Download Unbalanced translocation, a major chromosome alteration

Oncogene (2008) 27, 4788–4797
& 2008 Macmillan Publishers Limited All rights reserved 0950-9232/08 $30.00
www.nature.com/onc
ORIGINAL ARTICLE
Unbalanced translocation, a major chromosome alteration causing loss
of heterozygosity in human lung cancer
H Ogiwara, T Kohno, H Nakanishi, K Nagayama, M Sato and J Yokota
Biology Division, National Cancer Center Research Institute, Chuo-ku, Tokyo, Japan
Loss of heterozygosity (LOH) is a major genetic event
causing inactivation of tumor suppressor genes in human
carcinogenesis. To elucidate chromosomal mechanisms
causing LOH, 201 LOHs in 10 cases of human lung
cancer, which were detected by a genome-wide single
nucleotide polymorphism array analysis, were investigated
for responsible chromosome alterations by integrating
information on breakpoints for DNA copy number
changes obtained by array-comparative genome hybridization and on numerical and structural chromosomal
alterations obtained by spectral karyotyping. The majority (80%) of LOHs were partial chromosome LOHs
caused by structural chromosomal alterations, while the
remaining (20%) were whole chromosome LOHs caused
by whole chromosome deletions. Unbalanced translocation was defined as the most frequent alteration, and it
accounted for 30% of all LOHs. Three other structural
alterations—interstitial deletion (19%), mitotic recombination (9%) and gene conversion (6%)—also contributed
to the occurrence of LOH, while terminal deletion
contributed to only a small subset (1%). Since unbalanced
translocation is a common chromosomal alteration in lung
cancer cells, the results in the present study strongly
indicate that a considerable fraction of LOHs detected in
lung cancer cells are caused by unbalanced translocation.
Oncogene (2008) 27, 4788–4797; doi:10.1038/onc.2008.113;
published online 14 April 2008
Keywords: unbalanced translocation; loss of heterozygosity; lung adenocarcinoma; tumor suppressor gene;
mitotic recombination
Introduction
Loss of heterozygosity (LOH), that is, loss of one of the
parental alleles, is a major genetic alteration causing
inactivation of a tumor suppressor gene in multistage
human carcinogenesis (Cavenee et al., 1983; Weinberg,
2007). LOH is considered to be achieved through the
Correspondence: Dr J Yokota, Biology Division, National Cancer
Center Research Institute, 1-1, Tsukiji 5-chome, Chuo-ku, Tokyo 104-0045,
Japan.
E-mail: [email protected]
Received 9 November 2007; revised 26 February 2008; accepted 14
March 2008; published online 14 April 2008
loss of a whole chromosome due to an inappropriate
chromosomal segregation at mitosis and also through
genetic alterations that change chromosomal structures
of the cell (Lasko et al., 1991; Weinberg, 2007). In fact,
whole, terminal and interstitial chromosome deletions
were shown to cause LOH on several chromosomal loci
in human cancers (Naylor et al., 1987; Yokota et al.,
1987; Mori et al., 1989; Lasko et al., 1991). On the other
hand, mitotic recombination and gene conversion have
been indicated to be additional mechanisms causing
LOH (Cavenee et al., 1983; James et al., 1989; Mori
et al., 1989; Zhu et al., 1992; Adams et al., 2005).
Unbalanced translocation could also be an additional
responsible alteration due to the following reasons. This
alteration was shown to cause LOH of the VHL tumor
suppressor gene locus in individuals of a family
predisposed to renal cell cancer who carried a constitutional reciprocal chromosome translocation (Schmidt
et al., 1995). Cytogenetic studies have indicated unbalanced translocation as a frequent chromosome
alteration in a variety of human cancers (Mitelman,
2000; Roschke et al., 2003). In addition, it was reported
that a majority of partial chromosome LOH on
chromosomes 5, 8 and 17 in colorectal cancer is caused
by interchromosomal recombination (Thiagalingam
et al., 2001). Therefore, six chromosomal alterations
have been considered to be responsible for the occurrence of LOH in human carcinogenesis (Figure 1A).
However, contribution of each chromosome alteration
to the occurrence of LOH has been examined only for a
few chromosome regions of a few cancer cases, therefore, the significance of each chromosome alteration in
the occurrence of LOH remains unknown.
Tens of LOH are present on a variety of chromosomes in human lung cancer cells (Shiseki et al., 1996;
Virmani et al., 1998; Girard et al., 2000), and some of
them were shown to cause inactivation of tumor
suppressor genes, such as p16 and p53. As a result, it
is now widely accepted that LOH is a critical genetic
event for the development and progression of lung
cancer (Sekido et al., 2003; Yokota and Kohno, 2004;
Sato et al., 2007). Thus, the elucidation of chromosomal
mechanisms causing LOH in human lung cancer will
give us further understanding of lung carcinogenesis and
will also have preventive, diagnostic and therapeutic
implications in future. In this study, a comprehensive
genome-wide study was performed to elucidate
the prevalence and significance of each chromosome
LOH by unbalanced translocation in lung adenocarcinoma
H Ogiwara et al
4789
Whole chromosome LOH
Whole chromosome deletion
Partial chromosome LOH
Mitotic recombination
Mitotic gene conversion
Terminal deletion
Interstitial deletion
Unbalanced translocation
LOH
a
b
c
d
e
f
(201: 100%)
LOH
(40: 20%)
(161: 80%)
Partial chromosome LOH
Whole chromosome LOH
(130: 65%)
(31: 15%)
LOH without copy number change
LOH with copy number change
Monosomy
Multisomy
Undetermined
Mitotic
recombination
Mitotic
gene conversion
Terminal
deletion
Interstitial
deletion
Unbalanced
translocation
Complex
Undetermined
(2: 1%)
(23: 12%)
(15: 7%)
(14: 7%)
(17: 8%)
(2: 1%)
(26: 13%)
(42: 21%)
(23: 12%)
(37: 18%)
Figure 1 Classification of chromosome alterations causing loss of heterozygosity (LOH) in lung cancer cells. (A) A model for the
chromosome alterations causing LOH. Red and blue regions represent either of two homologous chromosomes, while a gray region
represents a nonhomologous chromosome. (B) Classification of LOH observed in 10 lung cancer cases. Whole and partial chromosome
LOHs were classified according to LOH regions determined by single nucleotide polymorphism (SNP) array analysis. Monosomy and
multisomy in whole chromosome LOHs were classified according to the number of chromosomes observed by spectral karyotyping
(SKY) analysis. LOHs with and without copy number changes were classified according to the changes detected at the boundary of
LOH regions by comparative genomic hybridization (CGH) array analysis. Mitotic recombination and mitotic gene conversion in
LOHs without copy number changes were defined if LOH was extended from a breakpoint within the chromosome to the telomere and
if LOH was found in an interstitial region, respectively. Unbalanced translocation and terminal deletion in LOHs with copy number
changes were classified if LOH was extended from a breakpoint within the chromosome to the telomere with or without translocation,
respectively. Interstitial deletion in LOHs with copy number changes was classified if LOH was found in an interstitial region with copy
number change. Complexes were defined if LOHs were too complex to explain by a simple event. Undetermined indicates 37 LOHs of
three surgical specimens not subjected to SKY analysis.
