| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Obstetrics and Gynecology, University of Connecticut Health Center, Farmington, Connecticut 06030
Address all correspondence and requests for reprints to: John J. Peluso, Ph.D., Department of Obstetrics/Gynecology, University of Connecticut Health Center, Farmington, Connecticut 06030.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As tyrosine kinase activity is involved in maintaining cell viability, the pattern of tyrosine-phosphorylated proteins was examined after culture in control (ascites) or N-cadherin antibody-supplemented medium. Exposure to the N-cadherin antibody altered the pattern of tyrosine-phosphorylated proteins, decreasing the tyrosine phosphorylation of proteins in the 130- to 180-kDa range and increasing the tyrosine phosphorylation of one or more proteins of about 50 kDa. The identity of the 50-kDa protein is unknown. However, immunoprecipitation studies demonstrated that the N-cadherin antibody reduced the amount of tyrosine-phosphorylated FGF receptor in both GCs and ROSE cells by 50%. This decrease corresponds to an increase in apoptosis among aggregated cells. Taken together, these data suggest that homophilic N-cadherin binding and bFGF-FGF receptor binding activate signal transduction pathways that converge at the level of the FGF receptor and subsequently promote the viability of both GC and ROSE cells.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GCs also undergo apoptosis in serum-free medium within 24 h (6, 7). In vitro GC apoptosis is suppressed by a diverse array of hormonal factors, including epidermal growth factor (6, 7), insulin-like growth factor I (8), insulin (8), basic fibroblast growth factor (bFGF) (6), and progesterone (7, 9). In addition to hormonal regulators, cell to cell contact inhibits GC apoptosis (9, 10). This antiapoptotic pathway is dependent on N-cadherin (10). N-Cadherin is a membrane-bound protein (11) that is expressed by GCs and ovarian surface epithelial cells (10, 12, 13, 14, 15) and is involved in specific cell to cell adhesion and tissue morphogenesis (11). N-Cadherin also appears to participate in signal transduction events, as evidenced by the ability to influence various cellular functions (16, 17). Because of the structural similarities between regions of the extracellular domains of N-cadherin and the receptors for bFGF (18), Walsh and Doherty proposed that N-cadherin interacts with the FGF receptor to regulate neurite outgrowth (19). As bFGF and N-cadherin are involved in maintaining GC viability, the present study was designed to determine whether bFGF and N-cadherin inhibit GC apoptosis by activating separate signaling pathways or converge at the level of the FGF receptor.
In addition to GCs, the ovarian surface epithelial cells associated with the stigmata of the ovulatory follicle undergo apoptosis as part of the ovulatory process, thereby creating an opening for the release of the oocyte (20). However, very little is known about the factors that regulate ovarian surface epithelial cell apoptosis. As GCs and ovarian surface epithelial cells both undergo apoptosis in vivo, experiments were also designed to investigate the putative role for N-cadherin in regulating surface epithelial cell apoptosis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Preparation of culture medium
RPMI 1640 without phenol red (Life Technologies, Grand Island,
NY) was used in all culture experiments involving GCs. It was
supplemented with penicillin (0.14 g/liter), streptomycin (0.27
g/liter), HEPES (4.76 g/liter), BSA (fraction V; 2 g/liter), sodium
selenite (5 ng/ml), transferrin (5 µg/ml), and sodium bicarbonate
(2.2 g/liter). The pH was adjusted to 7.4, and the medium was filtered
through a 0.2-µm filter. Depending on the experimental design, bFGF
(5 ng/ml), a goat antihuman bFGF antibody (100 µg/ml), a rabbit
antibody to bFGF receptor (3.75 µg/ml), and a mouse monoclonal
antibody to N-cadherin (144 µg/ml) were added to the cultures. bFGF
and its antibody were purchased from R&D (Minneapolis, MN); the
N-cadherin antibody was obtained from Sigma Chemical Co. (St. Louis,
MO). The neutralizing FGF receptor antibody was a rabbit antibody
directed against the chicken FGF receptor (Upstate Biotechnology, Lake
Placid, NY). For studies involving the N-cadherin antibody, the control
culture medium was supplemented with the appropriate amount of mouse
ascites fluid.
