Retinoic Acid Affects the EGF-R Signaling Pathway during Differentiation Induction of Human Endometrial Adenocarcinoma Cells

Experimental and Molecular Pathology 68, 170–186 (2000) doi:10.1006/exmp.2000.2301, available online at http://www.idealibrary.com on

Retinoic Acid Affects the EGF-R Signaling Pathway during Differentiation Induction of Human Endometrial Adenocarcinoma Cells1

Charleata A. Carter2 and Benjamin L. Shaw Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of Arkansas for Medical Sciences, Little Rock, Arkansas 72205

Received December 26, 1999

We have shown that moderately differentiated endometrial adenocarcinoma (RL95-2) cells differentiate in response to retinoic acid treatment, illustrated by their reorganization of actin filaments and cell enlargement (Carter et al., Anticancer Res. 16, 17–24, 1996). Tyrphostin, an inhibitor of epidermal growth factor receptor (EGF-R)-associated protein tyrosine kinases, caused a dramatic reorganization of actin filaments in RL95-2 cells, similar to retinoic-acid-treated cells (Carter and Bellido, J. Cell. Physiol. 178, 320–332, 1999). We evaluated the possibility that the differentiating effects of retinoids are due to retinoicacid-induced decreases in phosphorylation of EGF-R and changes in downstream effector proteins. Retinoic acid caused a decrease in tyrosine phosphorylation of EGF-R. Retinoic acid treatment induced a dramatic actin filament reorganization and cell enlargement. Treatment with EGF reversed this effect, because cells treated with retinoic acid followed by EGF only possessed disrupted actin aggregates and appeared small, thus resembling medium controls. Retinoic acid induced a relocalization and decrease in the amount of Shc protein, another actin-binding protein which is an adaptor protein for EGF-R signaling. In addition, retinoic acid induced a relocalization of gelsolin from the plasma membrane to the cytoplasm. Retinoic acid decreased cell detachment in detachment assays; one-half as many retinoic-acidtreated cells detached as in controls. These results are consistent with the idea that retinoic acid induces differentiation of RL95-2 cells by interfering with the EGF-R signaling pathway. q 2000 Academic Press

Key Words: retinoic acid; F-actin; EGF-R; gelsolin; Shc; endometrium.

INTRODUCTION

Retinoids are a group of natural and synthetic vitamin A derivatives that regulate cell growth and differentiation (DeLuca, 1991) and are potential targets for chemoprevention and treatment of human cancers (Hong and Itri, 1994). Induction of differentiation has been correlated with the antitumor effect of retinoids (Hong and Itri, 1994). In two recent studies of patients, b-carotene (the provitamin form of retinoic acid) in the diet conferred a significant protection against the development of endometrial carcinoma (Negri et al., 1996; Levi et al., 1993). Endometrial cancer is the most common form of gynecologic cancer in the United States and the fourth most common cancer in women. In 1998, approximately 36,100 women in the United States were diagnosed with endometrial carcinoma, resulting in 6300 deaths (American Cancer Society, 1998). All transretinoic acid combined with tamoxifen is being used in clinical trials to treat advanced breast cancer (Budd et al., 1998) and 13-cis-retinoic acid combined with a-interferon

1

Supported in part by an Intramural Pilot Grant to C.A.C. To whom correspondence should be addressed at Slot 518, Department of Ob/Gyn, University of Arkansas for Medical Sciences, 4301 West Markham Street, Little Rock, AR 72205. Fax: (501) 686-8107. E-mail: [email protected]. 2

170

0014-4800/00 $35.00 Copyright q 2000 by Academic Press All rights of reproduction in any form reserved.

RETINOIDS IN ENDOMETRIAL DIFFERENTIATION

has been used to treat metastatic endometrial cancer (Kudelka et al., 1993). Vitamin A is required for normal differentiation of reproductive epithelia (Underwood, 1984), but the role of retinoic acid in endometrial differentiation is still poorly understood. An organized cytoskeleton and an increase in cell size are associated with differentiation (Rao et al., 1990). Differentiated cultured cells have markedly higher F-actin levels than undifferentiated cells. Retinoic-acid-induced differentiation of HL-60 cells is associated with an increase in the amount of F-actin (Rao et al., 1990). The presence of actin filaments organized into stress fibers is characteristic of cells with a stationary phenotype that exhibit anchorage dependence, reduced mobility, and contact inhibition (Kolega, 1986). Upon transformation, cells lose stress fiber bundles and undergo a concomitant alteration in cell shape, loss of contact inhibition, and enhanced tumor-forming potential (Pienta et al., 1989). Consistent with the differentiating action of retinoids, we have previously shown that retinoic acid caused reversion of moderately differentiated human endometrial adenocarcinoma (RL95-2) cells toward the differentiated, stationary phenotype evidenced by retinoic-acid-induced Factin reorganization into stress fibers (Carter et al., 1996; Carter and Parham, 1997). Additionally, we demonstrated that retinoic acid induced the protein kinase C-alpha (PKCa)3 isoform to localize exclusively to the cytoplasm, indicating inactivation of this PKC isoform (Carter et al., 1998). The epidermal growth factor receptor (EGF-R), a membrane receptor for the polypeptide mitogen EGF, is a 170kDa tyrosine kinase whose activity and expression are regulated by multiple mechanisms (Schlessinger and Ullrich, 1992). Generally, binding of EGF to its receptor leads to increases in receptor autophosphorylation and receptor enzymatic activity, followed by rapid internalization and degradation of the ligand–receptor complex. EGF-Rs colocalize with cortical actin filaments in A431 human epidermal carcinoma cells as evidenced by immunogold labeling and electron microscopy (Wiegant et al., 1986). EGF treatment of A431 cells resulted in an increased colocalization of EGF-R and actin filaments compared to control cells (Rijken et al., 1991). Serum-starved Swiss 3T3 fibroblasts reorganized actin filaments upon addition of growth factors; this effect was blocked when the GTPase rho was inhibited (Ridley and Hall, 1992). Several studies have suggested that the cytoskeleton plays an important role in signal transduction. EGF-R

3

Abbreviations used: EGF-R, epidermal growth factor receptor; PBS, phosphate-buffered saline; TBS, Tris-buffered saline; BSA, bovine serum albumin; PKC, protein kinase C.

