[CANCER RESEARCH 59, 5283–5285, October 15, 1999]
The Relationship of DNA Ploidy to Chromosomal Instability in Primary Human
Colorectal Cancers1
Mutsuko Miyazaki, Tomoko Furuya, Akiko Shiraki, Toshihiko Sato, Atsunori Oga, and Kohsuke Sasaki2
Department of Pathology, Yamaguchi University School of Medicine, Ube 755-8505, Japan
ABSTRACT
The aim of this investigation was to corroborate the relationship between
DNA ploidy and chromosomal variation in surgically removed colorectal
cancers. For 101 specimens from 21 advanced colorectal cancers, the numerical variations in chromosomes 7, 17, and 18 among cells were measured by
fluorescence in situ hybridization using DNA probes specific for centromere
of each chromosome, and DNA ploidy was determined by laser scanning
cytometry or flow cytometry. DNA diploidy (DNA index 5 1.0) was linked
with minor variation in copy number of chromosomes 7, 17 and 18, whereas
DNA aneuploidy (DNA index > 1.2) was found exclusively in tumors with
large variations in centromere copy number for all chromosomes. There was
a significant difference in the degree of intercellular variations in chromosome copy number between diploid and aneuploid clones for all chromosomes examined (P < 0.001). In near-diploid clones (1.0 < DNA index < 1.2),
the numerical variation of chromosome 18 was significantly different from
that in diploid clones (P < 0.002), but it was not different from that in
aneuploid clones. These observations support the hypothesis that chromosomal instability is associated with DNA aneuploidy in colorectal cancers.
Additionally, they suggest that near-diploid tumors are also unstable at a
lower level than classic aneuploid tumors and that all chromosomes are not
affected equally in near-diploid cases.
aneuploidy but not with DNA diploidy even in surgically removed
colorectal cancers. Furthermore, we also show that near-diploid clones
manifest a characteristic pattern of numerical chromosome alterations,
suggesting an intermediate pattern between diploid and aneuploid populations.
MATERIALS AND METHODS
Specimens. We used 21 surgically removed advanced colorectal cancers
that were histologically well-differentiated adenocarcinoma. The patients consisted of 11 males and 10 females with a mean age of 63.0 years (range, 44 – 86
years). Family history was noncontributory for all patients. Usually, tumor
tissue specimens were taken from five different parts of the same tumor, and
as the control, an additional specimen was also obtained from the intact
mucosa far from the tumor. Totally, 101 samples were examined in this study.
The tissue specimens were stored at 280°C until use.
Touch Smear Preparations for FISH and LSC. At least four touch
smears were prepared by touching thawed tissue specimens to glass slides after
blood was wiped from the cut surface of the specimens with a paper towel. One
touch sample was dipped in 70% ethanol for fixation immediately after touch
smears were prepared for DNA measurement by LSC. The other samples were
dried well and fixed with 100% ethanol for FISH analysis.
FISH. The touch smears fixed in 100% ethanol were refixed in 0.2%
INTRODUCTION
paraformaldehyde/PBS at 4°C for 10 min as described previously (11, 12). We
It is widely accepted that malignant tumors intrinsically exhibit genetic examined numerical aberrations of chromosomes 7, 17, and 18 using biotinyinstability and that, consequently, genetic aberrations successively accu- lated alpha satellite DNA probes specific for the pericentromeric region of
mulate with tumor progression. Recently, genetic instability has been each chromosome (D7Z1, D17Z1, and D18Z1; Oncor Inc., Gaithersburg, MD),
shown to be divided into two types: MIN3 and CIN (1–3). MIN, which as described elsewhere (11, 12). Briefly, 10 ml of a hybridization mixture
is known as a genetic phenotype of nonpolyposis colorectal cancer, containing 1 mg/ml salmon sperm DNA (Sigma Chemical Co., St. Louis, MO),
55% formamide, 23 SSC (13 SSC, 0.15 M NaCl and 0.015 M sodium citrate)
results from abnormalities in the DNA mismatch repair pathway. The
and 10% dextran sulfate was heated in a water bath at 70°C for 5 min. The
mechanism of MIN has been partially elucidated by molecular investi- DNA mixture was applied to the slides, which were denatured for 2 min.
