| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Carcinogenesis |
Department of Internal Medicine, Division of Gastroenterology, Hepatology and Nutrition [J. H. S., J. X., A. P. M.] and Department of Integrative Biology [S. U., A. P. M.], The University of Texas Medical School-Houston, Houston, Texas 77030
ABSTRACT
ß-Catenin performs critical roles in development and cellular adhesion. More recently, an oncogenic role has been described. In colon cancer, decreased E-cadherin/ß-catenin association is causally linked to increased ß-catenin-regulated gene expression and increased cellular division. Whether the same pathway is active in native epithelia remains unknown. To address this question, we used the transmissible murine colonic hyperplasia model to measure changes in ß-catenin abundance, nuclear partitioning, target gene (c-myc and cyclin D1) expression, and subcellular distribution. Colonocyte hyperproliferation was associated with a 4.3 ± 0.56 (SD)-fold increase in total cellular ß-catenin protein content, whereas modest changes in 
-catenin and E-cadherin expression were recorded. The ß-catenin signal increased before changes in mucosal crypt length, a gross index of cellular proliferation/apoptosis. ß-Catenin detected in Triton X-100-soluble (cytosolic) cellular fractions was enriched 4.3 ± 0.9 (SD)-fold, whereas a modest decrease of 0.9 ± 0.09 (SD)-fold was recorded in Triton X-100-insoluble (cytoskeletal) fractions. After these changes, nuclear ß-catenin partitioning increased 2.4 ± 0.4 (SD)-fold, accompanied by 2.5 ± 0.4- and 4.0 ± 0.8-fold (SD) increases in cellular c-myc and cyclin D1 levels, respectively. Thus, increased cellular cytosolic and nuclear ß-catenin levels were associated with increased ß-catenin target protein expression. Significant alterations in ß-catenin subcellular distribution were also recorded immunohistochemically. Apical/lateral junctional labeling was observed in normal crypts with increased lateral membrane staining within the upper regions. During transmissible murine colonic hyperplasia, these gradients were dissipated, and basilar plaques were formed within a subset of basal crypt cells. These findings predict that an oncogenic signaling mechanism related to non-E-cadherin-bound ß-catenin is active in hyperproliferating native colonocytes and is similar to that recorded during the early stages of colon carcinogenesis.
INTRODUCTION
Cell adhesion, particularly the cadherin-catenin system, has been a focus of considerable interest because of its potential role in the development of colon cancer. E-cadherin is a transmembrane protein responsible for homophilic binding between epithelial cells and is necessary for establishing cellular polarity (1) . Mutations in E-cadherin are frequent in metastatic disease and may be a rate-limiting step in the progression from adenoma to carcinoma (2) . Chimeric transgenic mouse models that alter the expression of E-cadherin exhibit changes in proliferation, migration, and apoptosis (3, 4, 5) .
E-cadherin is linked to the cytoskeleton by a family of binding molecules, the catenins. ß-Catenin binds to either E-cadherin or APC,2 the gene product mutated in both familial adenomatous polyposis and most sporadic colon cancers (6) . The role of APC as a tumor suppressor gene may be related to its ability to regulate cellular ß-catenin; alterations in its physical interaction with ß-catenin are thought to control the amount of free cytosolic ß-catenin available for the modulation of effector genes controlling proliferation and differentiation. Increased cytoplasmic ß-catenin levels have been shown to cause the constitutive transcriptional activation of downstream target genes through heterotrimeric ß-catenin T-cell transcription factor/lymphoid enhancer-binding factor complexes (7) . This signaling mechanism has been proposed to facilitate tumorigenesis associated with either inactivation of the APC gene product (6) or loss of GSK-3ß-dependent phosphorylation of ß-catenin (8) . During embryogenesis, a related phenomenon occurs. Activation of the Wnt pathway results in inhibition of GSK-3ß activity, which, in turn, leads to stabilization of cytoplasmic ß-catenin and the activation ß-catenin signaling (9, 10, 11) .
The cellular distribution of ß-catenin reflects function. Localization of ß-catenin primarily to the apical-lateral membrane signifies its role in cell adhesion, whereas nuclear accumulation suggests enhanced transcription and activation of target genes. There appears to be a general correlation between the mutational background of cell lines, the degree of dedifferentiation, and increased free (noncytoskeletal) ß-catenin (12 , 13) . Furthermore, an inverse correlation has been observed between decreased membranous and increased nuclear staining of ß-catenin in adenomas and carcinomas (14, 15, 16) . Thus, a similar pattern emerges in both cell lines and clinical specimens, with a shift in distribution accompanying neoplastic transformation.
-Catenin (plakoglobin) is an adhesion molecule with molecular structure, cellular distribution, and protein binding partners similar to those of ß-catenin. In addition to functioning as a linking protein to the cytoskeleton in the z.a., it binds to E-cadherin, 
-catenin, APC, and nuclear transcription factors. However, it is not a functional twin to ß-catenin. -Catenin concentrates in the desmosomal area of cells and has different binding characteristics to APC and transcription factors than ß-catenin (17
, 18)
.