171 kinase activation induced phosphorylation of cytoskeletalassociated proteins including spectrin, ezrin, fodrin, and microtubule-associated protein-2 (Akiyama et al., 1986; Bretscher, 1989; Fava and Cohen, 1984). A negative feedback loop involving PKC has been shown to regulate EGFinduced morphological changes. Activated PKC catalzyed phosphorylation of EGF-R (Hunter et al., 1984; Cochet et al., 1984) and resulted in decreased affinity of the receptor for EGF which led to rapid physiological reduction of EGF signaling (Welsch et al., 1991). However, it has been suggested that EGF-induced actin polymerization into stress fibers is required for negative feedback regulation of EGFR tyrosine kinase activity by PKC, because when EGFinduced actin polymerization was inhibited by dihydrocytochalasin B, a marked increase in EGF-induced EGF-R tyrosine kinase activity occurred (Rijken et al., 1998). EGF-R and Src homologous and collagen protein (Shc) are phosphotyrosine substrates and actin-binding proteins (Wiegant et al., 1986; Thomas et al., 1995). Tyrosine phosphorylation of EGF-R causes the formation of a multiprotein complex consisting of a Shc/Grb2 interaction (Schlessinger and Ullrich, 1992). The SH2 domain in Shc binds to activated receptors by recognizing a phosphotyrosine residue within an amino acid sequence (Schlessinger, 1994). Nerve growth factor stimulation of PC12 (rat adrenal pheochromocytoma, neuronal) cells caused Shc to bind to F-actin and translocate to the cytoskeleton, but pretreatment with cytochalasin D abolished Shc translocation to the cytoskeleton. Thus, Shc plays a role in actin reorganization in response to growth factor stimulation (Thomas et al., 1995). Shc is an adaptor protein that couples EGF-R signaling to ras (Aronheim et al., 1994) and cytoplasmic signaling proteins. Members of the ras superfamily are involved in differentiation (ras) and cytoskeletal control (rho and rac) (Aronheim et al., 1994; Downward, 1990). EGF-induced membrane ruffling and lamellipodia formation are regulated by rac in A431 cells, while EGF-induced cortical actin polymerization and cell rounding are controlled by rho (Malliri et al., 1998). Several proteins participate in the regulation of growthfactor-induced actin remodeling, but many details with respect to the downstream signaling of these proteins are unknown. The actin-binding protein gelsolin regulates actin assembly and disassembly (Stossel et al., 1985; Janmey et al., 1985; Kwiatkowski, 1999) in response to changes in the concentrations of calcium and polyphosphoinositides in living cells (Yin, 1988). Gelsolin binds actin monomers, severs actin filaments, and blocks the fast-growing (barbed) ends but also can promote nucleation of polymerization (Yin, 1988). Interestingly, in tissues taken from rabbits, gelsolin

172 protein is higher in the uterus than in any other tissue (Yin et al., 1981). Endometrial cell differentiation is controlled by complex events induced by growth factors and steroid hormones. EGF-R appears to play an essential role in the regulation of endometrial cell function, because EGF-R is present in all major uterine cell types (Lin et al., 1988), endometrial cancer (Bauknecht et al., 1989), and various endometrial cell lines (Lelle et al., 1993). In this study, we examined whether the differentiating action of retinoic acid occurs through interfering with the EGF-R signaling pathway. Our results indicate that the differentiating effects of retinoids on endometrial cells are associated with decreases in tyrosine phosphorylation of EGF-R and changes in selected downstream proteins.

MATERIALS AND METHODS Cell Culture RL95-2 cells were obtained from the American Type Culture Collection (Rockville, MD) and were maintained in the laboratory in medium consisting of a 1:1 mixture of Medium 199 (Gibco BRL, Gaithersburg, MD) and Ham’s F12 (Gibco BRL) supplemented with 1% fetal bovine serum (Hyclone, Logan, UT), 3% bovine calf serum (Hyclone), 1% penicillin–streptomycin (Gibco BRL), 1% glutamine (Gibco BRL), and 0.1% insulin–transferrin–selenium (Collaborative Biomedical Products, Bedford, MA). Treatment with Retinoic Acid and EGF and F-Actin Staining RL95-2 cells were seeded onto 4-well Lab-Tek plastic chamber slides (Miles Scientific, Naperville, IL) at a density of 20,000 cells per well and allowed to attach for 24–48 h. Cells were left untreated (media controls) or treated with selected doses of retinoic acid (Sigma, St. Louis, MO) dissolved in ethanol and diluted in culture medium or treated with corresponding doses of ethanol alone in culture medium (vehicle controls) for 90 min at 378C. All retinoic acid exposures were performed in subdued light. Another set of cells was exposed to EGF purified from mouse salivary glands according to the methods of Savage and Cohen (1972). Specifically, cells were treated with 1 mM all trans or 13cis-retinoic acid or 50 ng/ml EGF for 2 h at 378C prior to F-actin staining. In order to investigate the effects of EGF

CARTER AND SHAW

following retinoic acid treatment, another set of cells was treated with 1 mM all trans or 13-cis-retinoic acid for 1 h followed by EGF (50 ng/ml) for 1 h at 378C. Cells were stained for F-actin according to the methods of Carter et al. (1991). Cells were rinsed with phosphate-buffered saline (PBS), fixed in 3.7% formaldehyde in PBS for 10 min at room temperature, and permeabilized for 5 min with acetone at 2208C. Then cells were incubated with fluorescein phalloidin (Molecular Probes, Eugene, OR) at a 1:20 dilution for 20 min at room temperature, rinsed, and mounted in medium described previously (Carter et al., 1991) consisting of 5% n-propyl gallate and 0.25% diazobicyclooctane in polyvinyl alcohol and observed with a Zeiss confocal laser scanning microscope. Confocal fluorescent images were obtained using an argon laser of 488-nm wavelength and a 520-nm long pass barrier filter. Hard copy images were produced by a Polaroid freeze-frame video printer. Cells were measured using proprietary software running on the Zeiss confocal microscope. The area of 100 cells was measured for each treatment group and the mean and standard error were calculated.

Western Blots For analysis of activated EGF-R, confluent RL95-2 cells in the exponential phase of growth (70–80% maximum density) on 100-mm dishes were exposed to medium (control), 1 mM of 13-cis- or all trans-retinoic acid, or 50 ng/ml EGF. Exposures were for 30 and 90 min at 378C. Another set of cells was also treated with 100 nM 12-O-tetradecanoylphorbol 13-acetate (TPA) (Sigma), a PKC activator, for 20 min. Whole cell lysate protein from these RL95-2 cells was isolated as described below. Cells were rinsed twice with PBS. All subsequent steps were done at 48C using ice-cold buffers. Cells were incubated in RIPA buffer (13 PBS, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) with freshly added inhibitors—1 mg/ml phenylmethylsulfonyl fluoride (Sigma) in isopropanol, aprotinin (Sigma) (30 ml/ml), 10 ml/ml sodium orthovanadate (Sigma)—and then scraped from the 100-mm dishes. Lysates were sheared twice through a 21gauge needle and transferred to a microcentrifuge tube. Each plate was washed with 0.3 ml of RIPA buffer, combined with the first lysate, and sheared twice through a 21-gauge needle. Combined lysates were incubated 30–60 min on ice in phenylmethylsulfonyl fluoride (Sigma). Lysates were microcentrifuged at 15,000g for 20 min at 48C and the supernatant was collected as the total cell lysate. Protein concentration was measured and aliquots of each sample containing 60 mg of protein were fractionated on a 7% polyacrylamide–

RETINOIDS IN ENDOMETRIAL DIFFERENTIATION

SDS gel and transferred electrophoretically to a PVDF Immobolin P transfer membrane (Millipore, Bedford, MA). Membranes were blocked with 10% bovine serum albumin (BSA) (Sigma) in Tris-buffered saline (TBS) containing 0.1% Tween for 2 h at room temperature. Then the membranes were incubated for 1 h with a 1:500 dilution of monoclonal anti-EGF-R antibody that reacts specifically with the activated or tyrosine phosphorylated form of EGFR (Transduction Laboratories, Lexington, KY) in 4% BSA.

