original article
Annals of Oncology 20: 441–448, 2009
doi:10.1093/annonc/mdn651
Published online 8 January 2009
Increased g-tubulin expression and P16INK4A promoter
methylation occur together in preinvasive lesions and
carcinomas of the breast
T. Liu, Y. Niu*, Y. Yu, Y. Liu & F. Zhang
Breast Cancer Research Key Laboratory of National Education Ministry, Cancer Institute and Hospital, Tianjin Medical University, Tianjin, China
Received 31 July 2008; accepted 1 September 2008
abnormalities in variant human mammary epithelial cells. c-Tubulin is a highly conserved component of centrosome in
most animal cells and c-tubulin protein overexpression could lead to centrosome aberration.
Materials and methods: A large series of breast premalignant lesions and carcinoma was analyzed. Real-time
quantitative PCR and immunohistochemistry were carried out to measure c-tubulin copy numbers and protein
expression. MethyLight and immunohistochemistry were carried out to determine p16INK4A methylation and protein
expression.
Results: c-Tubulin protein expression was concordant with gene amplification; both of them were found to increase
with atypical ductal hyperplasia–carcinoma sequence. The median value and positive rate of p16INK4a methylation
increased while protein expression displayed a decreasing trend. P16INK4a methylation showed a firm association with
c-tubulin gene amplification.
Conclusion: c-Tubulin gene amplification and the concomitant protein overexpression present not only in invasive
carcinoma but also in a significant fraction of atypical hyperplasia and in situ carcinomas. P16INK4a methylation and
c-tubulin gene amplification had a synergistic effect on tumor progression. The synergism might arise as a result of the
combined influence that p16INK4a and c-tubulin have on the G1–S cell cycle checkpoints and centrosome.
Key words: ADH–carcinoma sequence, immunohistochemistry, p16INK4A, real-time quantitative PCR, c-tubulin
introduction
Cell cycle checkpoints preserve genome integrity by monitoring
the fidelity of DNA replication and segregation. It requires the
close cooperation of cell cycle regulatory proteins and
cytoskeletal elements to sense spindle integrity. P16INK4A,
a negative regulator of G1–S checkpoint of cell cycle, plays a key
role in cell cycle progression by binding to cyclin-dependent
kinase-4(CDK4) and CDK6 and inhibiting the catalytic activity
of the CDK4–CDK6/cyclinD complex required for
retinoblastoma protein phosphorylation. Forced expression
of p16INK4A protein can induce a G1 arrest, thereby preventing
the transcription of cell cycle progression genes. The
methylation of p16INK4A gene locus is an effective way of
modulating gene expression [1]. P16INK4A protein expression is
frequently altered to various degrees in human cancers through
promoter methylation [2, 3].
Hannah Müller et al. [4] found that the molecular
requirements for a functional spindle checkpoint included
*Correspondence to: Prof. Y. Niu, Breast Cancer Research Key Laboratory of National
Education Ministry, Cancer Institute and Hospital, Tianjin Medical University, Huan Hu Xi
Road, Ti Yuan Bei, He Xi District, Tianjin 300060, China. Tel: +86-22-23340123-5221;
Fax: +86-22-23359337; E-mail: [email protected]
c-tubulin, and c-tubulin-depleted cells fail to form functional
spindles and arrest during nuclear division. c-Tubulin is a kind
of cytoskeletal element that exists in a macromolecular complex
called the c-tubulin ring complex (c-TuRC). c-TuRC exists in
a mass of amorphous electron-dense matrix termed the
pericentriolar material (PCM), which is the key element of the
centrosome. A centrosome consists of a pair of perpendicularly
positioned barrel-shaped centrioles surrounded by PCM. In
many studies, c-tubulin was used to localize and quantify
centrosomes [5, 6], and c-tubulin protein overexpression could
lead to centrosome aberration [7].
Loss of p16INK4A activity due to methylation presented in
variant human mammary epithelial cells (vHMECs) and thus
resulted in supernumerary centrosomes through centriole pair
splitting. Then generation of supernumerary centrosomes
was shown to nucleate multipolar spindles and directly drive
production of aneuploid daughter cells [8, 9]. c-Tubulin is
a highly conserved component of centrosomes in most animal
cells and is essential for coordinating mitotic events [10, 11].
Despite the extensive number of studies on the p16INK4A and
c-tubulin, we know of no other study, to date, that has analyzed
these two centrosome-associated factors in breast cancer
ª The Author 2009. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: [email protected]
original
article
Background: Loss of p16INK4A due to promoter hypermethylation is correlated with the ability to acquire centrosomal
original article
tissues or observed the relationship between these two critical
factors.
