Mucin-depleted foci have b-catenin gene mutations, altered expression
of its protein, and are dose- and time-dependent in the colon
of 1,2-dimethylhydrazine-treated rats
Angelo Pietro Femia1, Benedetta Bendinelli1, Augusto Giannini2, Maddalena Salvadori1, Pamela Pinzani3,
Piero Dolara1and Giovanna Caderni1*
1Department of Pharmacology, University of Florence, Firenze, Italy
2Department of Pathology, General Hospital of Prato-Azienda Sanitaria Locale 4, Prato, Italy
3Department of Clinical Physiopathology, University of Florence, Florence, Italy
Mucin-depleted foci (MDF) are purported preneoplastic lesions
that can be easily visualized in the unsectioned colon of carcino-
gen-treated rats stained with high-iron diamine alcian blue
(HID-AB). In F344 rats treated twice with 150 mg/kg of
1,2-dimethylhydrazine (DMH) and sacrificed after 5, 9, 13 and
28 weeks, MDF increased over time from 5 to 13 weeks,
whereas they decreased at 28 weeks, when tumors appear. MDF
multiplicity (crypts/MDF) linearly increased with time. Increas-
ing doses of DMH (100, 150 and 200 mg/kg 3 2 times) caused a
dose-related increase in MDF. Mutations in Ctnnb1 gene codify-
ing for b-catenin were identified with PCR amplification and
direct sequencing in 6/15 tumors (40%), 7/28 MDF (25%)
and 2/27 (7%) aberrant crypt foci (ACF) identified in HID-
AB-stained colon. All mutations in tumors and MDF caused
amino acid substitution, while one mutation in ACF was
silent. b-catenin detected at membrane level by immunohisto-
chemistry was markedly reduced in MDF and tumors and, to
a lesser extent, in ACF identified with HID-AB. By contrast,
nuclear localization of b-catenin was significantly increased in
MDF and tumors, while no variation was observed in ACF.
increased in MDF and tumors but to a lesser extent in ACF.
In conclusion, MDF are induced dose-dependently by DMH,
increase in size with time, have mutations in the b-catenin
gene and marked alterations in b-catenin cellular localization.
Since all these phenomena are considered specific steps for colon
tumorigenesis, these results further support the hypothesis that
MDF are cancer precursors and can be proposed as endpoints
in short-term carcinogenesis experiments.
' 2005 Wiley-Liss, Inc.
Key words: colon carcinogenesis; preneoplastic lesions; mucin-
depleted foci; b-catenin; 1,2-dimethylhydrazine
Colon carcinogenesis occurs through consecutive steps, starting
from the transformation of normal crypts into preneoplastic
lesions and their eventual change into adenomas and carcinomas.1
The formation of early neoplastic lesions with different degrees of
dysplasia is a crucial point in this process; therefore, many efforts
have been dedicated to the identification and characterization of
Mucin-depleted foci (MDF), characterized by absent or
scarce mucous production, have been recently described by our
group7in the colon of rats treated with azoxymethane (AOM).
MDF can be easily identified along the entire mucosal surface
of unembedded colon stained with high-iron diamine alcian
blue (HID-AB), a technique that highlights mucin production.
MDF are histologically dysplastic, and their number in a colon
is of the order of magnitude of tumors induced by the same
dosage of carcinogen.7,8MDF increased in rats treated with
known promoters of colon carcinogenesis, such as cholic acid8
and beef meat,9and decreased with chemopreventive agents
such as piroxicam and synbiotics.7,8On this basis, we sug-
gested that MDF represent preneoplastic lesions in the process
of colon carcinogenesis.
Abnormal activation of the APC/b-catenin/TCF/Lef pathway
(Wnt pathway) has been documented during various phases of
colon carcinogenesis. At a cellular level, this alteration is charac-
terized by a decrease in membrane b-catenin, concomitant with its
accumulation in the cytoplasm and in the nucleus,10–12where
b-catenin may function as a trascriptional activator.10The constit-
utive activation of the Wnt pathway may be caused by mutations
in the b-catenin gene (Ctnnb1 in the rat), affecting regulatory
phosphorylation sites responsible for protein degradation in the
Therefore, we studied Ctnnb1 mutations and cellular localization
of b-catenin protein in MDF induced by 1,2-dimethylhydrazine
(DMH) in rat colon. Cellular localization and gene mutations of
b-catenin were studied also in colonic tumors after DMH as well as
in aberrant crypt foci (ACF) identified in HID-AB stained colon.
Moreover, we studied the time-course and dose dependence of
MDF and we compared MDF to ACF, which have been extensively
characterized in rodents2,5,6and humans,3,4widely used for predict-
ing carcinogenesis,13with a few correlation problems.14,15
Material and methods
Animals and carcinogen treatment
F344 male rats aged 4–5 weeks (Nossan, Correzzana, Milan,
Italy) were housed according to the European Union Regulations
on the Care and Use of Laboratory Animals16and fed a high-fat
diet (23% corn oil w/w) whose composition is based on the
AIN-76 diet.17The experimental protocol was approved by the
Commission for Animal Experimentation of the Italian Ministry
of Health. Rats were allocated to 1 of 3 different groups that were
treated twice, 1 week apart with 2 s.c. injections of 1,2-dimethyl-
hydrazine-dihydrochloride (Sigma-Aldrich, Milan, Italy) at the
following dosages: 100 mg/kg (group 1; 8 rats), 150 mg/kg
(group 2; 30 rats) and 200 mg/kg (group 3; 8 rats). Therefore, the
total dosage of carcinogen administered to the 3 different groups
was 200 mg/kg (group 1), 300 mg/kg (group 2) and 400 mg/kg
(group 3). DMH was dissolved in sterile saline and buffered with
sodium hydroxide at neutral pH before injections. Rats treated
with 200 and 400 mg/kg were sacrificed 13 weeks after the first
DMH injection, whereas rats treated with 300 mg/kg of DMH
were sacrificed 5, 9, 13 and 28 weeks after the first injection of
carcinogen in order to study the sequential appearance of MDF
and the induction of tumors at later time points.