alteration in the occurrence of LOH in a set of 10 cases
of human lung cancer. All the cases were adenocarcinoma, the most prevalent histological type of lung
cancer (Colvy et al., 2004), and consisted of seven
cell lines and three surgical specimens, paired with
their corresponding noncancerous cells. LOHs were
searched for by genome-wide single nucleotide polymorphism (SNP) array analysis, and were classified
according to their responsible chromosome alterations
by integrating data of two other genome-wide analyses,
array-comparative genomic hybridization (array-CGH)
and spectral karyotyping (SKY). Unexpectedly, unbalanced translocation was the most frequent chromosomal alteration in the occurrence of LOH in lung
carcinogenesis.
Results
LOH, copy number changes and numerical and structural
chromosomal alterations in 10 cases of lung cancer
Genomic DNAs from seven lung cancer cell lines and
their corresponding EBV-immortalized B cell lines
(cases 1–7), and three surgical lung cancer specimens
and their corresponding normal lung tissues (cases 8–10)
were subjected to a SNP array analysis (Table 1).
Fractional allelic loss (FAL) for the 10 cases ranged
from 18 to 52%. A total of 215 regions were defined as
having LOH among the 10 cases based on the criteria
described in Materials and methods. However, 14 of
them were likely to be regions with spurious LOHs
caused by amplification/gain of one allele, since copy
numbers of these 14 regions were >2 times higher than
their surrounding regions. Therefore, the remaining 201
regions were defined as having LOH. These 201 LOHs
were dispersed on all autosomes (Supplementary Figure
1; Supplementary Table 1). Nine to 37 LOH regions
were detected in each of the 10 cases (Table 1).
Chromosome 17p regions containing the p53 tumor
suppressor gene were affected by LOH in all the 10
cases, being followed by chromosomes 1p, 2q, 8p, 9p
(containing the p16 gene), 18q and 19p (containing the
LKB1 gene) regions affected in 8 cases (Supplementary
Figure 1). LOH of these chromosome arms has been
frequently detected in lung cancer (Shiseki et al., 1996;
Virmani et al., 1998; Girard et al., 2000).
These 10 cases were then subjected to array-CGH
analysis for copy number changes along the genome,
and the seven cell lines were further subjected to SKY
analysis for numerical and structural chromosome
alterations. Representative results for the array-CGH
and SKY analyses are shown in Figure 2. The numbers
of breakpoints for copy number changes were approximately 10-times higher than those of LOH regions
(Table 1). Therefore, it was suggested that these lung
Oncogene
LOH by unbalanced translocation in lung adenocarcinoma
H Ogiwara et al
4790
Table 1
Sample
Case
Cell line
1
Name
No. of chromosomesa
Fraction of allelic
loss (%)b,c
No. of LOH
regionsb,d
No. of breakpointse
57–60
46
59–63
46–47
49–55
46–47
50–52
46–47
96–107
46
74–78
46
80–83
46
36.1
21
290
42.3
18
184
34.2
26
472
48.8
21
269
27.7
17
412
52.0
27
118
18.2
9
85
—
—
—
—
—
—
46.5
37
353
28.2
12
94
22.3
13
158
NCI-H1395
NCI-BL1395
NCI-H1437
NCI-BL1437
NCI-H2009
NCI-BL2009
NCI-H2087
NCI-BL2087
NCI-H2122
NCI-BL2122
NCI-H2126
NCI-BL2126
NCI-H2347
NCI-BL2347
2
3
4
5
6
7
Surgical specimen
Number of LOH regions and breakpoints in 10 lung adenocarcinomas
8
N211T
N212N
N2131T
N2132N
N2151T
N2152N
9
10
Abbreviation: LOH, loss of heterozygosity.
Assesed by SKY analysis.
b
Assesed by SNP analysis.
c
The fraction of probes for which noncancerous cell DNA was called as heterozygous and cancer cell DNA was called as homozygous.
d
Regions containing at least six consecutive ‘allelic loss’ loci were defined as LOH regions.
e
Assesed by CGH analysis.
a
LOH without copy number change (Mitotic recombination) H1395-Chr17
LOH
LOH without copy number change (Mitotic gene conversion) H2009-Chr10
LOH
LOH with copy number change (Terminal deletion) H2122-Chr18
LOH
del(18)
LOH with copy number change (Interstitial deletion) H2347-Chr5
LOH
del(5)
LOH with copy number change (Unbalanced translocation) H2347-Chr6
LOH
t(6; 8)
LOH with copy number change (Complex) H1395-Chr15
LOH
LOH
Figure 2 Representative patterns of partial losses of heterozygosity (LOHs) detected by single nucleotide polymorphism (SNP) array,
comparative genomic hybridization (CGH) array and spectral karyotyping (SKY) analyses.