Taken together, we extended our studies to analyze p16INK4A
methylation and c-tubulin expression in a large series of normal
breast tissue (NBT) and usual ductal hyperplasia (UDH),
atypical ductal hyperplasia (ADH), ductal carcinoma in situ
(DCIS), and infiltrative ductal carcinoma (IDC) samples of
breast. We expect to identify the molecular (causal) changes
that underlie progression of normal breast epithelial cells to
malignancy and with hopes that such information will provide
selective targets for effective treatment of breast cancer.
Annals of Oncology
manufacturer’s instructions. Quantitative DNA methylation analysis was
carried out through MethyLight, as previously described [13, 14].
GeneAmp5700 sequence detector was utilized to quantify p16INK4a
methylation. Primers and probes for p16INK4a [15] and b-actin [16] were
previously described, and b-actin was employed to normalize the amount of
input bisulfite-converted DNA. SssI methylase (New England Biolabs,
Peking, China)-treated human genomic DNA was utilized as a fully
methylated reference. The amount of methylated DNA (percentage of
methylated reference, PMR) [17] at p16INK4a was calculated by dividing the
p16INK4a: b-actin ratio of the sample by the p16INK4a: b-actin ratio of
SssI-treated human genomic DNA and multiplying by 100.
On the basis of the previously validated data [18], the p16INK4a gene was
considered methylated if the PMR value was >0.
materials and methods
tissue specimens
Tissue collection and analysis in this study were approved by the Ethical
Committee of Tianjin Medical University Tumor Hospital, China. All cases
of breast surgical specimens, formalin fixed and paraffin embedded were
anonymized after collection from the archival file of Breast Pathology
Department, Tianjin Tumor Hospital. Forty cases of NBT/UDH, 40 cases
of ADH, 60 cases of DCIS (24 low grade and 36 high grade), and 60 cases of
IDC (28 low grade and 32 high grade) were randomly selected from
1998 to 2006. The availability of enough tumor tissue in each case was
ensured for effective DNA isolation, excluding immunostaining. The
pathologic diagnosis was counterchecked by two senior pathologists
according to the 2003 World Health Organization histological classification
of breast tumors [12].
DNA isolation
Breast tissue samples utilized for real-time quantitative PCR (qPCR) and
methylation analysis were prepared by conducting microdissection of tissue
sections. Genomic DNA was isolated using an E.Z.N.A. Tissue DNA Kit
(Omega Bio-tek, Norcross GA), according to the manufacturer’s
instructions.
quantitative PCR
Specific primers were designed according to the published sequence of
c-tubulin gene in National Center for Biotechnology Information database
using primer design software programs Oligo 6.72. BLAST analysis was
carried out to exclude the influence of pseudogenes. c-Tubulin was
amplified using the forward (5#-GCG ATG CCG AGG GAA ATC ATC ACC
CT-3#) and reverse (5#-TGG GTG CCC CAG GAG ATG TAG TCT G-3#)
primers, which generate a fragment of 135 bp in length. a-Tubulin was used
as the external reference gene and was amplified with the following primers:
5#-TAT CGA GCG CCC AAC CTA CAC T-3# and 5#-CCT CAC CCT CTC
CTT CAA CAG AAT C-3# with generated fragments of 683 bp in length.
A GeneAmp 5700 Sequence Detection System (Applied Biosystems) was
used to perform qPCR amplification, PCR amplifications were carried out
in a final volume of 20 ll containing 10 ll SYBR premix (TaKaRa
Biotechnology Limited Company, Dalian, China), 0.8 ll of each primer,
and 80 ng of genomic DNA. Amplification condition was 94C for 5 min
then 40 cycles at 94C for 30s, 57C for 20s, and 72C for 20s. Negative
controls were carried out by omitting template or primers. Reactions
were carried out with different concentrations of one standard plasmid to
plot standard curve. Real-time PCR reactions were done in triplicate for
each sample, and absolute DNA quantification was calculated using the
standard curve method.
methylation analysis
The DNA was subjected to bisulfite treatment utilizing a Methylamp
DNA Modification Kit (Epigentek, Brooklyn, NY) according to the
442 | Liu et al.
