| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Short Communication |
-Catenin-Deficient F9 Cells Differentiate into Signet Ring Cells




From the Department of Cell Biology,*
Faculty of
Medicine, Kyoto University, Kyoto, Japan; the Department of Surgery
II,
Osaka University Medical School, Osaka,
Japan; the Department of Cardiology,
Heart
Institute of Japan, Tokyo Women's Medical College, Tokyo, Japan; and
the Department of Cell Biology,§
Cancer
Institute, Tokyo, Japan
| Abstract |
|---|
|
|
|---|
-catenin is frequently lost in
diffuse type adenocarcinomas. We have isolated
-catenin-deficient
mouse teratocarcinoma F9 cells by gene targeting. Wild-type F9 cell
aggregates cultured in the presence of retinoic acid differentiated
into embryoid bodies with an outer layer of epithelial cells. In
contrast, cell aggregates of
-catenin-deficient cells did
not develop outer layers under the same conditions. The outer surface
cells of
-catenin-deficient cell aggregates,
however, differentiated into epithelial cells as determined by
their expression of epithelial marker proteins. These differentiated
cells scattered from aggregates and showed signet ring cell
morphology, which is frequently observed in diffuse type
adenocarcinomas. We have provided clear evidence that a single mutation
in the
-catenin gene may be a direct cause not only of the scattered
properties of cells but also of signet ring cell formation in diffuse
type adenocarcinoma.
| Introduction |
|---|
|
|
|---|
-Catenin associates with the carboxy-terminal region of the
cadherin cytoplasmic domain via ß-catenin to form a functional
cadherin-catenin cell adhesion complex.1
It has been
demonstrated that
-catenin is frequently lost in diffuse type
adenocarcinomas.2-5
Loss of
-catenin expression has
also been observed in some human adenocarcinoma cell
lines.6-11
The epithelial morphology and rigid cell
adhesion activity were lost in these
-catenin-deficient carcinoma
cells but restored by the exogenous expression of
-catenin.11-14
These
observations strongly suggest that loss of
-catenin function is
involved in the scattered phenotype of diffuse type adenocarcinoma
cells.15
Signet ring cell carcinomas have been described in most but not all
series of diffuse type gastric carcinomas.16
Frequent loss
of
-catenin expression was observed in gastric carcinomas with
scattered cell growth.5
Loss of
-catenin expression was
also observed in a signet ring cell gastric carcinoma cell line.7, 17
However, the loss of
-catenin expression in colon cancer
cell lines did not cause a morphological change of cells to signet ring
cells.10
Thus, it is not clear whether loss of
-catenin
expression is involved in signet ring cell formation.
F9 is a clonal line of mouse teratocarcinoma-derived embryonal
carcinoma cells that shows no epithelial cell morphology under standard
culture conditions. However, F9 cultures treated with retinoic acid
when they are in the form of small aggregates differentiate epithelial
cells on the outer surface of aggregates.18
Due to these
properties, F9 cells are regarded as a good in vitro model
system for the study of epithelium formation mechanisms. The present
study was performed to explore the effects of loss of
-catenin
expression on epithelial cell morphogenesis using
-catenin-deficient
F9 cells.
| Materials and Methods |
|---|
|
|
|---|
The 17-kb mouse
-catenin genomic clone (p16-2) was isolated
from a 129/Sv mouse genomic library. For construction of the targeting
vector, the PSIBP trap-selection cassette replaced the
SmaI-XhoI fragment of the p16-2 clone, which
contains the 5' half of an exon encoding ß catenin binding site. In
the PSIBP trap-selection cassette, splicing acceptor,19
IRES,20
and ß-geo21
sequences are tandemly
connected and inserted between two lox P sequences.22
For
negative selection, the DTA gene replaced the 4.5-kb fragment following
unique BstPI site of the p16-2 clone.23
For construction of the
-catenin expression vector (pEF
FL), the
3.7-kb full-length
-catenin cDNA24
replaced the
XbaI-XbaI stuffer sequence of
pEF-MC1neo.25
Gene Targeting
Mouse F9 EC cells were cultured in Dulbecco's modified Eagle's medium containing 10% heat-inactivated fetal calf serum in gelatin-coated (0.2%) culture dishes. Cell aggregates, initially consisting of 2.5 x 103 cells per aggregate, were formed by the hanging drop culture method.26 For induction of visceral endoderm differentiation, 5 x 10-8 M retinoic acid were added to the medium from the beginning of aggregate formation.