Hyperproliferation is one of the earliest changes noted in the epithelium of those at high risk for colon cancer (19
, 20)
. Epigenetic signaling mechanisms active at this time have been particularly difficult to study because of the lack of suitable animal models. We have used a mouse model of hyperproliferation/hyperplasia (TMCH) to study changes in ß-catenin cell expression and signaling. TMCH, which was well described by Barthold et al. two decades ago (21, 22, 23)
, is induced in mouse colon by Citrobacter rodentium infection; it elicits a profound hyperplasia without a necroinflammatory response. Over a 2-week period, the crypts in the distal colon double in length. There is minimal change in the proximal colon and no change in the small intestine. Our initial studies have demonstrated sequential changes in specific growth factors (basic fibroblast growth factor and acidic fibroblast growth factor), mitogenic signaling molecules (protein kinase Cß and 
), and cystic fibrosis transmembrane conductance regulator anion channel expression, which is regulated by the proliferatory status of the tissue (24, 25, 26, 27)
.
In this study, we demonstrate by Western blotting that the amount of cytoplasmic ß-catenin increased significantly during TMCH, with only a modest decrease in the membrane-bound cytoskeletal pool. By immunohistochemistry, this up-regulation of ß-catenin protein expression in hyperproliferating colonocytes was coordinated with a major change in the subcellular cytoskeletal distribution of ß-catenin from the apical-lateral membrane region into both lateral and basal membranes. Increased nuclear ß-catenin expression levels were recorded by both biochemical and histological approaches, and accompanying increases in steady-state cyclin D1 and c-myc protein expression levels, downstream targets of ß-catenin-induced transcription, attest to the functional activation of a ß-catenin-dependent mitogenic signaling pathway in TMCH crypts. These changes suggest that altered function of adhesion molecules, particularly ß-catenin, is an early cellular event underlying hyperproliferation and hyperplasia in the premalignant mouse colon.
MATERIALS AND METHODS
TMCH.
As described previously (24, 25, 26, 27)
, Swiss-Webster mice were given an overnight culture of C. rodentium mixed with drinking water. Twelve to 14 days after exposure to C. rodentium, animals were euthanized, and the distal colon was removed. The characteristic findings of TMCH were invariably present: a grossly thickened distal colon, with no other changes noted in the remainder of the colon or within the peritoneal cavity. Microscopically, crypt length increased significantly with no obvious increase in epithelial or submucosal inflammatory cell numbers. Our prior studies have demonstrated an 8-fold increase in proliferation as measured by bromodeoxyuridine labeling (27)
.
Isolation of Crypts.
Crypts were harvested from distal colonic sheets by mechanical vibration into Parson’s solution [107 mM NaCl, 4.5 mM KCl, 0.2 mM Na2HPO4, 1.8 mM Na2HPO4, 25 mM NaHCO3, 10 mM EDTA, and 10 mM glucose (pH 7.4) at 27°C, gassed with 95% O2 and 5% CO2]. Crypts were then concentrated by centrifugation for protein analysis or fixed at room temperature in 3% formaldehyde (v/v in PBS) for immunohistology. In the latter instance, the crypts were incubated for 45 min in PBS-containing 50 mM NH4Cl and rinsed three times in PBS. They were then processed for immunohistochemistry or preserved by storage at 4°C in Cytospin Collection fluid containing 29% denatured ethanol, 3% poloxyetheylene (Carbowax), and 2% isopropanol (Shandon, Pittsburgh, PA).
Measurement of Crypt Length.
Preserved crypts (200 µg) were spun onto poly-L-lysine-coated glass coverslips and fixed with Cellfix (Shandon), and images were collected at x200 magnification with a 12-bit gray level charge-coupled device camera connected to an inverted microscope. Crypt length was measured and compared to a standard microscale etched onto a glass slide using Metamorph image analysis software (Universal Imaging Corp., Brandywine Parkway, PA).
Subcellular Fractionation and Protein Estimation.
Crude cellular homogenates were prepared from isolated crypts of normal and C. rodentium-infected mice by homogenization in buffer [50 mM Tris-HCl, 250 mM sucrose, 2 mM EDTA, 1 mM EGTA (pH 7.5), 10 mM 2-mercaptoethanol, 0.5% Triton X-100, plus protease inhibitors]. After a low-speed spin (15,000 x g for 15 min), the clear supernatant was saved as total solubilized protein cell extract. The remaining Triton X-100-insoluble pellet was then sonicated in radioimmunoprecipitation assay buffer (PBS, 1% NP40, 0.1% SDS, and 0.5% sodium deoxycholate) and saved as total non-Triton X-100-soluble cytoskeletal-associated protein.
