Carcinogenesis vol.24 no.5 pp.953±965, 2003
DOI: 10.1093/carcin/bgg024
Immortal, telomerase-negative cell lines derived from a Li-Fraumeni syndrome
patient exhibit telomere length variability and chromosomal and minisatellite
instabilities
Takeki Tsutsui1, Shin-ichi Kumakura1, Yukiko Tamura1,
Takeo W.Tsutsui1, Mizuki Sekiguchi1,
Tokihiro Higuchi2,4 and J.Carl Barrett3,5
1
Department of Pharmacology, The Nippon Dental University, School of
Dentistry at Tokyo, Tokyo, Japan, 2Molecular Pathology Group, TSL Inc.,
Tokyo, Japan and 3National Cancer Institute, National Institutes of Health,
Bethesda, Maryland, USA
5
To whom correspondence should be addressed
Email: [email protected]
Five immortal cell lines derived from a Li-Fraumeni syndrome patient (MDAH 087) with a germline mutant p53
allele were characterized with respect to telomere length
and genomic instability. The remaining wild-type p53 allele
is lost in the cell lines. Telomerase activity was undetectable
in all immortal cell lines. Five subclones of each cell line
and five re-subclones of each of the subclones also showed
undetectable telomerase activity. All five immortal cell
lines exhibited variability in the mean length of terminal
restriction fragments (TRFs). Subclones of each cell line,
and re-subclones of the subclones also showed TRF variability, indicating that the variability is owing to clonal
heterogeneity. Chromosome aberrations were observed at
high frequencies in these cell lines including the subclones
and re-subclones, and the principal types of aberrations
were breaks, double minute chromosomes and dicentric
chromosomes. In addition, minisatellite instability detected
by DNA fingerprints was observed in the immortal cell
lines. However, all of the cell lines were negative for microsatellite instability. As minisatellite sequences are considered recombinogenic in mammalian cells, these results
suggest that recombination rates can be increased in these
cell lines. Tumor-derived human cell lines, HT1080 cells
and HeLa cells that also lack p53 function, exhibited little
genomic instability involving chromosomal and minisatellite instabilities, indicating that chromosomal and minisatellite instabilities observed in the immortal cell lines
lacking telomerase activity could not result from loss of
p53 function.
Introduction
Normal human fibroblasts in culture have a finite proliferative
capacity. After a number of cell divisions, cells cease to divide
and arrest in a viable G0 /G1 state called senescence (1,2).
Telomeres, specialized structures at the ends of eukaryotic
chromosomes consisting of tandemly repeated short DNA
Abbreviations: AFB1 , aflatoxin B1 ; ALT, alternative lengthening of
telomeres; DM, double minute chromosomes LFS, Li-Fraumeni syndrome;
PD, population doubling; TRAP, telomeric repeat amplification protocol;
TRF, terminal restriction fragments.
953
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
4
Present address: Home Healthcare Business Unit, Teijin Ltd., Tokyo,
Japan
sequences (TTAGGG), in cultured somatic cells shorten as
a function of cell division (3). Normal human cells lose
proliferating capacity and undergo crisis when telomeric shortening reaches a critical length. Telomeres in germline cells and
in many immortalized cells and cancers are maintained by
telomerase, a cellular ribonucleoprotein reverse transcriptase
(4). Some human cell lines immortalized in vitro, and a number of both human tumors and tumor-derived cell lines exhibit
no detectable telomerase activity (5±9). A maintenance or
lengthening of telomeres in human cell lines lacking telomerase is presumably via an alternative telomere-lengthening
mechanism that is referred to by Bryan et al. (7) as ALT
(alternative lengthening of telomeres). Recently, telomerasenegative immortalized human cells and some human tumors,
both of which maintain their telomeres by an ALT mechanism,
were found to have promyelocytic leukemia (PML) bodies
containing PML protein, telomeric DNA, the telomere-binding
proteins TRF1 and TRF2, replication factor A, and the human
forms of RAD51 and RAD52 (10,11). Dunham et al. (12)
demonstrated that DNA sequences are copied from telomere
to telomere in an immortalized ALT cell line. These findings
suggest a possible involvement of homologous recombination
between telomeres in telomere maintenance or lengthening in
ALT cells.
