| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Short Communication |
From the Department of Molekulare Pathologie,*
Pathologisches Institut, Universität Erlangen-Nürnberg,
Erlangen, Germany; and the Department of Medical Biosciences and
Pathology,
Umeå University,
Umeå, Sweden
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-catenin.3
Thus, it helps to
maintain the epithelial phenotype. Second, dimers of ß-catenin and
one of the members of the TCF/LEF (T
cell factor/lymphoid enhancer factor) family function as a
transcription factor when localized in the nucleus. In the colon, the
relevant TCF/LEF family member appears to be TCF-4.4,5
Moreover, LEF-1 expression has been shown to be induced by ß-catenin
in colorectal carcinomas.6
Many other ß-catenin target
genes have been described in a variety of species, covering many
cellular functionalities (Stanford University:
http://www.stanford.edu/
rnusse/pathways/targets.html). However,
a correlation between nuclear ß-catenin accumulation and expression
in human tissues has been shown only for a minority of these genes.
Interestingly, in well-differentiated colorectal adenocarcinomas with
APC mutations, ß-catenin is not localized in the nuclei of the cells
by default. Instead, a pattern can be seen; at the invasion front in
most of these tumors, ß-catenin is localized in the nucleus, whereas
in central parts of the tumor, distant from the invasion front, it is
found at the plasma membrane or in the cytoplasm.7,8
Consequently, ß-catenin target genes such as
c-myc,9
matrilysin10,11
or
fibronectin12
are expressed in tumor cells with nuclear
ß-catenin.8,10,13
Another ß-catenin target gene is
cyclin D1.14,15
D-type cyclins
(D1, D2, and
D3) are important in overcoming the restriction
point (R) at the G1-S transition during the cell
cycle by activating cyclin D-dependent kinases, which in turn
phosphorylate the retinoblastoma protein.16
Additionally,
D-type cyclins can sequester the cell cycle inhibitor
p27Kip-1 allowing cyclin E to become fully
active.16
Thus, cyclin D1 is an
up-regulator of proliferation, and its gene is amplified in a variety
of tumors.17
As a ß-catenin target gene, one would
expect high expression of cyclin D1, and thus
proliferation, at the invasion front of well-differentiated colorectal
adenocarcinomas. However, we have shown previously that proliferation
of colorectal carcinomas is reduced at the invasion
front18,19
concomitantly with pronounced expression of the
cell cycle inhibitor p16INK4A.20
Furthermore, the expression of c-myc, another ß-catenin
target gene,9
also an accepted marker of proliferation, is
associated not with proliferation in large adenomas but with nuclear
ß-catenin expression.13
Therefore, we wanted to assess
the distribution of ß-catenin, its target gene cyclin
D1, and p16INK4A with
respect to proliferation. We document that the invasion front of
colorectal adenocarcinomas displays low proliferative activity with
highly significant nuclear co-localization of ß-catenin, cyclin
D1, and p16INK4A. |
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Well-differentiated, sporadic colorectal adenocarcinomas from 56 patients were investigated. All cases were drawn from the archives of the Institute for Pathology in Erlangen, Germany. Ages ranged from 44 to 88 years (mean 66.4 years). There were 28 women and 28 men. The distribution according to the TNM classification was as follows: 2 T1, 15 T2, 31 T3, and 8 T4 cases. 26 of the patients (46.4%) had lymph node metastases (N1, N2). 12 patients (21.4%) presented with distant metastases (M1, M2) in the liver (9) or lungs (3). According to the Unio Internationalis Contra Cancrum (International Union against Cancer, UICC) classification,21 15 patients each (26.8%) were staged UICC I or II and 13 patients each (23.2%) were classified UICC III or IV, respectively.
Immunohistochemistry
Tissue specimens were fixed in 10% buffered formaldehyde and embedded in paraffin according to routine methods. Sections 5 µm thick were mounted on Biogenex slides (Christian Sartori GmbH, Hamburg, Germany), deparaffinized, and hydrated using graded ethanols following routine protocols. Antigen retrieval and antibody dilutions were optimized individually as follows. A 1:750 dilution of a polyclonal rabbit serum specific for ß-catenin (C2206; Sigma, Deisenhof, Germany) or a 1:50 dilution of a mouse monoclonal antibody specific for the Ki-67 antigen (clone MIB-1 7240, DAKO, Hamburg, Germany) were applied overnight to serial sections after microwave treatment twice for 10 minutes in citrate buffer (10.5 g/L citric acid trihydrate-NaOH, pH 6.0) at 600 W and again at 450 W. For cyclin D1 (clone DCS-6; DAKO), sections were heated as described above in citrate buffer, pH 6.0 (DAKO), treated for 10 minutes in 3% (v/v) H2O2 at room temperature, and incubated overnight with a 1:750 dilution of the DCS-6 mouse monoclonal antibody. Developing was initiated by incubating the sections for 30 minutes with a 1:50 dilution of biotinylated goat anti-rabbit or rabbit anti-mouse secondary antibody (DAKO). For cyclin D1, an additional tyramide enhancement step was introduced by further incubating for a further 10 minutes with biotin-tyramide enhancer (0.000375% (w/v) tyramide, 0.00125% (w/v) biotin, 0.2% (v/v) H2O2 in Tris-HCl, pH 8.0).22 After 30 minutes of incubation with Strep-AB-complex (DAKO) and subsequent staining for 15 minutes with Fast Red (Sigma, Deisenhof, Germany), sections were counterstained with hemalaun (Merck, Darmstadt, Germany). p16INK4A staining was performed as described previously.20 Briefly, sections were microwave-treated in citrate retrieval buffer, pH 6.0 (Ventana Inc., Tucson, AZ). Staining was performed with a 1:25 dilution of mouse monoclonal antibody (clone G174–405, Pierce, Rockford, IL) and the Ventana DAB staining system using a Ventana semiautomated staining machine (all Ventana), following the manufacturer’s recommendations. For quality control of staining, positive and negative controls were included in each staining run. For cyclin D1, p16INK4A and Ki-67, sections of tonsils were used as positive controls. The cyclin D1 antibody strongly stained stratified squamous epithelium cells, whereas endothelial cells and histiocytes were stained weakly.23 p16INK4A antibody also stained suprabasal cells of the stratified squamous epithelium, but these were situated more apically. Lymphocytes were not stained. Ki-67 antibody reacted strongly with germinal center lymphocytes, whereas cells of the mantle zone remained negative. Sections of colorectal tumors served as positive controls for ß-catenin, with strong staining of epithelial cells but no staining of stroma cells. Staining in the absence of primary antibodies served as negative controls, in which no staining reaction was seen.
Scoring of Immunohistochemical Staining
Serial sections were used to analyze identical areas for the expression and subcellular localization of ß-catenin, cyclin D1, Ki-67 and p16INK4A. Tumor cells in 10 fields at high power magnification (400x) were counted independently by two observers (A. J. and M. S.).
As the functions of cyclin D1,
p16INK4A, ß-catenin as transcription factor and
of the Ki-67 antigen are associated with nuclear localization, the
percentage of cells showing nuclear staining was scored and rounded up
or down to the nearest 5%. The results of both observers were
comparable, and in two cases with conflicting results (deviation
> 5%), the final result was determined jointly. Initially, sections
were examined for expression of ß-catenin, cyclin
D1, and Ki-67 antigen only (set 1, represented in
Figure 1
). For analysis of
p16INK4A expression, a second series of sections
was subsequently cut from the same tissue blocks (set 2, represented in
Figure 2
). This set was also stained for
ß-catenin for comparison. Both sets showed virtually identical
results with regard to ß-catenin expression (Figure 1G
and Figure 2I
), demonstrating the high consistency of staining and analysis.
|
|
All statistical analyses were performed using SPSS version 10.0.5
software (SPSS Inc., Chicago, IL). The patient data were analyzed for
frequencies of age, gender, TNM status and UICC stage (see Patient
Data). Data are presented using box plots, a descriptive form of
statistical analysis. The box represents the interquartile range (75%
of all cases) and includes the median (bar in the box). The T-bars
represent so-called outliers (values deviating more than 1.5-fold from
the interquartile range), thus defining the total range of variation of
all values (100%). Correlations between parameters and resulting
p-values were calculated by applying the nonparametric Mann-Whitney
test or the 
2
test (Fisher’s exact test).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The subcellular distribution of ß-catenin was found to be
heterogeneous throughout each tumor, as described.7
At the
invasion front, single cells and small cell clusters showed strong
nuclear staining (Figure 1A
; Figure 2, A and E
) in varying numbers of
cells (Figure 1G
; 5 to 75%, median 45%; Figure 2I
; 15 to 75%, median
40%). Cells from central parts of the tumor rarely showed nuclear
staining (Figure 1G
, Figure 2I
; 0 to 10%, median 5%). Here, the
cytoplasm or the plasma membrane were stained prominently in some areas
of the tumor (Figure 1B
; Figure 2, B and F
). This subcellular
localization of ß-catenin and the spatial distribution within the
tumors correlated with high significance (Figure 1G
; Figure 2I
; set 1
and 2: Mann-Whitney test: P < 0.001).
Cyclin D1
As expected, in all cases, strong nuclear expression of cyclin D1
(Figure 1C)
was observed in those areas displaying nuclear ß-catenin,
so that mainly cells at the invasion front were positive (Figure 1G
; 5
to 60%, median 30%). In contrast, central parts of the tumor rarely
showed cyclin D1 staining (Figure 1, D and G
; 0
to 10%, median 5%). Nuclear localization of cyclin
D1 at the invasion front correlated with high
significance (Figure 1G
; Mann-Whitney test: P = 0.001),
as did co-localization of nuclear ß-catenin and nuclear cyclin
D1 (Figure 1G
; 2
test:
P = 0.001).
Proliferation (Ki-67 Antigen)
Taking cyclin D1 expression as a marker for
proliferation, one would expect proliferation mainly at the invasion
front. The Ki-67 antigen is expressed in all phases of the cell cycle
except G0 and is thus a generally accepted marker
of proliferation.24
Unexpectedly, in our collection of
cases, nuclear Ki-67 staining was prominent mainly in central parts of
the tumor, mostly in a large proportion of cells (Figure 1, F and G
; 5
to 80%, median 60%). In contrast, Ki-67 expression, and thus
proliferation, were much lower at the invasion front (Figure 1, E and G
; 0 to 30%, median 10%). Ki-67 expression and localization in
central parts of the tumor correlated strongly (Figure 1G
; Mann-Whitney
test: P < 0.001). In support of this, cyclin
D1 and ß-catenin showed a negative correlation
with Ki-67 expression, which was statistically highly significant for
both pairs (Figure 1G
; 2
test:
P < 0.001).
p16INK4A (Inhibitor of Kinase 4)
As the statistically significant negative correlation between
cyclin D1 and Ki-67 staining was unexpected, we
looked for expression of p16INK4A, as we had
previously shown that this cell cycle inhibitor is up-regulated at the
invasion front of colorectal adenocarcinomas.20
Strong
nuclear expression of p16INK4A was seen in areas
displaying nuclear ß-catenin (Figure 2, A and E)
, so that cells
mostly at the invasive front were positive (Figure 2, C, G, I
; 35 to
80%, median 55%). Central parts of the tumor with cytoplasmic or
membranous ß-catenin staining (Figure 2, B and F)
on the contrary
rarely showed nuclear p16INK4A staining (Figure 2, D, H, I
; 0 to 10%, median 5%). Notably, many of the tumors
(60%) showed residual p16INK4A expression in
the cytoplasm of cells in central parts of the tumors (Figure 2D)
,
whereas the rest (40%) showed only minimal or no staining at all
(Figure 2H)
. The absence of staining was not due to
methodological problems, as the same tumors displayed a strong
p16INK4A reactivity at the invasion front (Figure 2G)
.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our observation that cyclin D1 is predominantly
expressed in non-proliferating cells at the invasion front but not in
proliferating cells in central parts of colorectal adenocarcinomas was
unexpected. In elegant experiments, Tetsu and
McCormick14
showed that the human colorectal cancer
cell line, HCT 116 (mutant ß-catenin, wild type APC), ceased
proliferation after transduction with retroviral expression vectors
encoding dominant negative TCF-4. Dominant negative TCF-4 leads to
suppression of ß-catenin target genes, and consequently of cyclin
D1. Co-infection with a cyclin
D1-encoding retroviral expression vector restored
proliferation suggesting that cyclin D1 is
causative for proliferation in human colorectal cancer.14
Our results indicate, that within an intact histological context, a
different scenario may apply. This may relate to the fact that the cell
cycle inhibitor p16INK4A is expressed in tumor
cells at the invasion front together with cyclin
D1 and ß-catenin. Incidentally, we have data
showing that p16INK4A is another target gene for
ß-catenin (Wassermann S, Brabletz T, Palmqvist R, Frank P, Kirchner
T, Jung A, manuscript in preparation), which are in support with
the high correlation of ß-catenin and p16INK4A
co-expression. However, at least some cultured colorectal carcinoma
cell lines do not express p16INK4A 25 and show,
like 37 to 55% of human colorectal cancers, methylation of CpG islands
in the p16INK4A promoter/exon
1.25,26
Our observation that around 40% of investigated
tumors do not express p16INK4A in central parts
of the tumors (Figure 2H)
but re-express it at the invasion front
(Figure 2G)
may be due to promoter methylation, which is thought to
shut off gene expression.26
Thus, it will be interesting
to investigate first, the methylation status of tumor cells in central
parts and second, whether changes in the methylation status of the
p16INK4A promoter occur in colorectal tumor cells
at the invasion front.
Our observations raise questions regarding the possible function of cyclin D1 expression at the invasion front, and the role of the invasion front during the process of carcinogenesis of well-differentiated colorectal tumors. It could be envisaged that co-expression of cyclin D1 and p16INK4A maintains tumor cells in a non-proliferating state. However, at the same time this might allow tumor cells to react rapidly to environmental signals and to enter the cell cycle after degradation of p16INK4A. Moreover, it seems that cells displaying nuclear ß-catenin undergo a mesenchymal-like transdifferentiation, or possibly dedifferentiation, as suggested by changes in morphology8 and gene expression profile.8,10,11 If proliferation and differentiation are generally mutually exclusive functionalities of cells, and if co-expression of cyclin D1 and p16INK4A keeps cells at the invasion front of colorectal adenocarcinomas in a non-proliferating state, then such zones would allow cells to undergo differentiation and proliferation in close vicinity. Consequently, tumors could expand, and at the same time remodel the stroma, giving rise to highly regulated invasion. More work must be done comparing cells from the invasion front and central parts of the tumor, and developing in vitro systems reflecting the heterogeneity of these two areas in well-differentiated colorectal adenocarcinomas.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by Wilhelm-Sander-Stiftung, AZ. Grant 99.065.1 to A.J., T.B., and T.K.; The Swedish Cancer Society (2520-B99-13XAC), the Lion’s Cancer Research Foundation Umeå, and the Medical Faculty of Umeå University to R.P.
M. Schrauder and U. Oswald contributed equally to this work.
Accepted for publication July 27, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  P. T. Winnard Jr., A. P. Pathak, S. Dhara, S. Y. Cho, V. Raman, and M. G. Pomper Molecular Imaging of Metastatic Potential J. Nucl. Med., June 1, 2008; 49(Suppl_2): 96S - 112S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Mimeault and S. K. Batra Interplay of distinct growth factors during epithelial mesenchymal transition of cancer progenitor cells and molecular targeting as novel cancer therapies Ann. Onc., October 1, 2007; 18(10): 1605 - 1619. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Funasaka, H. Hu, T. Yanagawa, V. Hogan, and A. Raz Down-Regulation of Phosphoglucose Isomerase/Autocrine Motility Factor Results in Mesenchymal-to-Epithelial Transition of Human Lung Fibrosarcoma Cells Cancer Res., May 1, 2007; 67(9): 4236 - 4243. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Uchiyama, H Itoh, S Naganuma, K Nagaike, T Fukushima, H Tanaka, R Hamasuna, K Chijiiwa, and H Kataoka Enhanced expression of hepatocyte growth factor activator inhibitor type 2-related small peptide at the invasive front of colon cancers Gut, February 1, 2007; 56(2): 215 - 226. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. J. Sansom, K. R. Reed, M. van de Wetering, V. Muncan, D. J. Winton, H. Clevers, and A. R. Clarke Cyclin D1 Is Not an Immediate Target of {beta}-Catenin following Apc Loss in the Intestine J. Biol. Chem., August 5, 2005; 280(31): 28463 - 28467. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Barrallo-Gimeno and M. A. Nieto The Snail genes as inducers of cell movement and survival: implications in development and cancer Development, July 15, 2005; 132(14): 3151 - 3161. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Andreu, S. Colnot, C. Godard, S. Gad, P. Chafey, M. Niwa-Kawakita, P. Laurent-Puig, A. Kahn, S. Robine, C. Perret, et al. Crypt-restricted proliferation and commitment to the Paneth cell lineage following Apc loss in the mouse intestine Development, March 15, 2005; 132(6): 1443 - 1451. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Vega, A. V. Morales, O. H. Ocana, F. Valdes, I. Fabregat, and M. A. Nieto Snail blocks the cell cycle and confers resistance to cell death Genes & Dev., May 15, 2004; 18(10): 1131 - 1143. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Saegusa, M. Hashimura, T. Kuwata, M. Hamano, and I. Okayasu {beta}-Catenin Simultaneously Induces Activation of the p53-p21WAF1 Pathway and Overexpression of Cyclin D1 during Squamous Differentiation of Endometrial Carcinoma Cells Am. J. Pathol., May 1, 2004; 164(5): 1739 - 1749. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Hiendlmeyer, S. Regus, S. Wassermann, F. Hlubek, A. Haynl, A. Dimmler, C. Koch, C. Knoll, M. van Beest, U. Reuning, et al. {beta}-Catenin Up-Regulates the Expression of the Urokinase Plasminogen Activator in Human Colorectal Tumors Cancer Res., February 15, 2004; 64(4): 1209 - 1214. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Seemayer, S. Kuchen, P. Kuenzler, V. Rihoskova, J. Rethage, W. K. Aicher, B. A. Michel, R. E. Gay, D. Kyburz, M. Neidhart, et al. Cartilage Destruction Mediated by Synovial Fibroblasts Does Not Depend on Proliferation in Rheumatoid Arthritis Am. J. Pathol., May 1, 2003; 162(5): 1549 - 1557. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Svensson, K. Nilsson, A. Ringberg, and G. Landberg Invade or Proliferate? Two Contrasting Events in Malignant Behavior Governed by p16INK4a and an Intact Rb Pathway Illustrated by a Model System of Basal Cell Carcinoma Cancer Res., April 15, 2003; 63(8): 1737 - 1742. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J R Jass, M Barker, L Fraser, M D Walsh, V L J Whitehall, B Gabrielli, J Young, and B A Leggett APC mutation and tumour budding in colorectal cancer J. Clin. Pathol., January 1, 2003; 56(1): 69 - 73. [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]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