Grant sponsor: European Union Projects; Grant number: QLKI-1999-
00346 and QLRT 1999-00505; Grant sponsor: Ministero dell’ Universita ` e
della Ricerca Scientifica e Tecnologica, Italy; Grant sponsor: World
Cancer Research Fund, U.K.
*Correspondence to: Department of Pharmacology, University of
Florence, 50139 Firenze, Italy. Fax: þ39-055-4271-280.
Received 6 December 2004; Accepted after revision 14 January 2005
Published online 4 March 2005 in Wiley InterScience (www.interscience.
Int. J. Cancer: 116, 9–15 (2005)
' 2005 Wiley-Liss, Inc.
Publication of the International Union Against Cancer
Evaluation of ACF and MDF
At sacrifice, the colon was removed, washed with saline, longi-
tudinally opened and pinned flat on a polystyrene board as previ-
ously described.7,8After fixation in buffered formalin (at least
4 hr), the colon was removed and cut into 3 segments: proximal
(closer to the cecum and characterized by herring-bone folding of
the mucosa), mid and distal (obtained by cutting the remaining
colon into 2 equally long segments). ACF were determined
according to Bird2using methylene blue (MB) staining. After
ACF determination, colons were kept in formalin and then proc-
essed with HID-AB staining to visualize MDF as described.7,8
MDF are characterized by the absence or very limited production
of mucins. Moreover, MDF are focal lesions (i.e., there is a clear
distinction between normal surrounding crypts and the MDF).
To be defined as an MDF, a lesion has to show these 2 features:
the lack of mucins and the property to be a focus. Moreover, MDF
are often formed by crypts smaller than the normal ones and their
lumen is often distorted. Elevation of the lesion above the surface
of the colon and a multiplicity (i.e., the number of crypts forming
each focus) higher than 3 are also frequent features of MDF. How-
ever, especially at earlier time points (5 weeks after DMH), MDF
formed by only 2 crypts are also visible. Single crypts without
mucus are never considered as MDF.
The colons were coded and scored blindly by 2 observers. For
MDF determination, the 2 sets of data were first compared and if
the number and multiplicity of the lesions found were the same
between the 2 observations, those values were used. By contrast,
if discrepancies between the 2 observations were found, these
were discussed by reanalyzing the colon at the microscope until a
decision was taken. The size of ACF or MDF was determined with
a grid placed on the ocular of the microscope, calculating the area
occupied by the lesions at 100? magnification. We also deter-
mined the relative distribution of MDF along distal, mid and prox-
imal colon, calculating for each rat the number of MDF in each
segment of the colon/total MDF in that colon ? 100; a mean value
for all rats was then obtained. The same calculation was per-
formed to determine the relative distribution of ACF and tumors
along the different segments of the colon.
Appearance of MDF in MB-stained colons
After ACF determination, 9 methylene blue (MB)-stained
colons were reexamined under the microscope (40? magnifica-
tion). When a lesion suspected of being an MDF was observed, its
topographical localization was recorded and the lesion photo-
graphed. The same 9 colons were then stained with the HID-AB
technique and the number and position of MDF were compared
with the findings obtained in MB-stained colons.
Histopathologic evaluation of tumors
Histopathologic examination of the tumors was carried out as
Dissection of lesions and evaluation of immunostaining
For the study of b-catenin, 3 colons from the 300 mg/kg group,
sacrificed 13 weeks after the first DMH injection, were fixed in
formalin for not more than 4 hr and then immediately processed to
determine ACF and MDF as described above. During microscopic
observation, MDF and ACF identified in HID-AB-stained colon
were marked with permanent ink (Davidson Marking System;
Bradley Products, Bloomington, MN), dissected with a scalpel,
then embedded in paraffin in such a way that the crypts could be
sectioned longitudinally. Sample sections (4 mm) were mounted
on electrostatic-treated slides (Superfrost Plus, Medite, Italy) and
processed for immunohistochemistry as described before18using
the primary antibody against b-catenin (Transduction Laborato-
ries, Lexington, KY) diluted 1:1,000. The slides were weakly
counterstained with Harris’ hematoxylin and observed under a
microscope to evaluate the cellular localization of b-catenin. Each
slide presented in the same section both normal mucosa and the
lesions (ACF, MDF, or tumor). Starting from the normal mucosa,
b-catenin localization was evaluated, recording for each cell
whether the staining was localized at the cell membrane (in this
case the cell was classified as positive at the membrane, as in the
vast majority of normal mucosa cells), at the cytoplasm (positive
at the cytoplasm), or at the nucleus (positive at the nucleus).
