Changes in levels of mRNAs of transforming growth factor (TGF)-β1, -β2, -β3, TGF-β type II receptor and sulfated glycoprotein-2 during apoptosis of mouse uterine epithelium

ft. SteroidBiochern. Molec. Biol. Vol. 59, No. 5/6, pp. 367-375, 1996

Pergamon

PII: S 0 9 6 0 - 0 7 6 0 ( 9 6 ) 0 0 1 3 9 - 2

Copyright © 1996 Elsevier Science Ltd. All fights reserved Printed in Great Britain 0960-0760/96 $15.00 + 0.00

Changes in Levels of mRNAs of Transforming Growth Factor (TGF)-fll, -f12, -f13, TGF-fl type II Receptor and Sulfated Glycoprotein-2 During Apoptosis of Mouse Uterine Epithelium K a z u k o W a d a , ~* S h i n t a r o N o m u r a , 2 E i i c h i M o r i i , 2 Yukihiko Kitamura, 2 Yasuko Nishizawa, 3 Akira Miyake I and Nobuyuki Terada 3 'Department of ObstetriCs and Gynecology, Osaka University Medical School, Suita, Osaka 565, Japan; eDepartment of Pathology, Osaka University Medical School, Suita, Osaka 565, Japan and ~Department of Pathology, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka 537, Japan

To e x a m i n e the roles played by transforming growth factors (TGF)-fll, -//2, -f13, and TGF-fl type II receptors in the induction o f apoptosis in the m o u s e uterine epithelium after estrogen deprivation, we investigated the expression o f their m R N A s and the m R N A of sulfated glycoprotein-2 (SGP-2). Pellets containing 100 pg estradiol-17fl (E2) w e r e implanted into ovariectomized m i c e and r e m o v e d four days later. Apoptotic indices (percentage o f apoptotic cells) o f both luminal and glandular epithelia increased after E2 pellets were r e m o v e d , but administration o f progesterone (P), 5~-dihydrotestosterone ( D H T ) , or continued implantation o f E2 pellets suppressed this increase. Levels o f m R N A s o f TGF-fll, -f12, and -if3, and SGP-2 did not increase after estrogen deprivation. However, estrogen deprivation caused a gradual increase in the level o f TGF-fl type II receptor m R N A , and its level increased about six-fold six days later. Moreover, E2, P, and D H T markedly decreased the level o f TGF-fl type II receptor m R N A . In s i t u hybridization demonstrated that m R N A s of TGF-fll, -//2, -f13 and TGF-fl type II receptor were localized to the epithelium. E x o g e n o u s administration o f TGF-fll into the uterine stroma induced apoptosis in the epithelium, a finding that suggests that signals produced by TGF-fls can induce apoptosis. Therefore, the present results suggest that increased sensitivity o f uterine epithelial cells to TGF-fls, as d e m o n s t r a t e d by an increase in TGF-fl type II receptor m R N A , is involved in the induction o f apoptosis after estrogen deprivation, although signals p r o d u c e d by TGF-fls do not appear sufficient to induce apoptosis. Copyright © 1996 Elsevier Science Ltd.

J. Steroid Biochem. Molec. Biol., Vol. 59, No. 5/6, pp. 367-375, 1996

INTRODUCTION In the accessory sex organs, sex steroids first stimulate proliferation of epithelial cells and then maintain the proliferated epithelial cells; discontinuation of stimulation results in apoptosis of the epithelial cells [1-10]. *Correspondence to Kazuko Wada, D e p a r t m e n t of Public Health, E h i m e University School of Medicine, Shigenobu, E h i m e 79102, Japan. Fax: +81-89-964-4367. Received 10 Nov. 1995; accepted 3 Jul. 1996. 367

Transforming growth factor (TGF)-fl isoforms are a family of structurally homologous dimeric proteins; three isoforms TGF-fll, -f12, and -f13 have been identified in mammals [11, 12]. These TGF-fl isoforms have a variety of biological effects, including the regulation of cell growth, cell differentiation, and extracellular matrix formation, modulation of immune responses, and induction of apoptosis [11-14]. Recent studies of TGF-fl signalling have identified three types of TGF-fl binding components; TGF-fl types I, II, and III receptors, and have shown that

368

Kazuko Wada et al.