gations (4, 5). Mutations of mismatch repair genes such as hMSH2 and Incubation for hybridization was performed overnight at 37°C in a moist
loci are italicized. Please check throughout manuscript that all genes, chamber. The slides were rinsed in a washing solution containing 50% formalleles, and loci are italicized at all usages and hMLH1 have been found amide and 23 SSC at 45°C, and then processed immunologically to stain the
in some colorectal cancers, and they have been considered a cause of hybridized probe with FITC-avidin (Vector Laboratory, Burlingame, CA). The
MIN. Microsatellite instability can account for ;15% of sporadic colo- nuclei were counterstained by the addition of glycerol with propidium iodide
rectal cancers (6, 7). Colorectal cancers with MIN show characteristic (Sigma) and p-phenylenediamine dihydrochloride (Sigma).
Scoring of Hybridization Signals. Only those cells having a malignant
clinicopathological features (8, 9). In contrast, information concerning
CIN is still restricted (10). The instability associated with the CIN cytological appearance (especially large nuclei) were scored. Small, round
lymphocyte-like cells and overlapping or damaged nuclei were disregarded.
phenotype is likely due to missegregation of chromosomes during asymThe number of hybridization signals in each nucleus was determined by
metric cell divisions (1–3, 10). Accordingly, it is not difficult to anticipate observing more than 200 nuclei on a slide, using an epifluorescence microthat CIN results in distinct aneuploidy. In fact, this has been demonstrated scope equipped with a 3100 oil immersion objective (Olympus Co., Tokyo,
by in vitro experiments using colorectal cell lines (1). As far as we know, Japan). The percentages of cell populations with different signal counts were
however, there are no published reports that focus on the relationship determined for each slide, as shown in Fig. 2. The mode number of individual
between DNA ploidy and CIN in surgically removed tissue specimens. In chromosome represented the number of chromosomes of the tumor.
DNA Measurement by LSC. Samples were made and DNA ploidy was
this brief report, we validate the view that CIN is associated with DNA
determined by the procedures described previously (13–16). Briefly, the slides
fixed in 70% ethanol were dipped in a propidium iodide solution (25 mg/ml in
Received 1/27/99; accepted 8/19/99.
PBS) containing 0.1% RNase (Sigma). A coverslip was put on the slide and sealed
The costs of publication of this article were defrayed in part by the payment of page
with fingernail polish. DNA content was measured by a laser scanning cytometer
charges. This article must therefore be hereby marked advertisement in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
(LSC 101; Olympus). Usually, more than 5,000 cells were examined in each
1
Supported by Grant-in-Aid for Science Research 09670187 from the Ministry of
sample. A DNA histogram was generated, and DNA ploidy was determined.
Education, Science, Sports, and Culture of Japan, and by the New Energy and Industrial
DNA Measurement by FCM. Tumors in which nuclear DNA content was
Technology Development Organization (NEDO) of Japan.
2
To whom requests should be addressed, at Department of Pathology, Yamaguchi
not measured by LSC were subjected to FCM analysis. The remaining tissue
University School of Medicine, Ube 755-8505, Japan. Phone: 81-836-22-2221; Fax:
specimens were used for flow cytometric DNA ploidy analysis, which was
81-836-22-2223; E-mail: [email protected].
carried out according to the method reported previously (17–19). Briefly, the
3
The abbreviations used are: MIN, microsatellite instability; CIN, chromosomal
tissue specimens were minced with scissors and suspended in a PBS solution
instability; FISH, fluorescence in situ hybridization; LSC, laser scanning cytometry; FCM,
containing 0.2% Triton X-100. Single nuclear suspensions were prepared by
flow cytometry; DI, DNA index.
5283
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1999 American Association for Cancer
Research.
CHROMOSOMAL INSTABILITY IN COLORECTAL CANCERS
somes among parts of a tumor was regarded as showing intratumoral regional
heterogeneity. The difference in the DI value between two peaks was more
than .10% of the diploid DNA content, the tumor was considered to consist
of heterogeneous subpopulations with different DNA ploidy.