Determination of b-catenin localization in the normal mucosa was
carried out in crypts separated from the lesion by an area occupied
by at least 3 crypts, counting, when possible, at least 100 cells.
The same evaluation was then carried out for the cells of MDF or
ACF present in the same slide. Evaluation was performed at
1,000? magnification and the slides were coded. Results were
expressed as the number of cells positive at a given cellular dis-
trict/total number of cells scored ? 100 and expressed as % of cel-
lular localization. A mean of 432 cells were scored for each slide
(range, 171–914 cells; the minimum number of cells scored was
100 for normal mucosa and tumors and 70 for MDF or ACF).
Laser capture microdissection of lesions
Longitudinal sections (5 mm) from paraffin-embedded MDF,
ACF identified in HID-AB-stained colon and tumors were
mounted on apposite slides (PALM Membrane Slides) to allow
easy cut and catapulting of the sample during laser microdissec-
tion (PALM MicroBeam system). Slides were placed under the
microscope and areas containing epithelial cells from normal-
appearing crypts, MDF, ACF, or tumors were selected at 400?
(1–3 sections per lesion) and finally laser-microdissected (PALM
MicroBeam system) and catapulted to the top of a 0.5 ml tube con-
taining 5 ml of buffer ATL (Qiamp DNA Micro Kit; Qiagen,
Chatsworth, CA). DNA was then extracted using Qiamp DNA
Micro Kit (Qiagen) according to manufacturer’s instructions.
Exon 2 of the Ctnnb1 gene, comprising regulatory phosphoryla-
tion sites important for the degradation of the protein, was ampli-
fied by nested PCR to obtain enough product for sequencing. The
first round of PCR amplification was carried out using as forward
AGGCA19and RCATR primer GCTACTTGCTCTTGCGTG-
AA,20respectively. These 2 primers produce a 277 bp DNA
segment, which was then reamplified using nested primers:
RCATF primer GCTGACCTCATGGAGTTGGA20and R3 pri-
mer TCCACATCCTCTTCCTCAGG19as forward and reverse
primers, respectively. PCR reactions were carried out in a 25 ml
volume containing 1 ? PCR buffer, 1.5 mM MgCl2, 0.3 mM
dNTPs, 0.3 mM of each primers and 1.25 units of Taq polymerase
(Advanced Biotechnologies, Epsom, U.K.). Cycling conditions
were initial degradation for 5 min at 948C, 35 cycles of denatura-
tion for 45 sec at 948C, annealing for 45 sec at 578C and extension
728C for 2 min with a final extension step for 7 min at 728C. Final
products were separated on a 1.6% agarose gel and visualized by
ethidium bromide staining. Final PCR products (161 bp) were
purified using the Wizard SV Gel and PCR Clean-Up System
(Promega, Madison, WI) and directly sequenced in both directions
using the ABI Prism BigDye Terminator Cycle Sequencing Ready
Reaction Kit (Applied Biosystems, Foster City, CA); sequence
was analyzed by DNA sequencer (ABI Prism 310 Genetic Ana-
lyzer; Applied Biosystems) following the manufacturer’s protocol.
Differences between groups were analyzed with one-way
ANOVA; contrasts between means were calculated with Duncan’s
method for multiple comparisons. Difference between mutation
frequencies were calculated with Fisher’s exact test. The correla-
tions between time and MDF or ACF multiplicities were calcu-
lated with a linear regression model. Calculations were performed
using the Statgraphics Statistical Package (Statistical Graphic,
Rockville, MD). Differences were considered significant when
p was ? 0.05.
FEMIA ET AL.
Sequential analysis of MDF and ACF
Rats treated twice with 150 mg/kg of DMH were sacrificed 5, 9,
13 and 28 weeks after the first carcinogen injection. Before pro-
ceeding to MDF determination, colons were stained with MB to
determine ACF according to Bird.2The number of ACF/colon
(Fig. 1a) increased significantly (p < 0.01) after 5 weeks. ACF
multiplicity (the number of crypts forming each ACF) also
increased linearly with time (r ¼ 0.88; p < 0.0001; Fig. 1b).
Colons were then stained with HID-AB and MDF were deter-
mined. MDF increased up to 13 weeks from treatment (Fig. 1d).
However, 28 weeks after DMH, when macroscopic colon tumors
were present, the number of MDF/colon (5.87 6 1.23, mean 6
SE) was significantly lower than at 13 weeks (9.00 6 0.73,
mean 6 SE; p < 0.05). By contrast, the multiplicity of MDF
(Fig. 1e) linearly increased with time (r ¼ 0.81; p < 0.0001).
Small MDF were more frequent at earlier time points: the percent
of MDF with a multiplicity equal to 2 over the total number of
MDF was in fact 17%, 3%, 1% and 0% at 5, 9, 13 and 28 weeks,
respectively. The number of colonic tumors/rat 28 weeks after
DMH was 1.87 6 0.35 (mean 6 SE; n ¼ 8 rats). One tumor was
an adenoma; all the others were adenocarcinomas in situ.
Dose-related induction of MDF and ACF
Increasing dose of DMH from 200 to 300 mg/kg produced
no difference in ACF/colon (Fig. 1c), while at higher doses
(400 mg/kg) an increase was observed; the multiplicity of
ACF was not varied (data not shown). By contrast, MDF/
colon were linearly correlated with the dosage of carcinogen
administered (Fig. 1f), but not MDF multiplicity (crypts/MDF
were 9.8 6 1.7, 8.9 6 2.3 and 8.7 6 1.9, in rats treated
with 200, 300 and 400 mg/kg DMH; mean 6 SD).