Oncogene
LOH by unbalanced translocation in lung adenocarcinoma
H Ogiwara et al
4791
cancers also suffered chromosomal alterations associated with copy number changes but not with LOH.
Consistent with the result of array-CGH analysis, all the
seven cell lines were defined as being nondiploid with
multiple structural chromosome alterations by SKY
analysis.
Classification of LOHs by chromosome alterations
Chromosome alterations that can cause LOH are
illustrated in Figure 1A. LOHs can be classified into
two groups based on the underlying mechanisms: whole
chromosome LOH and partial chromosome LOH. The
former ((a) in Figure 1A) is considered to be caused by
whole chromosome deletion by illegitimate segregation
of chromosomes during mitosis, while the latter is by
intra- and interchromosomal recombination by illegitimate repair of DNA double strand breaks (DSBs).
Partial LOHs can be further classified into five groups
based on chromosome alterations; (b) mitotic recombination, (c) mitotic gene conversion, (d) terminal deletion, (e) interstitial deletion, and (f) unbalanced
translocation. The 201 LOHs detected in 10 lung cancer
cases were classified into these six groups to assess the
prevalence of chromosome alterations in the occurrence
of LOH in lung carcinogenesis.
Of 201 LOHs, 40 (20%) were judged as whole
chromosome LOHs since LOH occurred through the
whole chromosome. The remaining 161 LOHs (80%)
were detected in a part of chromosomes, thus, were
judged as partial chromosome LOHs (Figure 1B). Both
the whole and partial chromosome LOHs were observed
Classification of partial chromosome LOHs
by chromosome alterations
Next, 161 partial chromosome LOHs were classified.
Chromosome alterations, (b) and (c), are caused by
recombination between homologous chromosomes, and
therefore, do not leave a copy number change at
breakpoints for LOH, while (d), (e) and (f) do as a
result of intra- or interchromosome rearrangements.
Thirty-one LOHs (15%) were not accompanied by copy
Whole chromosome LOH (monosomy)
Whole chromosome LOH (multisomy)
Whole chromosome LOH (undetermined)
Partial chromosome LOH
20
10
Unbalanced translocation
Terminal deletion
Interstitial deletion
Complex
10
9
8
Number of LOH
Number of LOH
30
in all the 10 cases. Partial chromosome LOHs were more
frequent than whole chromosome LOH in nine cases
excluding case N2131T, in which the same number of
whole and partial chromosome LOHs was detected
(Figure 3a). The number of partial chromosome LOHs
(Mean±s.d ¼ 16.1±8.4) among the 10 cases was
significantly larger than that of whole chromosome
LOHs (Mean±s.d. ¼ 4.0±1.8, Po0.001 by t-test).
Thus, partial chromosome LOHs were indicated to be
more prevalent than whole chromosome LOHs,
although both types of LOHs occurred in lung cancer.
Of the 40 whole chromosome LOHs, 25 were detected
in cell lines, therefore, numerical alterations for chromosomes with LOH could be assessed based on SKY
data. Chromosomes for 23 LOHs (92%) were multisomy, and those for only two LOHs in H2087 cells (8%)
were monosomy (Figure 3a). Therefore, it was suggested
that whole chromosome LOH is mostly caused by
deletion of one chromosomal homologue followed by
multiplication of the other homologue (or gain of one
homologue followed by loss of the other homologue).
7
6
5
4
3
2
1
0
0
H1395 H1437 H2009 H2087 H2122 H2126 H2347 N211T N2131T N2151T
Number of LOH
30
LOH without copy number change (Mitotic recombination)
LOH without copy number change (Mitotic gene conversion)
LOH with copy number change
H1395
H1437
Mitotic recombination
6%
Mitotic gene conversion
9%
20
Complex
17%
10
Multisomy
17%
0
H2009
H2087
H2122
H2126
H2347
Monosomy
1%
Terminal deletion
1%
Unbalanced translocation
30%
Interstitial deletion
19%
H1395 H1437 H2009 H2087 H2122 H2126 H2347 N211T N2131T N2151T
Figure 3 Frequencies of losses of heterozygosity (LOHs) in seven cell lines and three surgical specimens. (a) Frequency of whole and
partial chromosome LOHs. (b) Frequency of LOHs with and without copy number changes at breakpoints. (c) Frequency of partial
chromosome LOHs with copy number change at breakpoints according to chromosome alterations. (d) Spectrum of chromosome
alterations causing LOH in seven cell lines.
Oncogene
LOH by unbalanced translocation in lung adenocarcinoma
H Ogiwara et al
4792
number changes at breakpoints, and were detected in
seven cases (70%; Figure 3b). The LOHs were further
classified into (b) mitotic recombination and (c) mitotic
gene conversion based on the region of LOHs (Figures
1B and 2). These alterations were detected in 14 and 17
LOHs without copy number changes, respectively. As
predicted, chromosome rearrangements were not detected for the chromosomal regions where mitotic
recombination or gene conversion was suggested to
have occurred (Figure 2).
Copy number changes were found at breakpoints for
the remaining 130 partial chromosome LOHs (65%),
therefore, occurrence of inter- and intrachromosomal
rearrangement was indicated. Of the 130 partial LOHs,
93 were detected in cell lines, therefore, structural
alterations causing LOH were further assessed by SKY
data. Unbalanced translocation was the most frequent
chromosomal alteration (42; 21%), and was observed in
all the seven cases (Figure 3c; Supplementary Figure 2).
Interstitial deletion was the second most frequent
alteration (26; 13%), and terminal deletion was responsible for only a small subset of LOHs (2; 1%). In
addition, a large portion of LOHs (23; 11%) could not
be explained simply by these three types of chromosome
alterations due to the fact that the regions with copy
number changes were not consistent with those with
LOHs (Figure 2), or multiple chromosome alterations
detected on a single chromosome prevented us from
identifying specific alterations causing LOH.
