Apoptosis of pregnancy-dependent mammary tumor and transplantable pregnancy-dependent mammary tumor in mice

CANCER LETTERS

ELSEVIER

Cancer Letters 110 (1996) 113-121

Apoptosis of pregnancy-dependent mammary tumor and transplantable pregnancy-dependent mammary tumor in mice Hiromi Kojima”, Yugo Fukazawaa, Tomomi Satoa, Masato Enari”, Akio Matsuzawab, Satoshi Tsunodac, Hiroshi Nagasawac, Yasuhiko Ohtad, Taisen Iguchi”,* “Graduate School of Integrated Science, Yokohama City University, 22-2 Seto, Kanazawa-ku, Yokohama 236, Japan Animal Research Center, Institute of Medical Science, University of Tokyo, 4-6-l Shirokanedai, Minato-ku, Tokyo 108, Japan “Experimental Animal Research Laboratory, Meiji University, Kawasaki, Kanagawa 214, Japan ‘Laboratory of Animal Science, Department of Veterinary Science, Faculty of Agriculture, Tottori lJniversity,Tottori 680. Japan

bLahoratory

Received 6 August 1996; revision received 20 August 1996; accepted 10 September 1996

Abstract Pregnancy-dependent mammary tumors (PDMT) of GR/A mice and transplantable PDMT (TPDMT-4 line) in DDD mice, are exceptionally stable in hormone dependence, continue to grow until parturition and regress soon after delivery. In order to study the regression mechanism of PDMT and TPDMT4, morphological and biochemical changes were examined in the tumors removed on day 18 (TPDMT-4) or day 20 (PDMT) of pregnancy, and on the expected parturient and the following postpartum days. DNA fragmentation occurred from day 18 (TPDMT-4) or day 20 (PDMT) of pregnancy to the day after parturition. Apoptotic cells were demonstrated by an in situ 3’-end labeling method, and the plateau of the number of apoptotic cells was observed on the parturient day in PDMT and on the day after parturition in TPDMT-4. Reverse transcriptase polymerase chain reaction showed that expression of Fas was slightly increased but that of bcl-2 was decreased during the process of involution of TPDMT-4 and PDMT. These results suggest that both an increase in expression of Fas and decrease in expression of bcl-2 are involved in the apoptosis of pregnancy-dependent mammary tumor cells after parturition. Keywords:

Pregnancy-dependent

mammary tumor; Apoptosis; Fas; bcl-2; Mouse

1. Introduction The GR/A mouse is characterized by the development of pregnancy-dependent mammary tumors (PDMT), which appear in the middle of pregnancy

and continue to grow until the end of pregnancy. PDMT regress or disappear soon after parturition

*Corresponding author. Tel.: +81 45 7872216, fax: +81 45 7872370; e-mail: [email protected]

0304-3835/96/$12X@ PII

SO304-3835(96)04469-2

regardlessof the onset of lactation, and appearagain at subsequentpregnancies. The PDMT often transform into hormone-independent, autonomousmammary tumors [l-3]. The TPDMT-4 tumors, a transplantable mouse mammary tumor line that is exceptionally stable in hormone dependence,were establishedfrom a virusinduced spontaneousmammary tumor in DDD mice and have been characterized by pregnancy-dependent growth. TPDMT4 grow during pregnancy and regressafter delivery, although higher growth peaks

Copyright 0 1996 Elsevier Science Ireland Ltd. All rights reserved

114

H. Kojima

et al. : Cancer

are observed in subsequent pregnancies. They remain dormant in virgins [4]. Estrogen and progesterone are known to be important hormones for mouse mammary tumorigenesis, including the induction of hormone-dependent mammary tumors in virgin DDD mice [5] and GR/A mice [2,6]. Many studies on the growth of PDMT and TPDMT4 [7-91 have been reported; however, there are only two reports on the regression of these tumors, one is that tumor necrosis factor-a! (TNF-a) somewhat inhibits the growth of PDMT [lo], and the other is that the regression of PDMT is caused by apoptosis in GR/ A mice [II]. In order to study the mechanism of regression of PDMT and TPDMT-4, DNA fragmentation, quantification of apoptotic cells and gene expression were studied by electrophoresis, in situ 3’-end labeling method [ 121 and reverse transcriptase polymerase chain reaction (RT-PCR).