GC isolation and culture
GCs were isolated according to the procedure of Luciano et
al. (7). Briefly, follicles were punctured with 20-gauge needles
and then incubated in EGTA-supplemented medium 199 for 5 min at 37 C in
a 5% CO2-air atmosphere. The ovaries were then incubated
with EGTA-sucrose-supplemented medium 199 for 10 min at 37 C in a 5%
CO2-air atmosphere. The ovaries were washed, resuspended
in fresh medium 199 containing 0.2% BSA, and then pressed to release
GCs. The cells were loaded onto the top of a 15–45% Percoll gradient.
Large GCs were collected from fractions 6–7, plated in either 60-mm
glass petri dishes at 1 x 106 GCs/2 ml for Western
blot procedures or glass eight-chamber Lab-Tek slides (Nunc,
Naperville, IL) at 7.5 x 104/400 µl for assessment
of apoptotic nuclei. GCs were subsequently cultured for 24 h in a
5% CO2-air atmosphere.
Rat ovarian surface epithelial (ROSE) cell culture
ROSE cells were generously provided by Dr. Robert Burghardt,
Texas A&M University (College Station, TX). These cells were maintained
in DMEM-Ham’s F-12 medium supplemented with 5% FBS (21). For
experimental procedures, ROSE cells were plated in either 60-mm glass
petri dishes at 1 x 106 cells/ml or glass
eight-chamber Lab-Tek slides at 1 x 105 cells/400
µl. The cultures were incubated in a 5% CO2/air
atmosphere for 30 min in serum-free medium supplemented with ascites
(control) or an antibody to either N-cadherin or the FGF receptor.
After 30 min, serum was added to a final concentration of 5%, and the
cultures were incubated for 24 h. This protocol was selected so
that the cells could be exposed to the antibody and then treated with
serum to facilitate their attachment to the culture dish.
Detection of N-cadherin antibody after 24 h of culture
To determine whether the antibody to N-cadherin could be
detected between adjacent surfaces of aggregated cells after culture,
both GCs and ROSE cells were plated with the N-cadherin antibody or
ascites fluid. After 24 h of culture, these cells were fixed in
formalin and then incubated with FITC-labeled goat antimouse IgG. The
cells were then observed under epifluorescent optics with the
fluorescein isothiocyanate filter set. The presence of the N-cadherin
antibody was revealed by bright green fluorescence. Cells not cultured
with the N-cadherin antibody were also processed as described above and
served as negative controls.
Assessment of apoptosis
The nuclear structure of GCs and ROSE cells was assessed after
24 h of culture by staining the DNA with hydroethidine
(Polysciences, Warrington, PA) as previously described (7). For each
treatment at least 100 cells were examined, and those cells that
possessed condensed and fragmented nuclei were considered apoptotic.
The percentage of apoptotic cells was then calculated. ROSE cells were
also examined under the electron microscope according to procedures
outlined previously (10).
Western blot analysis of tyrosine-phosphorylated proteins and FGF
receptors
Cell lysate was prepared according to the protocol provided by
Transduction Laboratories (Lexington, KY) with slight modifications.
Cells were rinsed in PBS and then lysed with the addition of 150 µl
boiling 2% SDS, 12.5 mM Tris-HCl buffer (pH 6.8), 20%
glycerol, and 0.005% 2-mercaptoethanol. The cells were placed on ice,
scraped from the dish using a cell scraper, and aspirated two or three
times into a syringe using a 26-gauge needle. The lysate was then
centrifuged in a microfuge at 13,000 x g at 4 C for 15
min. The supernatant was stored at -70 C.