Loss of p53 or mutations in mismatch repair proteins can
induce genetic instability (13±15). Mutations in p53 can
increase recombination rates (16). We have established previously several human cell lines immortalized by carcinogen
treatment of skin fibroblasts from a Li-Fraumeni syndrome
(LFS) patient (MDAH 087) with a mutated p53 allele
(17,18). The parental MDAH 087 cells and their immortal cell
lines show extensive karyotypic changes (17±19). Gollahon
et al. (20) reported that a spontaneously immortalized cell
line derived from MDAH 087 fibroblasts was negative for
telomerase activity and had long and heterogenous telomeres.
Spontaneously immortalized cell lines (IIICF) established
from a different LFS patient are also telomerase-negative and
have very long and heterogenous telomeres (7). Four of four
IIICF cell lines have PML bodies containing telomeric DNA,
telomere-specific binding proteins, and proteins involved in
recombination (11), suggesting that recombination may participate in telomere maintenance or lengthening in immortal
ALT cell lines established from LFS fibroblasts.
In order to study the association of p53 mutations or mutations in mismatch repair genes with telomere maintenance or
lengthening in ALT cell lines, we analyzed the status of telomeres and telomerase and also genomic instability involving
chromosomal and microsatellite instabilities of LFS fibroblast
cell lines. In addition, to examine the genomic recombination
in the cell lines, minisatellite instability in the cell lines was
studied using the DNA fingerprinting assay. Minisatellites,
known as variable number of tandem repeats, are hypervariable regions of DNA showing multiallelic variation (21). As
hypervariable minisatellite sequences are hotspots for meiotic
T.Tsutsui et al.
recombination within the mouse major histocompatibility
locus (22) and for homologous recombination in human cells
(23), they are considered recombinogenic in mammalian cells.
We found that five immortal cell lines established from
MDAH 087 cells are telomerase-negative but maintain variable and in some cases long telomeres. Telomere length variability and chromosomal and minisatellite instabilities but lack
of microsatellite instability are characteristic of the cell lines,
suggesting an increased recombination rate in the cells. Little
genomic instability involving chromosomal and minisatellite
instabilities were observed in p53 affected tumor-derived
human cell lines, indicating that chromosomal and minisatellite instabilities observed in the immortal cell lines lacking
telomerase activity could not result from loss of p53 function.
Cells and culture conditions
A human skin fibroblast strain (MDAH 087) derived from a LFS patient with a
germline mutant p53 allele. The strain has one wild-type p53 allele and one
allele with a mutation at codon 248 in exon 7 that changes an Arg codon (CGG)
to Trp (TGG) (14). A spontaneously immortalized cell line (designated LCSST) derived from this strain was generously provided by Dr M.A.Tainsky (The
University of Texas M. D. Anderson Cancer Center) (19). Immortal human cell
lines used were LCS-AF.1-2 cells, LCS-AF.1-3 cells and LCS-AF.3-1 cells, all
of which were immortalized by repeated treatments of MDAH 087 cells at 18
population doublings (PDs) and greater with 0.1 or 0.3 mg/ml of aflatoxin B1
(AFB1 ) one to three times in the presence of exogenous metabolic activation
(17). Other immortal cell lines used were LCS-4 2 cells, HT1080 cells and
HeLa cells. LCS-4 2 cells were derived from MDAH 087 cells and immortalized by periodical irradiation with X-ray at 4 Gy two times (18). The MDAH
087 cells at 16 or 21 PDs were subjected to the first or second irradiation,
respectively. Two variants of human fibrosarcoma-derived HT1080 cells were
used: one is HT1080 cells with two normal p53 alleles (designated wild
HT1080 cells), obtained from ATCC, and the other with two mutant alleles
of p53 at codon 245 in exon 7 and codon 277 in exon 8 (designated mutant
HT1080 cells), kindly provided by Dr B.Weissman (University of North
Carolina) (24). HeLa cells, which lack p53 function (25), were obtained from
the Japanese Cancer Research Resources Bank (Tokyo, Japan). The culture
medium used was Eagle's minimum essential medium containing 10% fetal
bovine serum, 0.2 mM serine, 0.1 mM aspartic acid, 1.0 mM pyruvate and
0.22% NaHCO3 . MDAH 087 cells and its immortal cell lines were subcultured
at split ratio of 1:2 or 1:4 by gentle trypsinization with 0.25% trypsin (Gibco,
Grand Island, NY) for 3 min at room temperature. HT1080 cells and HeLa cells
were subcultured at 1:4 or 1:8 following treatment with 0.1% trypsin at 37 C.