immunohistochemistry
Tissue sections (5 lm) were deparaffinized and hydrated utilizing standard
procedures. Immunostaining was carried out using the super-sensitivity S-P
immunohistochemistry (IHC) kit. Anti-c-tubulin (Santa Cruz
Biotechnology, CA, USA) and anti-p16INK4a’s (Beking Zhongshan
Biotechnology Limited Company, China) primary antibody was applied to
the sections at a dilution ratio of 1 : 500 and 1 : 40, respectively. Antigen
retrieval has been done for anti-p16INK4a antibody. Positive and negative
controls were included in every batch.
evaluation of g-tubulin staining
Positive signals of c-tubulin proteins were located in the cytoplasm
(Figure 1A–F). At least 10 power fields were chosen per case and >500 cells
were counted for each power field. Scoring system was modified and
used according to evaluation standard [19]. The percentage of the staining
cells (P) was scored as follows: 0 (staining of <5% of cells), 1 (5%–25%
of cells), 2 (25%–50%), 3 (50%–75%), and 4 (75%–100%). Staining
intensity (I) was graded as follows: 0 (no staining), 1 (weak staining),
2 (moderate staining), and 3 (intense staining). Samples in each power
field were evaluated for both factors, i.e. P multiplied by I. The scoring
of each case was a mean value of chosen power fields. Eventually the
sections were graded as follows: (2) scoring 0–2, 1+ scoring 2–5, 2+
scoring 5–8, and 3+ scoring >8.
evaluation of p16INK4a staining
Positive signals of p16INK4a proteins [20, 21] were located in the nuclei
(Figure 2A–F). At least 10 fields in each specimen were randomly selected
and examined under high-power magnification, and >500 cells were
counted to determine the percentage of positive cells. On the basis of the
criteria Geradts et al. [22, 23] established, with minor modification, the case
wherein the percentage of positive cells was ‡10% was considered as
p16INK4a positive.
statistical analysis
The data analysis was carried out with the SPSS13.0 software package.
One-way analysis of variance was used to compare quantitative data
in six groups. Tamhane and Games-Howell were used to do multiple
comparisons. Cochran–Mantel–Haenszel was used to analyze the
correlation. Chi-square tests were employed to analyze categorical data. All
P values were two sided, and statistical significance was set at P = 0.05.
results
IHC analysis of g-tubulin protein
In NBT/UDH, the scoring of c-tubulin staining showed
a mean value of 1.685 (0–6.4). In ADH, an increase to 3.34
(0–11.4) was observed. These changes were even more obvious
Volume 20 | No. 3 | March 2009
Annals of Oncology
original article
Figure 1. The expression of c-tubulin protein in the various lesions of the breast. (A) Negative expression of c-tubulin protein in normal breast tissue/usual
ductal hyperplasia (IHC-SP method ·100); (B) positive expression of c-tubulin protein observed as brown–yellow color in the cytoplasm in atypical ductal
hyperplasia (IHC-SP method ·100); (C) positive expression of c-tubulin protein observed as brown–yellow color in the cytoplasm in low-grade DCIS (IHCSP method ·100); (D) positive expression of c-tubulin protein observed as brown–yellow color in the cytoplasm in high-grade ductal carcinoma in situ
(IHC-SP method ·100); (E) positive expression of c-tubulin protein observed as brown–yellow color in the cytoplasm in low-grade infiltrative ductal
carcinoma (IDC) (IHC-SP method ·100); (F) positive expression of c-tubulin protein observed as brown–yellow color in the cytoplasm in high-grade IDC
(IHC-SP method ·200).
in low-grade ductal carcinoma in situ (LDCIS) and highgrade ductal carcinoma in situ (HDCIS), with a mean value
of 3.917 (0.2–12) and 4.594 (0.4–12), and low-grade
infiltrative ductal carcinoma (LIDC) and high-grade
infiltrative ductal carcinoma (HIDC), with a mean value of
5.036 (0.4–12) and 6.613 (1.2–12) (Figure 3).
Statistical analysis of the scoring of c-tubulin staining
revealed a significant differences among six groups
(P = 0.000). Compared with NBT/UDH, the other groups
showed statistical significance (P < 0.05). Besides, a significant
alteration occurred between (i) ADH and HIDC and
(ii) LDCIS and HIDC.
Also the positive rate of c-tubulin staining increased with the
NBT/UDH to ADH to DCIS to IDC sequence, and this increase
showed statistical significance (v2 = 25.381, P = 0.000)
(summarized in Table 1).