Nuclear extracts were prepared from freshly isolated crypts essentially as described by Zhang et al. (28) . Briefly, tissues were cut and rinsed in saline A [20 mM Tris-HCl (pH 7.0), 137 mM NaCl, and 5 mM KCl] and homogenized in buffer A [15 mM Tris-HCl (pH 7.0), 60 mM KCl, 15 mM NaCl, 2 mM EDTA, 0.5 mM EGTA, 1 mM DTT, 0.15 mM spermine, 0.5 mM spermidine, 0.4 mM PMSF, and 2 mM bezamidine], 0.25 M sucrose, and 1 µg/ml each of chymostatin, leupeptin, and pepstatin A. The homogenate was mixed with 2 volumes of buffer B (buffer A with 2.3 M sucrose), layered on top of buffer C (buffer A with 1.8 M sucrose), and centrifuged at 25,000 rpm for 60 min at 4°C in a SW27 rotor. The nuclear pellets were resuspended in buffer D [100 mM KCl, 10 mM Tris-HCl (pH 8.0), 2 mM MgCl2, 0.1 mM EDTA, 1 mM DTT, 0.4 mM PMSF, 2 mM bezamidine, and 1 µg/ml each of chymostatin, leupeptin, and pepstatin A]. The suspension was extracted with 0.1 volume of 4 M (NH4)2SO4 on a rotator for 30 min and then centrifuged at 30,000 rpm for 45 min in a SW40 rotor. The protein in the supernatant was precipitated with 0.3 gram/ml (NH4)2SO4; pelleted; resuspended in buffer E [20 mM HEPES (pH 7.8), 100 mM KCl, 0.2 mM EDTA, 0.5 mM DTT, 0.5 mM PMSF, and 2 mM benzamidine] and 1 µg/ml each of chymostatin, leupeptin, and pepstatin A; and dialyzed against buffer E for 4–6 h. The dialyzates were centrifuged for 5 min to remove the precipitates. Protein concentrations were determined, and extracts were frozen in liquid nitrogen and stored at -70°C.
Western Blotting.
Total crypt cellular extracts or subcellular fractions (30–100 µg protein/lane) were subjected to SDS-PAGE and electrotransferred to nitrocellulose membrane. The efficiency of electrotransfer was checked by back staining gels with Coomassie Blue and/or by reversible staining of the electrotransferred protein directly on the nitrocellulose membrane with Ponceau S solution. No variability in transfer was noted. De-stained membranes were blocked with 5% nonfat dried milk in TBS [20 mM Tris-HCl and 137 mM NaCl (pH 7.5)] for 1 h at room temperature and then overnight at 4°C. Immunoantigenicity was detected by incubating the membranes for 1–2 h with the appropriate primary antibodies [0.5–1.0 µg/ml in TBS containing 0.1% Tween 20 (TBS/Tween); Sigma Chemical Co.]. These antibodies were monoclonal anti-E-cadherin antibody (Sigma-Aldrich, St. Louis, MO), anti-ß-catenin and anti--catenin antibodies (Transduction Laboratories, San Diego, CA), and polyclonal anti-cyclin D1 and c-myc antibodies (Santa Cruz Biotechnology, Santa Cruz, CA). After washing, membranes were incubated with horseradish peroxidase-conjugated rabbit antigoat IgG or rat antimouse IgG and developed using the ECL detection system (Amersham Corp., Arlington Heights, IL) according to the manufacturer’s instructions.
Immunohistochemistry.
To prepare the specimens, 200 µl of freshly isolated or preserved crypt suspension were spun onto 1-ounce poly-L-lysine-coated glass coverslips using a concentrating centrifuge (Shandon Cytospin; Shandon). Adhered crypts were then postfixed at -20°C with ultrapure methanol (Polysciences, Inc., Warrington, PA) for 10 and 40 min, respectively. The longer time period required for preserved specimens facilitated recovery of antigenic sites masked by the Carbowax. The coverslips were washed with PBS three times for 5 min each and incubated with 3% sodium deoxycholate (w/v in deionized water) for 1–3 h at room temperature in a humidified chamber. An extended period of detergent permeabilization and extraction greatly facilitated antibody specificity and reduced background in preserved crypts. The coverslips were then rinsed with PBS three times for 5 min each at room temperature, and the crypts were processed for immunohistochemistry. They were blocked with 1% goat serum or 3% BSA in PBS for 20 min at room temperature followed by incubation overnight at 4°C with 25 µl of primary antibody (1:100). Crypts were then washed with three separate 10-min rinses of 1% goat serum/BSA in PBS at room temperature. For immunodetection of the monoclonal primary antibodies, the specimen was incubated at room temperature with 25 µl of fluorescence-conjugated goat antirat Bodipy FL [mouse cross-reactive; supplied by Molecular Probes (Eugene OR); 1:100] for 2 h in the absence of light. After rinsing three times for 10 min with 1% goat serum/BSA in PBS, the crypts were mounted on alcohol-cleaned slides in antifade media [1% phenylenediamine antioxidant in PBS (pH 9.0) in 90% glycerol]. Antibody controls included omission of the primary antibody, detection of endogenous IgG staining pattern with goat antimouse IgG (Calbiochem, San Diego, CA), and preabsorption of endogenous sites with the goat antimouse IgG antibody before specific primary antibodies.