Distribution of MDF, ACF and tumors along proximal, mid
and distal colon
MDF and tumors were similarly distributed along the colon,
both lesions being more prevalent along the distal part (Fig. 2).
By contrast, ACF were frequent in the mid colon and their relative
distribution at this location (i.e., the % of ACF in mid colon/total
FIGURE 1 – (a, b, d, e) ACF and MDF in rats treated with DMH (150 mg/kg ? 2; total dose, 300 mg/kg) and sacrificed at different times after
carcinogen. Values are mean 6 SE; n ¼ 7 (5 or 9 weeks) and 8 (13 or 28 weeks). (a) ACF/colon. (b) Multiplicity of ACF (number of crypts
forming each focus). (d) MDF/colon. (e) Multiplicity of MDF expressed as in (b). Points with a different superscript are significantly different
according to ANOVA: p < 0.01 in (a) and (d); p < 0.0001 in (b) and (e). (c, f) ACF and MDF in rats treated with different doses of DMH and
sacrificed after 13 weeks. Values represent mean 6 SE; n ¼ 8.
FIGURE 2 – Distribution of tumors, MDF and ACF along the colon.
The relative distribution of all lesions is shown (% of lesions in each
segment of the colon/total number of lesions in the whole colon):
proximal colon (black segments), mid colon (gray segments) or distal
colon (white segments). For each segment of the colon, values of dis-
tribution with a different letter are significantly different: p < 0.01 by
MUCIN-DEPLETED FOCI IN RAT COLON
ACF in the colon) was significantly different (p < 0.01) from that
of MDF or tumors (Fig. 2). No tumors were present in the proxi-
mal part of the colon, whereas some MDF and ACF were detect-
able there; however, fewer MDF were present in the proximal
colon (p < 0.05) relative to ACF (1.4% 6 3.6% and 8.7% 6 6.5%
for MDF and ACF, respectively; mean 6 SD).
b-catenin cellular localization
The cellular localization of b-catenin was studied in longitu-
dinal sections of MDF (17 samples; mean size, 0.067 mm2)
and ACF (21 samples; mean size, 0.096 mm2) identified in
HID-AB-stained colon from rats treated with 150 mg/kg ? 2
of DMH and sacrificed after 13 weeks. A correlation study car-
ried out in 31 rats showed that the number of ACF identified
in HID-AB-stained colons is significantly correlated with the
number of ACF determined in the same colons stained with
MB (correlation coefficient, r ¼ 0.83; slope of the straight line,
a ¼ 1.07; p < 0.001). Nine tumors harvested 28 weeks after
DMH were also evaluated for b-catenin localization.
While in normal mucosa b-catenin was almost exclusively
expressed at the cell membrane, MDF showed a reduced membrane
staining concomitant with an increase in cytoplasmic and nuclear
staining for b-catenin (Fig. 3a). In particular, while the membrane
localization of b-catenin in normal mucosa was present in 98% of
the cells, only 28% of cells in MDF showed membranous localiza-
tion (p < 0.001 when compared to normal mucosa; Fig. 4). In ACF,
membranous localization of b-catenin was also lower (p < 0.001)
than in normal mucosa, and, as expected, a dramatic reduction in
membranous b-catenin was observed in tumors (Fig. 4). The local-
ization of b-catenin in the cytoplasm was increased (p < 0.001) in
MDF and tumors and, to a lesser extent, in ACF. Nuclear localiza-
tion was increased in MDF and tumors (p < 0.001), but not in ACF
FIGURE 3 – (a) Expression of b-catenin in paraffin section of an MDF. In the normal crypt (black arrow), b-catenin is localized almost exclu-
sively at the border of the cells, while in MDF the membranous localization is reduced, and cytoplasmatic and nuclear (white arrows) localization
are evident (original magnification, 400?). (b) Appearance of an MDF (white arrow) in MB-stained colon; note that the MDF does not stand out
over normal mucosa. (c) The same MDF shown in (b) after staining with HID-AB. Original magnification: 100?.
FIGURE 4 – Diagram analysis of the scoring of cellular localization
of b-catenin in ACF identified in HID-AB colons, MDF, tumors and
their paired normal mucosa as detected by immunohistochemistry.
Bars represent the number of cells with localization at that particular
cell district/total number of cells scores ? 100. Values are mean 6
SE. For each cellular district, bars with different superscript are signif-
icantly different (p < 0.001 with ANOVA).
FEMIA ET AL.
b-catenin gene mutation analysis
b-catenin (Ctnnb1) gene mutations were studied in 15 colonic
tumors (28 weeks after DMH), in 28 MDF (mean size,
0.098 mm2) and 27 ACF identified in HID-AB-stained colon
(mean size, 0.096 mm2) from rats treated with 150 mg/kg ? 2 of
DMH and sacrificed 13 weeks after carcinogen. Six out of 15
tumors (40%) had mutations in the Ctnnb1 gene. Four of these
mutations were in codon 32 (GAT?AAT causing an Asp to
become Asn), one mutation was in codon 34 (GGA?GAA; Gly?
Glu) and one in codon 45 (TCC?TTC; Ser?Phe). The only
adenoma present was mutated (codon 34).