Spectrum of chromosome alterations causing LOH
in lung cancers
According to the classification described above, a
spectrum of chromosome alterations causing LOH was
determined as shown in Figure 3d. Data of 139 LOHs
from seven cell lines were used, because not only regions
of LOH and site of copy number change but also
numerical and structural alterations of chromosomes
were able to be assessed in these cases, and because
whole- and partial chromosome LOHs as well as partial
chromosome LOHs with and without copy number
changes were detected with similar frequencies between
cell lines and surgical specimens. Unbalanced translocation, accounting for 30% of LOHs, was the most
frequent alteration causing LOH. Interstitial deletion
(19%) and whole chromosome deletion (18%) (that is,
multisomy þ monosomy) were the next two frequent
alterations. Mitotic recombination and mitotic gene
conversion were observed less frequently (9 and 6%,
respectively), and terminal deletions were responsible for
only a small subset of LOHs (1%).
Chromosomal alterations causing LOH of the p53
tumor suppressor gene
The p53 gene is known to be frequently inactivated by
LOH and intragenic mutations in various human
cancers, including lung cancer (Robles et al., 2002;
Sengupta and Harris, 2005). In fact, six of the seven cell
lines carried homozygous p53 mutations, therefore, p53
is inactivated by LOH and mutation in these cell lines
Oncogene
(COSMIC; http://www.sanger.ac.uk/genetics/CGP/cosmic/).
Thus, we examined chromosome alterations causing LOH
of chromosome 17p, in which the p53 gene maps (Figure 4).
H2122 and H2087 cells had whole chromosome 17
LOH on chromosome 17. H2122 cells had four copies of
an apparently normal chromosome 17 homologue, while
H2087 cells had two copies of an apparently normal
chromosome 17 homologue and a copy of a 17p
isochromosome derived from the same homologue.
Therefore, deletion of a chromosome 17 homologue
was likely to be followed by multiplication of the
remaining chromosome 17 homologue in these two
cases. H1395 cells suffered mitotic recombination
resulting in LOH of a 15.5 Mb segment including
the p53 locus. Notably, mutation of the p53 gene has
not been detected in H1395 cells (COSMIC;
http://www.sanger.ac.uk/genetics/CGP/cosmic/). However, the possibility could not be ruled out that the cells
have a mutation in introns or the promoter region of the
p53 gene. Alternatively, the LOH might have lead to
inactivation of tumor suppressor gene(s) other than p53.
H2347, H1437 and H2126 cells had unbalanced
translocations between homologous and/or nonhomologous chromosomes that lead to LOH of chromosome
17p segments. All the three cell lines had homozygous
p53 mutations, therefore, these LOHs should have
contributed to inactivation of the other p53 allele in
each case. In summary, three types of chromosome
alterations, whole chromosome deletion, mitotic recombination and unbalanced translocation, were shown to
contribute to the occurrence of LOH on chromosome
17p including the p53 locus.
Discussion
Recent progress in the methods of genome-wide analysis
has made it possible to obtain comprehensive information on accumulated genetic alterations in cancer cells.
Particularly, SNP array and array-CGH analyses have
been widely used to detect LOH and decreases in copy
number of the cancer genome, respectively, to identify
tumor suppressor loci (Zhao et al., 2004; Sato et al.,
2005; Gaasenbeek et al., 2006). Here, we investigated the
prevalence and significance of several chromosomal
alterations causing LOH in lung carcinogenesis by
combining data obtained by these two methods with
data of chromosomal numerical and structural alterations obtained by SKY, a genome-wide cytogenetic
analysis. In total, 201 LOHs dispersed on all autosomes
in 10 cases of lung adenocarcinoma were classified,
based on responsible chromosome alterations. The
majority (80%) of LOHs were partial chromosome
LOHs, and they were found to be caused by several
structural chromosome alterations, including unbalanced translocation (30%), interstitial deletion (19%),
mitotic gene conversion (9%), mitotic recombination
(6%) and terminal deletion (1%). The remaining (20%)
were whole chromosome LOHs caused by whole
chromosome deletions. In fact, whole, interstitial
LOH by unbalanced translocation in lung adenocarcinoma
H Ogiwara et al
4793
Whole chromosome deletion
Mitotic recombination
H1395
H2087
H2122
Log2 ratio
Log2 ratio
Log2 ratio
-4 -2 0 +2 +4
-4 -2 0 +2 +4
-4 -2 0 +2 +4
mut mut mut mut
mut mut mut mut
WT WT WT
i(17p;17p)
Unbalanced translocation
H2347
H1437
H2126
Log2 ratio
Log2 ratio
Log2 ratio
-4 -2 0 +2 +4
-4 -2 0 +2 +4
-4 -2 0 +2 +4
mut mut
i(17q;17q)
mut mut
t(8;17)
8q
mut mut mut 14q
t(14;17)
Figure 4 The relationship between loss of heterozygosity (LOH) and mutation of the p53 gene on chromosome 17 in seven lung
cancer cell lines defined by single nucleotide polymorphism (SNP) array, comparative genomic hybridization (CGH) array and spectral
karyotyping (SKY) analyses. (a) Cases with LOH resulting from whole chromosome deletion. (b) A case with LOH resulting from
mitotic recombination. (c) Cases with LOH resulting from unbalanced translocation. Light green bars represent LOH regions detected
by SNP array analysis. Dots represent genome copy numbers determined by CGH array analysis. Red and blue regions on
chromosomes represent either of two chromosomes 17 homologues. Gray regions represent other nonhomologous chromosome
segments.
and terminal chromosome deletions and mitotic
recombination were shown to be causative chromosomal alterations inducing LOH on chromosomes 3p,
13q and 17p in human lung cancer (Naylor et al., 1987;
Yokota et al., 1987; Mori et al., 1989). However, a large
contribution of unbalanced translocation to the occurrence of LOH was first revealed in the present study to
the best of our knowledge. Unbalanced translocation
was shown to have caused LOH for the chromosome
17p segments containing the p53 tumor suppressor gene
in three cases of lung cancer cell lines. Thus, the present
genome-wide study led us not only to validate the
importance of several known chromosome alterations
but also to define unbalanced translocation as an
alteration contributing to the occurrence of LOH in
human lung carcinogenesis.