both TGF-fl types I and II receptors are essential for signalling of TGF-fls, whereas type III receptors regulate the access of TGF-fls to the signalling receptors [15-17]. Kyprianou and Isaacs [18] have shown a marked increase in the level of TGF-fll mRNA in the rat prostate, together with an increase in apoptosis of the epithelium after castration. Kyprianou and Isaacs [18] and Martikainen et al. [19] have also shown that TGF-fll induces epithelial apoptosis in prostatic tissue both in vivo and in vitro. TGF-fll has also been reported to induce apoptosis in cells of other types [20-22]. These results suggest that TGF-fll plays a role in the induction of apoptosis. A role for TGF-fl isoforms in apoptosis also has been suggested in the uterus. Moulton [23] suggests that TGF-fl2 induces apoptosis of the decidua in the mouse uterus during early pregnancy, and Rotello et al. [24] report that TGF-fll induces apoptosis of rabbit uterine epithelial cells in culture. In the uteri of ovariectomized mice, the withdrawal of estrogen stimulation results in apoptosis of the epithelium but not of the stroma [1, 2, 6, 7]. However, the roles played by signals produced by TGF-fl isoforms in apoptosis of the uterine epithelium, which occurs after estrogen deprivation, are still unclear. Therefore, we first examined the expression and localization of mRNAs of T G F- f l l , -f12, -f13, and TGF-fl type II receptor in the uterus of mature mice after estrogen deprivation. The changing pattern of these mRNA levels was compared with that of sulfated glycoprotein-2 (SGP-2), which is associated with apoptosis in various tissues [10, 25, 26]. Apoptosis of the uterine epithelium of mature mice after estrogen deprivation is inhibited by estrogen, progesterone, or androgen [1, 2, 6, 7], and is modulated by glucocorticoids [27]. Examination of the correlation between the effects of steroids on the expression of TGF-fll, f12, and -f13 mRNAs and of TGF-fl type II receptor mRNA and their inhibitory effects on apoptosis could thus provide an insight into the roles played by the signals produced by TGF-fls in apoptosis of the uterine epithelium. Therefore, we also examined the effects of various steroids on the expression of these mRNAs. In addition, using exogenous administration of TGF-fll into the uterine stoma of mature mice under stimulation by estrogen, we examined whether signals produced by TGF-fls can induce apoptosis of the epithelium.

MATERIALS AND METHODS

Mice

Female BALB/c mice, 8-12 weeks old and bred in our laboratory, were used. Mice were housed at 25°C under controlled lighting ( 1 2 h light/12h darkness) and allowed free access to water and pelleted food.

All mice underwent ovariectomy seven days before the experiment. All surgical procedures were performed under pentobarbital sodium anaesthesia. Injection of steroid

Progesterone (5 mg), 5~-dihydrotestosterone (DHT; 200 pg) or dexamethasone (20 #g) was suspended in 0.1 ml of vehicle (0.9% NaC1, 0.4% polysorbate 80, 0.5% carboxymethylcellulose, and 0.9% benzyl alcohol), and 0.1 ml of the solution was injected into the gluteal subcutaneous tissue. Treatment of mice

Cholesterol pellets containing 1% (w/w) estradiol17fl (E2), hereafter called E2 pellets, were prepared, and 10 mg pellets containing 100 pg E2 were implanted subcutaneously in the interscapular space of ovariectomized mice. Mice that received implants of E2 pellets were divided into five groups. Four days after implantation, E2 pellets were removed in four groups and were left in one group. The day of removal of E2 pellets was designated as day 0. Thereafter, each group without E2 pellets received daily subcutaneous injections of vehicle only, progesterone (5 mg/day), D H T (200 #g/day), or dexamethasone (20 #g/day) at 12 p.m. from day 0. Apoptotic index

Five mice of each group were killed by cervical dislocation at 10 a.m. every day, and their uteri were promptly removed. The uterus was fixed in 10% phosphate (0.01 M) buffered (pH7.2) formalin and embedded in paraffin. Transverse sections (5 pm thick) of the midportion of the uterus were prepared and stained with haematoxylin and eosin. About 1000 cells in the luminal or glandular epithelium were examined, and the percentage of apoptotic cells was defined as the apoptotic index. Apoptotic cells were identified by well-established criteria [28]. Cells with the following characteristics were regarded as apoptotic cells: condensation of cytoplasm and chromatin, and cytoplasmic fragments containing condensed chromatin (Fig. 1A). In situ D N A 3'-end labelling The uteri were fixed in 4% phosphate (0.01 M) buffered paraformaldehyde (pH 7.2) for 48h, and embedded in paraffin. Transverse sections (5/~m thick) of the midportion of the uterus were mounted on the poly-L-lysine-coated glass slides. In situ DNA end labelling was carried out according to the method of Gavrieli et al. [29]. Terminal deoxynucleotidyl transferase (GIBCO BRL, Grand Island, NY, U.S.A.) and biotin- 16-dUTP (Boehringer Mannheim Yamanouchi, Tokyo, Japan), were used at concentrations of 0.1 e.u./#l and 10/~M, respectively. The Vectastain Elite ABC kit (Vector, Burlingame, CA, U.S.A.) and diaminobenzidine tetrahydrochloride

TGF-fls in Uterine Apoptosis

369

Germany) according to the manufacturer's instructions.

F i g . 1. Apoptotic cells in uterine luminal epithelium 4 days

after removal o f E2 pellets. A. Apoptotic cells in a section stained by h a e m a t o x y l i n and e o s i n x255. B. Apoptotic cells stained by the in situ D N A 3'-end labelling m e t h o d )<255. Arrows show apoptotic c e l l s .

(DAB) were used for the detection of biotinylated dUTP.

Probes for Northern blot hybridization The T G F - f l l , -f12, and -f13 transcript-specific DNA probes were generously provided by Dr G. R. Cuhna (University of California, San Francisco, CA, U.S.A.). The specifici.ty of each probe was confirmed previously [30]. T he TGF-fl type II receptor DNA probe encoding nucleotides 827-1311 of the rat TGF-fl type II receptor complementary DNA (cDNA) was generously provided by Dr N. Nishi (Kagawa Medical School, Kidagun, Kagawa, Japan). The cDNA of SGP-2 was kindly provided by Dr A. Izawa (Tottori University Medical School, Tottori, Japan) [25]. Th e DNA probe of G A P D H for Northern blot analysis was purchased from Clontech Laboratories Inc. (Palo Alto, CA, U.S.A.). For Northern hybridization, mouse TGF-fll cDNA (1241-1519), mouse TGF-fl2 cDNA (236-737), and mouse TGF-fl3 cDNA (223-503) were used. These cDNA probes were labelled with [~-32p]dCTP (Amersham, Bucks., U.K.) with the Rediprime DNA labelling system (Amersham). T o generate the RNA probes for in situ hybridization, the nucleotide fragments from TGF-fll cDNA (1241-1519), TGF-fl2 cDNA (236-737), TGF-fl3 cDNA (223-503), and TGF-fl type II receptor cDNA (827-1311) were subcloned into Bluescript I KS(-) plasmids (Stratagene, La Jolla, CA, U.S.A.). These plasmids were linearized and transcribed with either T3 or T7 RNA polymerase with the D I G RNA labelling kit (Boehringer Mannheim, Mannheim, SBMB 59/5 f~ B