173 Following a 1-h rinse in several changes of Tris-buffered saline with Tween 20 (TBS-T), protein bands were detected by exposure for 30 min to a 1:5000 dilution of horseradishperoxidase-conjugated goat anti-mouse IgG secondary antibody (Bio-Rad, Hercules, CA) diluted in TBS-T and 4% BSA. Membranes were rinsed in several changes of TBST for 1 h and blots were developed by using the enhanced chemiluminescence (ECL) detection kit (Amersham, Arlington Heights, IL) and exposed to Kodak X-OMAT AR film for

B

FIG. 1. (A) Analysis of tyrosine phosphorylated EGF-R protein in RL95-2 cells by immunoblotting. Whole cell lysates were prepared from RL95-2 cells, 60 mg of protein/lane was applied to SDS/PAGE, and EGF-R was detected by Western blotting with an antibody that only recognizes the activated form of EGF-R. (B) A Fast Green stain of the membrane is shown to confirm equal protein loading. (C) Samples in each lane and integrated density values by densitometry are as follows: Lane 1, A431 cells—1 control; lane 2, Medium—30,411; lane 3, 1 mM all- trans—13,391; lane 4, 50 ng/ml EGF—8219; lane 5, 1 mM 13-cis-retinoic acid—10,065; lane 6, 100 nM TPA—14,080; lane 7 is a protein ladder and represents 160 kDa.

174 10 s to 60 min as required to achieve satisfactory exposure. Quantitation was performed by laser densitometry. The membrane was subesquently stained with Fast Green. Changes in Shc and gelsolin protein in particulate and cytosolic fractions of control and retinoic-acid-treated RL952 cells were analyzed by Western blot analysis. Confluent 100-mm dishes of RL95-2 cells in the exponential phase of growth (70–80% maximum density) were treated for 90 min with medium alone or 1 mM doses of all trans- or 13cis-retinoic acid. Particulate and cytosolic fractions of cell lysates were isolated as described previously (Carter et al., 1998). Aliquots of each sample containing 20 mg of protein (for Shc detection) and 50 mg of protein (for gelsolin detection) were fractionated on 8% polyacrylamide–SDS gels. Proteins were transferred electrophoretically to PVDF Immobilin P transfer membranes, (Millipore). Membranes were blocked with TBS-T containing 5% milk for 1 h at room temperature. Then, membranes were incubated for 1 h with a 1:1000 dilution of monoclonal IgG anti-Shc antibody (Santa Cruz Biotechnology, CA) or with a 1:500 dilution of monoclonal IgG anti-gelsolin antibody (Transduction Laboratories). Following a 25-min rinse in several changes of TBST, protein bands were detected by exposure for 30 min to a 1:5000 dilution of HRP-conjugated goat anti-mouse IgG secondary antibody (Bio-Rad) diluted in TBS-T and 1% milk. Membranes were then rinsed, processed, and analyzed as described above. Gelsolin and Shc Staining of RL95-2 Cells For gelsolin and Shc staining, cells were seeded onto 4well Lab-Tek plastic chamber slides (Miles Scientific) at a density of 20,000 cells per well and allowed to attach for 24–48 h. Cells were left untreated (medium controls) or treated with 1 mM of 13-cis-or all trans-retinoic acid (Sigma) dissolved in ethanol and diluted in culture medium or treated with corresponding doses of ethanol alone in culture medium (vehicle controls) for 90 min at 378C. Some cells used for Shc staining were exposed to 100 nM TPA, a PKC activator, for 20 min. For gelsolin staining, cells were rinsed twice with PBS, fixed in 3.7% formaldehyde for 10 min, rinsed twice in PBS, and permeabilized for 5 min in 2208C acetone. Following a PBS rinse, cells were incubated in a 1:100 dilution of monoclonal IgG anti-gelsolin antibody (Transduction Laboratories) in PBS containing 1% bovine serum albumin and 0.5% nonfat dry milk (Carnation) for 1 h at 378C. Cells were rinsed twice with PBS and exposed to a 1:256 dilution of fluorescein isothiocyanate (FITC)-labeled sheep anti-mouse IgG (Sigma). Finally, cells were rinsed three times in PBS, mounted, and observed as described

CARTER AND SHAW

above. For Shc staining cells were rinsed twice with PBS, fixed in 3.7% formaldehyde for 10 min, rinsed three times with PBS, and exposed to 0.2% Triton X-100 for 10 min at 48C. Following three rinses with PBS, cells were incubated overnight in a 1:50 dilution of polyclonal anti-Shc antibody (Transduction Laboratories) in 1% BSA. Following two rinses with PBS cells were exposed to a 1:80 dilution of FITC-labeled goat anti-rabbit IgG (Sigma) for 30 min at 378C. Cells were rinsed twice with PBS and mounted and observed as described above. For both gelsolin and Shc stains some cells received no primary antibody, but received only secondary antibody and served as staining controls. Detachment Assays RL95-2 were plated onto 35-mm plastic tissue culture dishes at 100,000 cells per dish and allowed to attach for 7 days. For analysis of the retinoic acid effects on cell attachment, cell detachment assays were performed using the methods of Guo et al. (1995). Cells were exposed to medium alone (controls), 1 mM 13-cis-retinoic acid, l mM all transretinoic acid, or corresponding doses of ethanol alone in culture medium alone (vehicle control) for 24 h at 378C prior to cell counting. In another detachment assay, 24 h prior to cell counting, cells were exposed to medium alone (controls), 0.1% serum, 10 ng/ml EGF, or 50 ng/ml EGF in medium containing 0.1% serum. The assays were performed in duplicate. Cells were rinsed twice with sterile Hank’s balanced salt solution (Gibco) and exposed to 0.5% trypsin/ EDTA (Gibco) for 10 min with constant agitation on a rocker set at 120 rpms. The medium containing the detached cells was collected. Detached cells from each dish were counted using a hemacytometer and the mean was calculated and graphed using Cricket Graph. Statistics were performed using StatMost (DataMost Corporation, Sandy, UT). Statistical analysis was performed using one way analysis of variance (ANOVA).

RESULTS Retinoic Acid Decreases the Amount of Activated EGF-R Protein in Human Endometrial Adenocarcinoma RL95-2 Cells Phosphorylated (activated) EGF-R protein was examined in RL95-2 cells exposed to medium, 1 mM 13-cis-, 1 mM all trans-, or 50 ng/ml EGF for 30 or 90 min. Because

RETINOIDS IN ENDOMETRIAL DIFFERENTIATION

175

FIG. 2. RL95-2 cells stained for F-actin. All treatments were at 378C. A. Media control cells appear small in size and display actin aggregates near the cell periphery (arrowheads). B. Treatment with 50 ng/ml of EGF for 2 h induced cell enlargement and caused some cells to reorganize actin filaments (arrows) and possess a nonlinear membrane with disrupted actin aggregates (arrowheads). C. When cells were treated for 2 h with 1 mM all trans-retinoic acid, actin filaments (arrows) became reorganized throughout the enlarged cells. D. Treatment with 1 mM all trans-retinoic acid for 1 h followed by 1 h of 50 ng/ml EGF caused actin filaments to disappear; cells reverted to a morphology similar to medium control cells. E. When cells were treated for 2 h with 1 mM 13-cis-retinoic acid, actin filaments reorganized (arrows) and cells enlarged. F. Treatment with 1 mM 13-cis-retinoic acid for 1 h followed by 1 h of 50 ng/ml EGF induced the disappearance of actin filaments. Original magnification 12003. Bar 5 50 mm.

activated PKC results in decreased EGF-R signaling (Welsch et al., 1991), cells were exposed to 100 nM TPA for 20 min to activate PKC; this experiment served as a positive control

for decreased EGF-R tyrosine phosphorylation. Medium controls exhibited more activated EGF-R protein than the 30-min-treated cells and TPA-treated cells (Figs. 1A and

176

CARTER AND SHAW

FIG. 3. Mean cell area (mm/cm2) of cells shown in Fig. 2. Note that cells treated with 13-cis- or all trans-retinoic acid enlarged the most while cells treated with EGF also enlarged compared to controls and cells treated with retinoic acid and then EGF. Measurements of cell perimeters were performed using proprietary software running on a Zeiss confocal laser scanning microscope. One hundred cells were measured for each treatment group and the mean and standard error calculated.