To determine the profile of tyrosine-phosphorylated proteins; equal amounts of cell lysate from each treatment group were loaded onto a 10% polyacrylamide gel and electrophoresed at 100 V. Proteins were then transferred to nitrocellulose and incubated with Tris-buffered saline with 0.1% Tween-20 and 5% dry milk for 1 h as previously described (10). The nitrocellulose was probed with an antibody to tyrosine-phosphorylated proteins at a dilution of 1:1000 (Upstate Biotechnology), washed twice with Tris-buffered saline, and incubated with a 1:25,000 dilution of a peroxidase-labeled goat antimouse IgG (Kirkegaard and Perry, Gaithersburg, MD). Tyrosine-phosphorylated proteins were detected by chemiluminescence using the detection system of Kirkegaard and Perry.
The relative amount of tyrosine-phosphorylated FGF receptor was assessed by first immunoprecipitating the FGF receptor using a monoclonal mouse antibody directed against the human FGF receptor (Upstate Biotechnology). Briefly, 100 µl cell lysate were mixed with 400 µl distilled water, 500 µl 2 x immunoprecipitation buffer [1% Triton X-100, 150 mM NaCl, 10 mM Tris (pH 7.4), 1 mM EDTA, 1 mM EGTA, 0.2 mM sodium vanadate, 0.5% Nonidet P-20, and 0.2 mM phenylmethylsulfonylfluoride] and 10 µg of an antibody to FGF receptor and then incubated on ice for 1 h. After this, 5 µg rabbit antimouse IgG were added, and the incubation was continued for an additional 30 min. Fifty microliters of a solution of 10% protein A-Sepharose in 0.1 M phosphate buffer (pH 7.4) were added to the mixture, and the incubation was continued for a final 30 min with agitation. This mixture was then centrifuged at 13,000 x g for 5 min and washed three times with immunoprecipitation buffer. The pellet was resuspended in 30 µl loading buffer, boiled for 5 min, centrifuged, and loaded onto a 10% polyacrylamide gel. After electrophoresis, the proteins were then transferred to nitrocellulose, and the presence of tyrosine-phosphorylated FGF receptor was detected as described previously.
Statistical analysis
All experiments in which apoptosis was assessed by nuclear
morphology were conducted in duplicate, and the entire experiment was
repeated two or three times. The data from these studies were pooled.
After determining that the percentages were normally distributed, the
data were analyzed by either a two- or one-way ANOVA followed by the
Student-Newman-Keuls multiple range test. Regardless of the statistical
test, only P 
0.05 was considered significant. For
experiments involving tyrosine phosphorylations, cell lysates from each
treatment group were electrophoresed as a single sample. These
experiments were repeated three to five times. The relative densities
of the bands obtained after detecting the tyrosine-phosphorylated FGF
receptors were determined by image densitometry using the imaging and
quantitation software of IP lab gel (Signal Analytics Corp., Vienna,
VA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The
percentage of aggregated apoptotic GCs was not reduced by the bFGF at
doses up to 10 ng/ml (Fig. 1). A neutralizing antibody directed against
bFGF did not alter the percentage of single apoptotic GCs compared to
that in the control group, but attenuated bFGF’s antiapoptotic action
(P < 0.05). The percentage of apoptotic aggregated GCs
remained relatively low even in the presence of the bFGF antibody (Fig. 2). Exposure to the N-cadherin antibody prevented bFGF
from maintaining the viability of single GCs (P <
0.05) and increased the rate of apoptosis of aggregated GCs
(P < 0.05; Fig. 3). The N-cadherin
antibody was detected on the cell surface of single GCs and
predominately at the junctional interface of aggregated GCs after
24 h of exposure (Fig. 3, inset). The detection of
N-cadherin antibody was specific, as GCs not previously treated with
the N-cadherin antibody did not fluoresce after incubation with
FITC-labeled goat antimouse IgG.
|
|
|
) and electron microscopic examination (Fig. 4B). Like
GCs, ROSE cells express N-cadherin (Fig. 5). When cultured in ascites
fluid-supplemented medium, ROSE cells aggregated, forming a monolayer.