Northern blot analysis After the mice were killed, their uteri were promptly removed and frozen in liquid nitrogen. Total RNA was extracted with TRIzol (total RNA isolation reagent; GIBCO BRL). Twenty micrograms of RNA of each sample was fractionated on a 1% agarose-formaldehyde gel and transferred to a Hybond N (+) membrane (Amersham). The blot was prehybridized with a rapid hybridization buffer (Amersham) and then hybridized at 62°C with cDNA probes labelled w i t h [0~-32p] dCTP. After hybridization, the blot was washed according to the manufacturer's instructions accompanying the Hybond N (+) membrane. For rehybridization, the blot was washed with 0.1 × SSC (SSC; 0.15 M NaC1, 0.015 M trisodium citrate)/0.1% sodium dodecyl sulfate for 5 min at 98°C. For radio image quantitation, autoradiographs were analysed with Image Quant T M (Molecular Dynamics, Sunnyvale, CA, U.S.A.). For normalization of signals, G A P D H signals were used. The amount of ribosomal RNA was correlated with the intensity of the G A P D H signal. In situ hybridization On days 0-3, the uteri of mice that had received daily injections of vehicle only were promptly removed, and frozen transverse sections (4 #m thick) of the midportion of the uterus were prepared. Sections of the uteri from days 0 to 3 were hybridized with the probes on a single slide so that the conditions of in situ hybridization were the same for all sections. In situ hybridization was performed as described by Nomura et al. [31].

Isolation of epithelial cells On day two, the uteri of mice that had received daily injections of vehicle only were promptly removed. The uterus was opened longitudinally and cut transversely into 3- to 4-mm segments. Uterine segments were incubated in Hank's balanced salt solution (HBSS) containing 1% bovine pancreatic trypsin (Sigma Chemical Co., St Louis, MO, U.S.A.) at 4°C for 90 rain. Pieces of the uterus were transferred to HBSS containing 10% fetal calf serum and then to HBSS. The uterine epithelium was separated from underlying tissue by several gentle pipettings with a Pasteur pipette. Pieces of underlying tissue were transferred to HBSS, and detached epithelial fragments were collected by centrifugation at 500 x g at 4°C for 5 min. Epithelial cells that still adhered to the underlying uterine tissue were separated by repeated vigorous pipettings with a Pasteur pipette. Pieces of the underlying uterine tissue without epithelium were then washed three times with HBSS and collected by centrifugation at 500 × g at 4°C for 5 min. Pieces of

Kazuko Wada et al.

370

the uterus were examined histologically to ensure that few or no epithelial cells were still attached. T h e 5/~g of total RNA extracted from epithelial cells or pieces of uterus was used for Northern blot analysis of T G F - f l l mRNA.

25 ---

(A) Luminal

- e - Control - ~ - Dexamethasone -~- Progesterone - ~ - DHT

20

10

.c

15

0

Effects of TGF-/31 on the apoptotic index of the luminal epithelium Spindle-shaped pellets, containing 200 ng of human recombinant TGF-/31 (King Brewing, Osaka, Japan) and 10 #g of bovine serum albumin (BSA), and spindle-shaped control pellets, containing 10 #g of BSA only, were prepared as described by Langer and Folkman [32]. T h e E2 pellets were implanted into ovariectomized mice. Four days later, a TGF-/31 pellet was inserted into the uterine stroma at the midportion along the longitudinal axis of one uterine horn in one group; a control pellet was inserted in the other group. F r o m the day after pellets were inserted (day one) to day four, five mice from each group were killed at 12 p.m., and the portion of the uterine horn containing the pellet was removed and fixed in 10% phosphate ( 0 . 0 1 M ) buffered ( p H 7 . 2 ) formalin. T h e apoptotic index of the luminal epithelium was determined for the half of the uterus in which the pellet had been implanted.

~.

10

0 0.

<

5 0

I

[

I

I

I

I

I

I

0

1

2

3

4

5

6

7

Days after estrogen deprivation

~,

10- (B) Glandular

X

- e - Control --o- Dexamethasone --*- Progesterone

1O =.m

o

6

~-

4

O D.

<¢

2 ^

0 0

i

i

I

i

1

2

3

4

I

5

i

6

i

7

Days after estrogen deprivation

Statistical analysis Statistical analysis was performed with Student's ttest. A P-value less than 0.05 was considered significant.