Confocal Microscopy.
Crypts were viewed using a Nikon Diaphot inverted microscope equipped with Nikon Fluor x20 and x40 objective lenses. Specimen scanning was achieved by an Odyssey Real Time Laser Confocal System (Noran Instruments, Inc., Middleton, WI). The fluorescent secondary antibodies were excited with the 488 nm line of an argon gas laser and emitted light detected after band pass at +515 nm before being digitally reconstructed into (512 x 512) 8- or 24-bit maps. Final image magnifications ranged between x200 and x800, depending on the numerical aperture of the objective lens. Sequential 0.1–4-µm Z-axis planes were collected over each crypt using a software-controlled Z-axis stepper motor attached to the focusing control of the inverted microscope. The resulting images were stored and analyzed on an IBM-PC loaded with Image-1 Metamorph Imaging Software (Universal Imaging Corp.).
RESULTS
Total Cellular ß-Catenin Protein Levels Increase during TMCH.
Elevated ß-catenin levels are proposed to promote early neoplastic change through oncogenic signaling within the cell, whereas mutations in E-cadherin may facilitate a later stage of neoplasia, metastasis. The role of -catenin has not been fully delineated. We therefore sought to determine whether a significant change in cellular ß-catenin expression occurred in vivo during TMCH. Infection with C. rodentium elicited a predictable response in mouse distal colon: gross thickening was accompanied microscopically by significant hyperplasia and a doubling of crypt length within 2 weeks (Fig. 1)
. Limited changes were noted in the proximal colon, and the small intestine was normal. Our previous studies have demonstrated significant increases in proliferation without parallel changes in apoptosis during the first 14 days after infection (24
, 26)
. When normal and hyperproliferative (day 12 after C. rodentium infection) total crypt extracts from distal colon were probed by Western blotting, a 4-fold increase in ß-catenin protein levels normalized to the housekeeping protein actin [4.3 ± 0.56 (SD); n = 6 mice] was recorded (Fig. 2A)
. In contrast, changes in -catenin and E-cadherin were not significant (1.2 ± 0.3- and 1.1 ± 0.4-fold, respectively; n = 6 mice; Fig. 2, ii and iii
).
|
|
. Sequential measurement of crypt length indicated a significant increase by day 3, preceding the gross morphological changes, which occurred at day 6 (26)
. However, cellular ß-catenin levels started to increase by day 1 after infection and continued to rise during the next 12 days of the mucosal regenerative response (Fig. 2B)
. In contrast, -catenin levels did not change (Fig. 2B)
. The early onset of increased cellular -catenin protein expression preceding increased crypt length suggested that this adhesion molecule may be pivotal to the hyperproliferative response.
Cytoplasmic and Nuclear ß-Catenin Pools Are Selectively Increased during TMCH.
Standard extraction techniques to fractionate the isolated crypt preparation into paired Triton-soluble (membrane/cytosolic) and Triton-insoluble (cytoskeletal) components were used to determine which subcellular pool of ß-catenin contributed to the rise in total cellular ß-catenin measured at day 12 of the hyperproliferative response (Fig. 3)
. There was a significant increase [4.3 ± 0.9-fold (SD); n = 6 animals] in the noncytoskeletal component of ß-catenin coupled to a minimal decrease (0.0 ± 0.09-fold; n = 6 animals) in the cytoskeletal pool (Fig. 3A)
. In comparison, changes in the Triton X-100-solubilized pool of -catenin were less pronounced (1.3 ± 0.4-fold), and no alterations in the Triton X-100-insoluble fractions of either -catenin or E-cadherin were recorded (Fig. 3, B and C
; n = 6 animals). Thus, mimicking both development and oncogenesis, the cytoplasmic pool of ß-catenin increased significantly during crypt hyperproliferation.
|
, demonstrated that a detectable amount of nuclear ß-catenin was present in the normal epithelium. During crypt hyperproliferation, an increase in the nuclear ß-catenin levels occurred (2.4 ± 0.4-fold relative to nuclear lamin B; n = 6 animals). Similarly, nuclear abundance of -catenin increased in hyperproliferating crypts (1.8 ± 0.6-fold relative to nuclear lamin B; n = 6 animals; Fig. 3E
). Nuclear lamin B levels, which were used as a standard for the measurement of nuclear protein levels, did not change during crypt hyperproliferation (Fig. 3D)
. The purity of the nuclear preparation was confirmed by the lack of cytoplasmic monomeric ß-actin immunoreactivity in either normal or hyperproliferative nuclear extracts (data not shown). Thus, significant nuclear ß-catenin translocation was recorded in hyperproliferating crypts. The fact that -catenin nuclear reactivity also increased during hyperproliferation suggested that this closely related molecule may likewise participate in the proliferatory signal.