2. Materials

and methods

2.1. Animals GR/A Mei mice and DDD mice were kept under a 12: 12 h light/dark cycle at 22-24°C. Standard laboratory feed (CE-2, CLEA, Tokyo) and tap water were given ad libitum. The mouse care and handling conformed to the NIH guidelines for Animal Research. The experimental protocol was approved by the Institutional Animal Care and Use Committee. GR/A mice maintained by strict brother-sister mating were obtained from the Experimental Animal Laboratory, Meiji University, Kawasaki, Japan. At 60-70 days of age, female mice were mated with males. Pregnant or lactating mice with palpable PDMT were sacrificed by decapitation under light ether anesthesia on day 20 of pregnancy, and on the expected parturient and the following postpartum days. At autopsy, PDMT were immediately removed, frozen in liquid nitrogen and stored at -80°C for later extraction of DNA and RNA or fixed in prepared 4% (w/v) paraformaldehyde for histological examination. DDD mice maintained in the mouse colony of the Laboratory Animal Research Center, Institute of Medical Science, University of Tokyo were used for the present study. To avoid confusion of spontaneously developed preneoplastic lesions with tumors

Letters

II0 (1996)

113-121

from the host gland, 3-week-old female DDDf mice free of milk-transmitted murine mammary tumor virus served as recipients for the observation of the tumor-producing capability in virgin mice These females were implanted with a piece of “: mm’ TPDMT-4 tumor into the right inguinal mammary fat pad and kept with males in order that they be impregnated [13]. The day on which the vaginal plug was found was designated as day 0 of pregnancy. TPDMT-4 appearing after midpregnancy reached the maximal size by the end of pregnancy, but disap peared quickly after parturition whether or not the mice underwent lactation [ 141. Volumes of TPDMT 4 were calculated as a rotation ellipsoid i&?/h; a,h = long and short diameters). Pregnant or lactating mice with palpable TPDMT-4 were sacrificed by decapitation under light ether anesthesia on day 1% of pregnancy, and on the expected parturient and the following postpartum days. At autopsy, TPDMT-4 were immediately removed, frozen in liquid nitrogen and stored at -80°C for later extraction of DNA and RNA or fixed in prepared 4% (w/v) paraformaldehyde for 4 days at 4°C for detection of apoptotic cells. Materials, unless otherwise mentioned, were obtained from Wako Pure Chemical Industries, Osaka, Japan. 2.2. DNA extraction und analysis Genomic DNA from PDMT on day 20 of prepnancy, and on the parturient and the following days in addition to that from TPDMT-4 on day IX of pregnancy, and on the parturient and the following days were prepared for examination of DNA fragmentation as an index of apoptosis. PDMT and TPDMT-4 frozen in liquid nitrogen were crushed into a line powder in liquid nitrogen and then incubated in a digestion but’. fer (100 mM Tris-HCl, 5 mM EDTA, 0.2% SDS, 200 mM NaCl, 100 pg/ml proteinase K) at 65°C until the tissue became totally dissolved. Samples were extracted with 1 vol. PCIAA (phenol/chlorofotm/isoaminoalcohol, 25:24: l), centrifuged for 5 nun at 3500 rpm, then treated with 10 mg/ml RNase A (Sigma Chemical Co., St. Louis, MO) and re-extracted, and genomic DNA was precipitated with 1 vol. of isopropan01 at -20°C for 60 min. Concentration and purity of DNA were determined by measuring the optical density at 260 and 280 nm. Ten pg DNA were subjected to electrophoresis on a 2% agarose gel with

H. Kojima Table

et al. I Cancer

Letters

II0

(1996)

1

PCR primers

for the detection

of various

gene expressions

used in this study

Gene

Genbank accession no. (ref.)