1C). The same pattern was seen in 90-min-treated cells (data not shown). A431 human epidermoid carcinoma cell lysates served as a positive control and appeared as a bright area around 170 kDa. Integrated density values are shown in Fig. 1C. These data are representative of three different experiments. Tyrosine phosphorylated proteins in the 170kDa region were decreased in all treated cells compared to medium controls when membranes were stripped and reprobed with an anti-phosphotyrosine antibody (Upstate Biotechnology) (data not shown). Equal protein loading was confirmed by staining the membrane with Fast Green (Fig. 1B). EGF Treatment Reverses the Retinoic-Acid-Induced F-Actin Reorganization and Cell Enlargement in RL95-2 Cells Control RL95-2 cells displayed only disrupted actin near the cell periphery (Fig. 2A). Treatment with 50 ng/ml EGF

for 2 h induced cell enlargement and caused some cells to possess organized actin filaments (Fig. 2B). Most cells possessed a nonlinear membrane that appeared to be ruffling. When cells were treated with 1 mM all trans-retinoic acid for 2 h, actin filaments reorganized throughout the enlarged cells (Fig. 2C). Cells exposed to corresponding doses of ethanol (vehicle control) displayed disrupted actin aggregates only near the cell periphery, similar to medium controls (data not shown). Treatment of RL95-2 cells with 1 mM all trans-retinoic acid for 1 h, followed by a 1-h treatment with 50 ng/ml EGF, caused a reversal of the effects induced by all trans-retinoic acid alone; cells were small similar to medium controls and possessed only disrupted actin aggregates near the cell periphery (Fig. 2D). Treatment with 1 mM 13-cis-retinoic acid for 2 h induced cell enlargement and the reorganization of actin filament into stress fibers traversing the cells (Fig. 2E). This effect was reversed when cells were treated with 1 mM 13-cis-retinoic acid for 1 h,

RETINOIDS IN ENDOMETRIAL DIFFERENTIATION

177

FIG. 4. RL95-2 cells stained for Shc. A. Medium control cells displayed Shc staining as punctate bright spots at the cell periphery (arrows). Perinuclear staining and a less obvious scattered cytoplasmic staining were also present. B and C. Cells treated with 1 mM all trans- (B) or 13cis-retinoic acid (C) for 90 min displayed decreased punctate bright spots at the cell periphery (arrows), while a perinuclear staining pattern was obvious. D. Cells treated for 20 min with TPA (100 nM) exhibited some punctate peripheral bright spots (arrows), a prominent perinuclear staining pattern, and obvious cytoplasmic staining. Original magnification 12003. Bar 5 50 mm.

followed by 50 ng/ml of EGF for 1 h (Fig. 2F). Figure 3 shows the cell area in control, retinoic-acid-, and EGFtreated cells. Note that cells enlarge upon treatment with retinoic acid or EGF alone compared to controls. Cells treated with retinoic acid were the largest. Treatment with retinoic acid followed by EGF resulted in a smaller cell size similar to medium controls. Retinoic Acid Reduces the Amount of Shc Protein in RL95-2 Cells Because Shc is an adaptor protein for EGF-R signaling as well as an actin-binding protein that is involved in actin filament reorganization in response to growth factor stimulation (Thomas et al., 1995), we investigated the effects of

retinoic acid on the amount and localization of Shc (Figs. 4 and 5). These data are representative of three different experiments. Medium control cells (Fig. 4A) and ethanol (vehicle) control cells (data not shown) exhibited Shc staining largely as bright punctate spots at the cell surface; some perinuclear staining was also apparent with a less obvious cytoplasmic staining. Peripheral bright spots were reduced upon 90-min treatment with 1 mM all trans- (Fig. 4B) or 1 mM 13-cis-retinoic acid (Fig. 4C), but perinuclear staining was predominant with some scattered cytoplasmic staining. Peripheral bright spots returned upon TPA treatment, but were less prominent than in control cells. TPA-treated cells possessed perinuclear staining and scattered cytoplasmic staining (Fig. 4D). Staining controls for Shc (receiving secondary antibody only) were negative (data not shown).

178 Consistent with the immunostaining, Western blot analysis of subcellular fractions showed that levels of Shc were decreased in retinoic-acid-treated cells compared to controls. Shc codes for three overlapping proteins of 66, 52, and 46 kDa (Pelicci et al., 1992). The 66-kDa isoform is the least common isoform; the 52- and 46-kDa isoforms appeared in RL95-2 cells. Medium control RL95-2 cells possessed the 52-kDa isoform and the 46-kDa isoform of Shc in the cytosolic fraction, while the 52-kDa isoform was prominent in the particulate fraction (Fig. 5A). Lysates from cells that were treated with 1 mM all trans-retinoic acid displayed little Shc protein except for a band of the 46-kDa isoform in the cytosolic fraction. Treatment with 13-cis-retinoic acid also led to a reduction in Shc protein, while the 46-kDa isoform band was prominent in the cytosolic fraction and the 52-kDa isoform was prominent in the particulate fraction. TPA induced less protein to be present in the cytosolic fraction while restoring the protein amount and pattern in the particulate fraction to a pattern similar to the particulate fraction of medium controls. Integrated density values for the 52- and 46-kDa Shc isoforms are shown in Figs. 5B and 5C, respectively.

Retinoic Acid Induces Gelsolin to Relocalize from the Plasma Membrane to the Cytoplasm Because EGF treatment of fibroblasts caused membranous gelsolin to decrease concurrent with a submembranous actin filament reorganization (Chen et al., 1996), we evaluated gelsolin localization in RL95-2 cells that had been treated with retinoids for 90 min (Figs. 6, 7A, and 7B). These data are representative of three different experiments. Medium control (Fig. 6A) and ethanol-treated (vehicle control) (Fig. 6B) RL95-2 cells display gelsolin staining mostly localized to the plasma membrane, although some diffuse cytoplasmic staining is present. Treatment with 1 mM 13-cis- (Fig. 6C) or all trans- (Fig. 6D) retinoic acid caused a decrease in plasma membrane staining along with a decrease in diffuse cytoplasmic staining. Cells that received no primary antibody but received secondary antibody only (staining controls) were negative (data not shown). Staining results were confirmed by Western blot analysis (Fig. 7A). In medium control RL95-2 cells, gelsolin was localized largely to the cytosolic fraction with a lesser apparent band in the particulate fraction. When RL95-2 cells were treated with all transretinoic acid there was a decrease in the amount of cytosolic and particulate gelsolin protein compared to controls. Treatment with 13-cis-retinoic acid caused a prominent decrease in gelsolin in the particulate fraction, but an increase in the

CARTER AND SHAW

FIG. 5. A. Analysis of the subcellular distribution of Shc by Western blot analysis in medium control and retinoic-acid-treated RL95-2 cells. Shc protein is decreased by retinoic acid treatment in protein from RL95-20 cells (2 mg/lane). Equal loading was confirmed by performing a Fast Green stain on the membrane after ECL detection (data not shown). The 52-kDa isoform is found mostly in the particulate fraction while the 46-kDa isoform is found mainly in the cytosolic fraction. B shows the integrated density values for Shc (52 kDa) and C shows the integrated density values for Shc (46-kDa isoform).

amount of gelsolin in the cytosolic fraction. We also evaluated the effects of TPA treatment on gelsolin and found that TPA causes a large decrease in the cytosolic fraction and an increase in gelsolin protein in the particulate fraction. Figure 7B shows the integrated density values of gelsolin in control and treated RL95-2 cells.