The results relative to MDF (Table I) show that 7/28 MDF
(25%) examined had mutations in the Ctnnb1 gene, a value not
different from the frequency found in tumors (p ¼ 0.33 with
Fisher’s exact test). Two out of 27 ACF (7%) were mutated; this
frequency was significantly lower than that found in tumors (p ¼
0.02 with Fisher’s exact test), while the difference with MDF did
not attain statistical significance (p ¼ 0.14 with Fisher’s exact
test). Moreover, while all the mutations harbored by MDF caused
amino acid substitution, 1 of the 2 mutations observed in ACF was
a silent mutation without amino acid substitution. Two of the
mutated MDF harbored 2 mutations each (codons 32 and 37 and
codons 32 and 33); therefore, the mean number of mutation/
lesions was 0.40 6 0.13, 0.32 6 0.11 and 0.07 6 0.05 (mean 6
SE) in tumors, MDF and ACF, respectively. The number of muta-
tions in tumors and ACF was significantly different (p < 0.05 with
ANOVA test), while the difference between MDF and ACF was
borderline significant (p ¼ 0.058).
MDF in methylene blue-stained colons
The appearance of MDF was studied in 9 MB-stained colons.
We identified 28 lesions that could have been MDF. Some of these
suspected MDF appeared as a spot not perfectly in focus
(see example in Fig. 3b), thus suggesting a microscopic elevation.
However, even focusing, the appearance of these lesions was not
always sharp and the crypt openings were not as evident as in nor-
mal crypts and do not always stain well with MB (Fig. 3b), thus
making difficult their identification. When the suspected MDF
(n ¼ 28) were restained with HID-AB and compared with the cor-
responding field in MB-stained colon, they were confirmed as
MDF (Fig. 3c). Beside these 28 MDF, we also identified another
18 MDF, up to a total of 46. Therefore, 39% (18/46) of all the
MDF were missed with MB staining.
MDF appear in the colon of DMH-treated rats 5 weeks after the
first treatment with DMH and increase in number up to 13 weeks;
at 28 weeks, when macroscopic tumors appear, they decrease.
Regression of preneoplastic lesions during carcinogenesis has
been reported previously in the colon for ACF,21,22but also in the
liver.23The lower number of MDF after 28 weeks might thus be
explained with a remodeling of some of the lesions into normal
crypts. However, considering that at this time macroscopic tumors
were also present, one might speculate that the reduction of MDF
number was due to their transformation into more advanced
It is also interesting to note that the multiplicity of MDF (i.e., the
number of crypts forming each focus), continued to increase over
time. Conversely, small MDF, formed by only 2 crypts, tend to
decrease over time and, after 9–13 weeks from carcinogen adminis-
tration, constitute a negligible fraction of total MDF (1–3%). The
increase in growth over time we already observed in AOM-induced
MDF7had been previously reported in ACF and in other purported
preneoplastic lesions such as b-catenin-accumulated crypts.5,24
Accordingly, during carcinogenesis, potentially preneoplastic crypts
expand and form multicryptal lesions that evolve into adenoma and
We also found that the number of MDF increased with the dos-
age of carcinogen administered. By contrast, the multiplicity of
MDF and ACF was not affected by the dose of DMH, a result in
agreement with McLellan et al.6in ACF induced by a single dose
MDF and tumors were more prevalent in the distal part of the
colon, while ACF were found more frequently in the mid colon.
Depending on the protocol used, the distribution of ACF and
tumors along the colon may differ.15,27–29For instance, high
dosages of DMH (20 mg ? 20 weeks) induce tumors predomi-
nantly in the mid part of the colon.28By contrast, a prevalence of
tumors in the distal part of the colon (as reported in this study)
was described by Glauert and Week29in rats treated with a single
dose of DMH or in mice treated with multiple DMH injections.15
In this latter work, the distribution of tumors was different from
ACF, a result that was used to disprove the correlation between
ACF and tumorigenesis. Accordingly, one may argue that MDF
and tumors, which in our study have a similar localization, are
Recently, Yoshimi et al.30reported that it is possible to identify
MDF in rat colon stained with an alcian blue (AB) solution that
highlights mucin production. The method is similar to the HID-
AB staining procedure that we7,8and others9use to identify MDF;
this method only omits the preliminary staining of the colon with
an high-iron diamine (HID) solution, a step that makes it possible
to distinguish between sialomucin and sulfomucins production.31
Although it is possible that Yoshimi’s method will give satisfying
results in the future, we think that more validation experiments
should be performed before adopting it to identify MDF. In fact, a
considerable fraction (about 24%) of the lesions identified as
MDF by these authors turned out to be normal-like crypts at the
histology.30It is possible, however, that the remaining 76% of the
MDF described by Yoshimi et al.30with dysplastic features at his-
tology are the same lesions as the MDF that we identified with the
We also determined the localization of b-catenin and the status
of its gene (Ctnnb1 gene) in MDF, in ACF identified in HID-
AB-stained colons and in tumors. In normal mature colonocytes,
b-catenin is a component of the cadherin-mediated cell adhesion
system and is localized mainly at the cellular membrane. b-catenin
TABLE I – b-CATENIN MUTATIONS IN MDF AND ACF IDENTIFIED IN HID-AB-STAINED COLON
Lesion DNAAmino acid Frequency1
MDFCodon 32 (GAT?AAT) and 37 (TCT?TTT)
Codon 32 (GAT?AAT)
Codon 45 (TCC?TTC)
Codon 32 (GAT?AAT) and 33 (TCT?TTT)
Codon 41 (ACC?ATC)
Codon 45 (TCC?TTC)
Codon 40 (ACC?ACT)
1Values refer to the number of lesions with a specific DNA sequence over the total number of MDF or
MUCIN-DEPLETED FOCI IN RAT COLON
is also a member of the APC/b-catenin/TCF/Lef pathway (Wnt
pathway). In embryonic colonocytes following Wnt signals,
b- catenin level rises and b-catenin binds to members of the TCF
family of transcription factors, activating gene transcription.10
Activation of this pathway, with increased levels of b-catenin in
the cytoplasm and nucleus, has been documented in both adeno-
mas and carcinomas,12,32–34but also in human ACF.11b-catenin is
also activated during hyperproliferation of colonic mucosa,35,36in
dysplastic ACF in Min/þ mice34and in b-catenin-accumulated
crypts in AOM-treated rats.19Recently, Yoshimi et al.30reported
that about 74% of MDF identified with the AB method dis-
cussed above showed b-catenin accumulation in cytoplasm or
nucleus. Here we also show that the localization of b-catenin in
the cytoplasm and nucleus is markedly increased in MDF as
well as in tumors. Moreover, in the present study, we also docu-
ment a concomitant decrease in membranous b-catenin, results
that further suggest the involvement of MDF in carcinogenesis.