Position of 5’ residue”

Sequence

TN&Y

X0261 1

996 1146 493 1067 13 324 339 573 457 797 1187 1513 1195 2390 1481 1840 1159 1356 911 091

TCCAACTCTGTGCTCAGAGC GAACTCAGGAATGGACATTCG GTGTGTAACTGCCATGCAGG AGGTCTGCATTGTCAGGACG ATACTGACACGGATCCTGCC CACAATTCCAACTGCCACG GCAGAAGATGCACACTCTGC ATGGTCAACAACCATAGGCG ACTCCGTGAGTTCACCAACC CCTCCA’ITAGCACCAGATCC GCCCTGGATACCAACTA’ITGC GCAGGAGCGCACAATCATG? GCCAGTGGTGATCAGAAAACT GAGAATGGTCAGTGTTCCAG CAGGGCAGTCAGAGGAAGAAG GCTGCACYTACACGACTTCAC GAAGTGCCATTGGTACCTGC AGGCTGGAAGGAGAAGATGC CACTCTTCCACCTTCGATGC CTCTTGCTCAGTGTCCTT’GC

TNF-R

115

113-121

(~55)

mKGF

MS8067 M76654-6 222703

Fas

M83649

Fas ligand

U06948

TGF-PI

Ml3177

TGF-02 TGF-/33 bcl-2

Ml6506

GAPDH

M32599

TNF-or, tumor necrosis factor-a; TNF-R 0; GAPDH, glyceraldehyde-3-phosphate “Numbers accession

besides the primer number is given.

sequences

(p55), 55 kDa TNF-cx receptor; dehydrogenase. indicate

5’-end

nucleotide

TBE (45 mM Tris, 45 mM boric acid, 1 mM EDTA, pH 8.0) and then stained with ethidium bromide. The 100 bp ladder (Pharmacia, Uppsala. Sweden) was used as a marker.

2.3. In situ 3’-end labeling method PDMT and TPDMT-4 fixed in 4% buffered paraformaldehyde were embedded in paraffin and sectioned at 8 pm. The sections mounted on slide glasses coated with 3-aminopropyltriethoxysilane (Shin-etsu, Nagano, Japan) were deparaffinized and hydrated. DNA nick end labeling of tissue sections were carried out following the method of Gavrieli et al. [ 121 with minor modifications. Nuclei of tissue sections were stripped from proteins by incubation with 20 pg/ml proteinase K in TE buffer (10 mM Tris-HCl, 10 mM EDTA) for 15 min at room temperature; the slides were washed in TE for 2 min. Endogenous peroxidase was inactivated by covering

(5’ + 3’)

KGF, keratinocyte of the primer

growth

corresponding

Amplified length (bp)

Annealing temp. for PCR

151

60

575

60

312

60

235

56

341

60

327

60

396

60

360

60

198

60

181

62

factor; TGF-P,

transforming

to the sequence

for which

growth

factor-

the Genbank

the sections with 2% Hz02 in TE for 5 min at room temperature. For a positive control of the labeling, sections were treated for 5 min at 37°C with 0.04% DNase I in a terminal deoxynucleotidyl transferase (TdT) buffer (30 mM Tris-HCl, 140 mM sodium cacodylate, 1 mM cobalt chloride, pH 7.2). The sections were rinsed twice with TdT buffer and preincubated for 30 min at room temperature. TdT (0.3 U/ml, Takara, Ohtsu, Japan) and biotinylated dATP (4 PM, GIBCO, Grand Island, NY) or dUTP (4 PM, Boehringer Mannheim GmbH, Germany) in TdT buffer were then added to cover the section, and then incubated in humidified atmosphere at 37°C for 90 min. For negative control, TdT enzyme was omitted from the reaction mixture. The slides were then transferred to the termination buffer (300 mM NaCl, 30 mM sodium citrate) for 15 min at room temperature. The sections were rinsed with 10 mM PBS (pH 7.4), covered with 2% bovine serum albumin (Sigma) for 10 min at room temperature, and then incubated in extra-avidin per-