The targeting vector (25 µg) was linearized with SalI and
electroporated at 250 V and 960 µF using a gene pulser into 2 x
107
F9 cells in 0.4 ml of HEPES-buffered saline. Cells were
subjected to G418 selection at 370 µg/ml for 1 week and at 100
µg/ml for another 12 days. For isolation of F9D
(-/-)
cells, the heterozygous clone F9S
(+/-) (clone 13) was
re-electroporated with the targeting vector as described above and then
subjected to G418 selection at 650 µg/ml for 4 days, 360 µg/ml for
5 days, and 150 µg/ml for 5 days. To obtain
F9D
(-/-) cells, 20 µg of pEF
FL expression
vector were electroporated into F9D
(-/-) as described
above, then subjected to G418 selection at 900 µg/ml for 14 days.
Resistant clones were picked after selection. DNA was isolated from
cultured cells as described27
and tested for integration at
the targeted locus by Southern blotting of HindIII
digests. The probe was a 1-kb BstPI-HindIII
fragment located immediately downstream of the region of homology of
the targeting vector (see Figure 1
).
|
The following primary antibodies were used: function-blocking rat
anti-mouse E-cadherin mAb (ECCD-1),28
rat anti-mouse
E-cadherin mAb (ECCD-2),29
rat anti-mouse
-catenin mAb
(
18)30; rat anti-mouse occludin mAb
(MOC37),27
mouse anti-ß-catenin mAb (Transduction Lab.,
Lexington, KY), and mouse anti-cytokeratin 18 mAb (Ks 18.04, Progen
Biotechnik, Heidelberg, Germany). As the secondary antibody,
FITC-conjugated goat anti-rat IgG (Tago, Inc., Burlingame, CA) and
rhodamine-conjugated goat anti-mouse IgG (Amersham Pharmacia Biotech,
Buckinghamshire, UK) were used. Immunoprecipitation, sodium
dodecyl sulfate-polyacrylamide gel electrophoresis, immunoblotting
analysis, and immunostaining were performed as described.24
Cell Dissociation and Aggregation Assay
Cell dissociation and aggregation assays were performed as described previously.30, 31 In some cell aggregation experiments, ECCD-1 was added to the culture medium.
Partial Purification of Signet Ring Cells and Histochemical Analysis
Day 10 embryoid body culture medium of F9D
(-/-)
cells was settled in a centrifuge tube to remove large cell aggregates
and the supernatant was then moved to another tube and centrifuged at
1000 x g for 5 minutes. The cell pellet with signet
ring cells was washed and resuspended in HEPES-buffered magnesium-free
saline. This signet ring cell-rich fraction was smeared on glass
slides, air-dried, and fixed with 10% formaldehyde. Slides were
stained with hematoxylin and periodic acid-Schiff.
Ultrathin Section Electron Microscopy
Embryoid bodies and pellets of signet ring cells were prepared for ultrathin sectioning as described previously.27 Ultrathin sections were cut with a diamond knife, double-stained with uranyl acetate and lead citrate, and then examined using a 1200EX electron microscope (JOEL, Tokyo, Japan) at an acceleration voltage of 100 kV.
