- Email: [email protected]
Steroid-induced proteins in human endometrium
Molecular and CellularEndocrinology, 23 (198 1) 321-33 1 Elsevier/NorthRolland Scientific Publishers, Ltd.
321
STEROID-INDUCED PROTEINS IN HUMAN ENDOMETRIUM S. IACOBELLI, P. MARCHETTI, E. BARTOCCIONI, V. NATOLI, G. SCAMBIA and A.M. KAYEa Laboraroriodi E~d~~nolo~ ~ole~olare, ~n~ver~t~&to&a S. Cuore, &X68 Roma (I~~~~ and a Department of Hormone Research, The WeizntannInstitute of Science, Rehovot (Israel) Received 19 December 1980; revision received 11 May 1981; accepted 3 June 1981
The synthesis of soluble proteins in human endometria at various phases of the rne~t~~ cycle was evaluated by polyacrylamide gel electrophoresis of [ 3 ’ S]methionine-labeled proteins. Densitometric analysis of the gels revealed alterations in the rate of synthesis of single protein bands throughout the cycle. Administration of conjugated estrogens (Premarin) to women undergoing hysterectomy, or exposure, in vitro, of the endometrial tissue to l’lr%stradiol produced an increased incorporation of [ 3SS]methionine into a specific protein which migrated on SDS-p~yacry~de gels at a molecular weight of about 55 000. Induction of this protein was observed only in those endometria showing a secretory histological appearance. The protein was resolved into at least 2 different spots in two-dimensional gel electrophoresis. An increase in the rate of synthesis of another endometrial protein with an apparent molecular weight of 51000 was observed in tissues exposed in vitro to medroxyprogesterone acetate. These steroid-induced proteins may be a useful marker for studying hormone action in both normal and neoplastic endometria. Keywords: human endome~um;
estrogen; progesterone; protein synthesis.
Estrogen and progesterone produce a progressive sequence of morpho-functional changes in uterine endometrium. Even though various aspects of cyclic endometrial modifications have been studied, at present only Iittle information is available regarding macromolecular synthesis in different stages of the menstrual cycle. The trophic action of estrogen on target tissues is mediated, at least in part, by specific increases in the synthesis of proteins. Much of the work on estrogen-dependent protein synthesis has been carried out on the uterus of both immature and mature ovariectomized rats. It was established that, in the rat uterus, most of the early responses to estrogen are blocked by i~ibitors of protein synthesis, suggesting that one of the important events in the mechanism of hormone action is the production of protein(s) that mediate the subsequent physiological responses. (For a review see Katzenellenbogen and Gorski, 1975 .) Furthermore, Notides and Gorski (1966) first found, by starch-gel electrophoresis of a rat uterus extract, a protein component called ‘Induced Protein’ (IP) whose rate of synthesis was enhanced by f7flestradiol. The synthesis of IP is detectable 30-60 min after estrogen stim~ation in viva or in 0303-7207/81/0000-0000/$02.50
0 1981 Elsevier/North-Holland
Scientific Publishers, Ltd.
S. Iacobelli et al.
322
vitro (Katzenellenbogen and Gorski, 1972) and represents one of the earliest physiological biochemical changes after hormone-receptor binding (Iacobelli, 1973). Villanueva and Leroy Heinrichs (1977) have presented preliminary evidence of rapidly synthesized estrogen-induced proteins in human endometrium. These proteins may represent a useful marker for studying, at the molecular level, estrogen action on human endometrial cells. In this paper, we describe the electrophoretic distribution of [35S]methionine-labeled proteins in human endometrium throughout the menstrual cycle. In addition, we show that the rate of synthesis of specific proteins can be enhanced after stimulation with estrogen or progestin either in vivo or in vitro.
MATERIALS
AND METHODS
[35S]Methionine (780 Ci/mmole) was obtained from the Radiochemical Centre, Amersham, Bucks (Great Britain). Medroxyprogesterone acetate (6cr-methyl-17ahydroxyprogesterane acetate) was a gift of the Upjohn Co., Kalamazoo, MI (U.S.A.). Other steroids were purchased from Sigma Chemical Co., Poole, Dorset (Great Britain). Premarin (conjugated equine estrogens, Ayerst Italiana S.p.A, Milano, Italy) was used as marketed. Hank’s balanced salt solution was obtained from Gibco Bio-Cult Ltd., Paisley (Scotland). Acrylamide and NJV’-methylenebisacrylamide were purchased from Eastman Kodak Co., Rochester, NY (U.S.A.), and recrystallized before use, All other chemicals were reagent-grade. Tissue and hormone treatments Endometrial samples were obtained either by curettage or after hysterectomy from patients with regular menstrual cycles. Informed consent was obtained from each patient whose tissue was used in the study. Samples were grouped according to their histological appearance (Dallenbach-Hellweg, 1975) and the phase of the cycle. For steroid treatment in vivo, the following protocol was used. First an endometrial tissue sample was collected by curettage of the anterior wall of the uterine cavity before the hysterectomy was begun. This tissue served as control. Then the patient was treated with 25 mg of Premarin injected i.v. A second sampling was taken from the posterior wall of the uterus immediately after its removal. The time of exposure to estrogen varied between 45 and 120 min. For steroid treatment in vitro, 3-5 fragments of endometrium were incubated in Hank’s balanced salt solution containing 50 nM of either 17&estradiol, 17a-estradiol or medroxyprogesterone acetate at 37°C for 2 h under an atmosphere of 95% 0s and 5% COZ. The final concentration of ethanol in the incubation was 0.5%; this same ethanol concentration was present in control incubations. Assay of radiolabeled methionine incorporation into endometrial proteins Endometrial tissue from steroid treatment, both in vivo and in vitro, was trans-