Telomerase assay
Telomerase activity in cells was detected by the telomeric repeat amplification
protocol (TRAP) developed and modified by Shay's group (26,27). Cells (106 )
in the logarithmic growth phase were suspended in 100 ml of ice-cold lysis
buffer [0.5% CHAPS, 10 mM Tris±HCl (pH 7.5), 1 mM MgCl2 , 1 mM EGTA,
5 mM b-mercaptoethanol, 0.1 mM AEBSF (a protease inhibitor) and 10%
glycerol], and kept on ice for 30 min with occasional tapping. The lysate was
centrifuged at 16 000 g for 20 min at 4 C. The supernatant (1 ml: equivalent to
104 cells) was assayed in 50 ml of reaction mixture containing 50 mM of each
deoxynucleoside triphosphate, 344 nM of the deoxyoligonucleotide primer
TS(50 -AATCCGTCGAGCAGAGTT-30 ), 0.5 mM T4 gene 32 protein
(Boehringer Mannheim, Mannheim, Germany), 4 mCi [a-32 P]deoxycytidine
triphosphate (sp. act. of ~3000 Ci/mmol, Amersham, Buckinghamshire, UK), 2
U of Taq DNA polymerase (Takara, Tokyo, Japan) and 5 ml of 10 PCR buffer
in a 0.5 ml tube that contained 344 nM of the deoxyoligonucleotide primer
CX(50 -CCTTACCCTTACCCTTACCCTAA-30 ) sequestered from the other
reaction components by a wax barrier. After 20 min incubation at room
temperature, the reaction mixture was amplified by PCR for 31 cycles in the
presence of 5 attograms of an internal TRAP assay standard (ITAS: generously
provided by Dr E.Hiyama, Hiroshima University, Japan). The PCR product
was subjected to electrophoresis on 10% polyacrylamide gels.
RNA expression
Total cellular RNAs were isolated from cultured cells by using the RNAzolÒ
B (Tel-Test, Friendswood, TX). For reverse transcription (RT)±PCR analyses,
total RNA (2 mg) was reverse-transcribed with the oligo(dT) primer by
using the You-Prime First-Strand BeadsÒ for first-strand cDNA synthesis
954
Terminal restriction fragment (TRF) analysis
Genomic DNA was extracted from cells by using a DNA extraction kit (DNA
Extractor WB kit, Wako Pure Chemical, Osaka, Japan). Two microgram of
DNA was digested with HinfI (BioLabs, Beverly, MA) and subjected to 0.6%
agarose gel electrophoresis. To ascertain that a comparable amount of DNA
from individual cells was loaded in each lane, the gel was stained with
ethidium bromide and examined by exposure to UV light. The gel was then
depurinated for 15 min in 0.25 M HCl, denatured for 20±30 min in 0.2 M
NaOH±0.6 M NaCl, and neutralized for 30±60 min in 0.2 M Tris±HCl (pH
7.4)±0.6 M NaCl. The gel was blotted to nitrocellulose membranes (BA-S85,
Scheicher & Schuell, Dassel, Germany). The membranes were pre-hybridized
at 65 C in the hybridization solution containing 1 M NaCl, 1 Denhardt's
solution, 50 mM Tris±HCl (pH 7.4), 10 mM Na2 EDTA (pH 7.4), 0.1% SDS
and salmon DNA (5 mg/ml), and hybridized overnight at 50 C with a 32 P-endlabeled (TTAGGG)4 telomeric probe. They were then washed at 55 C in 4
standard saline/citrate (SSC) and 0.1% SDS, and autoradiographed on X-ray
films (SR-H, Konica Co., Tokyo, Japan) at ÿ80 C. The mean length of TRFs,
which consist of the terminal genomic DNA fragments generated by HinfI
digestion and which contain telomeric sequences was estimated from a peak
migration distance of each lane autoradiographed, which was analyzed by
densitometry. Because the intensity of TRF signal varied with individual
cells (Figure 3), appropriate exposure time for autoradiography was chosen
for TRF length analyses. The kilobase size of the mean TRF length was
determined by comparison with the migration distance of fragments of
known DNA markers (1 kb DNA ladder and lambda bacteriophage DNA
digested with HindIII). Each TRF analysis was repeated two to three times.