| Results |
|---|
|
|
|---|
-catenin allele
(F9S
(+/-)) were generated by replacing part of the
-catenin genomic DNA with the PSIBP trap-selection cassette in sense
orientation (Figure 1a
(-/-)),
F9S
(+/-) cells were retransfected with the
-catenin-targeting vector by electroporation, then cultured in the
presence of increasing concentrations of G418. Heterozygous and
homozygous genotypes were demonstrated by Southern blotting (Figure 1b)
-catenin protein in
F9D
(-/-) cells was confirmed by immunoblotting with
the anti-
-catenin monoclonal antibody (mAb)
18 (Figure 1c)
(-/-) cells were comparable to those in parental F9
or F9S
(+/-) cells (Figure 1c)
-
and ß-catenin in wild-type F9 or F9S
(+/-) cells
(Figure 1d)
(-/-) cells, E-cadherin formed
a complex with ß-catenin despite the absence of
-catenin (Figure 1d)
(-/-) cells using the neo fragment as probe showed
that a single copy of the targeting vector was inserted in the
-catenin locus and random integration of the targeting vector did
not occur (data not shown). To confirm that F9D
(-/-)
cells have no severe mutations other than the lack of
-catenin
expression, we introduced an expression vector encoding full-length
-catenin into F9D
(-/-) cells, then isolated cells
re-expressing similar amounts of
-catenin to wild-type F9 cells,
F9D
(-/-). The expressed
-catenin was derived
from the introduced expression vector because the targeted locus in
F9D
(-/-) cells retained the homozygous genotype
(Figure 1b)
F9D
(-/-) cells, expressed
-catenin formed a complex with E-cadherin and ß-catenin
(Figure 1c)
F9D
(-/-) cells were
indistinguishable from the wild-type F9 cells in all assays
performed here (see below), one of the clones,
F9D
(-/-)-4-4, was used as a representative line of
normal F9 cells throughout subsequent experiments.
In low density monolayer culture,
F9D
(-/-) and
F9D
(-/-) cells showed different morphologies.
F9D
(-/-) cells formed small and compact colonies
(Figure 2a)
similar to those of parental
F9 cells. In contrast, F9D
(-/-) cells showed a
scattered morphology, suggesting dysfunction of their cadherin-catenin
adhesion system (Figure 2f)
. In fact, cell dissociation assay showed
that
F9D
(-/-) cells were hardly dissociated into
single cells, while F9D
(-/-) cells were readily
dissociated (Figure 2, b and g)
. Cell aggregation assay, however,
showed that both
F9D
(-/-) and
F9D
(-/-) cells formed cell aggregates of a similar
size and in similar numbers (Figure 2, c and h)
. After 1 day of
aggregation, both types of cells formed spherical aggregates
(data not shown). A function-blocking monoclonal anti-E-cadherin
antibody (ECCD-1) inhibited this aggregate formation (Figure 2, d and i)
. When cells were immunocytochemically stained for E-cadherin (Figure 2, e and j)
or ß-catenin (data not shown), the antigens were found to
be concentrated in the boundaries between cells. These results
demonstrated that the strong state of cell adhesion activity mediated
by E-cadherin-catenin complex was dependent on
-catenin, whereas the
weak state was not.
|
F9D
(-/-) and F9D
(-/-) cell
aggregates grown in the presence of retinoic acid. After culture in the
presence of 5 x 10-8 mol/L retinoic acid for 5 days
F9D
(-/-) cell aggregates were surrounded by an
outer layer with a smooth surface (Figure 3
(-/-) cell aggregates were irregularly surrounded
by round-shaped cells and did not develop a clear outer layer (Figure 3
F9D
(-/-) cell aggregates consisted of
polarized epithelial cells, characterized by numerous microvilli on the
outer (apical) surface and well-developed junctional complex at
cell-cell boundaries18
(Figure 3
(-/-) cell aggregates, however,
did not show such epithelial morphology (Figure 3
|
F9D
(-/-) cell aggregates expressed both cytokeratin
18 and occludin (Figure 3
(-/-) cell
aggregates (Figure 3
-Fetoprotein, a visceral
endoderm-specific product,18
was also detected in both
F9D
(-/-) and
F9D
(-/-) cell
aggregates after 8 days of culture in the presence of retinoic acid
(data not shown). These results indicated that, in the presence of
retinoic acid,
-catenin-deficient cells did not show epithelial
morphology but differentiated into visceral endoderm as determined by
the expression of epithelial marker proteins.