Due to the alteration in b-catenin localization in MDF, one
could literally define these lesions as b-catenin-accumulating
crypts (BCAC). In fact, it is possible that MDF and BCAC are
very similar lesions.
Alteration in the cellular localization of b-catenin has been
associated with a constitutive activation of the Wnt signaling path-
way, leading to increased transcription of oncogenes such as
cyclin D1 and c-myc.10Since these genes regulate cell growth, it
is possible that the constitutive activation of Wnt pathway in MDF
may lead to an increased proliferation, as observed during colon
carcinogenesis. It would also be possible that the activation of
Wnt pathway leads to the depletion of mucins. Accordingly, it has
been reported that abrogation of Wnt-signaling in colon cancer
cell lines upregulates MUC2, the most abundant secreted apomu-
cin in the colon.37
The localization of b-catenin was also determined in ACF iso-
lated in HID-AB-stained colon. Our previous work7,31and data
presented here indicate that ACF identified in HID-AB colon and
ACF as seen in MB-stained colon are strongly correlated. We
found that ACF isolated in HID-AB, though to a lesser extent than
MDF, have altered expression of b-catenin, in agreement with
data on human ACF,11although some authors did not find altera-
tion in b-catenin localization in rodent ACF.19
Activation of the Wnt pathway has been causally related to
mutations in the b-catenin gene, in particular in the GSK-3b phos-
phorylation consensus region of Ctnnb1 gene, which is important
for phosphorylation and degradation of b-catenin in the cyto-
plasm. Accordingly, 40% (6 out of 15) of the tumors analyzed by
us were mutated in this region of the Ctnnb1 gene. The mutations
we found affected important sites of phosphorylation, such as
codon 45, codifying a serine residue, or codons 32 or 34, which
flank codon 33, also codifying a serine.38,39The same mutations
found by us in colon tumors (i.e., at codons 32 and 45) were also
identified in MDF. In addition, MDF had mutations at codons 33,
37 and 41, codifying serine and threonine residues important for
GSK phosphorilation, and also found in colon tumors.20,39,40
Therefore, as seen for colon tumors in this and other studies,39all
the mutations we found in MDF converged at codons encoding
functionally important residues for b-catenin protein degradation,
thus indicating that these alterations are tumorigenic and that
MDF are precursor lesions of tumors. Two MDF harbored double
mutations; both MDF were mutated in codon 32. In addition,
one of these lesions was mutated in codon 37, while the other
one was mutated in codon 33. Double mutations in Ctnnb1 gene
were reported previously in colon cancer.40Also, 2/16 (7%)
ACF were mutated, but only 1 of these mutations caused an
amino acid substitution, the other 1 being a silent mutation
(codon 40 ACC?ACT).
Mutations in rat Ctnnb1 gene have been reported with variable
frequency during colon carcinogenesis. A high frequency
is observed inexperimentally
(36–69%)20,39,41and a lower frequency in b-catenin accumulated
crypts or ACF (46–15%).19,39,41In the present study, we found a
slightly lower frequency of mutations in MDF and a very low fre-
quency in ACF.
However, while we found that the majority of tumors and MDF
show altered cellular localization of b-catenin, only 40% of these
same tumors and even a lower percent (25%) of MDF harbored
mutations in the Ctnnb1 gene. A similar discrepancy between
Ctnnb1 status and altered cellular localization of its protein has
been reported in tumors and preneoplastic lesions,19,20indicating
that other components of the APC/b-catenin/TCF/Lef pathway
may be mutated or deregulated in these lesions.
Finally, we showed that MDF are not easily observed in MB-
stained colons since about 40% of MDF are missed with this stain-
ing. This fact, together with the abundance of the mucous layer
present in human colons, might render difficult a possible visual-
ization of MDF during colonoscopy in vivo.