Unbalanced translocation was defined as the most
frequent chromosome alteration causing LOH. The
alteration was observed in all the seven cell lines,
therefore, was indicated to be a common alteration
causing LOH. A recent SKY analysis of lung cancer cell
lines revealed that translocations are present in all 10
cases examined and the great majority (>90%) of the
translocations were unbalanced ones (Grigorova et al.,
2005). In fact, all the seven cell lines examined in the
present study carried translocations (4–15 translocaOncogene
LOH by unbalanced translocation in lung adenocarcinoma
H Ogiwara et al
4794
tions), and 114 translocations in total were detected in
these cases. Notably, 108 (95%) of them were unbalanced translocations. Therefore, unbalanced translocation is a major chromosome alteration in lung cancer
cells. Here, we revealed that 42 (39%) of the 108
unbalanced translocations caused LOH. Thus, the
present results strongly suggest that unbalanced translocation contributes to lung carcinogenesis as a major
chromosomal mechanism causing LOH. This is in
contrast to the findings that balanced translocations
causing activation of oncogenes function as major
genetic events for the development of leukemias
and sarcomas (Roschke et al., 2003; Mitelman Database
of Chromosome Aberrations in Cancer, http://
cgap.nci.nih.gov/Chromosomes/Mitelman). Thus, the
significance of chromosome translocation in tumorigenesis could be different among cancers, and, in lung
cancer, a considerable fraction of them is responsible
for the inactivation of tumor suppressor genes by
causing LOH.
Molecular processes of unbalanced translocation
remain unclear. Balanced translocation that occurred
in two different chromosomes followed by loss of one of
the chromosomes with balanced translocation is a
possible mechanism. If balanced translocation occurred
prior to unbalanced translocation, both of each
chromosome segment from two different chromosomes
consisting of balanced translocation would have shown
LOHs, and the regions of LOHs should have
corresponded to those of deleted chromosome segments.
Thus, we further assessed whether LOH occurred
in both of the chromosomes consisting of unbalanced
translocation or not. Partner chromosomes were
identified in 40 of the 42 LOHs caused by unbalanced
translocation, whereas they were not identified in
the remaining 2 LOHs. Ten (25%) of these 40 LOHs
were detected in the deleted regions of both
chromosomes consisting of unbalanced translocation
(that is, 5 pairs of LOHs), as expected. On the other
hand, the remaining 30 (75%) LOHs were detected only
in the deleted regions of one of the two chromosomes,
and not in those of partner chromosomes, with
unbalanced translocation. In addition, even in these
five pairs of LOHs, chromosomes with unbalanced
translocation had additional chromosomal rearrangements. Therefore, chromosomal mechanisms to induce
LOH accompanied by unbalanced translocation remain
unclear. However, a considerable fraction of unbalanced
translocations causing LOH might be generated by
chromosomal mechanisms other than loss of one of two
balanced translocation chromosomes, since partner
chromosomes did not show LOHs in any regions in 12
of 40 LOHs with unbalanced translocation. In fact,
recent studies indicate that unbalanced translocation
is also given rise to by several mechanisms without
generating balanced translocation, such as breakinduced replication (Bosco and Haber, 1998). Finally,
it should be noted that, even if we remove five LOHs
that could be redundantly counted from the five pairs of
LOHs, unbalanced translocation remains the most
frequent event causing LOH (that is, 37 LOHs, 27%).
Oncogene
Interstitial deletion was defined as the second most
frequent alteration causing LOH, and it was also
observed in all cell lines examined. Therefore, the
alteration was indicated to be another common alteration causing LOH in lung cancer. The process causing
interstitial deletion is better understood than that
causing unbalanced translocation. We and others
recently cloned various genomic fragments containing
breakpoint junctions for interstitial deletions causing
inactivation of the p16 tumor suppressor gene, and
revealed that traces of nonhomologous end joining
(NHEJ) of DSBs were commonly present at the
junctions (Florl and Schulz, 2003; Sasaki et al., 2003).
Namely, no significant homology was detected between
two DNA breakpoints and overlaps of two or more
nucleotides were frequently observed at the breakpoint
junctions. Breakpoints for interstitial deletion of other
chromosomal regions also have similar traces of NHEJ
as reviewed (Kohno and Yokota, 2006). Therefore,
NHEJ of two DNA ends in a single chromosome is
likely to cause partial chromosome LOH by interstitial
deletion.
Mitotic recombination and gene conversion were also
indicated to contribute to the occurrence of partial
chromosome LOHs. The alterations were likely to be
carried out by the machinery of homologous recombination repair for DSBs using an intact sister chromatid.
A considerable contribution of these alterations to the
occurrence of partial chromosome LOH was recently
indicated in colorectal cancer by integrating data of
genome-wide analyses using two platforms, SNP array
and array-CGH (Gaasenbeek et al., 2006). The present
study confirmed their contribution in lung cancer, and
further indicated that their contribution is less than that
of unbalanced translocation, at least in lung cancer, by
integrating data of SKY with those of SNP array and
array-CGH analyses. Mitotic recombination and gene
conversion were not detected in 3 of 10 cases in the
present study (Figure 3b). Interestingly, the contribution
of mitotic gene conversion was exclusively large in
one case, H2009. This result may indicate that a subset
of lung cancer cells have an intracellular condition
under which illegitimate homologous recombinational
repair predominantly occurs against DSBs. On the
other hand, terminal deletion accounted for only a
small subset of LOHs, although its contribution
was suggested in previous studies (Naylor et al., 1987;
Mori et al., 1989). Healing of broken chromosome
ends by de novo telomere addition has recently been
proposed as a mechanism for the occurrence of terminal
deletion (Pennaneach et al., 2006). However, the
present results indicated that at least in lung cancer,
terminal deletion has a limited role in the occurrence
of LOH.