Cellular Levels of Cyclin D1 and c-myc Protein Increase during TMCH.
Proliferation is controlled by a series of factors regulating progression through the cell cycle including cyclins and cyclin-dependent kinases. In both normal and neoplastic cells, cyclin D1 plays a critical role. ß-Catenin/T-cell transcription factor binding to the cyclin D1 promoter increases cyclin D1 expression and cellular transition through the G0-S phase of the cell cycle. Furthermore, sustained cyclin D1 expression is essential for maintained proliferatory status (30)
. A secondary target of this signaling pathway is c-myc, which, by inhibiting cyclin-dependent kinase inhibitors, promotes cellular division (31)
. Elevated cellular cyclin D1 and c-myc expression levels therefore represent putative mechanisms of the oncogenic effect of ß-catenin. We measured changes in the protein abundance of these two downstream targets of ß-catenin signaling during TMCH. By Western blotting, we found that total cellular levels of both cyclin D1 and c-myc were increased 4.0 ± 0.8- and 2.5 ± 0.4-fold, respectively (n = 6 animals) in crypt extracts at day 12 after C. rodentium infection when compared with normal conditions (Fig. 4)
. Clearly, TMCH-induced elevated cytoplasmic and nuclear ß-catenin expression levels correlated with increased levels of these cellular division-promoting proteins.
|
is a representative image showing the subcellular localization of ß-catenin in isolated crypts from normal (Figs. 5A
and 6A
) and day 12 hyperproliferative (Figs. 5B
and 6B
) mouse distal colon. Light collected from a single plane at the mid-Z axis, i.e., cut through the middle of the crypt lumen, provides a detailed view of the distribution of immunoreactivity in the presence or absence of primary antibody at x400 and x800 optical magnification. In normally proliferating crypts, punctate staining at the apical-lateral junction was prominent within individual cells at the crypt base but decreased in cells found in the mid-crypt region. Sixty-four percent of crypts exhibited apical-lateral puncta, whereas 36% demonstrated a more diffuse apical pole staining pattern. This pattern of staining was not merely a function of increased membrane area; the highly convoluted apical cell membrane tuft area failed to exhibit similar staining with controls in which the primary antibody was omitted (Fig. 5A, ii)
or substituted with a primary nonspecific goat antimouse IgG primary antibody (Fig. 6A, ii)
. There was faint lateral membrane staining and relatively little cytoplasmic staining in basal crypt regions. However, the intensity of basolateral membrane and cytoplasmic staining increases noticeably in crypt neck regions (n = 6–8 crypts/observation from a total of 12 animals). Nuclear/perinuclear staining was undetectable throughout the normal crypt.
|
|
and 6B
). The most striking was the diminished punctate apical-lateral staining pattern in basal crypt cells and increased diffuse cytoplasmic staining throughout the cellular axis of the crypt (n = 12 animals). At the optical resolution used for these figures (Figs. 5
and 6
), differences between nuclear and cytoplasmic ß-catenin levels (observed by Western blotting) were not readily apparent. In both normal and hyperproliferating crypts, the absence of primary antibody (Fig. 5, ii)
gave only background levels of immunoreactivity. A similar lack of staining was recorded when a nonspecific goat antimouse IgG was substituted for the primary antibody, and low levels of endogenous IgG were detected (Fig. 6, ii)
. The human colon cancer cell line SW480 exhibited lateral membrane staining and nuclear accumulation of ß-catenin as expected (Fig. 6C)
. Thus, in the normal crypt, ß-catenin was localized to puncta at the apical-lateral junction and exhibited a lateral membrane staining that increased in intensity in the upper crypt. TMCH caused a dissolution of the puncta at the apical/apical lateral junction (n = 6–8 crypts/observation from a total 12 animals).
ß-Catenin Exhibits Basilar Plaques in TMCH Colonocytes.
Basilar plaques, areas of intense staining in well-defined basal infranuclear regions of the individual cells, were observed in TMCH crypts (Fig. 7)
. These plaques were noted in a subset of cells at the base and in mid-crypt. When visualized in greater detail at x1200 magnification (Fig. 7)
, the plaques appeared as globular collections of ß-catenin signal between the nucleus and the basal membrane. In reassessing the normal crypts, faint basal staining could be seen occasionally, perhaps representing a form fruste of these plaques. Nuclear staining was generally higher in TMCH basal crypt regions, but there was no clear correlation among nuclear accumulation, loss of apical staining, and formation of basilar plaques.
|
-catenin (Fig. 8A)
. However, in contrast to ß-catenin, there is more basolateral membrane and cytoplasmic staining in cells at the base of the crypt. This more generalized staining lead to a less dramatic punctate appearance at the apical-lateral junction. Increased cellular E-cadherin staining occurred, in a diffuse cytoplasmic pattern, toward the top of the normal crypt. In the TMCH crypt (Fig. 8B)
, E-cadherin staining was modestly altered, with a more intense signal and a suggestion of enhanced accumulation in the granular subapical compartment. This localization pattern for E-cadherin has not been described previously. Thus, the subcellular distribution of E-cadherin was modestly altered during TMCH.