Volume 20 | No. 3 | March 2009
DNA copy numbers for g-tubulin
The measurement of DNA copy numbers in the six groups were
shown in Table 2. The mean value of DNA copy numbers
increased 1.64 times from NBT/UDH to ADH, 2.63 times
from NBT/UDH to LDCIS, 2.95 times from NBT/UDH to
HDCIS, 3.88 times from NBT/UDH to LIDC, and 4.93 times
from NBT/UDH to HIDC, and this increase was significant
(F = 34.596, P = 0.000). By multiple comparisons, there were
significant differences between (i) NBT/UDH and any other
groups, (ii) ADH and HDCIS, (iii) ADH and LIDC, (iv) ADH
and HIDC, (v) LDCIS and HIDC, and (vi) HDCIS and
HIDC. Most interestingly, different histological grade within
DCIS and IDC did not show statistical significances (P = 0.944
for LDCIS and HDCIS, P = 0.063 for LIDC and HIDC).
If the mean value of DNA copy number in NBT/UDH was
used as cut-off value, the percentage of cases with increased
doi:10.1093/annonc/mdn651 | 443
original article
Annals of Oncology
Figure 2. The expression of p16INK4a protein in the various lesions of the breast. (A) Positive expression of p16INK4a protein observed as brown–yellow color
in the nucleus in normal breast tissue/usual ductal hyperplasia (IHC-SP method ·100), (B) positive expression of p16INK4a protein observed as brown–
yellow color in the nucleus in atypical ductal hyperplasia (IHC-SP method ·200); (C) positive expression of p16INK4a protein observed as brown–yellow
color in the nucleus in low-grade ductal carcinoma in situ (DCIS) (IHC-SP method ·100); (D) positive expression of p16INK4a protein observed as brown–
yellow color in the nucleus in high-grade DCIS (IHC-SP method ·100); (E) positive expression of p16INK4a protein observed as brown–yellow color in the
nucleus in low-grade infiltrative ductal carcinoma (IDC) (IHC-SP method ·100); (F) negative expression of p16INK4a protein observed as brown–yellow
color in the nucleus in high-grade IDC (IHC-SP method ·100).
DNA copy numbers (c-tubulin gene amplification rate) was
30.0% in NBT/UDH. A sharp elevation to 55.0% was found in
ADH, with continuous increase to 62.5% in LDCIS, 72.2% in
HDCIS, 78.6% in LIDC, and 84.4% in HIDC (Figure 4). Similar
to DNA copy numbers, c-tubulin gene amplification rate showed
differences among six groups (v2 = 29.880, P = 0.000).
Moreover, c-tubulin gene amplification rate was positively
correlated with the positive rate of protein expression with the
NBT/UDH to ADH to DCIS to IDC sequence (r = 0.377,
P = 0.000) (Figure 5). Similarly, c-tubulin DNA copy
numbers were positively correlated with the scoring of
c-tubulin staining (Figure 6).
(Figure 7). It increased with the NBT/UDH to ADH to DCIS to
IDC sequence. There were significant differences among six
groups (F = 40.189, P = 0.000). By multiple comparisons,
statistical difference was found (i) between NBT/UDH and any
other groups, (ii) between ADH and HDCIS, LIDC, and HIDC,
(iii) between LDCIS and HDCIS, LIDC, and HIDC, and
(iv) between HDCIS and HIDC.
A total of 3 (7.5%) of 40 UDH, 14 (35%) of 40 ADH,
12 (50%) of 24 LDCIS, 24 (66.7%) of 36 HDCIS, 20 (71.4%) of
28 LIDC, and 24 (75%) of 32 HIDC were deemed methylated
by MethyLight, with significant differences found among six
groups (v2 = 49.507, P = 0.000).
MethyLight
The median PMR value showed 0.20 (0–4) in NBT/UDH, 5.48
(0–20) in ADH, 15.21 (0–36) in LDCIS, 32.83 (0–67) in
HDCIS, 44.61 (0–79) in LIDC, and 65.59 (0–98) in HIDC
IHC analysis of p16INK4a protein
P16INK4a protein expression was detected in the samples of
32 NBT/UDH, 19 ADH, 8 LDCIS, 11 HDCIS, 7 LIDC, and
6 HIDC, all of which were devoid of p16INK4a methylation
444 | Liu et al.
Volume 20 | No. 3 | March 2009
original article
Annals of Oncology
Figure 3. Summary of the numerical changes of c-tubulin staining along
the atypical ductal hyperplasia (ADH)–carcinoma sequence of breast
carcinoma. Box plots depict the score of c-tubulin staining for normal
epithelial cells, ADH, low- and high-grade ductal carcinoma in situ and
invasive carcinoma. Each individual sample is indicated by a dot, minimum
and maximum data points are indicated (bar, linked to the box by a line),
and one statistical ‘outsider’ (ADH) is shown by the closed circle.