RESULTS

Apoptotic index of the epithelium after estrogen deprivation, and effects of various steroids Apoptosis of the uterine epithelium appeared after removal of E2 pellets (Fig. 1A) but not when pellets remained in place (data not shown). Apoptotic cells identified with light microscopy were positively stained with the in situ D N A 3'-end labelling method (Fig. 1B) and showed D N A fragmentation. In luminal epithelium, the apoptotic index increased markedly on day one, reached a maximal level (about 20%) on day two, and thereafter decreased gradually (Fig. 2A). In glandular epithelium, the apoptotic index increased slightly on day one, reached a maximal level (about 4% to 5%) on days two and three, decreased slightly on day four, and remained constant thereafter (Fig. 2B). Both progesterone and D H T significantly decreased the apoptotic index in luminal epithelium after day one (Fig. 2A). T h e increase in the apoptotic index was almost completely suppressed from day three by progesterone and from day four by D H T . However, dexamethasone had little effect. In glandular epi-

Fig. 2. Apoptotic indices o f l u m i n a l a n d g l a n d u l a r epithelia after e s t r o g e n deprivation and the effects of various steroids on t h e m . Pellets c o n t a i n i n g 100 pg E2 were i m p l a n t e d i n t o o v a r i e c t o m i z e d m i c e and r e m o v e d f o u r d a y s later. T h e d a y on which E2 pellets were r e m o v e d was d e s i g n a t e d as d a y 0. T h e m i c e from w h i c h E 2 pellets h a d b e e n r e m o v e d were divided i n t o f o u r g r o u p s , e a c h o f w h i c h r e c e i v e d daily i n j e c t i o n s of vehicle only (control), d e x a m e t h a s o n e (20 #g), prog e s t e r o n e ( 5 m g ) , or D H T ( 2 0 0 # g ) f r o m d a y 0. S y m b o l s r e p r e s e n t m e a n s _+SE for five m i c e . * P < 0 . 0 5 , significant difference f r o m t h e v a l u e o f the control.

thelium both progesterone and D H T almost completely suppressed the increase in the apoptotic index from day two (Fig. 2B). T h e effects of dexamethasone on the apoptotic index were not uniform; the apoptotic index decreased on day two but increased on days five and seven.

Northern blot analysis of expression of mRNAs of TGFill, -/32, -/33, TGF-fl type H receptor, and SGP-2 T h e SGP-2 m R N A signals were highest on the day E2 pellets were removed, day 0 (Fig. 3). Thereafter SGP-2 m R N A signals were higher on days two, three, and four than on days one and six. T h e T G F - f l l m R N A signal was also highest on day 0, but was nearly as high as on day two. T h e m R N A signals of both TGF-fl2 and TGF-/33 did not increase after E2 pellets were removed.

371

TGF-fls in Uterine Apoptosis

Days after estrogen deprivation c

0

1

2

3

4

6

'l

O

E

¢1

SGP-2

.> o e-

TGF-~I

0

TGF-1~2

1

2

3

4

6

Days after estrogen deprivation Fig. 4. E x p r e s s i o n of u t e r i n e TGF-fl type II r e c e p t o r m R N A a f t e r estrogen deprivation. T h e density of a b a n d c o r r e s p o n d ing to TGF-fl type II r e c e p t o r rr~RNA s h o w n in Fig. 3 was m e a s u r e d with a d e n s i t o m e t e r , a n d the a m o u n t of the m R N A relative to G A P D H m R N A was calculated. T h e relative a m o u n t o f m R N A on day 0 was expressed as 1.0.

TGF-1~3 TGF-~ type II receptor

T h e effects of various steroids on the T G F - f l l m R N A levels on days two and four were similar to those on SGP-2 m R N A levels: T G F - f l l m R N A levels on days two and four were decreased by both pro-

GAPDH 28 S

day 2

18S Fig. 3. N o r t h e r n blot analysis of u t e r i n e SGP-2, TGF-fll, -f12, -f13, TGF-fl type II receptor, a n d G A P D H m R N A s a f t e r est r o g e n deprivation. E 2 pellets were i m p l a n t e d into ovarie c t o m i z e d m i c e a n d r e m o v e d four days l a t e r (designated day 0). Total RNA was e x t r a c t e d f r o m t h e u t e r i o n days 0-4 a n d six, a n d 20 pg was l o a d e d p e r lane. T h e expected sizes of TGF-fll (2.5 kb), TGF-ff2 (4.1 kb), TGF-ff3 (3.0 kb), TGF-fl type II r e c e p t o r (5.2 kb), a n d G A P D H (1.5 kb), respectively, were detected.

day 4

SGP-2

TGF-B 1

TGF-B 2

TGF- B 3 On the m R N A signals examined, only the signal of T G F - f l type II receptor m R N A increased continuously after E2 pellets were removed. When signals of T G F - f l type II receptor m R N A were normalized to G A P D H signals for quantitative comparison of m R N A levels after estrogen deprivation, levels of T G F - f l type II receptor m R N A increased continuously after Ez pellets were removed; the level on day six was about six times higher than that on day 0 (Fig. 4).

Effects of steroids on levels of TGF-fl isoforrns, TGF-fl type II receptor, and SGP-2 m R N A s The SGP-2 mRNA levels on days two and four were markedly decreased by progesterone and slightly d e c r e a s e d b y D H T ( F i g . 5 a n d F i g . 6). H o w e v e r , t h e SGP-2 mRNA levels on days two and four were not decreased by estrogen or dexamethasone.