|
Increased rates of proliferation have been described both as a precursor to cancer and as an integral part of the malignant transformation of the epithelium (19
, 20)
and are associated with a multitude of changes in cell signaling molecules and oncogenes (6
, 10
, 12
, 32)
. A major dilemma in assessing the significance of these changes is to determine their biological relevance, i.e., are the associations causally or casually related? During TMCH, hyperproliferation was associated, by Western blotting and immunohistochemistry, with (a) increased cytosolic ß-catenin protein abundance, but not that of -catenin or E-cadherin; (b) accumulation of both ß-catenin and -catenin in the nuclear compartment; (c) increased steady-state levels of two targets of ß-catenin-enhanced transcription, cyclin-D1 and c-myc; and (d) little change in the bound cytoskelatal fraction of all molecules, although the membrane staining pattern of ß-catenin changed from apical-lateral junction to lateral/basal membrane. The roles of ß-catenin in transformed cell lines and cancers have been well characterized, but little is known about how ß-catenin may function in nontransformed epithelium. To our knowledge, this study is the first report of such changes in a native, nonmalignant epithelium. The fact that sequential changes in ß-catenin abundance occurred before increased crypt length suggests that ß-catenin may be a pivotal factor in the regulation of epithelial proliferation in the colon and a major determinant of epithelial homeostasis preceding the early stages of neoplastic transformation.
Subcellular Changes in Homotypic ß-Catenin Expression Occur during Crypt Hyperproliferation in Vivo.
ß-Catenin has dual functions: adhesion and cell signaling (33
, 34)
. In general, a change in ß-catenin subcellular distribution reflects a change in cell function. Subcellular binding partners that regulate ß-catenin expression, other adhesion molecules (E-cadherin), molecules integral to the ß-catenin degradation pathway (axin, GSK-3ß, and Apc), and possibly tyrosine kinases therefore influence cellular proliferation. How cells determine the balance between these competing pathways is a complex but critical question. In cell lines or in chimeric mice, elevations in cellular ß-catenin levels occur after E-cadherin overexpression (3
, 5)
. However, this ß-catenin is primarily stable, is membrane bound, and does not exhibit increased nuclear translocation or entry into the degradative pathway (34)
. Under these conditions, in which a subcellular redistribution was not recorded, proliferatory status is unchanged and may in fact have been reduced. Recently, overexpression of a truncated ß-catenin that resists degradation has been shown to lead to enhanced proliferation in undifferentiated basal crypts cells but also to concomitant increases in apoptosis within the same population, leading to subtle changes in crypt dynamics (35
, 36)
. In uninvolved small intestinal mucosa of the heterozygous min mouse (wild-type APC/truncated APC), increased cellular ß-catenin expression has been shown to cause decreases in both proliferation and apoptosis. In this latter instance, truncated APC was proposed to oligomerize to cytoskeletal-associated wild-type APC and act as a dominant negative regulator of free ß-catenin (37)
. Thus, in both mouse models, enhanced cellular ß-catenin expression per se failed to change overall crypt proliferatory dynamics because of counteracting homeostatic mechanisms active in vivo. In contrast, both cancers of the colon associated with loss of ß-catenin binding to APC and the TMCH colon exhibit increased cytoplasmic ß-catenin levels correlated with both increased proliferation and significant hyperplasia. This suggests that similar counteracting homeostatic mechanisms are either inactive or ineffective under these conditions. In fact, we have already shown that apoptosis is unchanged in TMCH colonocytes (27)
. Our finding that significant increases in c-myc and cyclin D1 expression occur in ß-catenin-overexpressing crypts (Fig. 4)
provides a plausible mechanism for increased rates of cellular proliferation recorded during TMCH. Thus, altered ß-catenin signaling appears to be integral to hyperproliferation and/or hyperplasia in the TMCH mucosa.
Increased Cytoplasmic Expression of ß-Catenin Is Associated with Alterations in Immunocytochemical Distribution.
In the present study, we encountered a gradient in the distribution of ß-catenin along the crypt: surface axis in normal colon. Intense punctate staining was seen at the apical lateral junction, corresponding to the z.a. Occasionally, there was a more diffuse staining of the apical membrane. In basal crypt cells, there was minimal staining along the lateral membranes. As one ascends the crypt, there was an increase in the lateral membrane signal (Fig. 5)
. Cytoplasmic ß-catenin levels were difficult to quantify by immunofluorescence in both normal and TMCH crypts because of the accumulation of signal at these other subcellular locations.