Figure 4. c-Tubulin gene amplification rate in six groups.
Table 1. The positive rate of c-tubulin staining in six groups
Group
Cases
c-tubulin
2
+
Staining
++
+++
Positive
rate (%)
NBT/UDH
ADH
LDCIS
HDCIS
LIDC
HIDC
Total
40
40
24
36
28
32
200
26
17
9
10
7
4
73
1
7
6
10
9
15
48
35.0
57.5
62.5
72.2
75.0
87.5
63.5
13
14
8
12
9
5
61
0
2
1
4
3
8
18
Figure 5. The positive correlation of c-tubulin gene amplification rate and
c-tubulin protein-positive rate.
NBT/UDH, normal breast tissue/usual ductal hyperplasia; ADH, atypical
ductal hyperplasia; LDCIS, low-grade ductal carcinoma in situ; HDCIS,
high-grade ductal carcinoma in situ; LIDC, low-grade infiltrative ductal
carcinoma; HIDC, high-grade infiltrative ductal carcinoma.
Table 2. Summary of c-tubulin DNA copy numbers within the
ADH–carcinoma sequence
Group
Mean values of
DNA copy number
(·106/ng)
Range of DNA
copy number
(·106/ng)
Significance (P)
(relative to
normal)
NBT/UDH
ADH
LDCIS
HDCIS
LIDC
HIDC
1.84
3.01
4.84
5.43
7.13
9.07
1.78–2.26
1.80–7.90
1.80–9.62
1.80–10.24
1.82–14.72
1.83–14.89
0.000
0.000
0.000
0.000
0.000
NBT/UDH, normal breast tissue/usual ductal hyperplasia; ADH, atypical
ductal hyperplasia; LDCIS, low-grade ductal carcinoma in situ; HDCIS,
high-grade ductal carcinoma in situ; LIDC, low-grade infiltrative ductal
carcinoma; HIDC, high-grade infiltrative ductal carcinoma.
Volume 20 | No. 3 | March 2009
Figure 6. The positive correlation of c-tubulin DNA copy numbers
(·106/ng) and score of c-tubulin protein expression.
(Table 3). Otherwise, from NBT/UDH to ADH to DCIS and
to IDC, the positive expression of p16INK4a displayed a decline
(v2 = 37.412, P = 0.000). By multiple comparison, there
were significant differences between (i) NBT/UDH and ADH
doi:10.1093/annonc/mdn651 | 445
original article
Annals of Oncology
Figure 7. The median and range of percentage of methylated reference
value of p16INK4a methylation in six groups.
Figure 8. The negative correlation of p16INK4a protein expression and
p16INK4a methylation within six groups.
Table 3. The positive rate of p16INK4a protein expression in six groups
Group
Cases
P16INK4a protein expression
+
2
Positive
rate (%)
NBT/UDH
ADH
LDCIS
HDCIS
LIDC
HIDC
Total
40
40
24
36
28
32
200
32
19
8
11
7
6
83
80
47.5
33.3
30.6
25.0
18.8
41.5
8
21
16
25
21
26
117
NBT/UDH, normal breast tissue/usual ductal hyperplasia; ADH, atypical
ductal hyperplasia; LDCIS, low-grade ductal carcinoma in situ; HDCIS,
high-grade ductal carcinoma in situ; LIDC, low-grade infiltrative ductal
carcinoma; HIDC, high-grade infiltrative ductal carcinoma.
(v2 = 9.141, P = 0.002), (ii) NBT/UDH and LDCIS (v2 =
13.938, P = 0.000), (iii) NBT/UDH and HDCIS (v2 = 18.855,
P = 0.000), (4) NBT/UDH and LIDC (v2 = 20.370, P = 0.000),
and (v) NBT/UDH and HIDC (v2 = 26.760, P = 0.000).
The positive rate of p16INK4a protein expression was
negatively correlated with that of p16INK4a methylation within
six groups (r = 0.178, P = 0.12) (Figure 8).
correlation between c-tubulin and p16INK4a
P16INK4a methylation is positively related to c-tubulin gene
amplification by Cochran–Mantel–Haenszel statistical analysis
(v2 = 15.722, P = 0.000) (Figure 9). Two (5%) NBT/UDH,
10 (25%) ADH, 9 (37.5%) LDCIS, 20 (55.6%) HDCIS, 18
(64.3%) LIDC, and 22 (68.7%) HIDC simultaneously showed
p16INK4a methylation and c-tubulin gene amplification, while
27 (67.5%) NBT/UDH, 14 (35%) ADH, six (25%) LDCIS,
six (16.7%) HDCIS, four (14.3%) LIDC, and three (9.4%)
HIDC showed neither of them. One (2.5%) NBT/UDH, four
(10%) ADH, three (12.5%) LDCIS, four (11.1%) HDCIS,
two (7.1%) LIDC, and two (6.3%) HIDC showed p16INK4a
446 | Liu et al.