TGF- B type II receptor GAPDH C

E

P

T

D

C

E

P

T

D

Fig. 5. N o r t h e r n blot analysis o f effects o f steroids o n exp r e s s i o n o f u t e r i n e SGP-2, TGF-fll, -if2, -f13, a n d TGF-fl type II r e c e p t o r m R N A s . E2 pellets were i m p l a n t e d into five groups o f o v a r i e c t o m i z e d mice. F o u r days later, E2 pellets were r e m o v e d in four groups, e a c h of w h i c h received daily injections o f vehicle only, p r o g e s t e r o n e (5 mg), D H T (200/tg), o r d e x a m e t h a s o n e (20/zg) f r o m day 0 (the day E2 pellets were r e m o v e d ) . E 2 pellets were left in place i n one group. Total RNA was e x t r a c t e d f r o m t h e u t e r i of t h e s e m i c e o n days two a n d four. C, E, P, T, a n d D i n d i c a t e groups given vehicle only (control), i m p l a n t s of E 2 pellets, progesterone, D H T , a n d d e x a m e t h a s o n e , respectively. T w e n t y m i c r o g r a m s of total RNA was loaded in e a c h lane.

372

Kazuko Wada et al.

day 2

day 4

SGP-2 ~

2

TGF-~I

similar to the results of N o r t h e m blot analysis (Fig. 3 and Fig. 4). Sense strands of the probes did not show any significant signals (Fig. 7D). T o confirm the localization of mRNAs found by in situ hybridization, Northern blot hybridization was performed with total RNAs prepared from epithelial cells separated from stromal and myometrial cells (Fig. 8). T h e TGF-/31 m R N A was found in epithelial cells. Effects of exogenously administered TGF-/31 on the apopwtic index of the luminal epithelium In uteri in which control pellets were implanted, the apoptotic index of luminal epithelium was less than 0.2%. However, after TGF-/31 pellets were implanted, the apoptotic index of the epithelium increased significantly ( P < 0 . 0 1 ) to approximately 2.5% on day one and to about 6% on day three (Fig. 9).

TGF-p2

TGF-~3 2-

2DISCUSSION

TGF-~ type II receptor 2

C E

P T

D

Fig. 6. T h e effects of steroids on e x p r e s s i o n of uterine S G P 2, TGF-/~I, -~2, -j~3, and T G F - ~ type II r e c e p t o r m R N A s . T h e a m o u n t of each m R N A relative to G A P D H m R N A w a s calculated by m e a s u r i n g the density of a b a n d c o r r e s p o n d i n g to e a c h m R N A with a d e n s i t o m e t e r . T h e relative a m o u n t of e a c h m R N A for the control group was e x p r e s s e d as 1.0.

gesterone and D H T but not by estrogen or dexamethasone. Levels of TGF-/32 m R N A were not affected by any steroids on days two and four. Levels of TGF-/33 m R N A were not affected by any steroids on day two but were slightly decreased by estrogen, progesterone, and D H T on day four. T h e level of TGF-/3 type II receptor m R N A was decreased slightly on day two by progesterone and markedly decreased on day four by estrogen, progesterone, and D H T . In situ hybridization of TGF-BI, -/32, and -/33, and TGF-/3 type H receptor m R N A s Signals of TGF-/31, -/32, and -/33 mRNAs were found in both luminal and glandular epithelia but were stronger in luminal epithelium (Fig. 7A, B and C). This localization did not change after E2 pellets were removed (data not shown). T h e signal of TGF-/3 type II receptor m R N A was also localized in both luminal and glandular epithelia but was stronger in glandular epithelium. T h e intensity of the TGF-/3 type II receptor m R N A signal increased from day 0 to day three (Fig. 7E, F, G and H), a finding that was

In the present study, apoptosis occurred in the uterine epithelium after stimulation of the uterus by a supraphysiological dose of E2 was withdrawn. However, the pattern of apoptosis in the epithelium after estrogen deprivation was similar to that seen in previous studies in which apoptosis was induced after the uterus had been stimulated by three daily injections of E2 at a physiological dose (0.2 or 1 #g/day) [6, 7, 27]. Moreover, the effects of steroids on apoptosis were similar to those reported in previous studies [6, 7, 27]. Therefore, the mechanism by which apoptosis is induced after estrogen deprivation does not seem to be affected by the dose of E2 used to stimulate the uterus. Apoptotic cells identified with light microscopy were stained with the in situ D N A 3'-end labelling method. This finding implies that D N A in these cells is fragmented and provides further evidence for apoptosis. We have shown that internucleosomal fragmentation of D N A occurs after estrogen deprivation in the uteri of neonatal mice [33]. In situ hybridization showed that TGF-/31, -/~2, and -f13 mRNAs and T G F - f l type II receptor m R N A were localized in the epithelium, results that were confirmed by Northern blot analysis of T G F - f l l mRNA. In agreement with our results, T G F - f l l , -f12, and -f13 mRNAs were found by Takahashi et al. [35] to be localized in the uterine epithelium of prepubertal mice treated with estrogen or left untreated. T a m a d a et al. [36] and Das et al. [37] report that T G F - f l l and -f12 mRNAs are also expressed in the uterine epithelium of mice during the peri-implantation period but that TGF-fl3 m R N A is not. T h a t TGF-fl3 m R N A is not expressed in the epithelium during the peN-implantation period might be attributed to hormonal conditions during the peri-implantation period.