Chromosome analysis
Cells at 70±80% confluence were treated with 0.02 mg/ml Colcemid (Gibco)
for 3±4 h. After trypsinization, the cells were treated with 0.8% sodium
citrate at room temperature for 13 min and fixed with Carnoy's solution (3:1
methanol:acetic acid). The suspension of cells in fixative was dropped onto
glass slides and air-dried. The specimens were stained with Giemsa solution in
0.07 M phosphate buffer (pH 6.8) for 7 min. Only a restricted subset of
structural aberrations, such as gaps, breaks, exchanges, dicentric chromosomes, ring chromosomes and double minute chromosomes (DMs) was scored.
Achromatic lesions greater than the width of the chromatid were scored as gaps
unless there was displacement of the broken piece of chromatid. If there was
displacement, the lesions were recorded as breaks. Fifty well-spread metaphases were scored for each cell group.
Minisatellite alteration
Minisatellite alterations were examined with DNA multilocus fingerprinting, as
described previously (17). Genomic DNA isolated from cultured cells was
digested with HaeIII or HinfI. The DNA (2 mg) was electrophoresed in a 0.7%
agarose gel and then transferred by blotting to a Hybond-N membrane
(Amersham). The Southern blots were hybridized with alkaline phosphataselabeled multilocus probes (33.15 and 33.6) (NICE; Cellmark Diagnostics,
Abingdon, UK) according to the method described by Jeffreys et al. (21).
All DNA samples were analyzed in duplicate, and only reproducible bands
43 kb were scored. Alterations of minisatellite sequences were defined as
gain of new bands, because band losses are common in immortalized cell
lines due to deletion events involving loci that happen to contain minisatellite
DNA, whereas new bands are more likely to represent genuine minisatellite
alterations.
Microsatellite alteration
Microsatellite alterations were examined with a PCR detection assay using 69
microsatellite markers including D2S123, BAT25 and BAT26 (15). Microsatellites detectable by these markers were distributed to all chromosomes. The
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
Materials and methods
(Pharmacia Biotech, Tokyo, Japan). A part of the reaction (1 ml) was used as
a template for a 10-ml PCR amplification by using the Advantage cDNA PCR
kit (Clontech, Tokyo, Japan). For hTERC, the samples were subjected to 30
cycles of 94, 62 and 68 C for 30 s each with the primers 50 -CTA ACC CTA
ACT GAG AAG GGC GTA-30 and 50 -GTT TGC TCT AGA ATG AAC GGT
GGA AG-30 . For amplification of hTEP1, we used the primers TEP1-F
(50 -CTG TAC GGC TCT GGC AGG T-30 ) as a sense primer and TEP1-R
(50 -GGA GCC CAA TCC AGA CTT GT-30 ) as an antisense primer. The
following primers were used for hTERT amplification; TERT-F (50 -TGA
AAG CCA AGA ACG CAG GGA-30 ) as a sense primer and TERT-R
(50 -GGG AAG TGA AGA CGG CAG GT-30 ) as an antisense primer. The
control amplification of glyceraldehyde-3-phosphate dehydrogenase
(GAPDH) mRNA was performed by using the primers 50 -CCA TCT TCC
AGG AGC GAG A-30 and 50 -TGT CAT ACC AGG AAA TGA GC-30 under
the following conditions: 94 C for 1 min followed by 17 cycles of 94 C for
30 s, 60 C for 30 s and 68 C for 2 min. All the PCR products were analyzed by
gel electrophoresis in the appropriate percentage of agarose.
Genetic instability and telomere maintenance
primers for PCR amplification were purchased from Amersham (Tokyo,
Japan).