Endometrialproteinsynthesis
323
ferred to plastic tubes containing 75 $i of [35S]methionine per ml of Hank’s solution and incubated at 37°C for 2 h. After incubation, the samples were thoroughly rinsed and homogenized in 10 mM Tris-1 mM EDTA, pH 7.6 (about 0.4 ml per 50 mg tissue). Homogenates were centrifuged for 45 min at 175 000 X g at 1’C. For sodium dodecylsulfate (SDS) gel electrophoresis, lo-20 1.t1of cytosolic proteins (about 50 000 cpm) were mixed with an equal volume of buffer containing /3-mercaptoethanol (lo%), SDS (5%), glycerol (20%), Tris (100 mM), and Pyronin y (0.005%). Samples were heated at 100°C for 3 min and analyzed on 1.5-mm-thick 9% polyacrylamide-gel slabs with a 4% stacking gel (Laemli, 1971). Two-dimensional electrophoresis was carried out as described by O’Farrel (1975). Gels were processed for fluorography (Bonner and Laskey, 1974) and exposed to Kodak XR 1 film. The molecular weights of proteins were estimated by their mobilities relative to standard 14C-labeled proteins of known molecular weight. The percentage of [35S]methionine incorporation into the protein bands was estimated by scanning the films with a DD2 scanning densitometer equipped with BCl linear/log integrator (Italglas Ltd., Genova, Italy) and then quantitating each protein band as a percentage of the total band area.
RESULTS Electrophoretic profiles of protein synthesis in the endometrium throughout the cycle SDS-polyacrylamide gel electrophoresis separates soluble uterine proteins into a large number of bands with molecular weights corresponding to values between 12 000 and 120 000. To obtain a clearer resolution, the uterine extracts were fractionated by precipitation at 40-80% saturated ammonium sulfate before electrophoresis; this fraction of uterine proteins accounted for about two-thirds of the soluble neosynthesized proteins. Preliminary experiments showed that the O-40% ammonium sulfate fraction contained no proteins whose rates of synthesis were modified by steroids. Furthermore, it is in the 40-80% fraction that the presence of estrogen-induced protein(s) in rat uterus has been demonstrated (Somjen et al., 1973). For these reasons only the 40-80% fraction. was analyzed in our experiments. Examples of the electrophoretic distribution of newly synthesized proteins from endometria collected at various stages of the cycle are visualized fluorographitally in Fig. 1. In this figure, each of the major protein bands or groups of minor components has been numbered from 1 to 12 according to its molecular weight. Because the amount of labeled amino acid taken up was not constant for all proteins, the intensity level of the various bands in the fluorograms &mot be used as an absolute measurement of the quantity of neosynthesized proteins; however, it does represent a valuable index of the relative changes that take place during the cycle. Although the electrophoretic profile was basically the same throughout the
324
S. lacobelli et al.
;
3 4 5 6 7 8 9
IO 11
12 Fig. 1. SDS-polyac~~ide gels of soluble proteins synthesized in early (Pl), middle (P2) and late proliferative (p3), and secretory (S) endometria. Tissue fragments were incubated in Hank’s balanced salt solution in the presence of [35 Slmethionine for 2 h at 37°C. The tissues were then homogenized, and cytosols were analyzed on SDS-polyacrylamide gels as described in Methods.
cycle, there were some differences in the rate of synthesis of certain proteins. For example, the rate of synthesis of band 6 showed a progressive increase during the menstrual cycle, reaching its maximal intensity in late proliferative endometrium (P3) and remaining at the same level in the foIlowing secretory phase (S). Conversely, the rate of synthesis of band 4 decreased progressively from early (PI) to late (P3) proliferative endometrium, and in the secretory endometrium (S) increased to the previous intermediate proliferative rate (PZ). Moreover, the synthesis of band 5 proteins, which was approximately constant in the proliferative endometrium, showed a marked increase in the secretory tissue. Finally, the group of proteins in band 9 increased progressively in the proliferative endometrium and decreased in the secretory tissue. To obtain more precise information about these changes, the fluorograms of the newly synthesized endometrial proteins, collected during the various phases of the menstrual cycle, were analyzed by a densitometer equipped with a linear/log integrator. Data from different experiments showed
Endometrial protein synthesis
325
Table 1 Changes in endometrial protein synthesis throughout the menstrual cycle Molecular weight
% of total band area Proliferative (n = 15)
Secretory (n = 7)
66-120K
20.87 + 6.30
22.52 t 7.17
41-65K
26.36 f 3.80 a
36.32 + 1.32
31-40K
22.92 + 3.66
17.14 + 5.22
12-30K
29.85 f 5.90
24.12 * 4.80
Soluble endometrial proteins were electrophoresed and the fluorograms scanned as described in Methods. The values shown represent means + SD. of 3 determinations. a Value significantly different from corresponding group of proteins in secretory period @ < 0.01; analysis of variance).
e-96 -69
K K
t-55
K
c-43
K
c-30
K
c- 12.3 K
A
B
C
Fig. 2. SDS-polyacrylamide gels of soluble proteins synthesized in control (A), estrogen-treated (Premarin, 25 mg i.v. for 45 min) (B), and 17aestradiol-treated (C) (50 nM, in vitro for 2 h) endometria. After treatment, the tissue fragments were incubated for 2 h in the presence of [3 SS]methionine and the soluble proteins were analyzed on SDS-polyacrylamide gels as described in Methods. Endometrium was obtained on day 23 of the cycle.