Supplementing the medium with the N-cadherin antibody inhibited cell
aggregation to a large extent. Those ROSE cells that established cell
contact in the presence of the N-cadherin antibody were more likely to
be apoptotic than aggregated ROSE cells cultured in
ascites-supplemented medium (P < 0.05; Fig. 5). As with GCs, N-cadherin antibody was detected on the
surface membranes and at the junctional surfaces of aggregated ROSE
cells after 24 h of culture (data not shown).
|
|
). Similar results were obtained with
ROSE cells (Fig. 6B).
|
. Interestingly, treatment with the N-cadherin
antibody specifically reduced tyrosine phosphorylation of proteins in
the 130- to 180-kDa range, while stimulating tyrosine phosphorylation
of a protein(s) with a mol wt of approximately 50 kDa (Fig. 7).
|
, proteins with molecular masses of 140 and 150 kDa were identified as
FGF receptors. These two tyrosine-phosophorylated proteins were present
in both GC and ROSE cells after culture with ascites fluid (Fig. 8). This proteins represent the tyrosine-phosphorylated
state of the FGF receptor. Further, the amount of
tyrosine-phosphorylated FGF receptor was reduced by about 50% in both
GCs and ROSE cells that were cultured in medium supplemented with the
N-cadherin antibody compared to that in ascites fluid controls
(Fig. 8).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
bFGF functions as a survival factor for GCs (6) as well as several other cell types (26, 27, 28, 29), including ROSE cells (Peluso, J. J., unpublished data). Interestingly, bFGF only maintains the viability of single and not aggregated GCs. Although GCs synthesize bFGF (22, 23), it is unlikely that endogenous bFGF accounts for the viability of aggregated GCs. This is supported by the observations that 1) bFGF does not enhance the viability of aggregated GCs; and 2) a neutralizing antibody to bFGF blocks bFGF’s action on single cells, but does not influence the viability of aggregated GCs. Rather, the viability of aggregated cells appears to be due to the homophilic binding of N-cadherin molecules, as both GCs and ROSE cells express N-cadherin, and N-cadherin antibody increases apoptosis of aggregated cells. These results as well as previous studies (9, 10) suggest that N-cadherin-mediated cell contact activates a signal transduction pathway that inhibits apoptosis. Although not dependent on hormones, serum, or growth factors, the N-cadherin-mediated signal transduction pathway may converge with the bFGF pathway.
bFGF enhances GC survival via a tyrosine kinase-dependent mechanism (6). However, several proteins, including the FGF receptors, are tyrosine-phosphorylated in the absence of bFGF or serum. The presence of these tyrosine-phosphorylated proteins is probably due to the homophilic binding of N-cadherin between adjacent cells. This hypothesis is based on the following observations. First, in both GCs and ROSE cells, exposure to an N-cadherin antibody reduces the level of tyrosine-phosphorylated FGF receptor by 50%. This is associated with a corresponding increase in the rate of apoptosis among aggregated cells. Second, the N-cadherin antibody is detected at the junctional interface between cell contacts, demonstrating that this antibody does not completely block cell aggregation but, rather, specifically interferes with the homophilic binding of N-cadherin molecules of adjacent cells. Taken together, these data support the concept that homophilic binding of N-cadherin promotes the tyrosine phosphorylation of FGF receptors, thereby triggering a signal transduction pathway that prevents apoptosis. Further, homophilic N-cadherin binding and bFGF-FGF receptor binding appear to stimulate signal transduction cascades that ultimately activate FGF receptors. This concept is supported by the tyrosine phosphorylation studies and the FGF receptor antibody studies that demonstrate that FGF receptor is required for bFGF, serum, and N-cadherin to prevent apoptosis.