Statistical Analysis. The Student’s t test was used to compare two groups of
possible permutations of three categories of gastric cancers based on DNA indices.
Fig. 1. The percentage of tumors classified based on DNA ploidy in this series (n 5 21).
A, tumors consisting of diploid clones alone; B, tumors consisting of a mixture of near-diploid
and diploid clones; C, tumor consisting of only near-diploid clones; D, tumors consisting of a
mixture of aneuploid and (near)diploid clones; and E, tumors consisting of only aneuploid
clones. Numbers indicate the number of relevant tumors in this series.
Table 1 Population sizes (mean 6 SD) of cells with modal chromosome number for
three different DNA ploidy groups in colorectal cancersa
There is a significant difference in population size of cells with modal number of all
chromosomes between diploid and aneuploid clones (P , 0.0001). For chromosomes 7
and 17, the population of cells with modal chromosome number is significantly different
between near-diploid and aneuploid (P , 0.0001). For chromosome 18, however, it is
significantly different between diploid and near-diploid clones (P , 0.0002).
Population size (%)
b
Chromosome
Normal
7
17
18
94.1 6 2.8
93.5 6 2.9
92.9 6 2.7
Diploidc
Near-diploidd
Aneuploide
83.0 6 12.1
83.0 6 12.3
90.6 6 8.4
78.6 6 18.9
84.8 6 8.8
74.2 6 13.1
63.2 6 15.2
65.5 6 15.9
79.6 6 17.7
a
A list of primary data is available on request.
Normal mucosa.
DNA diploid clones (DI 5 1.0).
d
DNA near-diploid clones (1.0 , DI , 1.2).
e
Aneuploid clones (DI $ 1.2).
b
c
filtering the tissue suspensions through a nylon mesh. Nuclei were treated with
RNase (0.1%; Sigma), and then nuclear DNA was stained with propidium
iodide (50 mg/ml; Sigma). Nuclear DNA was measured by a FACScan flow
cytometer (Becton Dickinson Co., San Jose, CA). Usually 20,000 nuclei were
counted and DNA ploidy was expressed by the DI.
The DI was calculated according to the principles recommended by consensus (20). In this series, a population with 1.0 , DI , 1.2 was classified as
a near-diploid case and was separated from a DNA aneuploid tumor. On the
basis of DNA ploidy, we classified samples into three types: diploid, neardiploid, and aneuploid populations.
Intratumoral Heterogeneity. A tumor consisting of heterogeneous subpopulations with different DNA ploidies and/or different number of chromo-
RESULTS
In the mucosa with normal appearance, no aneuploid peaks were
detected by cytometric analysis, i.e., DI 5 1.0. FISH studies revealed
that 90.8 –96.4% (mean, 93.6%) of cells were disomic for each chromosome and that virtually no cells had more than three spots.
DNA Ploidy Analysis. DNA indices ranged from 1.0 to 2.01 in this
series of colorectal cancers. There were two cases (10% of tumors) in
which neither DNA aneuploid (DI $ 1.2) nor near-diploid
(1.0 , DI , 1.2) clones were detected, these tumors consisting of DNA
diploid (DI 5 1.0) clones alone. In other tumors (90%), DNA aneuploid
and/or near-diploid clones were found in at least one region within a
tumor. Of these 19 tumors, 14 exhibited DNA aneuploid clones
(DI $ 1.2) in at least one region within a tumor. One of the remaining
five tumors consisted of only near-diploid clones, and four tumors were
composed of a mixture of near-diploid and diploid clones. Four tumors
were devoid of diploid and near-diploid clones, i.e., these tumors consisted of only aneuploid clones, and 10 tumors consisted of a mixture of
aneuploid and (near)diploid clones (Fig. 1). From the viewpoint of DNA
ploidy, intratumoral regional heterogeneity was observed in 13 tumors
(62% of cases).
With respect to individual sample, 39, 16, and 49 specimens were
in diploid, near-diploid and aneuploid groups, respectively.