Results
Telomerase activity
Skin fibroblasts (MDAH 087 cells) derived from a LFS patient
and spontaneously, AFB1 -induced or X-ray-induced immortal
cell lines derived from MDAH 087 cells were analyzed for
telomerase activity. Parental MDAH 087 cells entered into a
growth crisis and senesced by 35 PDs (17). No telomerase
activity was detected in MDAH 087 cells or immortal cell
lines derived from these cells (Figure 1A). Although the
TRAP assay used was able to detect telomerase activity in
the equivalent of 102 HeLa cells (Figure 1B), no activity was
detected in the equivalent of 104 MDAH 087 cells or its
immortal cell lines. Telomerase activity was undetectable
even when the equivalent of 3 105 cells of each cell line
was assayed (data not shown), which is consistent with the
results by Gollahon et al. (20). The lack of detectable telomerase activity in MDAH 087 cells and its immortal cell lines was
not attributed to an inhibitor of the telomerase assay or Taq
polymerase because the telomerase activity in wild HT1080
cells and HeLa cells, which were abolished by preheating cell
lysates at 99 C for 10 min (Figure 1A), was not inhibited in a
mixture of HeLa cell lysate and cell lysate from the immortal
A
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
B
Fig. 1. No detectable telomerase activity either in MDAH 087 cells and their immortal cell lines (A) or in one of the immortal cell lines (LCS-4 2 cells at
482 PDs) and their subclones (B). MDAH 087 cells are skin fibroblasts from a LFS patient (19). LCS-ST cells are a spontaneously immortalized cell line
of MDAH 087 cells (17,19). LCS-AF.1-2, LCS-AF.1-3 and LCS-AF.3-1 cells are immortal cell lines induced by repeated treatments of MDAH 087 cells
with 0.1±0.3 mg/ml AFB1 (17). LCS-4 2 cells were an immortal cell line induced by repeated irradiations of MDAH 087 cells with X-rays (18). Cells
harvested at the indicated PDs were counted by a hemocytometer. Cell lysates were prepared from 106 cells using the CHAPS detergent lysis method (26), and
the equivalent of 104 , 103 or 102 cells were analyzed by the TRAP assay with the internal TRAP assay standard (ITAS) that is used to verify the efficiency
of amplification. Wild HT1080 cells and HeLa cells were used as positive controls, and lysis buffer was a negative control with no cell lysate. The position
of the ITAS is indicated. ‡ shows that cell lysates were pre-heated at 99 C for 10 min.
955
T.Tsutsui et al.
cell lines (data not shown). All subclones and re-subclones of
the immortal cell lines were also telomerase-negative in the
TRAP assay (Table I and Figure 1B).
Telomerase components
RT±PCR analysis showed that parental MDAH 087 cells and
all the immortal cell lines examined expressed hTEP1 mRNA
(Figure 2). Although there was variation in an expression level
among the cell lines, all the immortal cell lines expressed
hTERC RNA (Figure 2). In contrast, neither MDAH 087
cells at 21 PDs nor all the immortal cell lines expressed
hTERT mRNA (Figure 2).
Fig. 2. Expression of the genes encoding three components of the human
telomerase enzyme in MDAH 087 cells and their immortal cell lines analyzed
by RT±PCR. Wild HT1080 cells ( p53 ‡= ‡ ) and HeLa cells were used as
standards. GAPDH, glyceraldehyde-3-phosphate dehydrogenase used as an
amplification control.
Table I. The mean TRF length and telomerase status in cultured human cells
examined
Cells
MDAH 087 (PD 18)
MDAH 087 (PD 32)
LCS-ST (PD 318)
LCS-AF.1-2 (PD 481)
LCS-AF.1-3 (PD 494)
LCS-AF.3-1 (PD 438)
LCS-4 2 (PD 482)
Mean TRF length (kb)
Telomerase activitya
7.9
5.9
±
±
14.0
21.3
8.3
8.0
13.0
±
±
±
±
±
[Subclone of LCS-AF.1-2 (PD 481)]
LCS-AF.1-2 Cl 1
16.5
LCS-AF.1-2 Cl 2
13.9
LCS-AF.1-2 Cl 3
9.4
LCS-AF.1-2 Cl 4
5.4
LCS-AF.1-2 Cl 5
8.3
±
±
±
±
±
[Subclone of LCS-AF.1-3 (PD 494)]
LCS-AF.1-3 Cl 1
8.4
LCS-AF.1-3 Cl 2
10.2
LCS-AF.1-3 Cl 3
13.7
LCS-AF.1-3 Cl 4
7.9
LCS-AF.1-3 Cl 5
7.6
±
±
±
±
±
[Subclone of LCS-4 2 (PD 482)]
LCS-4 2 Cl 1
8.6
LCS-4 2 Cl 2
6.7
LCS-4 2 Cl 3
16.1
LCS-4 2 Cl 4
7.4
LCS-4 2 Cl 5
6.4
±
±
±
±
±
a
ÿ shows non-detectable-telomerase activity.
956
A
B
Fig. 3. Autoradiograph of radiolabeled telomere probe (TTAGGG)4 to
HinfI-digested genomic DNA either from MDAH 087 cells and their mmortal
cell lines (A) or from LCS-4 2 cells (PD 482) and their subclones (B).
Exposure times for autoradiograph were 3 days for (A) and 12 days for (B).
Size markers are indicated on the left.
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
TRF length
The mean TRF lengths of MDAH 087 cells at 18 PDs and 32
PDs were 7.9 and 5.9 kb, respectively (Table I). As shown in
Figure 3A and Table I, five immortal cell lines derived from
MDAH 087 cells exhibited variabilities in mean TRF lengths.
One cell line had a long TRF length (21.3 kb) and some had
intermediate (13.0 or 14.0 kb) or short TRF lengths (8.0 or 8.3
kb). Although the amount of DNA electrophoresed in the
agarose gel was the same in each lane when detected by UV
illumination (data not shown), the intensity of TRF signals
varied in each lane (Figure 3A), suggesting that the amount
of telomeric DNA varied among individual cell lines examined. The size and intensity of TRF signals decreased after
digestion with exonuclease Bal 31 (data not shown), indicating
that the TTAGGG-hybridizing fragments were telomeric.
To examine whether the variability in the TRF lengths
between the five immortal cell lines was due to clonal heterogeneity or heterogeneity of the parental cell population, we
isolated five subclones from each of three immortal cell lines,
which had either long (LCS-AF.1-2 cells), intermediate (LCS4 2 cells) or short TRF length (LCS-AF.1-3 cells). Analysis
of TRF lengths in these subclones revealed that the mean TRF
length of every subclone was different from each other for all
the immortal cell lines examined (Table I and Figure 3B). The
same results were obtained even when cells were digested with
either an excess amount of HinfI or HinfI for a long time. To
confirm the clonal variability in the TRF lengths, five resubclones were re-isolated from one subclone (LCS-AF.1-2
Cl 1), which had a long TRF length (16.5 kb). Most re-subclones exhibited similar levels of TRF length compared with
that of the parental subclone, but one re-subclone (LCS-AF.1-2
Cl 1-5) showed a short TRF length. The same results were
obtained with five re-subclones re-isolated from the other
subclone (LCS-AF.1-2 Cl 4) with a short TRF length (5.4 kb)
Genetic instability and telomere maintenance
(Table II). Estimated population doublings (EPDs) of the
subclones or re-subclones after subcloning or re-subcloning
were 22 to 24 PDs when they were analyzed for TRF length.
The EPDs were required for a single cell to grow to the number
of cells sufficient for the TRF length analysis. No senescence
or significant changes in cell growth determined by cell population doubling rates were observed between the parental cell
lines and their subclones and re-subclones. Particularly, LCSAF.1-2 Cl 4-3 cells, a re-subclone of LCS-AF.1-2 cells at 481
PDs, had short TRF length (5.4 kb), but showed no signs of
decreased growth over 100 PDs after re-subcloning (data not
Table II. The mean TRF length and telomerase status in re-subclones
re-isolated from subclones of immortal cell lines
LCS-AF.1-2 Cl 1
(TRF ˆ 16.5 kb)
LCS-AF.1-2 Cl 4
(TRF ˆ 5.4 kb)
a
b
Re-subclonea
Mean TRF
length (kb)
Telomerase
activityb
Cl
Cl
Cl
Cl
Cl
1-1
1-2
1-3
1-4
1-5
22.1
19.2
19.5
20.1
9.3
±
±
±
±
±
Cl
Cl
Cl
Cl
Cl
4-1
4-2
4-3
4-4
4-5
8.2
15.9
5.4
7.0
6.4
±
±
±
±
±
Estimated population doublings were 23 after re-subcloning.
ÿ shows nondetectable-telomerase activity.
Chromosome instability
Chromosome aberrations were detected at a very low frequency in a normal human fibroblast strain (WHE-7 cells)
(18) that was used as a negative control (Table III). In contrast,
various types of chromosome aberrations were observed at
high frequencies in MDAH 087 cells and its immortal derivatives as well as the subclones and re-subclones of the immortal
cell lines (Table III). In MDAH 087 cells at 19 PDs, 62% of the
metaphases exhibited various types of chromosome aberrations. Dicentric chromosomes and DMs were observed in 42
and 14% of the metaphases, respectively. The frequencies of
cells with aberrant chromosomes were higher in AFB1 induced immortal cell lines (LCS-AF.1-2, LCS-AF.1-3, and
LCS-AF.3-1) and an X-ray-induced cell line (LCS-4 2 cells)
than in the spontaneously immortalized cell line (LCS-ST
cells). In AFB1 - and X-ray-induced immortal cell lines,
dicentric chromosomes or DMs were observed in 410 or
46% of metaphases, respectively. Frequencies of aberrant
metaphases in the immortal cell lines were similar between
the same cell lines with different levels of PDs.