S. Iacobelli et al.
326
slight variations in the intensity of the individual protein bands from endometria collected at the same point in the cycle, which is probably due to several factors: differences in the plasma levels of endogenous hormones and/or in endometrial responsiveness of the specific patient; differences in the proportions of stromal and epithelial cells among the specimens; difficulty in pinpointing exact subdivisions early, middle, late) of the 2 principal phases of the menstrual cycle. Accordingly, the different protein peaks were divided into 4 molecular-weight groups with about equal proportions of total [35S]methionine-labeled proteins in each group. Endometria were classified as proliferative or secretory according to the histological appearance of the tissue. The results of 22 endometria analyzed are illustrated in Table 1. Only in the intermediate molecular-weight group (41K-65K) was there a significant difference in the rate of protein synthesis between the proliferative and the secretory endometria.
+_96 ,_.69
+._55
K K K
~~~43 K +_30K +_.I2.3 A
BC
K
D
Fig. 3. SDS-polyacrylamide gels of soluble proteins synthesized in control (A), 17@estradioltreated (B), testosterone-treated (C), and aldosterone-treated (D) endometria. Steroid treatment (50 nM for 2 h) was in vitro. After treatment, the tissue fragments were incubated for 2 h in the presence of [35S]methionine, and the soluble proteins were analyzed on SDS-polyacrylamide gels as described in the text.
Endometrial protein synthesis
321
Effects of steroids on the rate of synthesis of endometil proteins The modifications in the rate of synthesis of soluble endometrial proteins during different phases of the menstrual cycle are presumably due to changing levels of circulating ovarian steroids. Therefore, we studied the effects of estrogens and progestins on endometrial protein synthesis. In a first series of experiments, the hormone stimulation was carried out by administering conjugated estrogens to women (Premarin, 25 mg, iv). Under these conditions, the rate of synthesis of a 55K molecular-weight protein was increased. A representative example showing the induction of the 55K protein is illustrated in Fig. 2. Similar increases in the rate of synthesis of this specific protein band were obtained by endometrial stimulation in vitro with 17&estradiol (Fig. 3 and Table 2). No increase in the rate of synthesis of the 55K
41)
pH at 20% 5.0 6D
IO
69K 55K 43K 30K 123K
96K 69K 55K ‘?
43K 3OK 123 K
37 p estradiol
I .
Fig. 4. Two-dimensional gels of soluble proteins synthesized in control and 17pestradioLtreated (50 nM, in vitro for 2 h) endometria. Tissue fragments were incubated in the presence of [ 3s S]methionine and the soluble proteins were analyzed on two-dimensional gels as described in Methods. Endometrium was obtained on day 23 of the cycle.
S. Iacobelli et al.
328 Table 2 Magnitude of estrogen induction of the 55K protein in different endometria Sample No.
Fold increase in the 55K protein band area
1 2 3 4 5
1.31 1.68 1.25 1.40 1.79
+ 0.11 f 0.09 + 0.16 f 0.19 ?: 0.11
The fold increase was calculated by determining the increase in the percentage of the 55K band area from estrogen-treated endometria over the controls. Estrogen treatment was as follows: samples l-3, Premarin 25 mg iv. (time of exposure, 60-90 min); samples 4 and 5, 17p-estradiol 50 nM in vitro for 2 h. Soluble endometrial proteins were electrophoresed and the fluorograms scanned as described in Methods. The values shown represent means f S.D. of 3 separate analyses made on different fragments of the same endometrial sample.
-96 -69
K
K
-55K -43K -3OK
Fig. 5. SDS-polyacrylamide gels of soluble proteins synthesized in control (A), 17pestradioL treated (B), and medroxyprogesterone-acetate-treated (C) endometria. Steroid treatment (50 nM for 2 h) was in vitro. After treatment, the tissue fragments were incubated in the presence gels as of 135S]methionine and the soluble proteins were analyzed on SDS-polyacrylamide described in Methods. Endometrium was obtained on day 25 of the cycle.
Endometrtil protein synthesis
329
protein was obtained in samples exposed in vitro to 17cwestradiol (Fig. 2), testosterone, dexamethasone or aldosterone (Fig. 3). The synthesis of the 55K protein was increased by estrogen treatment in 5 of 22 endometria studied (Table 2). It is interesting to note that all the samples in which increased synthesis of specific protein(s) was detected by gel electrophoresis showed a secretory histological appearance and were obtained from patients on days 19-25 of the menstrual cycle. In a preliminary experiment, a secretory endometrial sample, which showed no increase in specific protein synthesis when analyzed by one-dimensional gel electrophoresis, was then re-analyzed on a two-dimensional gel. The results are shown in Fig. 4. Estrogen-induced proteins were seen and resolved into at least 2 proteins of molecular weight 55K and pI between 5.3 and 5.9. With regard to progestin, endometrium was exposed in vitro to the synthetic progestin medroxyprogesterone acetate at a concentration of 50 nM. Fig. 5 shows that this steroid stimulated the synthesis of a 51K molecular-weight protein while slightly inhibiting the synthesis of the 55K protein.