FISH Analysis. There were great difference in the population sizes of
cells with spots equivalent to modal chromosome number among samples. In DNA diploid clones, however, most tumor cells had two signals
for each chromosome, and cells with more than three spots were occasionally detected. On average, 83.0, 83.0, and 90.6% of tumor cells in
DNA diploid clones were disomic for chromosomes 7, 17, and 18,
respectively (Table 1). In contrast, DNA aneuploid clones showed a great
intercellular variation in the copy number of chromosomes (Fig. 2). The
Fig. 2. Intercellular numerical variation of chromosomes 7, 17 and 18 in representative cases of diploid (A), near-diploid (B), and aneuploid (C) colorectal cancer. Most cells were
disomic for all chromosomes examined, and aneusomy cells were seen occasionally in diploid tumors. In contrast, the fraction of cells with modal chromosomal number was smaller
in aneuploid tumors than in diploid tumors (P , 0.0001), and intercellular variation in chromosome number was distinct in aneuploid tumors. Near-diploid tumors, however, showed
an intermediate pattern between diploid and aneuploid tumors. Monosomy 18 was prominent in a near-diploid tumor shown here, but chromosomes 7 and 17 were still disomic.
Ordinate, the percentage of cells with different chromosome numbers. Abscissa, numbers indicate the areas within tumors from which tissue specimens were taken: 1, oral part of a
tumor; 2, anal part of a tumor; 3, anterior part of a tumor; 4, posterior part of a tumor.
5284
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1999 American Association for Cancer
Research.
CHROMOSOMAL INSTABILITY IN COLORECTAL CANCERS
mean percentage of cells with modal chromosomal number was significantly smaller in aneuploid tumors than in diploid tumors (P , 0.0001;
Table 1). This was also true of tumors consisting of a mixture of
(near)diploid and aneuploid populations. Polysomic nuclei that were
incidental in diploid clone were frequent in DNA aneuploid clones.
Near-diploid clones (1.0 , DI , 1.2) showed a different pattern
from diploid and aneuploid clones. The intercellular variation in the
copy number of chromosomes 7 and 17 was not different from that in
diploid clone, whereas for chromosome 18 it was significantly different between diploid and near-diploid clones (P , 0.0002; Table 1).
The modal number of chromosome 18 was frequently monosomic in
a near-diploid tumor. The intercellular variation in the copy number of
chromosomes 7 and 17 was significantly different between neardiploid and aneuploid clones as well as between diploid and aneuploid
clones (P , 0.0001). However, it was not significant for chromosome
18 between near-diploid and aneuploid clones.
MIN cells to the defective type characteristic of CIN cells (10). Taking
into account recent studies concerning the mitotic checkpoint in the cell
cycle, the present investigation suggests that a transient phenotype in
which some genes that check mitosis are involved may change diploid
cells into near-diploid cells in which intercellular numerical variation is
detected in only some of chromosomes. Subsequently, other mitotic
checkpoint genes are hit successively, and eventually extensive involvement of these genes results in distinct aneuploidy (11–23). In this series,
approximately half of tumors showed intratumoral heterogeneity consisting of a mixture of aneuploid and (near)diploid clones (Fig. 1D). Provided that virtually aneuploid clones cannot change into diploid clones, it
is natural to consider that aneuploid clones evolve in preexistent diploid
clones. The hypothesis offered here is compatible with the phenomenon
that CIN is a dominant phenotype in a fusion cell of CIN and MIN (1–3).
It is likely that near-diploid clones are in the transient state between two
distinct types of genetic instability.
DISCUSSION
REFERENCES
Recently, an interesting and attractive hypothesis has been proposed for genetic instability in colorectal cancers (1–3). The in vitro
experiments using established cell lines revealed that, whereas MIN
was linked with (near) diploidy, CIN was associated with aneuploidy
(1–3). Although instability is defined as a rate and cannot be assessed
from a single experiment (1–3), chromosome number variation can be
used as a surrogate marker for instability. This is a practical strategy
for elucidating CIN in surgically removed tissue specimens. In this
study, DNA aneuploid clones showed numerous variations in chromosomes in surgically removed cancer specimens as well as in cell
lines, whereas diploid clones showed fewer variations. DNA diploid
clone exhibited that on average 83.0, 83.0, and 90.6% of tumor cells
were disomic for chromosomes 7, 17, and 18, respectively. In vitro
results seem to also be true for surgical materials.