Prior studies examining ß-catenin distribution in normal colon are surprisingly limited. Senda et al. (38) described primarily lateral membrane binding of ß-catenin, with no punctate accumulation at the apical-lateral junction in the mouse. Their data suggest that ß-catenin staining is more intense at the luminal surface than at the base of the crypt. In normal mucosa from patients with familial adenomatous polyposis, the ß-catenin staining was seen primarily on the lateral membrane of colonocytes (16) , with the intensity increasing from the crypt to the surface. We and others have reported previously an increase in ß-catenin signal at the lateral membrane as cells migrate along the crypt villus axis in the mouse small intestine (5 , 39) . This is similar to the staining in isolated colonic crypts found here. Apical staining was observed in small intestinal crypts (5) with a pattern similar to that seen in a subset of our colonic crypts; no punctate accumulation at the apical lateral junction was noted in the small intestine, in contrast to that described in colonic crypts. However, the limited resolution available on conventional microscopy may have obscured the detection of puncta seen in colonic crypts visualized with the confocal light technique. The decrease in this apical signal in the small intestinal villus (5) parallels the present observation in the colonic crypt.
E-cadherin, ß-catenin, and -catenin (data not shown) exhibit similar punctate staining at the apical-lateral junction in the normal colonic crypt. The accumulation of three integral components of the z.a. at the apical-lateral junction lends credence to the concept that puncta represent the z.a. During TMCH, E-cadherin did not exhibit any major change in its subcellular distribution when assayed by either Western blotting (Fig. 3)
or immunohistochemistry (Fig. 8)
. However, the punctate staining pattern of ß-catenin (Figs. 5
and 6
) and -catenin (data not shown) at the apical-lateral border in the normal crypt was diminished during TMCH (Fig. 7)
. By Western blotting, the cytoskeletal pool of this molecule did not change dramatically in TMCH crypts when significant increases in both cytosolic and nuclear ß-catenin were detected. This finding indicates that during hyperproliferation, alterations in the adhesive function of the z.a. may have occurred, caused by a redistribution of cytoskeletal-bound ß-catenin away from the apical-lateral junction toward other cytoskeletal/membrane-associated structures associated with the lateral and basal membranes (Fig. 7)
.
Supporting this hypothesis, immunofluorescence studies made in TMCH crypts clearly show the accumulation of ß-catenin plaques within the cellular basal pole. Although the significance of this novel finding is uncertain at this time, we hypothesize that this accumulation represents an actively sequestered pool of ß-catenin undergoing ubiquitination for cellular removal. In a similar context, the increase in lateral membrane ß-catenin staining may also reflect the cell’s attempt to down-regulate the signaling effects of free cytoplasmic ß-catenin levels in both TMCH crypts and quiescent neck regions of normal crypts (Fig. 5)
. Both phenomena may represent a naturally occurring homeostatic process that maintains ß-catenin signaling in check with the needs of the cell.
Nuclear accumulation of both ß-catenin and -catenin has been described in tumors and in cancer cell lines; however, there are limited descriptions of this occurrence in native epithelia. In tumors, this has been demonstrated immunohistochemically rather than by selective Western blotting of cell fractions. During TMCH, immunofluorescent confocal images demonstrate some nuclear staining (see Fig. 7
), but the staining is not as marked as that in a cancer cell line control (Fig. 6C)
. Nevertheless, it is not unexpected that a premalignant epithelial cell may exhibit a different pattern than a cancer cell. Although there has been considerable focus on the role of nuclear ß-catenin, there is less understanding of the function of nuclear -catenin. Recent studies have suggested a role as a transcription factor or modulator of bcl-2 function (40
, 41)
.
TMCH in Context.
In summary, our studies have provided novel insights into the cellular expression and subcellular distribution of ß-catenin in the nonmalignant hyperproliferating crypt. The novel subcellular localization of ß-catenin in the cytoplasm and nucleus and the formation of cellular basilar pole plaques have not been observed in colonic epithelial cells in culture. These findings may represent a special aspect of hyperproliferation/hyperplasia with preserved colonic architecture. At this point, it is not possible to assign a functional role for either subapical E-cadherin or subbasilar ß-catenin. TMCH recapitulates increases in nuclear ß-catenin recorded in colon cancer cells, and the increased cellular expression levels of cyclin D1 and c-myc protein may explain why proliferation increases. Both findings are consistent with the proposed oncogenic role for ß-catenin in colon cancer. Our results therefore predict that a mitogenic signaling function related to non-E-cadherin-bound ß-catenin occurs in hyperproliferating native colonocytes, similar to that recorded during the early stages of colon carcinogenesis. To our knowledge, this is the first demonstration of this phenomenon in a reproducible model of native mucosal hyperproliferation. This model can provide an excellent opportunity to investigate in more detail the cell proliferatory role of ß-catenin in the normal colonic epithelium.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 To whom requests for reprints should be addressed, at Division of Gastroenterology, Hepatology and Nutrition, The University of Texas Medical School-Houston, 6431 Fannin, MSB 4.234, Houston, TX 77030. Phone: (713) 500-6677; Fax: (713) 500-6699. 