DISCUSSION The results of this study indicate that the rates of synthesis of many human endometrial proteins vary throughout the menstrual cycle. Moreover, some of these rates of synthesis are specifically increased by estrogen or progestin during the secretory period exclusively. In studies of progesterone effects, Hirsch et al. (1977) also found variations in the composition of human endometrial proteins and endometrial flushings at different stages of the menstrual cycle. Likewise, Shapiro and Forbes (1978) found an increase in the rate of synthesis of a protein in the immediate post-ovulatory period which could be mimicked by addition of proges‘ terone to endometrial explants. There may be several possible reasons for the different responsiveness to estrogen between the 2 phases of the menstrual cycle. The first one is based on the proportions of estrogen-responsive and unresponsive cells during the menstrual cycle (Prianishnikov, 1978). Secondly,‘because the experiments were based on the addition of exogenous estrogens, a response to these steroids may not be observed when circulating or endometrial estrogens are already at optimal concentration (SchmidtGollwitzer et al., 1978) if it is not counteracted by a high concentration of progesterone at the same time (Clark, 1979). Alternatively, there may be proteins whose maximal rate of synthesis depends on the synergistic effect of estrogens plus progesterone, a combination available only during the secretory period of the cycle. An interesting example of this in another system is the finding that optimal growth of mouse CR mammary tumors requires both estrogen and progesterone (Sluyser, 1979). While it is not yet possible to attribute a function to any of the specific proteins whose rate of synthesis is increased by estrogen, it is encouraging to note that the
S. Iacobelli et al.
330
rat uterine estrogen-induced protein first described by Notides and Gorski (1966) has recently been identified as the BB isozyme of creatine kinase (Reiss and Kaye, 1980). Perhaps some of the estrogen-responsive proteins detected in this study are also related to energy storage and/or metabolism (Hagerman and ViIlee, 1953; Hafez, 1968). The progesterone-stimulated protein with an apparent molecular weight of 51K detected in this study may be related to the protein of similar weight detected at a later time (24 h after progesterone) by Shapiro and Forbes (1978). Perhaps these progestin-induced effects in the post-ovulatory phase of the menstrual cycle, when the progesterone plasma levels are already elevated, could be pharmacological rather than physiological. The presence of proteins specifically induced by estrogen and progestin in human endometria may offer a valid tool for a better understanding of hormone action in human-steroid-sensitive cells, at the molecular level. Moreover, these proteins may represent a new marker for detecting hormone-sensitive neoplasia, such as endometrial and breast carcinoma. The recent evidence that estrogen can specifically induce the synthesis of secreted as well as cellular proteins in human breastcancer cell lines (Westley and Rochefort, 1979; Edwards et al., 1980) provides an additional incentive for pursuing these studies.
ACKNOWLEDGEMENTS This work was supported
by the C.N.R. Project ‘Biologia della Riproduzione’.
REFERENCES Bonner, W.M., and Laskey, R.A. (1974) Eur. J. Biochem. 46,83-88. Clark, J.H. (1979) Female Sex Steroids: Receptors and Function (Springer, New York). DaUenbach-Hellweg, G. (1975) Histopathology of theEndometrium (Springer, New York). Edwards, D.P., Adams, D.J., Savage, N., and McGuire, W.L. (1980) Biochim. Biophys. Res. Commun. 93,804-812. Hafez, E.S. (1968) in: Perspectives in Reproduction and Sexual Behaviour, Ed.: M. Diamond (Indiana University Press, Bloomington, IN) pp. 207-230. Hagerman, D.D., and VIUee, C.A. (1953) J. Biol. Chem. 203,425-431. Hirsch, P.J., Ferguson, I.L.C., and King, R.J.B. (1977) Ann. N.Y. Acad. Sci. 286,233-245. IacobeBi, S. (1973) Nature (London) New Biol. 245,154-155. Katzenellenbogen, B.A., and Gorski, J. (1972) J. Biol. Chem. 247,1299-1305. Katzenellenbogen, B.A., and Gorski, J. (1975) In: Biochemical Actions of Hormones, Vol. 3, Ed.: G. Litwack (Academic Press, New York) pp. 187-243. Laemli, U.K. (1971) Nature (London) 277,680-685. Notides, A., and Gorski, J. (1966) Proc. Natl. Acad. Sci. (U.S.A.) 56,230-235. O’Farrel, P.H. (1975) J. Biol. Chem. 10,4007-4021. Prianishnikov, V.A. (1978) Gynecol. Oncol. 6,420428.
Endometrial protein synthesis
331
Reiss, N., and Kaye, A.M. (1980) in: Proc. FEBS 13th Annu. Meeting, Jerusalem, p. 260 (Abstract). SchmidtGoBwitzer, M., Genz, T., SchmidtCoBwitzer, K., Pollow, B., and Pollow, K. (1978) in: Endometrial Cancer, Eds. M.G. Brush, R.W. Taylor and R.J.B. King (Bail&e Tindall, London) pp. 227-241. Shapiro, S.S., and Forbes, S.H. (1978) Fertil. Steril. 30, 175-180. Sluyser, M. (1979) Biochim. Biophys. Acta 560, 509-529. Somjen, D., Somjen, G., King, R.J.B., Kaye, A.M. and Lindner, H.R. (1973) Biochem. J. 136, 25-33. Villanueva, B., and Leroy Heinrichs, W. (1977) Gynecol. Oncol. 5,59-67. Westley, B., and Rochefort, H. (1979) Biochem. Biophys. Res. Commun. 90,410-416.