In conclusion, MDF grow and enlarge over time, are dose-
related, may harbor mutations in regulatory sites of b-catenin,
and show marked alteration in b-catenin cellular localization.
All these data strongly suggest that MDF are specific precan-
cerous lesions and an excellent tool for the study of colon
The authors thank M. Beni and P. Ceccatelli for expert technical
assistance in the management of the rats and Mary Forrest for
1.Chang WWL. Histogenesis of colon cancer in experimental animals.
Scand J Gastroenterol 1984;19:27–43.
Bird RP. Observation and quantification of aberrant crypts in the
murine colon treated with a colon carcinogen: preliminary findings.
Cancer Lett 1987;37:147–51.
Pretlow TP, Barrow BJ, Ashton WS, O’Riordan MA, Pretlow TG,
Jurcisek JA, Stellato TA. Aberrant crypts: putative preneoplastic foci
in human colonic mucosa. Cancer Res 1991;51:1564–7.
Roncucci L, Stamp D, Medline A, Cullen JB, Bruce WR. Identifica-
tion and quantification of aberrant crypt foci and microadenomas in
the human colon. Hum Pathol 1991;22:287–94.
McLellan EA, Medline A, Bird RP. Sequential analyses of the growth
and morphological characteristics of aberrant crypt foci: putative pre-
neoplastic lesions. Cancer Res 1991;51:5270–4.
McLellan EA, Medline A, Bird RP. Dose response and proliferative
characteristics of aberrant crypt foci: putative preneoplastic lesions in
rat colon. Carcinogenesis 1991;12:2095–8.
Caderni G, Femia AP, Giannini A, Favuzza A, Luceri C, Salvadori
M, Dolara P. Identification of mucin-depleted foci in the unsectioned
colon of azoxymethane-treated rats: correlation with carcinogenesis.
Cancer Res 2003;63:2388–92.
8. Femia AP, Dolara P, Caderni G. Mucin-depleted foci (MDF) in the
colon of rats treated with azoxymethane (AOM) are useful bio-
markers for colon carcinogenesis. Carcinogenesis 2004;25:277–81.
Pierre F, Freeman A, Tache ´ S, Van der Meer R, Corpet DE. Beef
meat and blood sausage promote azoxymethane-induced mucin-
depleted foci and aberrant crypt foci in rat colons. J Nutr
10. Polakis P. The oncogenic activation of b-catenin. Curr Opin Genet
11. Hao XP, Pretlow TG, Rao JS, Pretlow TP. b-catenin expression is
altered in human aberrant crypt foci. Cancer Res 2001;61:8085–8.
12. Herter P, Kuhnen C, Muller KM, Wittinghofer A, Muller O. Intracel-
lular distribution of b-catenin in colorectal adenomas, carcinomas, and
Peutz-Jeghers polyps. J Cancer Res Clin Oncol 1999;125:297–304.
13. Corpet DE, Tache ´ S. Most effective colon cancer chemopreventive
agents in rats: a review of aberrant crypt foci and tumour data, ranked
by potency. Nutr Cancer 2002;43:1–21.
14. Zheng Y, Kramer PM, Lubet RA, Steele VE, Keloff GJ, Pereira MA.
Effect of retinoids on AOM-induced colon cancer in rats: modulation
of cell proliferation, apoptosis and aberrant crypt foci. Carcinogenesis
FEMIA ET AL.
15. Carter JW, Lancaster HK, Hardman WE, Cameron IL. Distribution of Download full-text
intestine-associated lymphoid tissue, aberrant crypt foci, and tumours
in the large bowel of 1,2-dimethylhydrazine-treated mice. Cancer Res
16. European Community. European Community regulations on the care
and use of laboratory animals: law 86/609/EC. Luxembourg: Euro-
pean Community, 1986.
17. Caderni G, De Filippo C, Luceri C, Salvadori M, Giannini A,
Biggeri A, Remy S, Cheynier V, Dolara P. Effects of black tea, green
tea and wine extracts on intestinal carcinogenesis induced by azoxy-
methane (AOM) in F344 rats. Carcinogenesis 2000;21:1965–9.
18. Femia AP, Luceri C, Dolara P, Giannini A, Biggeri A, Salvadori M,
Clune Y, Collins KJ, Paglierani M, Caderni G. Antitumorigenic activ-
ity of the prebiotic inulin enriched with oligofructose in combination
with the probiotics Lactobacillus rhamnosus and Bifidobacterium lac-
tis on azoxymethane-induced colon carcinogenesis in rats. Carcino-
19. Yamada Y, Yoshimi N, Hirose Y, Kawabata K, Matsunaga K,
Shimizu M, Hara A, Mori H. Frequent b-catenin gene mutations and
accumulation of the protein in the putative preneoplastic lesions lack-
ing macroscopic aberrant crypt foci appearance, in rat colon carcino-
genesis. Cancer Res 2000;60:3323–7.
20. Koesters R, Hans MA, Benner A, Prosst R, Boehm J, Gahlen J,
Doeberitz MK. Predominant mutation of codon 41 of the beta-
catenin proto-oncogene in rat colon tumors induced by 1,2-dimethyl-
hydrazine using a complete carcinogenic protocol. Carcinogenesis
21. Magnusson BA, Carr I, Bird RP. Ability of aberrant crypt foci charac-
teristics to predict colonic tumor incidence in rats fed cholic acid.
Cancer Res 1993;53:4499–504.