TGF-fls in Uterine Apoptosis

373

iiiiiiii!!iiii!iiiii~iiil

Fig. 7. In s i t u h y b r i d i z a t i o n o f T G F - f l l , -if2, -f13, a n d T G F - f l t y p e II r e c e p t o r m R N A s . E2 p e l l e t s w e r e i m p l a n t e d into o v a r i e c t o m i z e d m i c e a n d r e m o v e d f o u r d a y s later. T h e d a y on w h i c h E 2 p e l l e t s w e r e r e m o v e d w a s d e s i g n a t e d as d a y 0. T r a n s v e r s e s e c t i o n s o f u t e r u s on d a y s 0, o n e , t w o , a n d t h r e e w e r e h y b r i d i z e d in s i t u w i t h a n t i s e n s e or s e n s e o l l g o n u c l e o t i d e p r o b e s s p e c i f i c for e a c h m R N A . A, B a n d C. I n s i t u h y b r i d i z a t i o n w i t h t h e a n t i s e n s e p r o b e s o f T G F - f l l , TGF-ff2, a n d TGF-fl3 m R N A s , r e s p e c t i v e l y , o n t h e s e c t i o n s o f t h e u t e r u s o n d a y 0. D . W i t h t h e s e n s e p r o b e o f T G F - f l 3 m R N A o n a s e c t i o n o f t h e u t e r u s o n d a y 0. E , F, G a n d H . W i t h t h e a n t i s e n s e p r o b e o f T G F - f l t y p e II r e c e p t o r m R N A o n s e c t i o n s o f t h e u t e r u s on d a y s 0, o n e , t w o , a n d t h r e e , r e s p e c t i v e l y . T h e b a r i n d i c a t e s 0.1 m m .

Levels of mRNAs of T G F - f l l , -f12, and -/33 did not increase after estrogen deprivation. Moreover, the effects of steroids on apoptosis of the epithelium were not necessarily correlated with their effects on m R N A levels of these TGF-/3 isoforms. These results suggest that changes in m R N A levels of TGF-/3 isoforms are unrelated to the mechanism of induction of apoptosis after estrogen deprivation.

RoteUo et al. [24] have shown that TGF-/31 induces apoptosis in rabbit uterine epithelial cells in culture. In agreement with their results, TGF-/31 implanted into the uterine stroma was found in the present study to induce apoptosis of the luminal epithelium. This result does not necessarily indicate a physiological role for TGF-/31 in the induction of apoptosis after estrogen deprivation. However, it strongly

374

Kazuko Wada et al. E

S+M

E

S+M

2.5kb I~

TGF-I~I

GAPDH

Fig. 8. Northern blot analysis of expression o f TGF-fll m R N A in epithelial cells, and stromal and myometrial cells of the uterus. E2 pellets were implanted into ovariectomized m i c e and removed four days later. Two days later, the m i c e were killed and their uteri were removed. Uterine epithelial cells were separated from stromal and myometrial cells, and Northern blot analysis of TGF-fll and G A P D H m R N A s was carried out with total RNA (5/~g) extracted from each cell fraction. E, Epithelial cells; S + M, stromal and myometrial cells.

suggests that signals produced by TGF-fls can induce apoptosis. Kyprianou and Isaacs [38] have shown, with a binding assay, that levels of T G F - f l receptors in the rat prostate increase after androgen deprivation, and that androgen inhibits both the increase in their levels and apoptosis after androgen deprivation. In the uterus, the level of T G F - f l type II receptor m R N A increased markedly after estrogen deprivation, and E2, progesterone and D H T decreased its level, consistent with their effects on apoptosis.

v,-Qx ':t "O

c

.2 O D.

,<

[]

Control

[]

TGF-~I

6-

42-

lj

day1

'

i day2

'

day3

'

day4

Days after implantation Fig. 9. T h e i n c r e a s e in the apoptotic i n d e x o f l u m i n a l epit h e l i u m b y e x o g e n o u s l y a d m i n i s t e r e d T G F - f l l . Mice that h a d u n d e r g o n e o v a r i e c t o m y r e c e i v e d i m p l a n t s o f E2 pellets, and 4 days later TGF-fll pellets (200 ng TGF-fll/pellet) or control pellets w e r e i m p l a n t e d into the uterine walls. Thereafter, the a p o p t o d c index of the l u m i n a l e p i t h e l i u m near the pellet w a s determined. T h e d a y after implantation o f TGF-fll or control pellets w a s designated as d a y one. T h e heights o f bars for contol (black) and TGF-fll groups (shaded) represent m e a n s +SE for five m i c e . * P < 0.01, significant difference from the value for m i c e w i t h control pellets.