321
STEROID-INDUCED PROTEINS IN HUMAN ENDOMETRIUM S. IACOBELLI, P. MARCHETTI, E. BARTOCCIONI, V. NATOLI, G. SCAMBIA and A.M. KAYEa Laboraroriodi E~d~~nolo~ ~ole~olare, ~n~ver~t~&to&a S. Cuore, &X68 Roma (I~~~~ and a Department of Hormone Research, The WeizntannInstitute of Science, Rehovot (Israel) Received 19 December 1980; revision received 11 May 1981; accepted 3 June 1981
The synthesis of soluble proteins in human endometria at various phases of the rne~t~~ cycle was evaluated by polyacrylamide gel electrophoresis of [ 3 ’ S]methionine-labeled proteins. Densitometric analysis of the gels revealed alterations in the rate of synthesis of single protein bands throughout the cycle. Administration of conjugated estrogens (Premarin) to women undergoing hysterectomy, or exposure, in vitro, of the endometrial tissue to l’lr%stradiol produced an increased incorporation of [ 3SS]methionine into a specific protein which migrated on SDS-p~yacry~de gels at a molecular weight of about 55 000. Induction of this protein was observed only in those endometria showing a secretory histological appearance. The protein was resolved into at least 2 different spots in two-dimensional gel electrophoresis. An increase in the rate of synthesis of another endometrial protein with an apparent molecular weight of 51000 was observed in tissues exposed in vitro to medroxyprogesterone acetate. These steroid-induced proteins may be a useful marker for studying hormone action in both normal and neoplastic endometria. Keywords: human endome~um;
estrogen; progesterone; protein synthesis.
Estrogen and progesterone produce a progressive sequence of morpho-functional changes in uterine endometrium. Even though various aspects of cyclic endometrial modifications have been studied, at present only Iittle information is available regarding macromolecular synthesis in different stages of the menstrual cycle. The trophic action of estrogen on target tissues is mediated, at least in part, by specific increases in the synthesis of proteins. Much of the work on estrogen-dependent protein synthesis has been carried out on the uterus of both immature and mature ovariectomized rats. It was established that, in the rat uterus, most of the early responses to estrogen are blocked by i~ibitors of protein synthesis, suggesting that one of the important events in the mechanism of hormone action is the production of protein(s) that mediate the subsequent physiological responses. (For a review see Katzenellenbogen and Gorski, 1975 .) Furthermore, Notides and Gorski (1966) first found, by starch-gel electrophoresis of a rat uterus extract, a protein component called ‘Induced Protein’ (IP) whose rate of synthesis was enhanced by f7flestradiol. The synthesis of IP is detectable 30-60 min after estrogen stim~ation in viva or in 0303-7207/81/0000-0000/$02.50
0 1981 Elsevier/North-Holland
Scientific Publishers, Ltd.
S. Iacobelli et al.
322
vitro (Katzenellenbogen and Gorski, 1972) and represents one of the earliest physiological biochemical changes after hormone-receptor binding (Iacobelli, 1973). Villanueva and Leroy Heinrichs (1977) have presented preliminary evidence of rapidly synthesized estrogen-induced proteins in human endometrium. These proteins may represent a useful marker for studying, at the molecular level, estrogen action on human endometrial cells. In this paper, we describe the electrophoretic distribution of [35S]methionine-labeled proteins in human endometrium throughout the menstrual cycle. In addition, we show that the rate of synthesis of specific proteins can be enhanced after stimulation with estrogen or progestin either in vivo or in vitro.
MATERIALS
AND METHODS
[35S]Methionine (780 Ci/mmole) was obtained from the Radiochemical Centre, Amersham, Bucks (Great Britain). Medroxyprogesterone acetate (6cr-methyl-17ahydroxyprogesterane acetate) was a gift of the Upjohn Co., Kalamazoo, MI (U.S.A.). Other steroids were purchased from Sigma Chemical Co., Poole, Dorset (Great Britain). Premarin (conjugated equine estrogens, Ayerst Italiana S.p.A, Milano, Italy) was used as marketed. Hank’s balanced salt solution was obtained from Gibco Bio-Cult Ltd., Paisley (Scotland). Acrylamide and NJV’-methylenebisacrylamide were purchased from Eastman Kodak Co., Rochester, NY (U.S.A.), and recrystallized before use, All other chemicals were reagent-grade. Tissue and hormone treatments Endometrial samples were obtained either by curettage or after hysterectomy from patients with regular menstrual cycles. Informed consent was obtained from each patient whose tissue was used in the study. Samples were grouped according to their histological appearance (Dallenbach-Hellweg, 1975) and the phase of the cycle. For steroid treatment in vivo, the following protocol was used. First an endometrial tissue sample was collected by curettage of the anterior wall of the uterine cavity before the hysterectomy was begun. This tissue served as control. Then the patient was treated with 25 mg of Premarin injected i.v. A second sampling was taken from the posterior wall of the uterus immediately after its removal. The time of exposure to estrogen varied between 45 and 120 min. For steroid treatment in vitro, 3-5 fragments of endometrium were incubated in Hank’s balanced salt solution containing 50 nM of either 17&estradiol, 17a-estradiol or medroxyprogesterone acetate at 37°C for 2 h under an atmosphere of 95% 0s and 5% COZ. The final concentration of ethanol in the incubation was 0.5%; this same ethanol concentration was present in control incubations. Assay of radiolabeled methionine incorporation into endometrial proteins Endometrial tissue from steroid treatment, both in vivo and in vitro, was trans-