Aneuploid clones consisted of heterogeneous cells with different chromosome copy number as anticipated. The fraction of cells with the modal
chromosome number was much smaller in aneuploid clones than in
diploid clones (P , 0.0001). Contrary to the reports by Lengauer et al.
(1–3), however, their principle did not hold for near-diploid clones in
surgically removed cancers. In near-diploid clones, the numerical chromosome variations were intermediate between two types of genetic
instability. Although only three kinds of chromosome were examined in
this study, the degree of intercellular numerical variation was different
among chromosomes in DNA near-diploid clones. Chromosome 18
showed greater intercellular numerical variation than other chromosomes
in near-diploid clones (P , 0.0002). Chromosome 18 exhibited large
variations in copy number even in near-diploid clones, whereas the copy
number variation of chromosomes 7 and 17 remained at the level of
diploid clone. We emphasize that all chromosomes are not equally
affected, at least in near-diploid clones. The intriguing changes seen in
near-diploid clones raise the possibility that near-diploidy is a transient
condition from diploidy to aneuploidy, and that, in such a condition, all
chromosomes are not affected equally. Because monosomy 18 is detected
in specific early colorectal cancers (11, 12), it is likely that chromosome
18 is intrinsically affected more than others in the early stage of oncogenesis in the colon and rectum. DNA aneuploid tumors may have
multiple mutational hits.
MIN invokes successive aberrations in many regions of the genome
with tumor progression (1–3). Some of them are genes relevant to control
of chromosomal division, e.g., mitosis checkpoint genes, and aberrations
in such genes result in CIN (1–3). Defective repair of mismatched bases
may provoke aberrations of genes checking mitosis in cells with intact
mitotic checkpoints, and eventually CIN is established. For example, the
expression of hBUB1 mutants converts the normal checkpoint status of
1. Lengauer, C., Kinzler, K. W., and Vogelstein, B. Genetic instability in colorectal
cancers. Nature (Lond.), 386: 623– 627, 1997.
2. Lengauer, C., Kinzler, K. W., and Vogelstein, B. DNA methylation and genetic instability
in colorectal cancer cells. Proc. Natl. Acad. Sci. USA, 94: 2545–2550, 1997.
3. Laungauer, C., Kinzler, K. W., and Vogelstein, B. Genetic instabilities in human
cancers. Nature (Lond.), 396: 643– 649, 1998.
4. Perucho, M. Cancer of the microsatellite mutator phenotype. Biol. Chem., 377:
675– 684, 1996.
5. Toft, N. J., and Arends, M. J. DNA mismatch repair and colorectal cancer. J. Pathol.,
185: 123–129, 1998.
6. Thibodeau, S. N., Bren, G., and Schaid, D. Microsatellite instability in cancer of the
proximal colon. Science (Washington DC), 260: 816 – 819, 1993.
7. Ionov, Y., Peinado, M. A., Malkhosyan, S., Shibata, D., and Perucho, M. Ubiquitous
somatic mutations in simple repetitive sequences reveal a new mechanism for colonic
carcinogenesis. Nature (Lond.), 363: 558 –561, 1993.
8. Kim, H., Jen, J., Vogelstein, B., and Hamilton, S. R. Clinical and pathological
characteristics of sporadic colorectal carcinomas with DNA replication errors in
microsatellite sequences. Am. J. Pathol., 145: 148 –156, 1994.
9. Lothe, R. A., Peltomaki, P., Meling, G. E., Aaltonen, L. A., Nystroem-Lahti, M.,
Pylkkanen, L., Heimdal, K., Andersen, T. I., Moller, P., Rognum, T. O., Fossa, S. D.,
Haldorsen, T., Langmark, F., Brogger, A., de la Chapelle, A., and Borresen, A. L.
Genomic instability in colorectal cancer: relationship to clinicopathological variables
and family history. Cancer Res., 53: 5849 –5852, 1993.