I I6

A (3 E

2500 2000

TPDMT-4

J5 5 1500 Q1 2

1000

6E

500

c'

0

13 14 15 16 17 18

Days after parturition

Days in pregnancy Fig. 1. The volume of TPDh4’T-4 during turition in DDD mice. Three representative

0 1 2 3

pregnancy TPDMT-4

and after pararc depicted.

oxidase (Sigma) diluted 1:1000 in PBS for 30 min at 37°C. Diaminobenzidine tetrahydrochloride (Sigma) reaction with imidazole was carried out for the detection of biotinylated nucleotide. The slideswere counterstained with 0.5% methyl green (Merck, Darmstadt, Germany). The apoptotic index was estimated by counting the stainedapoptotic cells per 500 tumor cells in five randomly-selected sections. Data were expressedasthe mean and standarderror and the significance of difference was analyzed by anova or the student t-test.

[ 151with minor modifications. An optimai condition for the RT-PCR was examined by changing initial amounts of total RNA (150-600 ng) and/or PCR cycles (26-30 cycles) so as to be in the linear range. Three hundred ng of total RNA were amphtied with RNA PCR kit with AMV reverse transcriptase (RTI (Tdkara) according to the manufacturer’i protocol with minor modifications. The RT reaction M’BSperformed at 42°C for 30 min using3.5 $vl oligo(dT) I?1X (GIBCO BRI,), and PCR was subsequently prrformed with Program Temp Control System PC‘-700 (ASTEC, Fukuoka, Japan)operatedfor 26 cycles with the following profile: I min for denaturdtion at 04’(Y‘, 2 min for annealing and 2 min for extension at 7X‘ Annealing temperatures were varied a\ \hown WI Table 1. Finally, the reaction mixtur-c ~a\ kept ,tt 72°C for 5 min to achieve complete cxtcns~on. F0r detection of the dig-labeled RT-PCR products, 1‘5

P18 0 1

2.4. mRNA isolation and RT-PCR Total RNA was isolated from PDMT and TPDMT4 using a RNeasy total RNA kit (Quiagen: Chatsworth, CA). Forty mg tissueswere lysed and homogenized under highly denaturing conditions to inactivate RNase and to ensure isolation of intact RNA. The sampleswere adjusted to the appropriate binding conditions and applied to RNeasy spin columns. Total RNA was eluted in 40 ~1 diethylpyrocarbonate-treated water. The concentrations and quality of extracted RNA were estimated by measurementof 260 nm and checked by denaturing aga.rose/formaldehyde gel electrophoresis, respectively, then stored at -80°C until used.The digoxygenin (dig)-labeled RTPCR products from various mRNA molecules were semi-quantitated by the method reported previously

Fig. 2. Analysis of DX.4 fragmentation by electrophorcsis GI TPDMT-4 during pregnancy and after parturition. PI 8, 0 and I indicate days 1X of pregnancy, and the expected parturienr and the following postpartum days, respectively.

H. Kojima

et al. I Cancer

Letters

110 (1996)

117

113-121

gel or sequencing by ligation into TA cloning vector (Invitrogen, San Diego, CA). Blots were scanned with GT-8500 scanner (Seiko-Epson, Nagano, Japan) and arbitrary densitometric units (ADU) of each band were integrated by means of a computer software program, ‘NIH image’ (Mr. Wayne Rasband, NIH, Bethesda, MD). Each ADU was normalized by the glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) signal of the same sample, followed by conversion to relative expression to control. All semi-quantitative experiments were repeated at least three times. 3. Results Both PDMT and TPDMT-4 appeared on days 1315 of pregnancy; the volume of tumors increased until parturition and decreased soon after parturition. Parturition in DDD and GR/A mice occurred on days 19 and 2 1 of pregnancy, respectively. The time course of volume change was similar between TPDMT-4 (Fig. 1) and PDMT (see Goto et al. [l 11). DNA fragmentation, as revealed by electrophoresis, is shown in Fig. 2. In PDMT, the ladder pattern of DNA degradation into nucleosomal oligomers was observed on day 20 of pregnancy, and on the 1st and