On culture in the presence of retinoic acid, many cells were scattered
from F9D
(-/-) cell aggregates. Most of these scattered
cells expressed epithelial marker proteins (data not shown, see Figure 3
(bottom) d). Some of the scattered cells showed a unique signet ring
cellular configuration (Figure 3
(bottom) a). Electron microscopic
examination demonstrated that these cells had an intracytoplasmic lumen
pushing the nucleus to one side (Figure 3
(bottom) b). Numerous
microvilli were often observed on the surface of this intracytoplasmic
lumen. Signet ring cells have been reported to be present in
mucin-producing carcinomas. Signet ring cells derived from
F9D
(-/-) cell aggregates showed positive staining with
periodic acid-Schiff, a routine mucin staining method (Figure 3
(bottom) c), and, for cytokeratin 18 and occludin, epithelial marker
proteins (Figure 3
(bottom) d). Signet ring cells were hardly produced
from F9D
(-/-) cell aggregates cultured in the absence
of retinoic acid (data not shown). These observations indicated that
the signet ring cells originated from F9D
(-/-) cells
differentiated into visceral endoderm.
| Discussion |
|---|
|
|
|---|
-catenin expression is thought to be one of direct
causes of the scattered phenotype of diffuse type adenocarcinoma
cells.15
Consistent with this hypothesis,
-catenin-deficient F9 cells lost their adhesiveness and scattered
from cell aggregates when cultured under conditions suitable for
epithelial differentiation. Because these scattered cells expressed
epithelial marker proteins, they were differentiated into epithelial
cells but failed to form epithelium. Undifferentiated F9 cells,
however, showed the weak state of cadherin-dependent cell adhesion
activity and formed spherical aggregates in suspension culture. These
observations suggested that E-cadherin/ß-catenin complex without
-catenin is not sufficient for the maintenance of cell-cell adhesion
between F9 cells differentiated into epithelial cells, although it
mediates the weak state of cell adhesion between undifferentiated F9
cells.
-Catenin plays critical roles in junctional complex formation
between epithelial cells.14
This junctional complex
formation may be important for the maintenance of cell-cell adhesion
between epithelial cells.
Morphological analysis showed that
-catenin-deficient F9 cells
formed a signet ring cell configuration when they were differentiated
into epithelial cells. Because these cells were positively stained with
periodic acid-Schiff, they were thought to resemble a type of
traditional signet ring cell carcinoma. Numerous microvilli were often
observed on the intracytoplasmic luminal surface of these signet ring
cells. This observation suggested the maintenance of epithelial cell
polarity in these cells, even though they grew as single round cells.
This is consistent with a previous observation obtained by electron
microscopic analysis of signet ring adenocarcinoma cells in serous
effusions.16
Several human gastric carcinoma cell lines with signet ring cell
properties, such as HSC-4317
and HSC-39,34
have
been established. In these cell lines, abnormalities of the
cadherin-catenin system were frequently observed. For example, the
expression of
-catenin was not detected in HSC-43.7
In
HSC-39, ß-catenin molecules lacking the
-catenin-binding domain
were expressed and E-cadherin/ß-catenin complex did not interact with
-catenin.35
The present results also demonstrated that
the exogenous expression of
-catenin in
-catenin-deficient F9
cells prevented signet ring cell formation. These observations strongly
suggested that dysfunction of
-catenin in epithelial cells is a
direct cause of signet ring cell formation. It remains unclear how
-catenin-deficient cells form signet ring cellular configurations.
Further studies addressing this question may provide important
information regarding the oncogenic mechanisms of signet ring cell
formation and the molecular mechanisms of epithelial cell
morphogenesis.
We have provided clear evidence that a single mutation in the
-catenin gene may be a direct cause not only of the scattered
properties of cells but also of signet ring cell formation in diffuse
type adenocarcinoma. Because signet ring cell formation was regulated
by addition of retinoic acid in the
-catenin-deficient F9 cell
system, these cells are a promising new in vitro model
system for studying oncogenic mechanisms and for diagnosis of signet
ring cell carcinoma.
| Acknowledgements |
|---|
| Footnotes |
|---|
Supported by a Grant-in-Aid for Cancer Research and a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan, and by Special Coordination Funds for Promoting Science and Technology from the Science and Technology Agency of Japan.
