpide E.: Induction of human endometrial estradiol dehydrogenase by progestins. Endocrinology 97 (1975) 8255833. Tseng L. and Gurpide E.: Effects of progestins on estradiol receptor levels in human endometrium. J. c&n. Endocr. Metab. 41 (1975) 402-404. Galand P., Flandroy L. ‘and Mairesse N.: Relationship between the oestrogen induced protein IP and other parameters of estrogenic stimulation. A hypothesis. Life Sci. 22 (1978) 217-238. Manak R.. Wertz N., Slabaugh M., Denari H., Jen-Te Wang and Gorski J.: Purification and characterization of the estrogen-induced protein (IP) of the rat uterus. Mol. cell. Endocr. 17 (1980) 119-132. Iacobelli S.. Merchetti P., Bartoccioni E., Natoli V.. Scambia G. and Kaye A. M.: Steroid-induced proteins in human endometrium. Mol. cell. Endow. 23 (1981) 321-331. Hirsch P. J.. Fergusson I. L. C. and King R. J. B.: Protein composition of human endometrium and its secretion at different stages of the menstrual cycle. Ann. N.Y Acad. Sci. 286 (1977) 233-248. Shapiro S. S. and Forbes S. H.: Alterations in human endometrial protein synthesis during the menstrual cycle and in progesterone stimulated organ culture. Fertil. Steril. 30 (1978) 175-180. Joshi S. G.. Ebert K. M. and Schwartz D. P.: Detection and synthesis of a progesterone-dependent protein in human endometrium. J. Reprod. Pert. 59 (1980) 273-285. Edwards D. P., Adams D. J., Savage N. and McGuire W. L.: Estrogen induced synthesis of specific proteins in human breast cancer cells. Biochem. biophys. Res. Comnun. 93 (1980) 804812. Mairesse N., Devleeschouwer N.. Leclerq G. and Galand P.: Estrogen-induced protein in the human breast cancer cell line MCF-7. Biochmi. biophys. Res. Commun. 97 (1980) 1251l1257. Szikely J. A.. Kornyei J. and Vtrtes M.: Estrogen induced synthesis of specific proteins in human endometria. In Abstr. of III. World Congress of Human Reproduction, Berlin (198 1). pp. 339. Lowry 0. H.. Rosebrough N. J., Farr A. L. and Randall d. J.: Protein measurement with the Folin phenol reagent. J. biol. Chem. 193 (1951) 265-275. KaFzenellenbogen B. S. and Gorski J.: Estrogen action in rho. J. hiol. Chern. 247 (1972) 1299-1305. Weber K. and Osborn M.: The reliability of molecular

16.

17.

18.

19.

20.

21.

weight determinations by dodecyl-sulfate polyacr!Iamide. .I. hiol. C/IL’I~. 244 (I 969) 44064412. Sanborn B. M.. Kuo H. S. and Held B.: Estrogen and progesterone binding site concentrations in human endometrium and cervix throughout the menstrual cycle and in tissue from women taking oral contraceptives. J. \trr-oitl Biochr. 9 i19781 95 I-955. Spona J.. Ulm R.. Biegelmayer G. and Husslein P.: Hormone serum levels and hormone receptor contents of rndometria in women with normal menstrual cycles and patients bearing endometrial carcinoma. Gynrcol. o/?.\r. 1nrc.\t. 10 (1979) 71-80. Notides C. A.: Conformational forms of the estrogen receptor. In: Receptors and hormone action (Edited by B. W. O’Malley and L. Birnbaumer). Academic Press. New York. Vol. II. (1978) 33-81. Feigelson M.. Noske I. G.. Goswami A. K. and Kay E.: Reproductive tract fluid proteins and their hormonal control. Ann. N.Y Acad. Sci. 286 (1977) 273-285. Katzenellenbogen B. S.: Synthesis and inducibility of the uterine estrogen-induced protein, IP, during the rat estrous cycle: clues to uterine estrogen sensitivity. Endocrinology 96 (1975) 289-297. Iacobelli S.. King R. J. B. and Vokaer A.: Antibody to estrogen induced protein (IP) and quantification of the

‘2

23.

24.

25.

26.

protein in rat uterus by radioimmunoassay. Rio&r/?i. hionl~,~ Rex Corwmtr~. 76 ( 1977) 123&l 237. S&idt-Gollwitzer M.. Eiletr J.. Genr T. and Pollow K.: Determination of rstradiol, estrone and progesterone in serum and myometrium: correlation with the content of sex steroid receptors and 17[{-hydroxysteroid dehydrogenase activity throughout the menstrual cvcle. J. cli,l. Endocr. Mrtuh. 49 (19791 37@376. E;ans R. W. and Leavitt W. W.: Progesterone action in hamster uterus: rapid inhibition of 3H-estradiol retention by the nuclear fraction. Elldocrirwlogy 107 (1980) 1261L1263. Gurpide E. and Marks C.: Influence of endometriai 17/I-hydroxysteroid dehydrogenase activity on the binding of estradiol to receptors. J. c/in. Endow. Metab. 52 (1981) 252-255. Reiss N. A. and Kaye A. M.: Identification of the major component of the estrogen-induced protein of rat uterus as the BB isozyme of creatine kinase. J. biol. Chem. 256 (1981) 5741-5749. Kaye A. M., Reiss N.. Shaer A.. Sluyser M., Iacobelii S., Amroch D. and Soffer Y.: Estrogen responsive creatine kinase in normal and neoplastic cells. J. sreroid Biochem. 15 (1981) 69-75.