Fig. 3. In situ 3’-end labeling of fragmented DNA in PDMT. PDMT was collected in GR/A mice on day 18 of pregnancy (A), and on the parturient (B) and the following days (C) x 400.

p. a

10 8 6 4

nM of dig-dUTP (Boehringer-Mannheim) was added to the PCR reaction mixture. Ten ~1 aliquots of RTPCR products were resolved on 2.5% agarose gel, blotted and fixed to a positively charged nylon membrane (Amersham, Buckinghamshire, UK), and then detected immunologically with a digoxygenin luminescence detection kit (Boehringer-Mannheim). The detected band was confirmed by its size on agarose

2 0

18 Days in pregnancy

0

1

Days after parturition

Fig. 4. Apoptotic index of PDMT and TPDMT on day 18 or 20 of pregnancy, and on the parturient day and on the following days. Five tumors were used in each point. Mean + SE: a, P < 0.01, vs. the days 18 or 20 of pregnancy.

II8

Total

A

150

The apoptotic index showedthe highest vahte on thri parturient day in PDMT and on the day after parturition in TPDMT-4 (Fig. 4). In the presentstudy, expressionof genesthought to be associatedwith apoptosis in reproductive accersory glands were investigated by RT-PCR. The optimal condition for RT-PCR was 300 ng RNA and 26 cycles (Fig. 5). In PDMT, the expressionof transforming growth factor-& (TGF-&), Fas and Pas ligand showed a slight increase on the parturient day. then it decreasedon the day after parturition, showing no drastic change thereafter (Fig. 6). The expression ot mRNA of keratinocyte growth factor (KGFJ mRNA wasincreasedon the day of parturition, and decreased on the 1st postpartum day. The expression of bcl-2 and TNF-ol was decreased on the 1st postpartum day. The expression of TNF receptor showedno sig, nificant change (Fig. 6). In TPDMT-4, the expression of TGF-gI WI> slightly increasedbut TGF-/32 was slightly decreased on the day after parturition; TGF-/3.? showed itc! change (Fig. 6 and Fig. 7). The expressionsof KGF.

RNA (qj) 300

450

600

B

TPlMT-4

PDMT -0

100

200

300

400

500

P20

600 TGF-@l

Total RNA (ng)

TGF-02 Fig. merit

5. Semi-quantitatlvc of TGF-PI)

in PDMT

analysir of CR/A

of PCR mice

products

(.360

bp l’rq-

on day 20 01 pregnant)

by RT-PCR. The PCR procedure wa\ standardized for amount 01 RNA and number of cycles so that the condition\ used wcrc u, Ihe linear range. Dig-labeled PCR products wcrc separated in ?.5’,i Seakem ME agarose. blotted and fixed to a nylon membrane. chcmiluminescence was exposed to the X-ray film. The intensity each band was analyzed by NIH image program and plotted. the standard curve was estimated template RNA and PCR cycles.

in the various

concentration\

and ot and of.

2nd postpartum days when the most intense DNA fragmentation occurred (see Goto et al. 1I I I). Such intense DNA fragmentation in TPDMT-4 was observed on the parturient and the following days (Fig. 2). Apoptotic cells were detected by in situ i/-end cells. In PDMT and labeling method in tumor TPDMT-4, apoptotic cells were found during pregnancy and after parturition (Fig. 3); however, apoptotic cells were encountered on the day of parturition in PDMT and on the 1st postpartum day in TPDMT-4.