10. Cahill, D. P., Lengauer, C., Yu, J., Riggins, G. J., Willson, J. K. V., Markowitz, S. D.,
Kinzler, K. W., and Vogelstein, B. Mutations of mitotic checkpoint genes in human
cancers. Nature (Lond.), 392: 300 –303, 1998.
11. Sasaki, K., Sato, T., Kurose, A., Uesugi, N., and Ikeda, E. Monosomy of chromosome
18 detected by fluorescence in situ hybridization in colorectal tumors. Cancer (Phila.),
76: 1132–1138, 1995.
12. Sato, T., Oga, A., Ikeda, E., Todoroki, T., Furuya, T., and Sasaki, K. Cytogenetic
aberrations detected by FCM and FISH in colorectal cancers: two cytogenetic
pathways in colorectal carcinogenesis. Oncology (Basel), 57: 63– 69, 1999.
13. Sasaki, K., Kurose, A., Miura, Y., Sato, T., and Ikeda, E. DNA ploidy analysis by
laser scanning cytometry (LSC) in colorectal cancers and comparison with flow
cytometry. Cytometry, 23: 106 –109, 1996.
14. Kamada, T., Sasaki, K., Tsuji, T., Takahashi, M., and Kurose, A. Sample preparation
from paraffin-embedded tissue specimens for laser scanning cytometric DNA analysis. Cytometry, 27: 290 –294, 1997.
15. Furuya, T., Kamada, T., Murakami, T., Kurose, A., and Sasaki, K. Laser scanning
cytometry allows detection of cell death with morphological features of apoptosis in
cells stained with PI. Cytometry, 29: 1713–1717, 1997.
16. Kawasaki, M., Sasaki, K., Satoh, T., Kurose, A., Kamada, T., Furuya, T., Murakami,
T., and Todoroki, T. Laser scanning cytometry (LSC) allows detailed analysis of the cell
cycle in PI stained human fibroblasts (TIG-7). Cell Proliferation, 30: 139–147, 1997.
17. Sasaki, K., Hashimoto, T., Kawachino, K., and Takahashi, M. Intratumoral regional differences in DNA ploidy of gastrointestinal carcinomas. Cancer (Phila.), 62: 2569–2575, 1988.
18. Sasaki, K., Takahashi, M., Hashimoto, T., and Kawachino, K. Flow cytometric DNA
measurement of gastric cancers: clinico-pathological implication of DNA ploidy.
Pathol. Res. Pract., 184: 561–566, 1989.
19. Fujimaki, E., Sasaki, K., Nakano, O., Chiba, S., Tazawa, H., Yamashiki, H., Orii, S.,
and Sugai, T. DNA ploidy heterogeneity in early and advanced gastric cancers.
Cytometry, 26: 131–136, 1996.
20. Hiddemann, W., Schumann, J., Andreeff, M., Barlogie, B., Herman, C. J., Leif, R. C.,
Mayall, B. H., Murphy, R. F., and Sandberg, A. A. Convention on nomenclature for
DNA cytometry. Cytometry, 5: 445– 446, 1984.
21. Paulovich, A. G., Toczyski, D. P., and Hartwell, L. H. When checkpoints fail. Cell,
88: 315–321, 1997.
22. Elledge, S. J. Mitotic arrest: Mad2 prevents sleepy from waking up the APC. Science
(Washington DC), 279: 999 –1000, 1998.
23. Pennisi, E. Cell division gatekeepers identified. Science (Washington DC), 279:
477– 478, 1998.
5285
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1999 American Association for Cancer
Research.
The Relationship of DNA Ploidy to Chromosomal Instability in
Primary Human Colorectal Cancers
Mutsuko Miyazaki, Tomoko Furuya, Akiko Shiraki, et al.
Cancer Res 1999;59:5283-5285.
Updated version
Cited articles
Citing articles
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/59/20/5283
This article cites 22 articles, 5 of which you can access for free at:
http://cancerres.aacrjournals.org/content/59/20/5283.full#ref-list-1
This article has been cited by 11 HighWire-hosted articles. Access the articles at:
http://cancerres.aacrjournals.org/content/59/20/5283.full#related-urls
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1999 American Association for Cancer
Research.