Two transmembrane serine/threonine kinases, T G F - f l type I and type II receptors, are essential for signal transduction of TGF-fls [15-17]. When T G F - f l type II receptor, a constitutively active kinase, directly binds TGF-fl, T G F - f l type I receptors are recruited and phosphorylated to produce a heteromeric signalling complex [15-17]. Choi and Ballermann [22] report that the action of T G F - f l to induce apoptosis in glomerular capillary endothelial cells is transmitted by this signal-transducing system. Because most of the T G F - f l type II receptor m R N A was located on the uterine luminal and glandular epithelia, the increase in the level of T G F - f l type II receptor m R N A after estrogen deprivation suggests that the uterine epithelium becomes more sensitive to TGF-fls after estrogen deprivation. Furthermore, apoptosis of uterine epithelial cells can be induced by TGF-fls, as described above. An increase in sensitivity to TGF-fls may thus be involved in the induction of apoptosis after estrogen deprivation. However, not all epithelial cells with a high level of T G F - f l type II receptor m R N A underwent apoptosis. Furthermore, levels of T G F - f l type II receptor m R N A increased more slowly than did the rate of apoptosis. Therefore, another factor or factors other than T G F - f l type II receptors appear to play central roles in the induction of apoptosis of uterine epithelial cells. In the present study, levels of T G F - f l type II receptor m R N A continued to increase after estrogen deprivation even after the apoptotic index of the epithelium had decreased. Because most T G F - f l type II receptor m R N A was located in the epithelium, the continuous increase in T G F - f l type II receptor m R N A suggests that surviving epithelial cells have more T G F - f l type II receptors. Therefore, T G F - f l type II receptors in these cells might play another role in addition to the induction of apoptosis. Furthermore, heterogeneous populations of epithelial cells might show different responses to TGF-fls. T h e SGP-2 protein was first described as a constitutively expressed secretory protein of Sertoli and epididymal epithelial cells, and is identical to testosterone-regressed prostate message-2 and clusterin [10, 26]. T h e SGP-2 gene is induced in various tissues undergoing apoptosis, but the precise role of SGP-2 in apoptosis is not known [10, 26]. In the present study, the SGP-2 gene was expressed before estrogen deprivation, and its expression after estrogen deprivation did not appear to be correlated with the apoptotic index in the epithelium. This finding suggests that most SGP-2 protein have functions other than that in apoptosis in the uterine epithelium. In conclusion, the present results support the possibility that a T G F - f l signal-transducing system through T G F - f l type II receptors is involved in the induction of apoptosis of the uterine epithelium after estrogen deprivation, along with other factors that may play more central roles.

T G F - f l s in U t e r i n e Apoptosis Acknowledgements--This work was in part supported by a grant from the Osaka Cancer Fotmdation.

REFERENCES 1. Martin L., Pollard J. W. and Fagg B.: Oestriol, oestradiol-17fl and the proliferation and death of uterine ceils..7. Endocr. 69 (1976) 103-105. 2. Finn C. A. and Publicover M.: Hormonal control of cell death in the luminal epithelium of the mouse uterus..7. Endocr. 91 (1981) 335-340. 3. Terada N., Ogasawara Y., Yamane T., Matsumoto K. and Kitamura Y.: Heterogeneity in mouse seminal vesicle epithelial cells responding to androgen as evaluated by incorporation of [L25I] iododeoxyuridine. Endocrinology 116 (1985) 1466-1472. 4. Kyprianou N. and Isaacs J. T.: Activation of programmed cell death in the rat ventral prostate after castration. Endocrinology 122 (1988) 552-562. 5. Rotello R. J., Hocker M. B. and Gerschenson L. E.: Biochemical evidence for programmed cell death in rabbit uterine epithelium. Am..7. Pathol. 134 (1989) 491-495. 6. Terada N., Yamamoto R., Takada T., Miyake T., Terakawa N., Wakimoto H., Taniguchi H., Li W., Kitamura Y. and Matsumoto K.: Inhibitory effect of progesterone on cell death of mouse uterine epithelium. J. Steroid Biochem. 33 (1989) 1091-1096. 7. Terada N., Yamamoto R., Takada T., Taniguchi H., Terakawa N., Li W., Kitamura Y. and Matsumoto K.: Inhibitory effect of androgen on cell death of mouse uterine epithelium. J. Steroid Biochem. 36 (1990) 305-310. 8. Rotello R. J., Lieberman R. C., Lepoff R. B. and Gerschenson L. E.: Characterization of uterine epithelium apoptotic cell death kinetics and re~lation by progesterone and RU 486. Am. _7. Pathol. 140 (19o2) 449-456. 9. Thompson E. B.: Apoptosis and steroid hormones. Molec. Endocr. 8 (1994) 665-673. 10. Tenniswood M. P., Grenette R. S., Lakins J., Mooibroeck M., Wong P. and Welsh J.-E.: Active cell death in hormone-dependent tissues. Cancer Metast. Rev. 11 (1992) 197-220. 11. Spore M. B., RobeIx A. B., Wakefield L. M. and de Crombrugghe B.: Some recent advances in the chemistry and biology of transforming growth factor-ft, ft. Cell. BioL 105 (1987) 1039-1045. 12. Massagud J.: The transforming growth factor-]/ family. Annu. Rev. Cell. Biol. 6 (1990) 597-641. 13. Shull M. M., Ormsby I., Kier A. B., Pawlowski S., Diebold R. J., Yin M., Allen R., Sidman C., Proetzel G., Calvin D., Annunziata N. and Doetschman T.: Targeted disruption of the mouse transforming growth factor-ftl gene results in multifocal inflammatory disease. Nature 359 (1992) 693-699. 14. Sellheyer K., Bickenbazh J. R., Rothnagel J. A., Bundman D., Longley M. A., Krieg T., Roche N. S., Roberts A. B. and Roop D. R.: Inhibitior, of skin development by overexpression of transforming growth factor ftl in epidermis of transgenic mice. Proc. Natl. Acad. ScL U.S.A. 90 (1993) 5237-5241. 15. Franzen P., Dijke P. T., Ichijo H., Yamashita H., Shulz P., Heldin C.-H. and Miyazono K.: Cloning of a TGFft type I receptor that forms a heteromeric complex with the TGFft type II receptor. Cell 75 (1993) 681-692. 16. Massague J., Attisano L. and Wrana J. L.: The TGF-ft family and its composite receptors. Trends Cell. Biol. 4 (1994) 172178. 17. Wrana J. L., Attisano L., Wieser R., Ventura F. and Massague J.: Mechanism of actiwttion of the TGF-ft receptor. Nature 370 (1994) 341-347. 18. Kyprianou N. and Isaacs J. T.: Expression of transforming growth factor-]/ in the rat ventral prostate during castrationinduced programmed cell death. Molec. Endocr. 3 (1989) 1515-1522. 19. Martikainen P., Kyprianou N. and Isaacs J. T.: Effect of transforming growth factor-ill on proliferation and death of rat prostatic cells. Endocrinolo~,,y 127 (1990) 2963-2968.