Whole chromosome deletion, which has been considered to be led by inappropriate chromosomal
segregation at mitosis, was found to cause a subset of
LOH in lung cancer. Interestingly, multiplication of
the remaining homologous chromosome was observed
in most cases (92%), and was also considered to
be caused by inappropriate chromosomal segregation.
LOH by unbalanced translocation in lung adenocarcinoma
H Ogiwara et al
4795
By considering that most lung cancer cells are aneuploids, the cells (or their precursor cells) might
commonly have a defect in the regulation of chromosomal segregation, leading to the occurrence of LOH.
It was previously reported that B40% of lung
cancer cells exhibit mitotic spindle checkpoint defects
(Takahashi et al., 1999; Masuda and Takahashi, 2002).
Thus, defects in mitotic spindle and other checkpoints in
lung cancer cells might cause whole chromosome LOH
by deletion of whole chromosomes.
The spectrum of chromosome alterations presented
here emphasizes the significance of illegitimate homologous and nonhomologous DSB repairs that cause
inter- and intrachromosomal rearrangements in the
occurrence of LOH in lung carcinogenesis. However, it
remains unknown how the illegitimate DSB repairs were
induced in lung cancer (or precursor) cells. Recently, it
was reported that activation of oncogenes triggers DSBs
by causing DNA replication (Bartkova et al., 2005;
Gorgoulis et al., 2005) and/or oxidative stresses (Vafa
et al., 2002). In fact, the amounts of DSBs in
premalignant and malignant cells for human lung
cancers were reported to be much higher than those in
proliferating noncancerous cells (Bartkova et al., 2005;
Gorgoulis et al., 2005). We recently indicated that an
increase in the intracellular amounts of DSBs causes the
predominance of an illegitimate NHEJ, microhomology
mediated-NHEJ (Katsura et al., 2007), which was
indicated to be a repair pathway prone to cause
chromosome translocations (Zhu et al., 2002). In
addition, a fraction of DSBs occurring in vivo are
indicated to have structures of DNA ends that cannot be
directly joined (Lieber et al., 2003). Thus, a large
amount of and/or the nature of DSBs could lead to
the occurrence of illegitimate repairs that cause
LOH. In addition, lung cancer cells (or their precursors)
might have defects in DSB repair and DNA
damage checkpoint that accelerate the illegitimate
repairs.
Definition of LOH regions and their breakpoints by SNP
array analysis
Seven lung cancer cell lines, three surgically resected lung
cancers and their corresponding noncancerous cells were
genotyped for 115 553 autosomal SNPs using a Mapping
100-k array (Affymetrix Inc., Santa Clara, CA, USA) and the
GeneChip Genotyping Analysis Software (Version 4.0).
Genotype calls were obtained in 91–99% (average ¼ 96%) of
the SNP sites on the array; 1985–9117 loci were informative
(that is, heterozygous in noncancerous cells) for detection of
LOH in the tumors. FAL for each tumor sample was
calculated as the fraction of SNP probes for which noncancerous cell DNA was called as heterozygous and cancer cell DNA
was called as homozygous. The fraction of error calls for each
tumor sample was calculated as the fraction of SNP probes
for which noncancerous cell DNA was called as homozygous
and cancer cell DNA was called as heterozygous, and ranged
from 0.39 to 4.53%. When a locus was called ‘homozygous’ in
tumor DNA and ‘heterozygous’ in the corresponding normal
tissue DNA, such a locus was judged as being a ‘LOH’ in
the tumor. On the other hand, when a locus was called
‘heterozygous’ both in the tumor and corresponding
normal tissue DNA, such a locus was judged as being
‘retained’ in the tumor. Regions containing six consecutive
LOH and retained loci were judged as LOH and retained
regions, respectively to avoid misdetection of LOH and
retained regions by call error, since the appearance of such
regions by the call error was far less than one, even for a case
with the largest informative loci and the highest probability of
call error (that is, 9117 informative loci (0.0453)6 ¼ 0.002).
Breakpoints for LOH regions were mapped between LOH
and the retained regions. Five of the seven cell lines, H1395,
H2009, H2087, H2122 and H2347, used in this study
were previously analysed for LOH using 400 microsatellite
markers placed at 10-Mb distance in the human genome
(Girard et al., 2000). These five cell lines contained 13 sets
of trisomic chromosomes with no visible rearrangements in
total. In the microsatellite analysis, ‘complete or nearly
complete loss of one allele was scored as LOH’. Under this
criterion, 12 of them were judged as retention of heterozygosity and the remaining 1 as LOH in their study. The
judgment of LOH for these 13 trisomic chromosomes was
completely the same in the present study, supporting that the
algorithm used in this study was appropriate and trisomic
regions without LOH were discriminated from LOH regions in
the present study.
Materials and methods
Cell lines and surgical specimens
A total of 10 lung adenocarcinomas, including 7 cell lines and
3 surgical specimens, were used. Seven cell lines, H1395,
H1437, H2009, H2087, H2122, H2126 and H2347, and
their corresponding EBV-immortalized B-lymphoblasts
(Phelps et al., 1996) were obtained from the American Type
Culture Collection (Manassas, VA, USA). Three metastatic
lung cancers to the brain and their corresponding normal
lung tissues were obtained at surgery from three patients
treated at the National Cancer Center Hospital, Tokyo,
Japan (NCCH). All lung cancer and lymphoblast cell
lines were cultured in RPMI-1640 medium containing 10%
fetal bovine serum. Genomic DNA was isolated from cell lines
and surgical specimens as described (Sakamoto et al., 1986).
SKY analysis of seven lung cancer cell lines and their
corresponding lymphoblast cell lines was performed as
described (Schrock et al., 1996; Abdel-Rahman et al., 2001).
This study was approved by the institutional review board
of NCCH.