TGF-03 Faa ligand Fas TNF-c( TNFR bcl-2 KGF GAPDH

0

P

P18

0

I

H. Kojima

et al. I Cancer

Letters

110 (1996)

119

113-121

+

TGF-fil

+

TGF- j32

+

TGF-P3

+

Fas ligand

+

Fas

+

TNF-a

+

TNFR

--(I

bcl-2

+

KGF

o3 2 1

”

20

Days in pregnancy

0

18

1

Days after parturition

Days in pregnancy

0

Days after parturition

Fig. 7. Relative expression of several mRNAs in PDMT of GR/A mice and TPDMT-4 scanning of intensity was normalized on the basis of GAPDH expression.

TNF-a! and Fas ligand mRNAs were increasedon the parturient day, and decreasedon the day after parturition. The expressionof Fas was increasedon the parturient and the following days. The expressionof bcl2 was decreased on the 1st postpartum day. The expression of TNF receptor showed no significant change (Fig. 6 and Fig. 7). 4. Discussion PDMT appearedon day 15 of pregnancy in GR/A mice and grew rapidly until day 20 of pregnancy but regressedon the parturient day asreported previously [l]-[3,11], while in DDD mice, TPDMT-4 appeared on day 13 of pregnancy and regressedon the expected parturient day after growing until day 19of pregnancy [4]. PDMT and TPDMT4 are hormone-dependent mammary tumors [5,7,9]. In this study, oligonucleosomal DNA fragmentation, an apoptosis indicator, was detected by ethidium bromide gel staining. In PDMT, DNA fragmentation was observed from day 20 of pregnancy, and on the parturient and the following postpartum days as reported previously [ll]. Apoptotic cells in PDMT also determined by in situ 3’-end labeling method increasedon the parturient and

1

of DDD

mice by RT-PCR.

A semi-quantitative

the following days compared with that on day 20 of pregnancy. In TPDMT4, DNA fragmentation was observed from day 18 of pregnancy to the day after parturition, indicating that the initial rapid regressionof TPDMT4 was causedby the apoptotic cell death. Apoptotic cells in TPDMT-4 determinedby in situ 3’-end labeling method increasedon the parturient andthe following days when compared with that on day 18 of pregnancy, indicating the DNA fragmentation. TGF-/3s are known to be involved in apoptotic cell death of epithelial cells in prostate and uterus [ 16,171 and tissueremodeling of mammary gland [18,19]. In PDMT and TPDMT-4, only a slight change in gene expressionof the TGF-/3,, -& and -& was found in the presentstudy, suggestingthat thesefactors may not be involved in the regressionof TPDMT-4 and PDMT. Fas ligand, a cell surface molecule belonging to TNF family, binds to its receptor Fas, thus inducing apoptosis of Fas-bearingcells. In the immune system, Fas and Fas ligand are involved in down-regulation of immunoreaction as well as in T cell-mediated cytotoxicity [20,21]. The present study showed that gene expressionof Fas and Fas ligand was increasedon the parturient day, suggesting that these factors are