375

20. Picht G., Hundertmark N., Shmitt C. P. and Bauer G.: Clonal analysis of the effect of TGF-ft on the apoptosis-inducing activity of normal cells. Exp. Cell. Res. 218 (1995) 71-78. 21. Bursh W., Oberhammer F., Jirtle R. L., Askari M., Sedivy R., Grasl-Kraupp B., Purchio A. F. and Schutle-Hermann R.: Transforming growth factor-ftl as a signal for induction of cell death by apoptosis. Br. J. Cancer 67 (1993) 531-536. 22. Choi M. E. and Ballermann B. J.: Inhibition of capillary morphogenesis and associated apoptosis by dominant negative mutant transforming growth factor-]? receptors..7. Biol. Chem. 270 (1995) 21144-21150. 23. Moulton B. C.: Transforming growth factor-ft stimulates endometrial stromal apoptosis in vitro. Endocrinology 134 (1994) 1055-1060. 24. Rotello R. J., Lieberman R. C., Purchio A. F. and Gerschenson L. E.: Coordinated regulation of apoptosis and cell proliferation by transforming growth factor ftl in cultured uterine epithelial cells. Proc. Natl. Acad. Sci. U.S.A. 88 (1991) 3412-3415. 25. Izawa M.: Expression of sulfated glycoprotein 2 and pSvr-1 genes and involution of steroid hormone-dependent rat tissues. Endocr. Jpn. 38 (1991) 61-66. 26. Sensibar J. A., Qian Y., Griswold M. D., Sylverster S. R., Bardin C. W., Cheng C. Y. and Lee C.: Localization and molecular heterogeneity of sulfated glycoprotein-2 (clusterin) among ventral prostate, seminal vesicle, testis, and epididymis of rats. BioL Reprod. 49 (1993) 233-242. 27. Terada N., Yamamoto R., Yamamoto T., Nishizawa Y., Taniguchi H., Terakawa N., Kitamura Y. and Matsumoto K.: Effect of dexamethasone on uterine cell death..7. Steroid Biochem. Molec. Biol. 38 (1991) 111-115. 28. Wyllie A. H., Kerr J. F. R. and Currie A. R.: Cell death: the significance of apoptosis. Int. Rev. Cytol. 68 (1980) 251-306. 29. Gavrieli Y., Sherman Y. and Ben-Sasson S. A.: Identification of programmed cell death in situ via specific labeling nuclear DNA fragmentation. J. Cell. Biol. 119 (1992) 493-501. 30. Robinson S. D., Silberstein G. B., Roberts A. B., Flanders K. C. and Daniel C. W.: Regulated expression and growth inhibitory effects of transforming growth factor-beta isoforms in mouse mammary gland development. Development 113 (1991) 867-878. 31. Nomura S., Hirota S., Morii E., Ito A. and Kitamura Y.: Methods for detecting expression of c-kit receptor tyrosine kinase and ligand in adult mouse and rat brain. Neuroprotocols 1 (1992) 256-262. 32. Langer R. and Folkrnan J.: Polymers for the sustained release of proteins and other macromolecules. Nature 263 (1976) 797800. 33. Jo T., Terada N., Saji F. and Tanizawa O.: Inhibitory effects of estrogen, progesterone, androgen and glucocorticoid on death of neonatal mouse uterine epithelial cells induced to proliferate by estrogen. J. Steroid Biochem. Molec. Biol. 46 (1993) 25-32. 35. Takahashi T., Eizman B., Bossert N. L., Walmer D., Sparrow K., Flanders K. C., McLachlan J. and Nelson K. G.: Transforming growth factors ftl, ]/2, and ft3 messenger RNA and protein expression in mouse uterus and vagina during estrogen-induced growth: a comparison to other estrogen-regulated genes. Cell Growth Diff. 5 (1994) 919-935. 36. Tamada H., McMaster M. T., Flanders K. C., Andrews G. K. Dey S. K.: Cell type-specific expression of transforming growth factor-ftl in the mouse uterus during the periimplantation period. Molec. Endocr. 4 (1990) 965-972. 37. Das S. K., Flanders K. C., Andrews G. K. and Dey S. K.: Expression of transforming growth factor-ft isoforrns (ft2 and ft3) in the mouse uterus: analysis of the periimplantation period and effects of ovarian steroids. Endocrinology 130 (1992) 34593466. 38. Kyprianou M. and Isaacs J. T.: Identification of a cellular receptor for transforming growth factor-ft in rat ventral prostate and its negative regulation by androgens. Endocrinology 123 (1988) 2124-2131.