Figure 9. The correlation of p16INK4a methylation presentation and
c-tubulin gene amplification in six groups.
methylation, without c-tubulin gene amplification. Ten (25%)
NBT/UDH, 12 (30%) ADH, six (25%) LDCIS, six (16.6%)
HDCIS, four (14.3%) LIDC, and five (15.6%) HIDC showed
c-tubulin gene amplification, without p16INK4a methylation.
In contrast, p16INK4a protein expression is negatively
related to c-tubulin protein expression (v2 = 7.447, P = 0.006)
(Table 4).
discussion
In this study, the median PMR value and positive rate of
p16INK4a methylation increased from NBT/UDH, to ADH, to
DCIS, to IDC, with both displaying significant differences.
Correspondingly, p16INK4a protein expression displayed
a decreasing trend, being consistent with methylation
presentation. Compared with NBT/UDH, ADH, DCIS, and
IDC showed significant differences in p16INK4A protein
expression and p16INK4A methylation. Results of this study
support the argument that the loss of p16INK4A protein
Volume 20 | No. 3 | March 2009
original article
Annals of Oncology
Table 4. Association of c-tubulin protein expression with p16INK4a protein expression in six groups
p16INK4a protein expression
+
2
c-tubulin
NBT/UDH
+
2
ADH
+
10
4
17
6
22
4
2
Protein
LDCIS
+
2
15
7
8
2
HDCIS
+
1
8
9
17
2
Expression
LIDC
+
2
HIDC
+
2
2
8
6
15
5
23
1
3
1
6
NBT/UDH, normal breast tissue/usual ductal hyperplasia; ADH, atypical ductal hyperplasia; LDCIS, low-grade ductal carcinoma in situ; HDCIS, high-grade
ductal carcinoma in situ; LIDC, low-grade infiltrative ductal carcinoma; HIDC, high-grade infiltrative ductal carcinoma.
expression due to methylation played an important role in the
evolution of breast cancer and might contribute to the earliest
stage of cancer development.
c-Tubulin was an essential protein for cell growth and
organizes microtubule arrays in the nucleus and cytoplasm.
Many studies profiled c-tubulin expression from the point of
protein level to assess centrosome abnormality [24, 25], and
alteration in the levels of c-tubulin protein could lead to
abnormal centrosome [8]. In this study, we determined
c-tubulin expression by simultaneously measuring its DNA
copy numbers and protein expression. A new method of PCR
quantification, real-time qPCR, was carried out because it
allows us to actually view the increase in the amount of DNA.
Our results showed that c-tubulin protein expression was
concordant with DNA copy numbers, both of them were found
to increase from NBT/UDH to ADH to DCIS and IDC, with
both displaying significant differences. This suggested that
c-tubulin gene amplification led to protein overexpression.
c-Tubulin gene amplification is present not only in invasive
carcinoma but also in a significant fraction of atypical
hyperplasia and in situ carcinomas. c-Tubulin gene
amplification and subsequent protein overexpression might
greatly contribute to the evolution of breast epithelial from
normal ductal epithelial to atypical hyperplasia, to in situ
carcinoma, and to invasive carcinoma.
Tumor grade has been identified as the most powerful
prognostic indicator of disease-free survival, with high-grade
tumors having poorer prognosis. Centrosome defect occur in
in situ carcinoma and invasive carcinoma and correlated with
the histological/cytologic grade in literature [8]. Although
c-tubulin was considered as the most important component of
the centrosome; c-tubulin DNA copy numbers and protein
expression did not show differences between different
histological grades in our studies. This might be due to that
there were other factors influencing centrosome abnormality,
such as posttranslational modifications of tubulin [26], changes
in proteins involved in cell cycle control, centrosome structure
or function, and DNA repair, e.g. mutation or elimination of
p53 or p53 downstream effectors/regulators [27].
A significant correlation was found between c-tubulin gene
amplification and p16INK4a methylation. With the NBT/UDH
to ADH to DCIS to IDC sequence, p16INK4a methylation and
c-tubulin gene amplification simultaneously showed an
increasing trend. This might suggest that p16INK4a methylation
and c-tubulin gene amplification had a synergistic effect on
breast tumor progression and simultaneously contributed to
the earliest stage of cancer development. However, the
Volume 20 | No. 3 | March 2009
mechanism for the synergistic effect is unknown. A functional
link may exist between p16 activity and p53 protein levels [28].