involved in the regression of PDMT and TPDMT-4. KGF is thought to be an important paracrine mediator of uterine epithelial cell proliferation in vivo 1221. No significant change in gene expression of TNF-a and TNF receptor was observed, indicating that these factors are not involved in the regression of PDMT and TPDMT-4. The previous reports that increases in the levels of bcl-2 protein causes resistance to the induction of apoptosis (231 accord with the present result that the gene expression of bcl-2 was gradually decreased on the day after parturition. In conclusion, the present findings reveal that apoptosis occurs in the regressing PDMT and TPDMT-4 on day 20 or 18 of pregnancy, and on the parturient and the following days, respectively, showing that regression is associated with the expression of Fas in PDMT and of Fas and TNF-a! in TPDMT-4. Acknowledgements The authors thank Dr. Raphael Guzman of the Department of Molecular and Cell Biology and Cancer Research Laboratory at the Ilniversity of California at Berkeley and Emeritus Professor Noboru Takasugi of Yokohama City University for their valuable advice and critical reading of this manuscript. This work was supported by a Grant-inAid for Scientific Research from the Ministry of Education, Science and Culture of Japan, a Research Grant from Kihara Science Foundation, a Grant for Support of the Promotion of Research at Yokohama City University (T.I.), and Fellowships of the Japan Society for the Promotion of Science for Japanese Junior Scientists (M.E., T.S.). References [ 1 I Miihlbock, 0. ( 196.5) Note on a new inbred mouse strain GK/ A, Eur. J. Cancer. 1, 123.-124. (21 Van Nie, R. and Dux, A. (1971) Biological and morphological characteristics of mammary tumors in GR mice, J. Natl. Cancer Inst., 46, X85-897. [3I Nagasawa, II., Suzuki, M., Yamamuro, Y.. Sensui, N., hIaba. T. and Mori, J. (1989) Suppression of lactation by pregnancydependent mammary tumors in GR/A mice, P.S.E.B.M.. 192, 31-34. 141 Matsurawa, A. and Yamamoto, 1‘. (1974) A transplantable pregnancy-dependent mammary tumor line (TPDMT-4) in strain DDD mice, Gann, 55, 307-315.

1.71 MatsuLawa.

A., Kaneko, I, and Ike&d, Y. ( 1% 4 t i~celcrateti to autonomy Of a pregnancy-dcpcndcnr mou%’ mammary tumor (TPDMT4) by hormones. (‘:LIIC~I- Kc% 43, 22x3 1289. [hl Sluyscr. M. and Van Nie. K. , 1974) Estrogen rcq)ti,~ contc‘:i~ and hormone-responsive growth of mouse mamm;tr\ tnmor-.. Cancer Res.. 34, 325%32S’? 17) Harbcll, J.W.. Papkoff. J.S. and Daniel. C.kV h it!X?) flo! monr requirements of the prcgnancv-dependcn~ mammar\ tumor of’ GR/A mice: an in vitro 51udq. J V,ni <‘,~ncr‘r lost.. 69, 1391 -1402. 1x1 Matsu/awa, A. and Yamamoto. 1. ( 19X.2) t:nhancc:tl a:iii rcvcrscd prowth in viva of ;I (,rcgnancy-dcpend~~~~i n~>u~ mammary tumor (TPDM7‘.4, by s gonadotn!pin-releasir!~ hormone agonist analog, liut .I. (‘anccr (“tin. <)n~ol.. j I 4os- SOT. [u] %kii, S.. Yamdmoto, K., A~hara, Vi.. Sti:uhf. ht. a~:
H. Kojima

et al. i Cancer

in transforming growth factor-fll-inhibited mouse mammary gland, J. Cell Biol., 110, 2209-2219. [ZO] Enari, M., Hug, H. and Nagata, S. (1995) Involvement of ICE-like protease in Fas-mediated apoptosis, Nature, 375, 78-8 1. [21] Nagata, S. and Golstein, P. (1995) The Fas death factor, Science, 267, 1449-1456. 1221 Koji, T., Chedid, M., Rubin, J.S., Slayden, O.D., Csaky, K.G., Aaronson, S.A. and Brenner, R.M. (1994) Progesterone-

Letters

110 (1996)

113-121

121

dependent expression of keratinocyte growth factor mRNA in stromal cells of the primate endometrium: keratinocyte growth factor as a progestomedin, J. Cell Biol., 125, 393401. [23] Liu, Y.-J., Mason. D.Y., Johnson, G.D., Abbot, S., Gregory, C.D., Hardie, D.L., Gordon, J. and MacLennan, I.C.M. (1991) Germinal center cells express bcl-2 protein after activation by signals which prevent their entry into apoptosis, Eur. J. Immunol., 21, 1905-1910.