2 The abbreviations used are: APC, adenomatous polyposis coli; TMCH, transmissible murine colonic hyperplasia; GSK, glycogen synthetase kinase; z.a., zonula adherens; PMSF, phenylmethylsulfonyl fluoride; TBS, Tris-buffered saline.
Received 7/26/00. Accepted 1/31/01.
REFERENCES
during mouse colonocyte hyperproliferation. Am. J. Physiol., 279: G223-G237, 2000.
-fetoprotein gene. J. Biol. Chem., 265: 3382-3391, 1990.-Catenin is regulated by the APC tumor suppressor and its oncogenic activity is distinct from that of ß-catenin. Genes Dev., 14: 1319-1331, 2000.This article has been cited by other articles:
|
|
 |
|
  T. Imai, K. Fukuta, M. Hasumura, Y.-M. Cho, Y. Ota, S. Takami, H. Nakagama, and M. Hirose Significance of inflammation-associated regenerative mucosa characterized by Paneth cell metaplasia and {beta}-catenin accumulation for the onset of colorectal carcinogenesis in rats initiated with 1,2-dimethylhydrazine Carcinogenesis, October 1, 2007; 28(10): 2199 - 2206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Umar, Y. Wang, A. P. Morris, and J. H. Sellin Dual alterations in casein kinase I-{epsilon} and GSK-3beta modulate beta-catenin stability in hyperproliferating colonic epithelia Am J Physiol Gastrointest Liver Physiol, February 1, 2007; 292(2): G599 - G607. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Plateroti, E. Kress, J. I. Mori, and J. Samarut Thyroid Hormone Receptor {alpha}1 Directly Controls Transcription of the {beta}-Catenin Gene in Intestinal Epithelial Cells Mol. Cell. Biol., April 15, 2006; 26(8): 3204 - 3214. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ju, J. Hong, J.-n. Zhou, Z. Pan, M. Bose, J. Liao, G.-y. Yang, Y. Y. Liu, Z. Hou, Y. Lin, et al. Inhibition of Intestinal Tumorigenesis in Apcmin/+ Mice by (-)-Epigallocatechin-3-Gallate, the Major Catechin in Green Tea Cancer Res., November 15, 2005; 65(22): 10623 - 10631. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-M. A. Klapproth, M. Sasaki, M. Sherman, B. Babbin, M. S. Donnenberg, P. J. Fernandes, I. C. A. Scaletsky, D. Kalman, A. Nusrat, and I. R. Williams Citrobacter rodentium lifA/efa1 Is Essential for Colonic Colonization and Crypt Cell Hyperplasia In Vivo Infect. Immun., March 1, 2005; 73(3): 1441 - 1451. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Barandon, T. Couffinhal, J. Ezan, P. Dufourcq, P. Costet, P. Alzieu, L. Leroux, C. Moreau, D. Dare, and C. Duplaa Reduction of Infarct Size and Prevention of Cardiac Rupture in Transgenic Mice Overexpressing FrzA Circulation, November 4, 2003; 108(18): 2282 - 2289. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. H. Song, B. Rana, J. R. Wolfe, G. Crimmins, C. Choi, C. Albanese, T. C. Wang, R. G. Pestell, and M. M. Wolfe Gastrin-induced gastric adenocarcinoma growth is mediated through cyclin D1 Am J Physiol Gastrointest Liver Physiol, June 9, 2003; 285(1): G217 - G222. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Rajakangas, S. Basu, I. Salminen, and M. Mutanen Adenoma Growth Stimulation by the trans-10, cis-12 Isomer of Conjugated Linoleic Acid (CLA) Is Associated with Changes in Mucosal NF-{kappa}B and Cyclin D1 Protein Levels in the Min Mouse J. Nutr., June 1, 2003; 133(6): 1943 - 1948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kotsinas, K. Evangelou, P. Zacharatos, C. Kittas, and V. G. Gorgoulis Proliferation, but Not Apoptosis, Is Associated with Distinct {beta}-Catenin Expression Patterns in Non-Small-Cell Lung Carcinomas : Relationship with Adenomatous Polyposis Coli and G1-to S-Phase Cell-Cycle Regulators Am. J. Pathol., November 1, 2002; 161(5): 1619 - 1634. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Shah, M. J. Pishvaian, V. Easwaran, P. H. Brown, and S. W. Byers The Role of Cadherin, beta -Catenin, and AP-1 in Retinoid-regulated Carcinoma Cell Differentiation and Proliferation J. Biol. Chem., July 5, 2002; 277(28): 25313 - 25322. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. C. Abraham, D. S. Klimstra, R. E. Wilentz, C. J. Yeo, K. Conlon, M. Brennan, J. L. Cameron, T.-T. Wu, and R. H. Hruban Solid-Pseudopapillary Tumors of the Pancreas Are Genetically Distinct from Pancreatic Ductal Adenocarcinomas and Almost Always Harbor {beta}-catenin Mutations Am. J. Pathol., April 1, 2002; 160(4): 1361 - 1369. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