Definition of breakpoints for copy number change by array
CGH analysis
Copy number changes in genomic DNAs of 10 lung cancers
were assessed using a Human CGH 185-k array covering
181 988 loci and Agilent CGH Analytics Software, Version 3.3
(Agilent Technologies, Santa Clara, CA, USA). Genomic
DNAs from the seven lung cancer cell lines, three surgical
specimens and their corresponding noncancerous cells were
assayed according to the manufacturer’s protocol using a
human normal genomic DNA mix (Promega, Madison, WI,
USA) as a reference. Firstly, data for probes that were located
in copy number variable regions deposited in the UCSC
genome database and the ones that showed log2 ratios o2.5
or >2.5 in 10 noncancerous DNAs in the present analyses
were removed to mask copy number variable regions. Next,
the copy number along the genome of 10 lung cancers was
inferred by the Aberration Detection Method 2 (ADM2)
algorithm. Breakpoints were defined as regions where the
difference in the inferred copy number between two adjacently
Oncogene
LOH by unbalanced translocation in lung adenocarcinoma
H Ogiwara et al
4796
located regions was >DLR (derivative log ratio; a cut off
value for the copy number changes in the ADM2 algorithm).
Acknowledgements
This work was supported by Grants-in-Aid from the Ministry
of Health, Labor and Welfare of Japan for the 3rd-term
Comprehensive 10-year Strategy for Cancer Control and for
Cancer Research (16-1 and 19-9), and from the Program for
Promotion of Fundamental Studies in Health Sciences of the
National Institute of Biomedical Innovation (NiBio). We
thank Kaho Minoura and Yayoi Fukuoka of Agilent
Technologies Japan for technical assistance in array-CGH
analysis.
References
Abdel-Rahman WM, Katsura K, Rens W, Gorman PA, Sheer D,
Bicknell D et al. (2001). Spectral karyotyping suggests additional
subsets of colorectal cancers characterized by pattern of chromosome rearrangement. Proc Natl Acad Sci USA 98: 2538–2543.
Adams J, Williams SV, Aveyard JS, Knowles MA. (2005). Loss of
heterozygosity analysis and DNA copy number measurement on 8p
in bladder cancer reveals two mechanisms of allelic loss. Cancer Res
65: 66–75.
Bartkova J, Horejsi Z, Koed K, Kramer A, Tort F, Zieger K et al.
(2005). DNA damage response as a candidate anti-cancer barrier in
early human tumorigenesis. Nature 434: 864–870.
Bosco G, Haber JE. (1998). Chromosome break-induced DNA
replication leads to nonreciprocal translocations and telomere
capture. Genetics 150: 1037–1047.
Cavenee WK, Dryja TP, Phillips RA, Benedict WF, Godbout R,
Gallie BL et al. (1983). Expression of recessive alleles by
chromosomal mechanisms in retinoblastoma. Nature 305: 779–784.
Colvy T, Noguch IM, Henschk EC, Vazquez M, Geisinger K, Yokose T
et al. (2004). Adenocarcinoma. In: Travis WD, Brambilla E,
Muller-Hermelink HK, Harris CC (eds). World Health Organization
Classification of Tumors: Pathology and Genetics, Tumours of
Lung, Pleura, Thymus and Heart. IARC Press: Lyon (France).
pp 31–34.
Florl AR, Schulz WA. (2003). Peculiar structure and location of 9p21
homozygous deletion breakpoints in human cancer cells. Genes
Chromosomes Cancer 37: 141–148.
Gaasenbeek M, Howarth K, Rowan AJ, Gorman PA, Jones A,
Chaplin T et al. (2006). Combined array-comparative genomic
hybridization and single-nucleotide polymorphism-loss of heterozygosity analysis reveals complex changes and multiple forms
of chromosomal instability in colorectal cancers. Cancer Res 66:
3471–3479.
Girard L, Zochbauer-Muller S, Virmani AK, Gazdar AF, Minna JD.
(2000). Genome-wide allelotyping of lung cancer identifies new
regions of allelic loss, differences between small cell lung cancer
and non-small cell lung cancer, and loci clustering. Cancer Res 60:
4894–4906.
Gorgoulis VG, Vassiliou LV, Karakaidos P, Zacharatos P, Kotsinas
A, Liloglou T et al. (2005). Activation of the DNA damage
checkpoint and genomic instability in human precancerous lesions.
Nature 434: 907–913.
Grigorova M, Lyman RC, Caldas C, Edwards PA. (2005). Chromosome abnormalities in 10 lung cancer cell lines of the NCI-H
series analyzed with spectral karyotyping. Cancer Genet Cytogenet
162: 1–9.
James CD, Carlbom E, Nordenskjold M, Collins VP, Cavenee WK.
(1989). Mitotic recombination of chromosome 17 in astrocytomas.
Proc Natl Acad Sci USA 86: 2858–2862.
Katsura Y, Sasaki S, Sato M, Yamaoka K, Suzukawa K, Nagasawa T
et al. (2007). Involvement of Ku80 in microhomology-mediated end
joining for DNA double-strand breaks in vivo. DNA Repair (Amst)
6: 639–648.
Kohno T, Yokota J. (2006). Molecular processes of chromosome 9p21
deletions causing inactivation of the p16 tumor suppressor gene in
human cancer: deduction from structural analysis of breakpoints for
deletions. DNA Repair (Amst) 5: 1273–1281.
Lasko D, Cavenee W, Nordenskjold M. (1991). Loss of constitutional
heterozygosity in human cancer. Annu Rev Genet 25: 281–314.
Oncogene
Lieber MR, Ma Y, Pannicke U, Schwarz K. (2003). Mechanism and
regulation of human non-homologous DNA end-joining. Nat Rev
Mol Cell Biol 4: 712–720.
Masuda A, Takahashi T. (2002). Chromosome instability in human
lung cancers: possible underlying mechanisms and potential
consequences in the pathogenesis. Oncogene 21: 6884–6897.