Endometrialproteinsynthesis
323
ferred to plastic tubes containing 75 $i of [35S]methionine per ml of Hank’s solution and incubated at 37°C for 2 h. After incubation, the samples were thoroughly rinsed and homogenized in 10 mM Tris-1 mM EDTA, pH 7.6 (about 0.4 ml per 50 mg tissue). Homogenates were centrifuged for 45 min at 175 000 X g at 1’C. For sodium dodecylsulfate (SDS) gel electrophoresis, lo-20 1.t1of cytosolic proteins (about 50 000 cpm) were mixed with an equal volume of buffer containing /3-mercaptoethanol (lo%), SDS (5%), glycerol (20%), Tris (100 mM), and Pyronin y (0.005%). Samples were heated at 100°C for 3 min and analyzed on 1.5-mm-thick 9% polyacrylamide-gel slabs with a 4% stacking gel (Laemli, 1971). Two-dimensional electrophoresis was carried out as described by O’Farrel (1975). Gels were processed for fluorography (Bonner and Laskey, 1974) and exposed to Kodak XR 1 film. The molecular weights of proteins were estimated by their mobilities relative to standard 14C-labeled proteins of known molecular weight. The percentage of [35S]methionine incorporation into the protein bands was estimated by scanning the films with a DD2 scanning densitometer equipped with BCl linear/log integrator (Italglas Ltd., Genova, Italy) and then quantitating each protein band as a percentage of the total band area.
RESULTS Electrophoretic profiles of protein synthesis in the endometrium throughout the cycle SDS-polyacrylamide gel electrophoresis separates soluble uterine proteins into a large number of bands with molecular weights corresponding to values between 12 000 and 120 000. To obtain a clearer resolution, the uterine extracts were fractionated by precipitation at 40-80% saturated ammonium sulfate before electrophoresis; this fraction of uterine proteins accounted for about two-thirds of the soluble neosynthesized proteins. Preliminary experiments showed that the O-40% ammonium sulfate fraction contained no proteins whose rates of synthesis were modified by steroids. Furthermore, it is in the 40-80% fraction that the presence of estrogen-induced protein(s) in rat uterus has been demonstrated (Somjen et al., 1973). For these reasons only the 40-80% fraction. was analyzed in our experiments. Examples of the electrophoretic distribution of newly synthesized proteins from endometria collected at various stages of the cycle are visualized fluorographitally in Fig. 1. In this figure, each of the major protein bands or groups of minor components has been numbered from 1 to 12 according to its molecular weight. Because the amount of labeled amino acid taken up was not constant for all proteins, the intensity level of the various bands in the fluorograms &mot be used as an absolute measurement of the quantity of neosynthesized proteins; however, it does represent a valuable index of the relative changes that take place during the cycle. Although the electrophoretic profile was basically the same throughout the
324
S. lacobelli et al.
;
3 4 5 6 7 8 9
IO 11
12 Fig. 1. SDS-polyac~~ide gels of soluble proteins synthesized in early (Pl), middle (P2) and late proliferative (p3), and secretory (S) endometria. Tissue fragments were incubated in Hank’s balanced salt solution in the presence of [35 Slmethionine for 2 h at 37°C. The tissues were then homogenized, and cytosols were analyzed on SDS-polyacrylamide gels as described in Methods.
cycle, there were some differences in the rate of synthesis of certain proteins. For example, the rate of synthesis of band 6 showed a progressive increase during the menstrual cycle, reaching its maximal intensity in late proliferative endometrium (P3) and remaining at the same level in the foIlowing secretory phase (S). Conversely, the rate of synthesis of band 4 decreased progressively from early (PI) to late (P3) proliferative endometrium, and in the secretory endometrium (S) increased to the previous intermediate proliferative rate (PZ). Moreover, the synthesis of band 5 proteins, which was approximately constant in the proliferative endometrium, showed a marked increase in the secretory tissue. Finally, the group of proteins in band 9 increased progressively in the proliferative endometrium and decreased in the secretory tissue. To obtain more precise information about these changes, the fluorograms of the newly synthesized endometrial proteins, collected during the various phases of the menstrual cycle, were analyzed by a densitometer equipped with a linear/log integrator. Data from different experiments showed
Endometrial protein synthesis
325
Table 1 Changes in endometrial protein synthesis throughout the menstrual cycle Molecular weight
% of total band area Proliferative (n = 15)
Secretory (n = 7)
66-120K
20.87 + 6.30
22.52 t 7.17
41-65K
26.36 f 3.80 a
36.32 + 1.32
31-40K
22.92 + 3.66
17.14 + 5.22
12-30K
29.85 f 5.90
24.12 * 4.80
Soluble endometrial proteins were electrophoresed and the fluorograms scanned as described in Methods. The values shown represent means + SD. of 3 determinations. a Value significantly different from corresponding group of proteins in secretory period @ < 0.01; analysis of variance).
e-96 -69
K K
t-55
K
c-43
K
c-30
K
c- 12.3 K
A
B
C
Fig. 2. SDS-polyacrylamide gels of soluble proteins synthesized in control (A), estrogen-treated (Premarin, 25 mg i.v. for 45 min) (B), and 17aestradiol-treated (C) (50 nM, in vitro for 2 h) endometria. After treatment, the tissue fragments were incubated for 2 h in the presence of [3 SS]methionine and the soluble proteins were analyzed on SDS-polyacrylamide gels as described in Methods. Endometrium was obtained on day 23 of the cycle.