The concept that N-cadherin-mediated cell contact controls various cellular functions by activating FGF receptors is important and may be relevant to other cellular systems. For example, Walsh and Doherty (19) were the first to propose that FGF receptors are an essential part of the mechanism through which cell contact mediates neurite growth. Their hypothesis is based in part on blocking antibody studies similar to ours. In addition, Walsh and Doherty showed that cell contact stimulates neurite growth by a tyrosine kinase-dependent process, but reagents that inhibit nonreceptor cytoplasmic tyrosine kinases do not prevent neurite growth. This suggests that the tyrosine kinase involved in signaling the response to cell adhesion is more closely related to the transmembrane receptor tyrosine kinases (for review, see Ref.17). Although antibody and pharmacological studies support the concept that homophilic binding of N-cadherin molecules stimulates FGF receptor function, the results of the present study are the first to demonstrate a direct link between N-cadherin and the activation (i.e. tyrosine phosphorylation) of FGF receptors.
Although it was anticipated that the N-cadherin antibody would reduce the viability of aggregated cells, it is surprising that N-cadherin antibody interferes with bFGF’s ability to maintain single GCs. Several different mechanisms could account for this observation. First, the N-cadherin antibody could inhibit bFGF from binding to its receptor. This is unlikely because a synthetic N-cadherin peptide that also attenuates N-cadherin-mediated cell survival (10) does not influence bFGF binding (19). Rather, recent studies have proposed that ligand binding results in the FGF receptor forming a complex with N-cadherin molecules within the same cell (30). This putative interaction may be required for bFGF to activate the receptor tyrosine kinase activity of its receptor. The N-cadherin antibody and synthetic N-cadherin peptide are likely to interfere with the formation of the putative FGF receptor/N-cadherin complex. This would account for the ability of these reagents to prevent bFGF from maintaining the viability of single GCs. This type of N-cadherin/FGF receptor interaction could also be induced by N-cadherin-mediated cell contact.
Finally, exposure to N-cadherin antibody induces an approximately 50-kDa protein(s) that is tyrosine phosphorylated in both GCs (31) and ROSE cells. The identity of this protein(s) is unknown, but it may represent an important step in the process of apoptosis. This protein could be an src-related tyrosine kinase such as blk or the cytoskeletal protein, cytokeratin. Both of these proteins are about 50 kDa and are tyrosine phosphorylated in apoptotic cells (32). A more complete biochemical and molecular biological approach will be required to determine both the identity of this protein(s) and its potential role in regulating GC and ROSE cell apoptosis.
|
Acknowledgments |
|---|
Received June 26, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Liebner, U. Cavallaro, and E. Dejana The Multiple Languages of Endothelial Cell-to-Cell Communication Arterioscler. Thromb. Vasc. Biol., July 1, 2006; 26(7): 1431 - 1438. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Berisha, M. Steffl, W. Amselgruber, and D. Schams Changes in fibroblast growth factor 2 and its receptors in bovine follicles before and after GnRH application and after ovulation Reproduction, February 1, 2006; 131(2): 319 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. L. Pon, N. Auersperg, and A. S. T. Wong Gonadotropins Regulate N-cadherin-mediated Human Ovarian Surface Epithelial Cell Survival at Both Post-translational and Transcriptional Levels through a Cyclic AMP/Protein Kinase A Pathway J. Biol. Chem., April 15, 2005; 280(15): 15438 - 15448. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ricken*, P. Lochhead, M. Kontogiannea, and R. Farookhi Wnt Signaling in the Ovary: Identification and Compartmentalized Expression of wnt-2, wnt-2b, and Frizzled-4 mRNAs Endocrinology, July 1, 2002; 143(7): 2741 - 2749. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Williams, E.-J. Williams, and P. Doherty Dimeric Versions of Two Short N-cadherin Binding Motifs (HAVDI and INPISG) Function as N-cadherin Agonists J. Biol. Chem., February 1, 2002; 277(6): 4361 - 4367. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E.-J. Williams, G. Williams, F. V. Howell, S. D. Skaper, F. S. Walsh, and P. Doherty Identification of an N-cadherin Motif That Can Interact with the Fibroblast Growth Factor Receptor and Is Required for Axonal Growth J. Biol. Chem., November 16, 2001; 276(47): 43879 - 43886. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Peluso, A. Pappalardo, and G. Fernandez Basic Fibroblast Growth Factor Maintains Calcium Homeostasis and Granulosa Cell Viability by Stimulating Calcium Efflux via a PKC{delta}-Dependent Pathway Endocrinology, October 1, 2001; 142(10): 4203 - 4211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Auersperg, A. S. T. Wong, K.-C. Choi, S. K. Kang, and P. C. K. Leung Ovarian Surface Epithelium: Biology, Endocrinology, and Pathology Endocr. Rev., April 1, 2001; 22(2): 255 - 288. [Abstract] [Full Text]
|
|
|
|
 |
|
 J.J. Peluso, A. Pappalardo, and G. Fernandez E-Cadherin-Mediated Cell Contact Prevents Apoptosis of Spontaneously Immortalized Granulosa Cells by Regulating Akt Kinase Activity Biol Reprod, April 1, 2001; 64(4): 1183 - 1190. [Abstract] [Full Text]
|
|
|
|
 |
|
 Y Luo, M Ferreira-Cornwell, H Baldwin, I Kostetskii, J Lenox, M Lieberman, and G Radice Rescuing the N-cadherin knockout by cardiac-specific expression of N- or E-cadherin Development, January 2, 2001; 128(4): 459 - 469. [Abstract] [PDF]
|
|
|
|
|
|
A.M. Luciano, S. Modina, F. Gandolfi, A. Lauria, and D.T. Armstrong Effect of Cell-to-Cell Contact on In Vitro Deoxyribonucleic Acid Synthesis and Apoptosis Responses of Bovine Granulosa Cells to Insulin-Like Growth Factor-I and Epidermal Growth Factor Biol Reprod, December 1, 2000; 63(6): 1580 - 1585. [Abstract] [Full Text]
|
|
|
|
|
|
N. H. Machell, O. W. Blaschuk, and R. Farookhi Developmental Expression and Distribution of N- and E-Cadherin in the Rat Ovary Biol Reprod, September 1, 2000; 63(3): 797 - 804. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. MAKRIGIANNAKIS, G. COUKOS, O. BLASCHUK, and C. COUTIFARIS Follicular Atresia and Luteolysis Evidence of a Role for N-Cadherin Ann. N.Y. Acad. Sci., April 1, 2000; 900(1): 46 - 55. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hess, R. Gulati, and J. J. Peluso Hepatocyte Growth Factor Induces Rat Ovarian Surface Epithelial Cell Mitosis or Apoptosis Depending on the Presence or Absence of an Extracellular Matrix Endocrinology, June 1, 1999; 140(6): 2908 - 2916. [Abstract] [Full Text]
|
|
|
|
|
|
J.J. Peluso and A. Pappalardo Progesterone Maintains Large Rat Granulosa Cell Viability Indirectly by Stimulating Small Granulosa Cells to Synthesize Basic Fibroblast Growth Factor Biol Reprod, February 1, 1999; 60(2): 290 - 296. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. S. Kantak and R. H. Kramer E-cadherin Regulates Anchorage-independent Growth and Survival in Oral Squamous Cell Carcinoma Cells J. Biol. Chem., July 3, 1998; 273(27): 16953 - 16961. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Gulati and J. J. Peluso Opposing Actions of Hepatocyte Growth Factor and Basic Fibroblast Growth Factor on Cell Contact, Intracellular Free Calcium Levels, and Rat Ovarian Surface Epithelial Cell Viability Endocrinology, May 1, 1997; 138(5): 1847 - 1856. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
130 kDa). Nonspecific staining
is shown in the lane marked with a minus sign. The effect of the
N-cadherin antibody (Ab-N-cad) on the percentage of aggregated
apoptotic ROSE cells is shown in the lower panel.