Mitelman F. (2000). Recurrent chromosome aberrations in cancer.
Mutat Res 462: 247–253.
Mori N, Yokota J, Oshimura M, Cavenee WK, Mizoguchi H,
Noguchi M et al. (1989). Concordant deletions of chromosome 3p
and loss of heterozygosity for chromosomes 13 and 17 in small cell
lung carcinoma. Cancer Res 49: 5130–5135.
Naylor SL, Johnson BE, Minna JD, Sakaguchi AY. (1987). Loss of
heterozygosity of chromosome 3p markers in small-cell lung cancer.
Nature 329: 451–454.
Pennaneach V, Putnam CD, Kolodner RD. (2006). Chromosome
healing by de novo telomere addition in Saccharomyces cerevisiae.
Mol Microbiol 59: 1357–1368.
Phelps RM, Johnson BE, Ihde DC, Gazdar AF, Carbone DP,
McClintock PR et al. (1996). NCI-Navy Medical Oncology Branch
cell line data base. J Cell Biochem Suppl 24: 32–91.
Robles AI, Linke SP, Harris CC. (2002). The p53 network in lung
carcinogenesis. Oncogene 21: 6898–6907.
Roschke AV, Tonon G, Gehlhaus KS, McTyre N, Bussey KJ,
Lababidi S et al. (2003). Karyotypic complexity of the NCI-60 drugscreening panel. Cancer Res 63: 8634–8647.
Sakamoto H, Mori M, Taira M, Yoshida T, Matsukawa S, Shimizu K
et al. (1986). Transforming gene from human stomach cancers and a
noncancerous portion of stomach mucosa. Proc Natl Acad Sci USA
83: 3997–4001.
Sasaki S, Kitagawa Y, Sekido Y, Minna JD, Kuwano H, Yokota J
et al. (2003). Molecular processes of chromosome 9p21 deletions in
human cancers. Oncogene 22: 3792–3798.
Sato M, Shames DS, Gazdar AF, Minna JD. (2007). A translational
view of the molecular pathogenesis of lung cancer. J Thorac Oncol 2:
327–343.
Sato M, Takahashi K, Nagayama K, Arai Y, Ito N, Okada M et al.
(2005). Identification of chromosome arm 9p as the most frequent
target of homozygous deletions in lung cancer. Genes Chromosomes
Cancer 44: 405–414.
Schmidt L, Li F, Brown RS, Berg S, Chen F, Wei MH et al. (1995).
Mechanism of tumorigenesis of renal carcinomas associated with the
constitutional chromosome 3;8 translocation. Cancer J Sci Am 1:
191–195.
Schrock E, du Manoir S, Veldman T, Schoell B, Wienberg J,
Ferguson-Smith MA et al. (1996). Multicolor spectral karyotyping
of human chromosomes. Science 273: 494–497.
Sekido Y, Fong KM, Minna JD. (2003). Molecular genetics of lung
cancer. Annu Rev Med 54: 73–87.
Sengupta S, Harris CC. (2005). p53: traffic cop at the crossroads
of DNA repair and recombination. Nat Rev Mol Cell Biol 6:
44–55.
Shiseki M, Kohno T, Adachi J, Okazaki T, Otsuka T, Mizoguchi H
et al. (1996). Comparative allelotype of early and advanced stage nonsmall cell lung carcinomas. Genes Chromosomes Cancer 17: 71–77.
Takahashi T, Haruki N, Nomoto S, Masuda A, Saji S, Osada H et al.
(1999). Identification of frequent impairment of the mitotic
checkpoint and molecular analysis of the mitotic checkpoint
LOH by unbalanced translocation in lung adenocarcinoma
H Ogiwara et al
4797
genes, hsMAD2 and p55CDC, in human lung cancers. Oncogene 18:
4295–4300.
Thiagalingam S, Laken S, Willson JK, Markowitz SD, Kinzler KW,
Vogelstein B et al. (2001). Mechanisms underlying losses of
heterozygosity in human colorectal cancers. Proc Natl Acad Sci
USA 98: 2698–2702.
Vafa O, Wade M, Kern S, Beeche M, Pandita TK, Hampton GM et al.
(2002). c-Myc can induce DNA damage, increase reactive oxygen
species, and mitigate p53 function: a mechanism for oncogeneinduced genetic instability. Mol Cell 9: 1031–1044.
Virmani AK, Fong KM, Kodagoda D, McIntire D, Hung J, Tonk V
et al. (1998). Allelotyping demonstrates common and distinct
patterns of chromosomal loss in human lung cancer types. Genes
Chromosomes Cancer 21: 308–319.
Weinberg R. (2007) Tumor suppressor gene. In: Weinberg R (ed). The
Biology of Cancer. Garland Science, Taylor and Francis Group,
LLC: New York. pp 209–254.
Yokota J, Kohno T. (2004). Molecular footprints of human lung
cancer progression. Cancer Sci 95: 197–204.
Yokota J, Wada M, Shimosato Y, Terada M, Sugimura T. (1987).
Loss of heterozygosity on chromosomes 3, 13, and 17 in small-cell
carcinoma and on chromosome 3 in adenocarcinoma of the lung.
Proc Natl Acad Sci USA 84: 9252–9256.
Zhao X, Li C, Paez JG, Chin K, Janne PA, Chen TH et al. (2004). An
integrated view of copy number and allelic alterations in the cancer
genome using single nucleotide polymorphism arrays. Cancer Res
64: 3060–3071.
Zhu C, Mills KD, Ferguson DO, Lee C, Manis J, Fleming J et al.
(2002). Unrepaired DNA breaks in p53-deficient cells lead to
oncogenic gene amplification subsequent to translocations. Cell 109:
811–821.
Zhu X, Dunn JM, Goddard AD, Squire JA, Becker A, Phillips RA
et al. (1992). Mechanisms of loss of heterozygosity in retinoblastoma. Cytogenet Cell Genet 59: 248–252.
Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc).
Oncogene