S. Iacobelli et al.
326
slight variations in the intensity of the individual protein bands from endometria collected at the same point in the cycle, which is probably due to several factors: differences in the plasma levels of endogenous hormones and/or in endometrial responsiveness of the specific patient; differences in the proportions of stromal and epithelial cells among the specimens; difficulty in pinpointing exact subdivisions early, middle, late) of the 2 principal phases of the menstrual cycle. Accordingly, the different protein peaks were divided into 4 molecular-weight groups with about equal proportions of total [35S]methionine-labeled proteins in each group. Endometria were classified as proliferative or secretory according to the histological appearance of the tissue. The results of 22 endometria analyzed are illustrated in Table 1. Only in the intermediate molecular-weight group (41K-65K) was there a significant difference in the rate of protein synthesis between the proliferative and the secretory endometria.
+_96 ,_.69
+._55
K K K
~~~43 K +_30K +_.I2.3 A
BC
K
D
Fig. 3. SDS-polyacrylamide gels of soluble proteins synthesized in control (A), 17@estradioltreated (B), testosterone-treated (C), and aldosterone-treated (D) endometria. Steroid treatment (50 nM for 2 h) was in vitro. After treatment, the tissue fragments were incubated for 2 h in the presence of [35S]methionine, and the soluble proteins were analyzed on SDS-polyacrylamide gels as described in the text.
Endometrial protein synthesis
321
Effects of steroids on the rate of synthesis of endometil proteins The modifications in the rate of synthesis of soluble endometrial proteins during different phases of the menstrual cycle are presumably due to changing levels of circulating ovarian steroids. Therefore, we studied the effects of estrogens and progestins on endometrial protein synthesis. In a first series of experiments, the hormone stimulation was carried out by administering conjugated estrogens to women (Premarin, 25 mg, iv). Under these conditions, the rate of synthesis of a 55K molecular-weight protein was increased. A representative example showing the induction of the 55K protein is illustrated in Fig. 2. Similar increases in the rate of synthesis of this specific protein band were obtained by endometrial stimulation in vitro with 17&estradiol (Fig. 3 and Table 2). No increase in the rate of synthesis of the 55K
41)
pH at 20% 5.0 6D
IO
69K 55K 43K 30K 123K
96K 69K 55K ‘?
43K 3OK 123 K
37 p estradiol
I .
Fig. 4. Two-dimensional gels of soluble proteins synthesized in control and 17pestradioLtreated (50 nM, in vitro for 2 h) endometria. Tissue fragments were incubated in the presence of [ 3s S]methionine and the soluble proteins were analyzed on two-dimensional gels as described in Methods. Endometrium was obtained on day 23 of the cycle.
S. Iacobelli et al.
328 Table 2 Magnitude of estrogen induction of the 55K protein in different endometria Sample No.
Fold increase in the 55K protein band area
1 2 3 4 5
1.31 1.68 1.25 1.40 1.79
+ 0.11 f 0.09 + 0.16 f 0.19 ?: 0.11
The fold increase was calculated by determining the increase in the percentage of the 55K band area from estrogen-treated endometria over the controls. Estrogen treatment was as follows: samples l-3, Premarin 25 mg iv. (time of exposure, 60-90 min); samples 4 and 5, 17p-estradiol 50 nM in vitro for 2 h. Soluble endometrial proteins were electrophoresed and the fluorograms scanned as described in Methods. The values shown represent means f S.D. of 3 separate analyses made on different fragments of the same endometrial sample.
-96 -69
K
K
-55K -43K -3OK
Fig. 5. SDS-polyacrylamide gels of soluble proteins synthesized in control (A), 17pestradioL treated (B), and medroxyprogesterone-acetate-treated (C) endometria. Steroid treatment (50 nM for 2 h) was in vitro. After treatment, the tissue fragments were incubated in the presence gels as of 135S]methionine and the soluble proteins were analyzed on SDS-polyacrylamide described in Methods. Endometrium was obtained on day 25 of the cycle.
Endometrtil protein synthesis
329
protein was obtained in samples exposed in vitro to 17cwestradiol (Fig. 2), testosterone, dexamethasone or aldosterone (Fig. 3). The synthesis of the 55K protein was increased by estrogen treatment in 5 of 22 endometria studied (Table 2). It is interesting to note that all the samples in which increased synthesis of specific protein(s) was detected by gel electrophoresis showed a secretory histological appearance and were obtained from patients on days 19-25 of the menstrual cycle. In a preliminary experiment, a secretory endometrial sample, which showed no increase in specific protein synthesis when analyzed by one-dimensional gel electrophoresis, was then re-analyzed on a two-dimensional gel. The results are shown in Fig. 4. Estrogen-induced proteins were seen and resolved into at least 2 proteins of molecular weight 55K and pI between 5.3 and 5.9. With regard to progestin, endometrium was exposed in vitro to the synthetic progestin medroxyprogesterone acetate at a concentration of 50 nM. Fig. 5 shows that this steroid stimulated the synthesis of a 51K molecular-weight protein while slightly inhibiting the synthesis of the 55K protein.