Retinoic Acid Reduces the Number of Cells That Detach from a Plastic Substrate In order to investigate cell function, cell detachment assays of control, retinoic-acid-, and EGF-treated cells were performed using the methods of Guo et al. (1995). Approximately twice as many cells detached in controls as in retinoic-acid-treated cells (Fig. 8), demonstrating that retinoic acid promotes cell attachment of RL95-2 cells. Statistical analysis using ANOVA showed that there was no significant difference between the ethanol and medium control cells. There was a statistically significant difference in the number of detached cells between medium and ethanol controls compared to both types of retinoic acid (P , 0.05). EGF treatment decreased cell detachment by around 40% compared to controls (Fig. 9). There was a statistically significant difference in the number of detached cells between medium

179

RETINOIDS IN ENDOMETRIAL DIFFERENTIATION

FIG. 5—Continued

180

CARTER AND SHAW

FIG. 6. RL95-2 cells stained for gelsolin. Medium (control) (A) and ethanol (vehicle control) (B) cells displayed gelsolin staining at the plasma membrane (arrowheads) with some scattered cytoplasmic staining. Cells treated with 1 mM 13-cis- (C) or all trans-(D) retinoic acid displayed decreased plasma membrane staining. Original magnification 12003. Bar 5 50 mm.

and 0.1% serum controls compared to both doses of EGF (P , 0.05). These data demonstrate that retinoic acid or EGF treatment induced a decrease in cell detachment of RL95-2 cells from their substrate compared to controls. Figure 10 shows a model of the effects of retinoic acid on selected proteins involved in cell differentiation concurrent with actin organization.

DISCUSSION The state of differentiation is a major determinant of prognosis in endometrial cancer (Tornus and Elvio, 1993). Studies provide data indicating that EGF-R does not directly correlate with endometrial cancer, making the role of EGFR in endometrial cancer unclear. One study shows that EGFR was present in 12 normal uteri and in 27 uteri with adenocarcinoma (Zarcone et al., 1995). Another study showed

that EGF-R immunoreactivities were observed in 58.3% of normal endometrial specimens, 100% of endometrial hyperplasia specimens, and 67.5% of endometrial carcinoma specimens (Niikura et al., 1996). However, it was reported that there was a progressive decrease in EGF receptors in cancers of increasing grade (Reynolds et al., 1990), suggesting that EGF-R may be protective in endometrial cancer. In this study, a survey of 37 samples of patients in Michigan revealed that EGF-Rs decreased 34% in grade 1 and 2 adenocarcinomas, 90% in grade 3 adenocarcinomas, and 72% in sarcomas over normal, control tissues (Reynolds et al., 1990). It has been suggested that the EGF-R gene is a target of retinoid action (Thompson and Rosner, 1989). Retinoic acid increased transcription of the EGF-R gene in normal rat kidney fibroblasts, but synergized with EGF and TGF-b to stimulate anchorage-independent growth (Thompson and Rosner, 1989). Retinoic acid also induced an increase in EGF-R in LLC-PK1 kidney cells (Mantzuoris et al., 1993).

RETINOIDS IN ENDOMETRIAL DIFFERENTIATION

181

FIG. 7. A. Analysis of the subcellular distribution of gelsolin by Western blot analysis of medium control and retinoic-acid-treated RL95-2 cells. Gelsolin protein in the particulate fraction is decreased by retinoic acid treatment in protein from RL95-2 cells (50 mg/lane). Equal loading was confirmed by Fast Green staining of the membrane after ECL detection (data not shown). Densitometric quantitation is shown in B.

PKC regulated EGF binding in kidney cells; when PKC was increased, decreases in EGF-R binding were seen and vice versa. However, in human trophoblast cells, retinoic acid decreased EGF-R tyrosine phosphorylation, the amount of EGF-R protein, and EGF-R mRNA levels (Roulier et al., 1994). We previously showed that retinoic acid decreased tyrosine phosphorylation of proteins in RL95-2 cells and that tyrphostin, a potent inhibitor of EGF-R tyrosine kinases,

induced a reorganization of actin filaments similar to the effects of retinoic acid treatment (Carter and Bellido, 1999). We hypothesized that retinoic acid decreased phosphorylation of EGF-R in RL95-2 cells causing changes in proteins in the EGF-R signaling pathway contributing to reorganization of actin filaments, causing a decrease in cell detachment. We examined the amount of EGF-R protein in three adenocarcinoma cell lines and found that RL95-2 cells contained more EGF-R protein than the KLE and Ishikawa cells

182

CARTER AND SHAW

FIG. 8. Mean number of detached cells in medium (control), ethanol- (vehicle control), 1 mM 13-cis–and 1 mM all trans-retinoic-acid-treated RL95-2 cells. This experiment was performed in duplicate and is consistent with data from two additional experiments. Approximately twice as many cells detached in controls as in retinoic-acid-treated cells.

FIG. 9. Mean number of detached cells in medium control, 0.1% serum-, 10 ng/ml EGF-, and 50 ng/ml EGF-treated RL95-2 cells. This experiment was performed in duplicate and is consistent with data from two separate experiments. Approximately 40% more cells detached in controls as in EGF-treated cells.

RETINOIDS IN ENDOMETRIAL DIFFERENTIATION

183

FIG. 10. Schematic representation of the proposed mechanism of retinoid-induced differentiation in RL95-2 cells. In untreated cells, EGF-R is tyrosine phosphorylated, Shc and Grb2 form part of a multiprotein complex near the plasma membrane, gelsolin is located near the plasma membrane, and actin filaments are absent. Upon treatment with retinoic acid, tyrosine phosphorylation of EGF-R is reduced, Shc and gelsolin move away from the plasma membrane to the cytoplasm, and actin filaments become reorganized throughout the cells.

(data not shown). Similarly, an earlier study showed that RL95-2 cells had a high number of binding sites for EGF (3.8 pg/mg DNA), while KLE cells had a low number of binding sites (0.3 pg/mg DNA) (Lelle et al., 1993). Because we have shown previously that RL95-2 cells are more responsive to the differentiating effects on retinoic acid treatment than KLE or Ishikawa cells (Carter and Parham, 1997), we investigated the retinoic acid effects of the EGF-R signaling pathway in RL95-2 cells. Retinoic acid caused a decrease in EGF-R tyrosine phosphorylation, a decrease in Shc protein, and a relocalization of Shc and gelsolin away from the plasma membrane. Additionally, in an assay designed to evaluate cell function, both retinoic acid and EGF decreased cell detachment of RL95-2 cells, with retinoic acid being more effective. Actin filaments organized into stress fibers are a marker for differentiation (Lehtonen et al., 1983; Rao et al., 1990) and are also associated with a stationary phenotype (Kolega, 1986; Pienta et al., 1989). F-actin increased 86–96% following differentiation of human myeloid cells (Meyer and Howard, 1983). Retinoic acid (1 mM for 5 days) caused reorganization of F-actin in melanoma cells, promoted cell adhesion, and decreased invasion (Helige et al., 1993). Retinoids also enhanced adhesion of F9 cells (Ross et al., 1994). Our results are consistent with the above studies because retinoic acid treatment of RL95-2 cells caused differentiation and reversion to the stationary phenotype evidenced by a dramatic reorganization of actin filaments and cell enlargement, concomitant with a significant decrease in cell detachment.