22. Shpitz B, Hay K, Medline A, Bruce WR, Bull SB, Gallinger S,
Stern H. Natural history of aberrant crypt foci, a surgical approach.
Dis Colon Rectum 1996;39:763–7.
23. Pitot HC, Dragan YP. Chemical carcinogenesis. In: Klaasen CD,
ed. Casarett and Doull’s toxicology: the basic science of poisons.
New York: McGraw-Hill, 2001. 241–320.
24. Yamada Y, Yoshimi N, Hirose Y, Matsunaga K, Katayama M,
Sakata K, Shimizu M, Kuno T, Mori H. Sequential analysis of
morphological and biological properties of b-catenin accumulated
crypts, provable premalignant lesions independent of aberrant
crypt foci in rat colon carcinogenesis. Cancer Res 2001;61:
25. Preston SL, Wong WM, Chan AO, Poulsom R, Jeffery R, Goodlad
RA, Mandir N, Elia G, Novelli M, Bodmer WF, Tomlinson IP, Wright
NA. Bottom-up histogenesis of colorectal adenomas: origin in the
monocryptal adenoma and initial expansion by crypt fission. Cancer
26. Shih IM, Wang TL, Traverso G, Romans K, Hamilton SR,
Ben-Sasson S, Kinzler KW, Vogelstein B. Top down morphogenesis
of colorectal tumors. Proc Natl Acad Sci USA 2001;98:2640–5.
27. Tudek B, Bird RP, Bruce WR. Foci of aberrant crypts in the colons of
mice and rats exposed to carcinogens associated with foods. Cancer
28. Maskens AP. Histogenesis and growth pattern of 1,2-dimethyl-
hydrazine-induced rat colon adenocarcinoma. Cancer Res 1976;36:
29. Glauert HP, Week JA. Dose- and time-response of colon carcinogene-
sis in Fisher-344 rats after a single dose of 1,2-dimethylhydrazine.
Toxicol Lett 1989;48:283–7.
30. Yoshimi N, Morioka T, Kinjo T, Inamine M, Kaneshiro T, Shimizu
T, Suzui M, Yamada Y, Mori H. Histological and immunohistochem-
ical observations of mucin-depleted foci (MDF) stained with Alcian
blue, in rat colon carcinogenesis induced with 1,2-dimethylhydrazine
dihydrochloride. Cancer Sci 2004;95:792–7.
31. Caderni G, Giannini A, Lancioni L, Luceri C, Biggeri A, Dolara P.
Characterisation of aberrant crypt foci in carcinogen-treated rats:
association with intestinal carcinogenesis. Br J Cancer 1995;71:
32. Brabletz T, Herrmann K, Jung A, Faller G, Kirchner T. Expression of
nuclear b-catenin and c-myc is correlated with tumor size but not
with proliferative activity of colorectal adenomas. Am J Pathol 2000;
33. Hao X, Tomlinson I, Ilyas M, Palazzo JP, Talbot IC. Reciprocity
between membranous and nuclear expression of b-catenin in colorec-
tal tumours. Virch Arch 1997;431:167–72.
34. Paulsen JE, Steffensen IL, Loberg EM, Husoy T, Namork E,
Alexander J. Qualitative and quantitative relationship between dys-
plastic aberrant crypt foci and tumorigenesis in the Min/þ mouse
colon. Cancer Res 2001;61:5010–5.
35. Iwamoto M, Ahnen DJ, Franklin WA, Maltzman TH. Expression of
b-catenin and full-length APC protein in normal and neoplastic
colonic tissues. Carcinogenesis 2000;21:1935–40.
36. Sellin JH, Umar S, Xiao J, Morris AP. Increased b-catenin expression
and nuclear translocation accompany cellular hyperproliferation
in vivo. Cancer Res 2001;61:2899–906.
37. Van de Wetering M, Sancho E, Verweij C, de Lau W, Oving I,
Hurlstone A, van der Horn K, Batlle E, Coudreuse D, Haramis AP,
Tjon-Pon-Fong M, Moerer P, et al. The beta-catenin/TCF-4 complex
imposes a crypt progenitor phenotype on colorectal cancer cells. Cell
38. Blum CA, Xu M, Orner GA, Fong AT, Bailey GS, Stoner GD,
Horio DT, Dashwood RH. b-catenin mutation in rat colon tumors ini-
tiated by 1,2-dimethylhydrazine and 2-amino-3-methylimidazo[4,5-
f]quinoline, and the effect of post-initiation treatment with chloro-
phyllin and indole-3-carbinol. Carcinogenesis 2001;22:315–20.
39. Yamada Y, Oyama T, Hirose Y, Hara A, Sugie S, Yoshida K,
Yoshimi N, Mori H. b-catenin mutation is selected during malignant
transformation in colon carcinogenesis. Carcinogenesis 2003;24:
40. Takahashi M, Fukuda K, Sugimura T, Wakabayashi K. b-catenin is
frequently mutated and demonstrates altered cellular location in
azoxymethane-induced rat colon tumors. Cancer Res 1998;58:42–6.
41. Takahashi M, Mutoh M, Kawamori T, Sugimura T, Wakabayashi K.
Altered expression of b-catenin, inducible nitric oxide synthase and
cyclooxygenase-2 in azoxymethane-induced rat colon carcinogenesis.
MUCIN-DEPLETED FOCI IN RAT COLON