P16INK4a gene activity inversely modulates p53 status and
function in primary human mammary epithelial cells. Reduced
levels of p16 can increase p53 protein level and functions.
However, sustained upregulation of p53 activity may also
increase the selective pressure to inactivate p53. Loss of p53
might result in deregulation of centrosome duplication and
lead to functionally amplified centrosomes [5]. On the other
hand, loss of p16INK4a generates supernumerary centrosomes
through centriole pair splitting in cultured vHMECs. c-Tubulin
localizes to the centrosome and might normally function to
accelerate the kinetics of centriole assembly. The synergism
between p16INK4a and c-tubulin could be a result of p16INK4a
methylation and c-tubulin gene amplification, which influence
their ability to carry out their roles in proper maintenance and
duplication of the centrosome.
Another possible mechanism by which p16INK4a and
c-tubulin could have a synergistic effect on tumor progression
is through their influence on the G1–S checkpoint. Disruption
of the p16-CDK4/cyclin D1-pRB pathway occurs frequently
in breast premalignant and cancer [29]. Although c-tubulin is
thought to act primarily at the centrosome, in most cells the
majority of c-tubulin is in the cytoplasm. Analysis of c-tubulin
mutants in fungi suggests that c-tubulin might also have an
independent role in cell cycle control. Cells treated with
double-stranded RNA for c-tubulin showed a clear increase in
the number of mitotic cells in Drosophila cells as opposed to
the control cells, supporting the argument that c-tubulin is
thought to be required for a G1-related checkpoint pathway
and spindle formation [4, 30]. The synergism might therefore
arise as a result of the combined influence that p16INK4a and
c-tubulin have on the G1–S cell cycle checkpoints.
In summary, we have shown that p16INK4a methylation
and c-tubulin gene amplification are early events in the
ADH–carcinoma sequence of breast carcinoma. The aberration
of these two factors might not only contribute to loss of G1 cell
cycle checkpoint controls, but also induce normal ductal
epithelia to accumulate chromosomal abnormalities and
centrosome abnormality and express phenotypes that are
critical to malignant progression.
funding
National Science Fund of China (30471967). It is also
supported by Program for Changjing Scholars and Innovative
Research Team in University (PCSIRT).
doi:10.1093/annonc/mdn651 | 447
original article
references
1. Berman H, Zhang J, Crawford YG et al. Genetic and epigenetic changes in
mammary epithelial cells identify a subpopulation of cells involved in early
carcinogenesis. Cold Spring Harb Symp Quant Biol 2005; 70: 317–327.
2. Dessy E, Rossi E, Berenzi A et al. Chromosome 9 instability and alterations of
p16 gene in squamous cell carcinoma of the lung and in adjacent normal
bronchi: FISH and immunohistochemical study. Histopathology 2008; 52:
475–482.
3. Bean GR, Bryson AD, Pilie PG et al. Morphologically normal-appearing mammary
epithelial cells obtained from high-risk women exhibit methylation silencing of
INK4a/ARF. Clin Cancer Res 2007; 13: 6834–6841.
4. Müller H, Fogeron ML, Lehmann V et al. A centrosome-independent role for
gamma-TuRC proteins in the spindle assembly checkpoint. Science 2006; 314:
654–657.
5. Salisbury JL, D’Assoro AB, Lingle WL. Centrosome amplification and the origin of
chromosomal instability in breast cancer. J Mammary Gland Biol Neoplasia
2004; 9: 275–283.
6. Sankaran S, Parvin JD. Centrosome function in normal and tumor cells. J Cell
Biochem 2006; 99: 1240–1250.
7. Pihan GA, Wallace J, Zhou Y et al. Centrosome abnormalities and chromosome
instability occur together in pre-invasive carcinomas. Cancer Res 2003; 63:
1398–1404.
8. McDermott KM, Zhang J, Holst CR et al. p16(INK4a) prevents centrosome
dysfunction and genomic instability in primary cells. PLoS Biol 2006; 4: e51.
9. Berman H, Zhang J, Crawford YG et al. Genetic and epigenetic changes in
mammary epithelial cells identify a subpopulation of cells involved in early
carcinogenesis. Cold Spring Harb Symp Quant Biol 2005; 70: 317–327.