EGF treatment caused a heterogenous actin filament reorganization, homogenous cell enlargement, and a significant decrease in cell detachment. The EGF-induced decrease in cell detachment was less pronounced than the retinoid effects, but correlates well with the lesser actin filament reorganization induced by EGF treatment. EGF decreased tyrosine phosphorylation of EGF-R in RL95-2 cells after 30 min of treatment. This is not surprising since EGF binding itself induces an internalization and subsequent down-modulation of the receptor (Thompson and Rosner, 1989). EGF induced rapid remodeling of the actin microfilament system in a variety of cells (Rijken et al., 1991; Peppelenbosch et al., 1993; Welsch et al., 1991). Newly assembled actin filaments localize to the tyrosine-phosphorylated EGF-R in the plasma membrane of A431 cells; actin polymerization is not observed at the internalized tyrosine-phosphorylated EGF-R (Rijken et al., 1995). In RL95-2 cells, EGF reversed the retinoic-acid-induced differentiation, because treatment of RL95-2 cells with retinoic acid, followed by EGF treatment, caused cells to display disrupted actin aggregates similar to untreated cells. We propose that the differences in the effect of EGF treatment on control or retinoic-acid-treated cells are at least partly due to the fact that EGF is acting on an activated receptor in control cells and an inactivated receptor in retinoic-acid-treated cells. Shc is an actin-binding protein that plays a role in actin reorganization (Thomas et al., 1995) and is necessary for growth-factor-induced differentiation of PC12 cells (Rozakis-Adcock et al., 1992). NIH 3T3 mouse fibroblasts that

184 constitutively overexpress Shc acquired a transformed phenotype in culture and formed tumors in nude mice (Pelicci et al., 1992). Retinoic acid treatment of RL95-2 cells caused a decrease in the amount of Shc protein and a decrease in peripheral punctate staining. Perinuclear staining occurred in all cells, but was more prevalent in retinoic-acid-treated cells. This is consistent with an earlier study which showed that upon tyrosine kinase activation of EGF-R Shc was associated with the cytosolic surface of the plasma membrane, while in unstimulated cells Shc localized to the perinuclear area and was associated mostly with the rough endoplasmic reticulum (Lotti et al., 1996). The results of several investigations support gelsolin as a tumor suppressor gene, because down-regulation of gelsolin is frequently found in several types of transformed cells and tumors. It has been suggested that gelsolin is a tumor suppressor gene in in vivo studies of bladder cancer, breast cancer, and non-small-cell lung cancer (Tanaka et al., 1995; Asch et al., 1996; Dosaka-Akita et al., 1998). Gelsolin protein and RNA were absent or markedly reduced in human breast cancer cell lines or in human, mouse, and rat mammary tumors relative to control tissues or cells (Asch et al., 1996). Additionally, in most breast cancer cells, reduced gelsolin protein expression was associated with decreased actin filaments (Asch et al., 1996). Gelsolin is also downregulated in hyperplastic and neoplastic prostate lesions (Lee et al., 1999). Radicicol-induced reversion of transformed fibroblasts to flat cells containing organized actin filaments occurred concomitantly with an increase in gelsolin (Kwon et al., 1997). Gelsolin increased during differentiation of PC-13 cells (Dieffenbach et al., 1989) and myeloid cells (Kwiatkowski et al., 1988). In this study, we show that retinoic acid treatment of RL95-2 cells caused a relocalization of gelsolin away from the plasma membrane to the cytoplasm, concurrent with reorganization of actin filaments throughout the cells. The overall amount of gelsolin protein in RL95-2 cells does not dramatically change upon retinoic acid treatment, although treatment with 13-cis-retinoic acid reduced the amount of gelsolin in the particulate fraction while causing an increase in the cytosolic fraction. A dramatic reduction in gelsolin protein in the cytosolic fraction is observed when cells are treated with TPA, while a slight increase was obvious in the particulate fraction. These changes in gelsolin amount and localization are consistent with disrupted actin filaments, because we have previously shown that when RL95-2 cells are treated with TPA, disrupted actin aggregates occur near the cell periphery (Carter et al., 1998). Taken together, these data lead us to suggest that retinoic acid in the diet may promote differentiation of malignant

CARTER AND SHAW

endometrial cells containing elevated EGF-R protein. It is possible that cells expressing higher levels of EGF-R protein are more responsive to differentiating agents; this could help explain how higher levels of EGF-Rs may be protective in endometrial cancer. Further investigation of the proteins involved in the regulation of growth-factor-induced actin remodeling should lead to a greater understanding of the mechanism by which monomeric actin is converted into filamentous actin and lead to a greater understanding of the EGF-R and gelsolin association in endometrial biology.

ACKNOWLEDGMENTS

We thank Dr. Richard Kurten for helpful comments and for the gift of EGF and Dr. Teresita Bellido for critical reading of the manuscript. We are grateful to Ms. Donna Montague for assistance with statistical analysis. This work was supported in part by an Intramural Pilot Grant to C.A.C.

REFERENCES

Akiyama, T., Kadowaki, T., Nishida, E., Kadooka, T., Ogawara, H., Fukama, Y., Sakai, H., Takaku, F., and Kasuga, M. (1986). Substrate specificities of tyrosine-specific protein kinases toward cytoskeletal proteins in vitro. J. Biol. Chem. 261, 14797–14803. American Cancer Society. (1998). Cancer Facts and Figures. Atlanta, GA. Aronheim, A., Engelberg, D., Li, N., Al-Alawi, N., Schlessinger, J., and Karin, M. (1994). Membrane targeting of the nucleotide exchange factor SOS is sufficient for activating the Ras signaling pathway. Cell 78, 949–961. Asch, H. L., Head, K., Dong, Y., Natoli, F., Winston, J. S., Connolly, J. L., and Asch, B. B. (1996). Widespread loss of gelsolin in breast cancers of humans, mice, and rats. Cancer Res. 56, 4841–4845. Bauknecht, T., Kohler, M., Janz, I., and Pfleiderer, A. (1989). The occurrence of epidermal growth factor receptors and the characterization of EGF-like factors in human ovarian, endometrial, cervical and breast cancer. J. Cancer Res. Clin. Oncol. 115, 193–199. Bretscher A. (1989). Rapid phosphorylation and reorganization of ezrin and spectrin accompany morphological changes induced in A431 cells by epidermal growth factor. J. Cell Biol. 108, 921–930. Budd, G. T., Adamson, P. C., Gupta, M., Homayoun, P., Sandstrom, S. K., Murphy, R. F., McLain, D., Tuason, L., Peereboom, D., and Ganapathi, R. D. (1998). Phase I/II trial of all trans-retinoic acid and tamoxifen in patients with advanced breast cancer. Clin. Cancer Res. 4, 635–642. Carter, C. A., Rinehart, C. A., Bagnell, C. R., and Kaufman, D. G. (1991). Fluorescent laser scanning microscopy of F-actin disruption