Accepted for publication February 16, 1999.
| References |
|---|
|
|
|---|
-catenin expression in human esophageal cancer. Cancer Res 1994, 54:291-296
-catenin, ß-catenin and plakoglobin in esophageal carcinomas and its prognostic significance: immunohistochemical analysis of 96 lesions. Oncology 1997, 54:158-165[Medline]
-catenin expression in human cancers. Am J Pathol 1994, 144:667-674[Abstract]
catenin expression in scirrhous carcinomas with scattered cell growth. Jpn J Cancer Res 1994, 85:266-273[Medline]
-catenin expression in reduced cell-cell adhesiveness. Cancer Res 1992, 52:5770-5774
- and ß-catenins in human gastric carcinomas: correlation with histology and tumor progression. Oncology Reports 1995, 2:111-117
-catenin cDNA and its aberrant mRNA in a human cancer cell line. Biochem Biophys Res Commun 1993, 193:897-904[Medline]
-catenin gene in human prostate cancer cells. Cancer Res 1993, 53:3585-3590
-catenin in human colon cancer cells. Cancer Res 1995, 55:4722-4728
-catenin protein in cells with a mutant
-catenin gene restores both growth regulation and tumor suppressor activities. Mol Cell Biol 1997, 17:4501-4508[Abstract]
-catenin-negative human colon cancer cell variants via induction of desmosomes. J Cell Biol 1997, 137:1103-1116
-catenin as a key regulator of cadherin function and multicellular organization. Cell 1992, 70:293-301[Medline]
catenin fusion molecules. J Cell Biol 1994, 127:235-245
-catenin: a cause of loss of intercellular adhesiveness in human cancer cell lines. Cancer Res 1994, 54:6282-6287This article has been cited by other articles:
![]() |
J. Tominaga, Y. Fukunaga, E. Abelardo, and A. Nagafuchi Defining the function of beta-catenin tyrosine phosphorylation in cadherin-mediated cell-cell adhesion. Genes Cells, January 1, 2008; 13(1): 67 - 77. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Adachi, A. Inoko, M. Hata, K. Furuse, K. Umeda, M. Itoh, and S. Tsukita Normal Establishment of Epithelial Tight Junctions in Mice and Cultured Cells Lacking Expression of ZO-3, a Tight-Junction MAGUK Protein Mol. Cell. Biol., December 1, 2006; 26(23): 9003 - 9015. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Komiya, M. Shimizu, J. Ikenouchi, S. Yonemura, T. Matsui, Y. Fukunaga, H. Liu, F. Endo, S. Tsukita, and A. Nagafuchi Apical membrane and junctional complex formation during simple epithelial cell differentiation of F9 cells Genes Cells, November 1, 2005; 10(11): 1065 - 1080. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Matsubara and M. Ozawa Expression of {alpha}-catenin in {alpha}-catenin-deficient cells increases resistance to sphingosine-induced apoptosis J. Cell Biol., August 6, 2001; 154(3): 573 - 584. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Yokoyama, K. Tachibana, H. Nakanishi, Y. Yamamoto, K. Irie, K. Mandai, A. Nagafuchi, M. Monden, and Y. Takai {alpha}-Catenin-independent Recruitment of ZO-1 to Nectin-based Cell-Cell Adhesion Sites through Afadin Mol. Biol. Cell, June 1, 2001; 12(6): 1595 - 1609. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Fukata, M. Nakagawa, N. Itoh, A. Kawajiri, M. Yamaga, S. Kuroda, and K. Kaibuchi Involvement of IQGAP1, an Effector of Rac1 and Cdc42 GTPases, in Cell-Cell Dissociation during Cell Scattering Mol. Cell. Biol., March 15, 2001; 21(6): 2165 - 2183. [Abstract] [Full Text] |
||||
![]() |
K. Tachibana, H. Nakanishi, K. Mandai, K. Ozaki, W. Ikeda, Y. Yamamoto, A. Nagafuchi, S. Tsukita, and Y. Takai Two Cell Adhesion Molecules, Nectin and Cadherin, Interact through Their Cytoplasmic Domain-associated Proteins J. Cell Biol., September 5, 2000; 150(5): 1161 - 1176. [Abstract] [Full Text] [PDF] |
||||
![]() |
M Fukata, M Nakagawa, S Kuroda, and K Kaibuchi Cell adhesion and Rho small GTPases J. Cell Sci., January 12, 1999; 112(24): 4491 - 4500. [Abstract] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |