Full text of DOI:10.1093/carcin/21.5.313

Carcinogenesis vol.21 no.7 pp.1313–1317, 2000
Expression of Jun family members in human colorectal
adenocarcinoma
Hanlin Wang1, Mark Birkenbach and John Hart
AP-1-dependent genes is primarily regulated by Jun protein
expression levels since Jun proteins are absolutely required
for AP-1 functioning (9).
Department of Pathology, The University of Chicago Hospitals,
AMB S-352, MC 3083, 5841 South Maryland Avenue, Chicago,
IL 60637, USA
Although c-Jun, JunB and JunD are closely related structur-
ally and share many biochemical properties (5–7), they differ
significantly in their biological functions. In many cell types,
c-jun and junB are generally found to function as proliferation-
promoting genes and their activation (especially c-jun) is
required for cell cycle progression and neoplastic transforma-
tion (9–13). In contrast, the function of JunD is much less
clear and several studies have suggested that it may actually
function to inhibit cell proliferation (14–16). In this regard, it
is significant that c-jun, but not junB or junD, has recently
been shown to be a target gene of the β-catenin/T cell factor
(Tcf) complex whose activity is negatively regulated by the
APC tumor suppressor (17).
Only a few studies have examined c-Jun expression in
human colorectal tumors (18,19) and no study has investigated
the expression of JunB and JunD in these tumors. In the
experiments described in this report, the expression levels
of all three Jun proteins were studied in human colorectal
adenocarcinomas by western immunoblotting analysis and
immunohistochemical staining. The results show that both
c-Jun and JunB are upregulated in colorectal carcinomas
compared with normal colonic mucosa, whereas JunD expres-
sion levels are relatively unaffected by the process of colorectal
tumorigenesis.
1To whom correspondence should be addressed
Email: hwang@mort.bsd.uchicago.edu
The products of the Jun family genes, c-Jun, JunB and
JunD, are essential components of the activating protein-1
transcription factor complexes that are critically important
in the control of cell growth, differentiation and neoplastic
transformation. Although increased c-Jun expression has
been reported in human colorectal tumors, expression of
JunB and JunD in these tumors has not previously been
characterized. In the current study, we examined 24 cases
of human colorectal adenocarcinoma by western immuno-
blotting analysis and immunohistochemical staining for the
expression of c-Jun, JunB and JunD proteins. Normal-
appearing colonic mucosa distant from the tumors in the
same colectomy specimens were used as a reference for
comparison. The results showed that both c-Jun and JunB
proteins were undetectable or barely detectable in normal
mucosa but their expression levels were significantly
increased in human colorectal adenocarcinomas. In con-
trast, JunD protein was present at high levels in normal
mucosa and only showed a minimal increase in adenocarcin-
omas. These observations suggest that different Jun
proteins may serve different roles in regulating colonic
epithelial cell growth and in colorectal tumorigenesis.
Materials and methods
Patient samples
Introduction
Fresh tissue samples from 24 primary sporadic human colorectal adenocarcin-
omas were collected. Histologic examination showed five tumors to be well
differentiated, 15 moderately differentiated and four poorly differentiated.
Tumor size ranged from 2.5 to 11 cm (mean size 4.9 cm). Thirteen patients
had regional lymph node metastases and seven had distant metastases mostly
to the liver. Three of the tumors exhibited mucinous differentiation. Normal-
appearing mucosa distant from the tumor (at least 5 cm from the edge of the
tumor) in the same colectomy specimens were collected as controls. The fresh
tissue samples removed from the specimens were immediately snap-frozen in
liquid nitrogen and stored at –80°C until analysis. Precautions were taken
during sampling to avoid necrotic or hemorrhagic areas and to minimize
stromal contamination.
Colorectal tumorigenesis is a multistep process involving
genetic alterations in both oncogenes and tumor suppressor
genes (1–3). These include activating mutations in ras onco-
genes seen in nearly 50% of colorectal adenocarcinomas (1,2)
and loss-of-function mutations in the adenomatous polyposis
coli (APC) tumor suppressor detected in Ͼ80% of human
sporadic colorectal neoplastic epithelial lesions (4). Many
studies have demonstrated that activated ras-induced neoplastic
transformation is mediated through activation of a group of
nuclear regulatory proteins, the activating protein-1 (AP-1)
transcription factors, which are primarily composed of two
protooncogene families, Jun and Fos (5,6). The members of
the Jun family, c-Jun, JunB and JunD, function by forming
either homodimers or heterodimers among themselves or with
Fos proteins to regulate the transcriptional activity of AP-1-
dependent genes (5–7). Although Jun–Fos heterodimers bind
DNA at the AP-1 site more tightly and are more potent
regulators of transcription than Jun–Jun homodimers, Fos
proteins cannot form homodimers and thus cannot independ-
ently bind DNA (8). As a consequence, transcription of
Preparation of soluble cell lysates
Frozen tissues were washed twice in 4°C phosphate-buffered saline (pH 7.4)
and homogenized in 0.3–0.7 ml of 4°C lysis buffer that contains 50 mM
Tris–HCl (pH 7.4), 120 mM NaCl, 0.5% Triton X-100 and 10 µg/ml
phenylmethylsulfonyl fluoride. After sonication for 20 s, the tissue samples
were centrifuged at 2000 g for 10 min at 4°C. The supernatant was collected,
aliquoted and frozen at –80°C before use. Protein concentrations were
determined by using a protein assay kit obtained from Bio-Rad Laboratories
(Hercules, CA) according to the supplier’s instructions.
Western immunoblotting analysis
Fifty micrograms of cellular protein from human tissues were mixed with
SDS sample buffer, boiled for 10 min and then subjected to electrophoresis
on 10% SDS–polyacrylamide gels. Western immunoblotting procedures were
performed as described previously (16). After the transfer of the proteins onto
nitrocellulose filters, immunological evaluation was performed using an ECL
western blotting analysis kit (Amersham Corp., Arlington Heights, IL) with
Abbreviations: AP-1, activating protein-1; APC, adenomatous polyposis coli;
Tcf, T cell factor.
© Oxford University Press
1313

H.Wang, M.Birkenbach and J.Hart
Fig. 3. Lack of correlation of c-Jun, JunB and JunD expression levels with
clinical stages. The protein expression levels determined by western
immunoblotting analysis were densitometrically quantitated and normalized
with the expression levels of β-actin. The cases were divided into four
groups according to the TNM staging system: stage I (n ϭ 4), stage II (n ϭ
7), stage III (n ϭ 6) and stage IV (n ϭ 7). The data are presented as means
Ϯ SD of relative fold expression using the expression levels in normal
colonic mucosa as a reference (designated as 1).
Immunohistochemistry
Immunohistochemical staining was performed using the DAKO LSAB2
horseradish peroxidase system and DAKO AEC substrate system (DAKO
Corp., Carpinteria, CA) according to the manufacturer’s instructions with
slight modifications. Briefly, deparaffinized tissue sections were first treated
with 3% H2O2 for 10 min to inhibit endogenous peroxidase followed by
incubation with blocking serum for 20 min. The sections were then incubated
with primary rabbit polyclonal antibodies against c-Jun, JunB or JunD at
room temperature for 1 h at final antibody concentrations of 2 µg/ml diluted
in blocking serum solution. After further incubation with biotinylated link
antibody and peroxidase-labelled streptavidin, the staining was developed by
reaction with AEC substrate–chromogen solution followed by counterstaining
with hematoxylin. In each experiment, a negative control was included where
the primary antibodies for c-Jun, JunB or JunD were replaced by a non-
human reactive rabbit IgG (Santa Cruz Biotechnology, Inc.).
Fig. 1. Expression of c-Jun, JunB and JunD in human colorectal
adenocarcinomas (T) and paired normal control mucosa (N) from six
representative cases determined by western immunoblotting analysis (A).
Fifty micrograms of soluble protein from each tissue sample were used.
The expression levels of β-actin were used as loading controls. Equal
protein loading was also monitored by staining the protein gels with
Coomassie blue as exemplified in (B). Positions of molecular mass markers
(in kDa) are shown to the left.
Results
Twenty-four primary sporadic human colorectal adenocarcin-
omas were examined for the expression of c-Jun, JunB and
JunD proteins by western immunoblotting analysis in compar-
ison with paired normal colonic mucosa using soluble cell
lysates prepared from surgical colectomy specimens. Shown
in Figure 1A are six representative cases, which demonstrated
that in normal control mucosa the expression levels of c-Jun
and JunB were essentially undetectable or barely detectable,
but their levels were significantly increased in colorectal
carcinomas. In contrast, JunD protein was present at relatively
high levels in normal mucosa and was only modestly elevated
in carcinomas. Similar results were also observed with lysates
prepared from 18 other cases, with all tumors exhibiting
elevated c-Jun and JunB protein levels relative to control
normal mucosa. In these experiments, the protein levels of β-
actin were also examined as loading controls. However, since
the β-actin levels varied somewhat among different cases
(Figure 1A), protein loading was also monitored by staining
the protein gels with Coomassie blue as shown in Figure 1B.
Figure 2 summarizes the quantitative data from 24 cases
studied by western immunoblotting analysis and shows that,
compared with normal colonic mucosa, adenocarcinomas
showed ~13- and ~5-fold increases in the expression levels of
c-Jun and JunB, respectively (P Ͻ 0.05 by paired t-test). The
Fig. 2. Comparison of c-Jun, JunB and JunD expression in adenocarcinomas
and normal control mucosa from 24 cases. The protein expression levels
determined by western immunoblotting analysis were densitometrically
quantitated and normalized with the expression levels of β-actin. The data
are presented as means Ϯ SD of relative fold expression where the
expression levels of c-Jun, JunB and JunD in normal control mucosa are
normalized to 1. *P Ͻ 0.05 by paired t-test.
rabbit polyclonal antibodies against c-Jun (N), JunB (N-17) or JunD (329)
obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), or a mouse
monoclonal antibody against β-actin (AC-15) obtained from Sigma Chemical
Co. (St Louis, MO). The signals generated by chemiluminescence on the
X-ray films with linear exposure were quantitated by densitometric analysis.
1314

Jun expression in human colorectal carcinoma
Fig. 4. Comparison of c-Jun, JunB and JunD expression in human colorectal adenocarcinoma and normal control mucosa by immunohistochemistry (original
magnification ϫ85). Higher power views are highlighted in inserts (original magnification ϫ470). Six cases were studied with similar results.
expression levels of JunD in adenocarcinomas only showed
~1.5-fold increase, which was statistically insignificant (P Ͼ
0.05). Interestingly, the abundance of c-Jun and JunB proteins
varied significantly among different tumors, but no consistent
correlation was observed between these two proteins. In more
than one-third of the tumors examined, the increase in JunB
expression levels was not proportionate to that of c-Jun (see
Figure 1A for examples).
There was no correlation between clinical TNM stages of
the tumors and the expression levels of any of the three Jun
proteins (Figure 3). Other pathologic features, such as tumor
size, the degree of histologic differentiation or mucin produc-
tion, also did not correlate with the expression levels of c-Jun,
JunB or JunD. The TNM staging system is defined as follows:
T, the extent of the primary tumor; N, the absence or presence
of regional lymph node metastasis; M, the absence or presence
of distant metastasis.
staining than the tumor cells. More specifically, immuno-
staining for c-Jun revealed only scattered epithelial cells in
the normal mucosa to be weakly positive in the nucleus and
immunostaining for JunB was essentially negative, in contrast
to a more uniform and stronger nuclear staining seen in the
neoplastic cells (Figure 4, see inserts for higher magnification).
In addition, an altered subcellular distribution of JunD protein
was noted in tumor cells, in that both nucleus and cytoplasm
were diffusely stained (Figure 4F), whereas in normal mucosa
JunD was primarily localized to the nucleus (Figure 4E).
Discussion
Regulation of the Jun/AP-1 family genes has been proposed
to be critically important in the control of cell growth and
neoplastic transformation (5,6,15). For example, cells over-
expressing c-jun were found to grow in low-serum-containing
medium and to develop colonies in agar, whereas cells over-
expressing junB grew in agar with a reduced efficiency
but did not grow in low-serum-containing medium (10).
Overexpression of junD, on the other hand, either had no
Immunohistochemical staining confirmed immunoblotting
results and demonstrated that Jun protein expression is largely
restricted to the neoplastic epithelial components of the tumors
(Figure 4). Normal mucosa showed comparatively weaker
1315

H.Wang, M.Birkenbach and J.Hart
effect on cell growth (10) or had resulted in slower cell growth
and an increase in the percentage of cells in G0/G1 phase of
the cell cycle (14). JunD was also found to partially suppress
neoplastic transformation induced by an activated ras oncogene
in murine fibroblasts (14). In human ovarian tumors derived
from surface epithelium, the expression levels of junD were
found to be downregulated when compared with normal
ovarian surface epithelial cells. In contrast, the expression
levels of c-jun and junB were comparable with the normal
cells although high junB levels in tumors seemed to correlate
with a more malignant phenotype (20).
JunB. These findings are consistent with the current hypothesis
that JunD may function to suppress cell growth (14).
In summary, we demonstrate that in addition to c-Jun, JunB
is also markedly upregulated in human colorectal carcinomas.
Both c-Jun and JunB proteins are essential components of the
AP-1 transcription factor complexes that may serve important
roles in mediating the transforming effects of β-catenin and/
or mutated ras. JunD is present at relatively high levels in
normal colonic epithelium and neoplastic transformation does
not seem to alter its expression level drastically, suggesting
that it may serve a role different from c-Jun and JunB
in regulating colonic epithelial cell growth and colorectal
tumorigenesis.
Our results presented in this report are in accord with the
few published studies that have examined c-Jun expression
in human colorectal tumors. Using immunohistochemistry,
Magrisso et al. (18) found that the expression of c-Jun was
infrequent in normal colonic mucosa but common in colorectal
adenocarcinomas. Using western immunoblotting analysis,
Pandey et al. (19) demonstrated elevated levels of c-Jun
in colorectal adenocarcinomas and in adjacent histologically
normal-appearing mucosa up to 4 cm from tumor margins.
More importantly, our data provide further evidence to support
the recent report (17) showing that c-jun is a downstream
target gene controlled by the β-catenin/Tcf pathway as demon-
strated by transfection assays using colorectal cell lines. This
is so because in the same 24 tumors reported here (where
c-Jun was found to be upregulated), we have also documented
increased expression of β-catenin (compared with normal
control mucosa), presumably as a result of APC and/or
β-catenin mutations (21). On the other hand, our results suggest
that upregulation of c-Jun expression in human colorectal
carcinomas may not be entirely ras-dependent because the
reported rate of ras mutations in colorectal tumors is Ͻ50%
(1,2). In this regard, it has been shown recently that in
rat intestinal tumors induced by 1,2-dimethylhydrazine, the
presence of K-ras mutations did not correlate with the kinase
activity of either c-Jun N-terminal kinase (JNK) or extracellular
signal regulated kinase (ERK), two important positive regu-
lators of the AP-1 activity (22). In that study, all the adenomas
and carcinomas examined displayed elevated JNK and ERK
activity but only 56% of the tumors contained a mutation in
K-ras.
Acknowledgements
The authors thank Dr Shih-Fan Kuan for his help with immunohistochemistry.
This work was supported by a research fund from the Department of Pathology,
the University of Chicago Hospitals.
References
1.Fearon,E.R. and Vogelstein,B. (1990) A genetic model for colorectal
tumorigenesis. Cell, 61, 759–767.
2.Cho,K.R. and Vogelstein,B. (1992) Genetic alterations in the adenoma–
carcinoma sequence. Cancer, 70, 1727–1731.
3.Vogelstein,B. and Kinzler,K.W. (1993) The multistep nature of cancer.
Trends Genet., 9, 138–141.
4.Kinzler,K.W. and Vogelstein,B. (1996) Lessons from hereditary colorectal
cancer. Cell, 87, 159–170.
5.Angel,P. and Karin,M. (1991) The role of Jun, Fos and the AP-1 complex
in cell-proliferation and transformation. Biochim. Biophys. Acta, 1072,
129–157.
6.Karin,M., Liu,Z.-g. and Zandi,E. (1997) AP-1 function and regulation.
Curr. Opin. Cell Biol., 9, 240–246.
7.Nakabeppu,Y., Ryder,K. and Nathans,D. (1988) DNA binding activities of
three murine Jun proteins. Cell, 55, 907–915.
8.Abel,T. and Maniatis,T. (1989) Action of leucine zippers. Nature, 341,
24–25.
9.Hughes,M., Sehgal,A., Hadman,M. and Bos,T. (1992) Heterodimerization
with c-Fos is not required for cell transformation of chicken embryo
fibroblasts by Jun. Cell Growth Differ., 3, 889–897.
10.Castellazzi,M., Spyrou,G., La Vista,N., Dangy,J.-P., Piu,F., Yaniv,M. and
Brun,G. (1991) Overexpression of c-jun, junB, or junD affects cell growth
differently. Proc. Natl Acad. Sci. USA, 88, 8890–8894.
11.Riabowol,K., Schiff,J. and Gilman,M.Z. (1992) Transcription factor AP-1
activity is required for initiation of DNA synthesis and is lost during
cellular aging. Proc. Natl Acad. Sci. USA, 89, 157–161.
12.Wang,H. and Scott,R.E. (1994) Adipocyte differentiation selectively
represses the serum inducibility of c-jun and junB by reversible
transcription-dependent mechanisms. Proc. Natl Acad. Sci. USA, 91,
4649–4653.
13.Dong,Z., Watts,R.G., Sun,Y., Zhan,S.-N. and Colburn,N.H. (1995)
Progressive elevation of AP-1 activity during preneoplastic-to-neoplastic
progression as modeled in mouse JB6 cell variants. Int. J. Oncol., 7,
359–364.
14.Pfarr,C.M., Mechta,F., Spyrou,G., Lallemand,D., Carillo,S. and Yaniv,M.
(1994) Mouse JunD negatively regulates fibroblast growth and antagonizes
transformation by ras. Cell, 76, 747–760.
15.Wang,H., Xie,Z. and Scott,R.E. (1996) Differentiation modulates the
balance of positive and negative Jun/AP-1 DNA binding activities to
regulate cellular proliferative potential: different effects in nontransformed
and transformed cells. J. Cell. Biol., 135, 1151–1162.
16.Wang,H., Xie,Z. and Scott,R.E. (1996) JunD phosphorylation and
expression of AP-1 DNA binding activity modulated by serum growth
factors in quiescent murine 3T3T cells. Oncogene, 13, 2639–2647.
17.Mann,B., Gelos,M., Siedow,A., Hanski,M.L., Gratchev,A., Ilyas,M.,
Bodmer,W.F., Moyer,M.P., Riecken,E.O., Buhr,H.J. and Hanski,C. (1999)
Target genes of β-catenin-T cell-factor/lymphoid-enhancer-factor signaling
in human colorectal carcinomas. Proc. Natl Acad. Sci. USA, 96, 1603–1608.
18.Magrisso,I.J., Richmond,R.E., Carter,J.H., Pross,C.B., Gilfillen,R.A. and
Carter,H.W. (1993) Immunohistochemical detection of RAS, JUN, FOS
and p53 oncoprotein expression in human colorectal adenomas and
carcinomas. Lab. Invest., 69, 674–681.
The expression of JunB in human colorectal adenocarcin-
omas has not previously been examined. In this report, we
demonstrated for the first time that in addition to c-Jun, JunB
is also upregulated in human colorectal carcinomas. This is
not unexpected, however, since junB, like c-jun, is generally
considered as a proliferation-promoting gene (10,12,15),
although it may be less potent than c-jun in transforming cells
(10). Whether junB is also regulated in a manner similar to
c-jun in human colorectal neoplasms (i.e. controlled by the
β-catenin/Tcf pathway and/or mutated ras gene) is currently
unknown. However, the fact that the expression levels of JunB
were not closely correlated with c-Jun levels suggests that
while JunB induction is also characteristic of neoplastic colonic
epithelium, its expression may be regulated independently of
c-Jun in these cells.
In contrast to c-Jun and JunB, JunD expression appears to
be regulated differently in human colorectal epithelium. Our
results demonstrate that normal colonic epithelial cells express
JunD at much higher level than c-Jun and JunB. Moreover,
carcinoma cells show only a slight increase in JunD protein
levels, in contrast to the marked increases seen in c-Jun and
1316

Jun expression in human colorectal carcinoma
19.Pandey,S., Gordon,P.H. and Wang,E. (1995) Expression of proliferation-
specific genes in the mucosa adjacent to colon carcinoma. Dis. Colon
Rectum, 38, 462–467.
20.Neyns,B., Katesuwanasing,D., Vermeij,J., Bourgain,C., Vandamme,B.,
Amfo,K., Lissens,W., De Sutter,P., Hooghe-Peters,E. and De Greve,J.
(1996) Expression of the jun family of genes in human ovarian cancer
and normal ovarian surface epithelium. Oncogene, 12, 1247–1257.
21.Wang,H.L., Birkenbach,M. and Hart,J. (1999) Upregulation of glycogen
synthase kinase-3β in human colorectal adenocarcinomas correlates with
accumulation of β-catenin. Lab. Invest., 79, 86A.
22.Licato,L.L., Keku,T.O., Wurzelmann,J.I., Murray,S.C., Woosley,J.T.,
Sandler,R.S. and Brenner,D.A. (1997) In vivo activation of mitogen-
activated protein kinases in rat intestinal neoplasia. Gastroenterology, 113,
1589–1598.
Received December 8, 1999; revised March 22, 2000; accepted March 29, 2000
1317