RETINOIDS IN ENDOMETRIAL DIFFERENTIATION

in human endometrial stromal cells expressing the SV40 large T antigen and the EJ ras oncogene. Pathobiology 59, 36–45. Carter, C. A., Pogribny, M., Davidson, A., Jackson, C. D., McGarrity, L. J., and Morris, S. M. (1996). Effects of retinoic acid on cell differentiation and reversion toward normal in human endometrial adenocarcinoma (RL95-2) cells. Anticancer Res. 16, 17–24. Carter, C. A., and Parham, G. P. (1997). State of differentiation affects the response of endometrial adenocarcinoma cells to retinoic acid. Anticancer Res. 17, 1973–1984. Carter, C. A., Parham, G. P., and Chambers, T. C. (1998). Cytoskeletal reorganization induced by retinoic acid treatment of human endometrial adenocarcinoma (RL95-2) cells is correlated with alterations in protein kinase C-a. Pathobiology 66, 284–292. Carter, C. A., and Bellido, T. (1999). Decrease in protein tyrosine phosphorylation is associated with F-actin reorganization by retinoic acid in human endometrial adenocarcinoma (RL95-2) cells. J. Cellular Physiol. 178, 320–332. Chen, P., Murphy-Ullrich, J. E., and Wells, A. (1996). A role for gelsolin in actuating epidermal growth factor receptor-mediated cell motility. J. Cell Biol. 134, 689–698. Cochet, C., Gill, G. N., Meisenhelder, J., Cooper, J. A., and Hunter, T. (1984). C-kinase phosphorylates the epidermal growth factor receptor and reduces its epidermal growth factor-stimulated tyrosine protein kinase activity. J. Biol. Chem. 259, 2553–2558. DeLuca, L. M. (1991). Retinoids and their receptors in differentiation, embryogenesis and neoplasia. FASEB J. 5, 2924–2933. Dieffenbach, C. W., SenGupta, D., Krause, D., Sawzak, D., and Silverman, R. H. 1989. Cloning of murine gelsolin and its regulation during differentiation of embryonal carcinoma cells. J. Biol. Chem. 264, 13281–13288. Dosaka-Akita, H., Hommura, F., Fujita, H., Kinoshita, I., Nishi, M., Morikawa, T., Katoh, H., Kawakami, Y., and Kuzumaki, N. (1998). Frequent loss of gelsolin expression in non-small cell lung cancers of heavy smokers. Cancer Res. 58, 322–327. Downward, J. (1990). The ras superfamily of small GTP-binding proteins. Trends Biochem. Sci. 15, 469–472. Fava, R. A., and Cohen, S. (1984). Isolation of a calcium-dependent 35-kilodalton substrate for the epidermal growth factor receptor kinase from A431 cells. J. Biol. Chem. 265, 16725–16728. Guo, W. X., Gill, P. S., and Antakly, T. (1995). Inhibition of AIDSKaposi’s sarcoma cell proliferation following retinoic acid receptor activation. Cancer Res. 55, 823–829. Helige, C., Smolle, J., Zellnig, G., Hartmann, E., Fink-Puches, R., Kerl, H., and Tritthart, H. A. (1993). Inhibition of K1735-M2 melanoma cell invasion in vitro by retinoic acid. Clin. Exp. Metast. 11, 409–418. Hong, W. K., and Itri, L. M. (1994). Retinoids and human cancer. In The Retinoids: Biology, Chemistry, and Medicine. (Sporn, M. B., Roberts, A. B., and Goodman, D. S., Eds.), 2nd ed., Raven Press, New York. Hunter, T., Ling, N., and Cooper, J. A. (1984). Protein kinase C phosphorylation of the EGF receptor at a threonine residue close to the cytoplaslic face of the plasma membrane. Nature (Lond.) 311, 480–483. Janmey, P. A., Chaponnier, C., Lind, S. E., Zaner, K. S., Stossel, T. P., and Yin, H. L. (1985). Interactions of gelsolin and gelsolin

185 actin complexes with actin. Effects of calcium on actin nucleation, filament severing and end blocking. Biochemistry 24, 3714–3723. Kolega, J. (1986). Effects of mechanical tension on protrusive activity and microfilament and intermediate filament organization in an epidermal epithelium moving in culture. J. Cell Biol. 102, 1400–1411. Kore, M., Haussler, C. A., and Trookman, N. S. (1987). Divergent effects of epidermal growth factor and transforming growth factors on a human endometrial carcinoma cell line. Cancer Res. 47, 4909– 4914. Kudelka, A. P., Freedman, R. S., Edwards, C. L., Lippman, S. M., Tornos, C. S., Krakoff, I. H., and Kavanagh, J. J. (1993). Metastatic adenocarcinoma of the endometrium treated with 13-cis-retinoic acid plus interferon-a. Anti-Cancer Drugs 4, 335–337. Kwiatkowski, D. J., Mehl, R., and Yin, H. L. (1988). Genomic organization and biosynthesis of secreted and cytoplasmic forms of gelsolin. J. Cell Biol. 106, 375–384. Kwiatkowski, D. J. (1999). Functions of gelsolin: Motility, signaling, apoptosis, cancer. Curr. Opin. Cell Biol. 11, 103–108. Kwon, H. J., Yoshida, M., Nagaoka, R., Obinata, T., Beppu, T., and Horinouchi, S. (1997). Suppression of morphological transformation by radicicol is accompanied by enhanced gelsolin expression. Oncogene 15, 2625–2631. Lee, H. K., Driscoll, D., Asch, H., Asch, B., and Zhang, P. J. (1999). Downregulated gelsolin expression in hyperplastic and neoplastic lesions of the prostate. Prostate 40, 14–19. Lehtonen, E., Lehto, V-P., Badley, R. A., and Virtanen I. (1983). Formation of vinculin plaques precedes other cytoskeletal changes during retinoic acid-induced teratocarcinoma cell differentiation. Exp. Cell Res. 144, 191–197. Lelle, R. J., Talavera, F., Gretz, H., Roberts, J. A., and Menon, K. M. (1993). Epidermal growth factor receptor expression in three different human endometrial cancer cell lines. Cancer 72, 519–525. Levi, F., Franceschi, S., Negri, E., and La Vecchia, C. (1993). Dietary factors and the risk of endometrial cancer. Cancer 71, 3575–3581. Lin, T. H., Mukku, V. R., Verner, G., Kirkland, J. L., and Stancel, G. M. (1988). Autoradiographic localization of epidermal growth factor receptors to all major uterine cell types. Biol. Reprod. 38, 403–411. Lotti, L. V., Lanfrancone, L., Migliaccio, E., Zompetta, C., Pelicci, G., Salcini, A. E., Falini, B., Pelicci, P. G., and Torrisi, M. R. (1996). Shc proteins are localized on endoplasmic reticulum membranes and are redistributed after tyrosine kinase receptor activation. Mol. Cell Biol. 16, 1946–1954. Malliri, A., Symons, M., Hennigen, R. F., Hurlstone, A. F., Lamb, R. F., Wheeler, T., and Ozanne, B. W. (1998). The transcription factor AP-1 is required for EGF-induced activation of rho-like GTPases, cytoskeletal rearrangements, motility, and in vitro invasion of A431 cells. J. Cell Biol. 143, 1087–1099. Mantzuoris, N. M., Kaplan, J., Clark, G., and Hise, M. K. (1993). Regulation of epidermal growth factor receptor expression and growth by protein kinase C and retinoic acid in LLC-PK1 cells. Am. J. Kidney Dis. 22, 858–864. Meyer, W. H., and Howard, T. H. (1983). Changes in actin content during induced myeloid maturation of human promyelocytes. Blood 62, 308–314. Negri, E., Vacchia, C. L., Franceschi, S., Levi, F., and Parazzini, F. (1996). Intake of selected micronutrients and the risk of endometrial carcinoma. Cancer 77, 917–923.