10. Vérollet C, Colombié N, Daubon T et al. Drosophila melanogaster gamma-TuRC
is dispensable for targeting gamma-tubulin to the centrosome and microtubule
nucleation. J Cell Biol 2006; 172: 517–528.
11. Colombié N, Vérollet C, Sampaio P et al. The Drosophila gamma-tubulin small
complex subunit Dgrip84 is required for structural and functional integrity of the
spindle apparatus. Mol Biol Cell 2006; 17: 272–282.
12. Tavassoli FA, Devilee P. The WHO classification of tumors of the breast and
female genital organs. Lyon, France: IARC Press 2003.
13. Eads CA, Danenberg KD, Kawakami K et al. Methylight: a high-throughput assay
to measure DNA methylation. Nucleic Acids Res 2000; 28: E32.
14. Ogino S, Kawasaki T, Brahmandam M et al. Precision and performance
characteristics of bisulfite conversion and real-time PCR (MethyLight) for
quantitative DNA methylation analysis. J Mol Diagn 2006; 8: 209–217.
15. Widschwendter M, Siegmund KD, Müller HM et al. Association of breast cancer
DNA methylation profiles with hormone receptor status and response to
tamoxifen. Cancer Res 2004; 64: 3807–3813.
448 | Liu et al.
Annals of Oncology
16. Eads CA, Lord RV, Kurumboor SK et al. Fields of aberrant CpG island
hypermethylation in Barrett’s esophagus and associated adenocarcinoma.
Cancer Res 2000; 60: 5021–5026.
17. Eads CA, Lord RV, Wickramasinghe K et al. Epigenetic patterns in the
progression of esophageal adenocarcinoma. Cancer Res 2001; 61:
3410–3418.
18. Ogino S, Cantor M, Kawasaki T et al. CpG island methylator phenotype (CIMP) of
colorectal cancer is best characterized by quantitative DNA methylation analysis
and prospective cohort studies. Gut 2006; 55: 1000–1006.
19. Tanaka K, Iwamoto S, Gon G et al. Expression of survivin and its relationship
to loss of apoptosis in breast carcinomas. Clin Cancer Res 2000; 6:
127–134.
20. Tsujie M, Yamamoto H, Tomita N et al. Expression of tumor suppressor gene
p16(INK4) products in primary gastric cancer. Oncology 2000; 58: 126–136.
21. Umekita Y, Ohi Y, Sagara Y et al. Overexpression of cyclinD1 predicts for poor
prognosis in estrogen receptor-negative breast cancer patients. Int J Cancer
2002; 98: 415–418.
22. Geradts J, Wilentz RE, Roberts H. Immunohistochemical [corrected] detection of
the alternate INK4a-encoded tumor suppressor protein p14(ARF) in archival
human cancers and cell lines using commercial antibodies: correlation with
p16(INK4a) expression. Mod Pathol 2001; 14: 1162–1168.
23. Geradts J, Ingram CD. Abnormal expression of cell cycle regulatory proteins
in ductal and lobular carcinomas of the breast. Mod Pathol 2000; 13:
945–953.
24. Löffler H, Lukas J, Bartek J et al. Structure meets function—centrosomes,
genome maintenance and the DNA damage response. Exp Cell Res 2006; 312:
2633–2640.
25. Szebenyi G, Hall B, Yu R et al. Hook2 localizes to the centrosome, binds directly
to centriolin/CEP110 and contributes to centrosomal function. Traffic 2007; 8:
32–46.
26. Hammond JW, Cai D, Verhey KJ. Tubulin modifications and their cellular
functions. Curr Opin Cell Biol 2008; 20: 71–76.
27. D’Assoro AB, Busby R, Acu ID et al. Impaired p53 function leads to
centrosome amplification, acquired ERalpha phenotypic heterogeneity and
distant metastases in breast cancer MCF-7 xenografts. Oncogene 2008;
26(27): 3901–3911.
28. Zhang J, Pickering CR, Holst CR et al. p16INK4a modulates p53 in primary
human mammary epithelial cells. Cancer Res 2006; 66: 10325–10331.
29. Niu Y, Liu TJ, Li Y et al. The clinical significance of the expression of four cell
cycle regulators in breast papillomotosis. Zhonghua Yi Xue Za Zhi 2003; 83:
1769–1773.
30. Vérollet C, Colombié N, Daubon T et al. Drosophila melanogaster gamma-TuRC
is dispensable for targeting gamma-tubulin to the centrosome and microtubule
nucleation. J Cell Biol 2006; 172: 517–528.
Volume 20 | No. 3 | March 2009