DISCUSSION The results of this study indicate that the rates of synthesis of many human endometrial proteins vary throughout the menstrual cycle. Moreover, some of these rates of synthesis are specifically increased by estrogen or progestin during the secretory period exclusively. In studies of progesterone effects, Hirsch et al. (1977) also found variations in the composition of human endometrial proteins and endometrial flushings at different stages of the menstrual cycle. Likewise, Shapiro and Forbes (1978) found an increase in the rate of synthesis of a protein in the immediate post-ovulatory period which could be mimicked by addition of proges‘ terone to endometrial explants. There may be several possible reasons for the different responsiveness to estrogen between the 2 phases of the menstrual cycle. The first one is based on the proportions of estrogen-responsive and unresponsive cells during the menstrual cycle (Prianishnikov, 1978). Secondly,‘because the experiments were based on the addition of exogenous estrogens, a response to these steroids may not be observed when circulating or endometrial estrogens are already at optimal concentration (SchmidtGollwitzer et al., 1978) if it is not counteracted by a high concentration of progesterone at the same time (Clark, 1979). Alternatively, there may be proteins whose maximal rate of synthesis depends on the synergistic effect of estrogens plus progesterone, a combination available only during the secretory period of the cycle. An interesting example of this in another system is the finding that optimal growth of mouse CR mammary tumors requires both estrogen and progesterone (Sluyser, 1979). While it is not yet possible to attribute a function to any of the specific proteins whose rate of synthesis is increased by estrogen, it is encouraging to note that the
S. Iacobelli et al.
330
rat uterine estrogen-induced protein first described by Notides and Gorski (1966) has recently been identified as the BB isozyme of creatine kinase (Reiss and Kaye, 1980). Perhaps some of the estrogen-responsive proteins detected in this study are also related to energy storage and/or metabolism (Hagerman and ViIlee, 1953; Hafez, 1968). The progesterone-stimulated protein with an apparent molecular weight of 51K detected in this study may be related to the protein of similar weight detected at a later time (24 h after progesterone) by Shapiro and Forbes (1978). Perhaps these progestin-induced effects in the post-ovulatory phase of the menstrual cycle, when the progesterone plasma levels are already elevated, could be pharmacological rather than physiological. The presence of proteins specifically induced by estrogen and progestin in human endometria may offer a valid tool for a better understanding of hormone action in human-steroid-sensitive cells, at the molecular level. Moreover, these proteins may represent a new marker for detecting hormone-sensitive neoplasia, such as endometrial and breast carcinoma. The recent evidence that estrogen can specifically induce the synthesis of secreted as well as cellular proteins in human breastcancer cell lines (Westley and Rochefort, 1979; Edwards et al., 1980) provides an additional incentive for pursuing these studies.
ACKNOWLEDGEMENTS This work was supported
by the C.N.R. Project ‘Biologia della Riproduzione’.
REFERENCES Bonner, W.M., and Laskey, R.A. (1974) Eur. J. Biochem. 46,83-88. Clark, J.H. (1979) Female Sex Steroids: Receptors and Function (Springer, New York). DaUenbach-Hellweg, G. (1975) Histopathology of theEndometrium (Springer, New York). Edwards, D.P., Adams, D.J., Savage, N., and McGuire, W.L. (1980) Biochim. Biophys. Res. Commun. 93,804-812. Hafez, E.S. (1968) in: Perspectives in Reproduction and Sexual Behaviour, Ed.: M. Diamond (Indiana University Press, Bloomington, IN) pp. 207-230. Hagerman, D.D., and VIUee, C.A. (1953) J. Biol. Chem. 203,425-431. Hirsch, P.J., Ferguson, I.L.C., and King, R.J.B. (1977) Ann. N.Y. Acad. Sci. 286,233-245. IacobeBi, S. (1973) Nature (London) New Biol. 245,154-155. Katzenellenbogen, B.A., and Gorski, J. (1972) J. Biol. Chem. 247,1299-1305. Katzenellenbogen, B.A., and Gorski, J. (1975) In: Biochemical Actions of Hormones, Vol. 3, Ed.: G. Litwack (Academic Press, New York) pp. 187-243. Laemli, U.K. (1971) Nature (London) 277,680-685. Notides, A., and Gorski, J. (1966) Proc. Natl. Acad. Sci. (U.S.A.) 56,230-235. O’Farrel, P.H. (1975) J. Biol. Chem. 10,4007-4021. Prianishnikov, V.A. (1978) Gynecol. Oncol. 6,420428.
Endometrial protein synthesis
331
Reiss, N., and Kaye, A.M. (1980) in: Proc. FEBS 13th Annu. Meeting, Jerusalem, p. 260 (Abstract). SchmidtGoBwitzer, M., Genz, T., SchmidtCoBwitzer, K., Pollow, B., and Pollow, K. (1978) in: Endometrial Cancer, Eds. M.G. Brush, R.W. Taylor and R.J.B. King (Bail&e Tindall, London) pp. 227-241. Shapiro, S.S., and Forbes, S.H. (1978) Fertil. Steril. 30, 175-180. Sluyser, M. (1979) Biochim. Biophys. Acta 560, 509-529. Somjen, D., Somjen, G., King, R.J.B., Kaye, A.M. and Lindner, H.R. (1973) Biochem. J. 136, 25-33. Villanueva, B., and Leroy Heinrichs, W. (1977) Gynecol. Oncol. 5,59-67. Westley, B., and Rochefort, H. (1979) Biochem. Biophys. Res. Commun. 90,410-416.