186 Niikura, H., Sasano, H., Kaga, K., Sato, S., and Yajima, A. (1996). Expression of epidermal growth factor family proteins and epidermal growth factor receptor in human endometrium. Hum. Pathol. 27, 282–290. Pelicci, G., Lanfrancone, L., Grignani, F., McGlade, J., Cavallo, F., Forni, G., Nicoletti, I., Grihnani, F., Pawson, T., and Pelicci, P. G. (1992). A novel transforming protein (SHC) with an SH2 domain implicated in mitogenic signal transduction. Cell 70, 93–104. Peppelenbosch, M. P., Tertoolen, L. G. J., Hage, W. J., and de Laat, S. W. (1993). Epidermal growth factor-induced actin remodeling is regulated by 5-lipoxygenase and cyclooxygenase products. Cell 74, 565–575. Pienta, K. J., Partin, A. W., and Coffey, D. S. (1989). Cancer as a disease of DNA organization and dynamic cell structure. Cancer Res. 49, 2525–2532. Rao, J. Y., Hurst, R. E., Bales, W. D., Jones, P. L., Bass, R. A., Archer, L. T., Bell, P. B., and Hemstreet G. P. (1990). Cellular F-actin levels as a marker for cellular transformation: Relationship to cell division and differentiation. Cancer Res. 50, 215–2220. Reynolds, R. K., Talavera, F., Roberts, J. A., Hopkins, M. P., and Menon, K. M. J. (1990). Characterization of epidermal growth factor receptor in normal and neoplastic endometrium. Cancer 66, 1967– 1974. Ridley, A. J., and Hall, A. (1992). The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 70, 389–399. Rijken, P. J., Hage, W. J., Van Bergen En Henegouwen, P. M. P., Verkleij, A. J., and Boonstra, J. (1991). Epidermal growth factor induces rapid reorganization of the actin microfilament system in human A431 cells. J. Cell Sci. 100, 491–499. Rijken, P. J., Post, S. M., Hage, W. J., Van Bergen En Henegouwen, P. M. P., Verkleij, A. J., and Boonstra, J. (1995). Actin polymerization localizes to the activated epidermal growth factor receptor in the plasma membrane, independent of the cytosolic free calcium transient. Exp. Cell Res. 218, 223–232. Rijken, P. J., van Hal, G. J., van der Heyden, M. A. G., Verkleij, A. J., and Boonstra, J. (1998). Actin polymerization is required for negative feedback regulation of epidermal growth factor-induced signal transduction. Exp. Cell Res. 243, 254–262. Ross, S. A., Ahrens, R. A., and DeLuca, L. M. (1994). Retinoic acid enhances adhesiveness, laminin and integrin b1 synthesis and retinoic acid receptor expression in F9 teratocarcinoma cells. J. Cell Physiol. 159, 263–273. Roulier, S., Rochette-Egly, C., Rebut-Bonneton, C., Porquet, D., and Evain-Brion, D. (1994). Nuclear retinoic acid receptor characterization in cultured human trophoblast cells: Effect of retinoic acid on epidermal growth factor receptor expression. Mol. Cell Endocrinol. 105, 165–173. Rozakis-Adcock, M., McGlade, J., Mbamalu, G., Pelicci, G., Daly, R.,

CARTER AND SHAW

Li,W., Batzer, A., Thomas, S., Brugge, J., Pelicci, P. G., Schlessinger, J., and Pawson, T. (1992). Association of the Shc and Grb2/Sem5 SH2-containing proteins is implicated in activation of the Ras pathway by tyrosine kinases. Nature 360, 689–692. Savage, C. R., and Cohen, S. (1972). Epidermal growth factor and a new derivative. Rapid isolation procedures and biological and chemical characterization. J. Biol. Chem. 247, 7609–7611. Schlessinger, J., and Ullrich, A. (1992). Growth factor signalling by receptor tyrosine kinases. Neuron 9, 383–391. Schlessinger, J. (1994). SH2/SH3 signaling proteins. Curr. Opin. Genet. Dev. 4, 25–30. Stossel, T. P., Chaponnier, C., Ezell, R. M., Hartwig, J. H., Janmey, P. A., Kwiatkowski, D. J., Lind, S. E., Smith, D. B., Southwick, F. S., Yin, H. L., and Zaner, K. S. (1985). Nonmuscle actin-binding proteins. Annu. Rev. Cell Biol. 1, 353–402. Tanaka, M., Mu¨llauer, L., Ogiso, Y., Fujita, H., Moriya, S., Furuuchi, K., Harabayash, T., Shinohara, N., Koyanagi, T., and Kuzumaki, N. (1995). Gelsolin: A candidate for suppressor of human bladder cancer. Cancer Res. 55, 3228–3232. Thomas, D., Patterson, S. D., and Bradshaw, R. A. (1995). Src homologous and collagen (Shc) protein binds to F-actin and translocates to the cytoskeleton upon nerve growth factor stimulation in PC12 cells. J. Biol Chem. 270, 28924–28931. Thompson, K. L., and Rosner, M. R. (1989). Regulation of epidermal growth factor receptor gene expression by retinoic acid and epidermal growth factor. J. Biol. Chem. 263, 3230–3234. Tornus, C., and Elvio, S. (1993). Pathology and grading of endometrial adenocarcinomas. Clin. Consult. Obstet. Gynecol. 5, 87–95. Underwood, B. A. (1984). Vitamin A in animal and human nutrition. In The Retinoids (Sporn, M. B., Roberts, A. B., and Goodman, D. S., Eds.), vol. 1. Academic Press, London. Welsch, J. B., Gill, G. N., Rosenfeld, M. G., and Wells, A. (1991). A negative feedback loop attenuates EGF-induced morphological changes. J. Cell Biol. 114, 533–543. Wiegant, F. A. C., Blok, F. J., Defize, L. H. K., Linnemens, W. A. M., Verkley, A. J., and Boonstra, J. (1986). Epidermal growth factor receptors associated to cytoskeletal elements of epidermoid carcinoma (A431) cells. J. Cell Biol. 103, 87–94. Yin, H. L., Albrech, J., and Fattoum, A. (1981). Identification of gelsolin, a Ca21-dependent regulatory protein of actin gel-sol transformation: Its intracellular distribution in a variety of cells and tissues. J. Cell Biol. 9, 901–906. Yin, H. L. (1988). Gelsolin: Calcium- and polyphosphosphoinositideregulated modulating protein. Bioessays 7, 176–179. Zarcone, R., Bellini, P., Cardone, G., Vicinanza, G., and Cardone, A. (1995). Epidermal growth factor receptor expression: Is it the same in normal and malignant endometria? Clin. Exp. Obstet. Gynecol. 22, 298–300.