To examine the generation of chromosome aberrations in
individual cells composing the population of immortal cell
lines, five subclones were isolated from each immortal cell
line and scored for chromosome aberrations when the
Table III. Chromosome aberrations in MDAH 087 cells, their immortal cell lines, and human tumor-derived cell lines
PDb
Cells
Type of aberrations (%)c
G
WHE-7a
MDAH 087
LCS-ST
B
E
D
R
DM
Aberrant metaphasesd
(%)
LCS-AF.3-1
LCS-4 2
20
19
108
309
69
469
74
477
439
481
3
4
2
0
6
0
0
2
0
6
0
16
4
10
20
32
36
20
12
22
0
6
0
0
0
0
2
0
0
2
0
42
18
0
18
10
16
32
26
10
0
0
2
0
0
0
0
0
0
0
0
14
2
18
56
46
66
58
84
50
3
62
22
26
64
64
80
74
94
74
[Subclone of LCS-AF.1-2 (PD 481)]
LCS-AF.1-2 Cl 1
LCS-AF.1-2 Cl 2
LCS-AF.1-2 Cl 3
LCS-AF.1-2 Cl 4
LCS-AF.1-2 Cl 5
22e
23e
23e
22e
23e
10
2
8
4
2
12
24
14
24
24
0
4
1
0
2
6
24
16
36
24
0
0
0
0
0
60
82
82
90
84
74
94
90
94
90
22e
8
16
0
18
0
52
78
22e
6
34
0
44
0
88
96
4
1
3
2
0
2
2
4
0
0
0
0
0
0
0
0
0
0
0
0
1
4
0
0
5
6
6
6
LCS-AF.1-2
LCS-AF.1-3
[Re-subclone of LCS-AF.1-2 Cl 1]
LCS-AF.1-2 Cl 1-1
[Re-subclone of LCS-AF.1-2 Cl 4]
LCS-AF.1-2 Cl 4-1
‡= ‡
)
Wild HT1080 cells ( p53
Mutant HT1080 cells ( p53 ÿ= ÿ )
HeLa cells with a low level of PDsf
HeLa cells with a high level of PDsf
a
A normal human fibroblast strain as described previously (17).
Population doubling.
c
G, gaps; B, breaks; E, exchanges; D, dicentric chromosomes; R, ring chromosomes; DM, double minute chromosomes.
d
Only a restricted subset of aberrations represented in the column of type of aberrations was scored.
e
Estimated population doublings after subcloning or re-subcloning.
f
Three hundred and two difference in PD between both the HeLa cells.
b
957
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
Parental clone
shown). Cell viability was ~99% in all cells when assayed by
the dye exclusion test with trypan blue (data not shown). In
addition, very few cells underwent apoptosis, as characterized
by nuclear fragmentation and/or chromatin condensation (data
not shown) (28).
T.Tsutsui et al.
subclones were at 22 or 23 PDs after subcloning. As shown in
Table III, high percentages of chromosome aberrations were
detected in each subclone of LCS-AF.1-2 cells. The principal
types of aberrations were breaks, dicentric chromosomes and
DMs, as observed in the parental cell line, LCS-AF.1-2 cells.
Similar results were obtained from all subclones of other
immortal cell lines (LCS-ST, LCS-AF.1-3, LCS-AF.3-1, and
LCS-4 2) (data not shown). The generation of chromosome
aberrations was also observed in re-subclones re-isolated from
subclones of LCS-AF.1-2 cells.
Table IV. Number of new bands present in human immortal cell lines with
different levels of PDsa
Cell line
958
33.15b
33.6b
HaeIIIc
HinfIc
HaeIIIc
HinfIc
LCS-ST
109
318
2
2d
2
2d
0
0
0
0
LCS-AF.1-2
81
491
0
0
1
0e
0
0
1
0e
LCS-AF.1-3
105
499
1
1d
1
1d
2
3f
2
2d
LCS-AF.3-1
120
466
1
0e
1
0e
0
2g
0
1g
LCS-4 2
120
451
0
0
0
0
0
1g
0
0
a
Fig. 4. DNA fingerprints of MDAH 087 cells and their immortal cell lines
with the early levels of population doublings obtained with either the
multilocus probe 33.15 after digestion with HaeIII (A) or the multilocus probe
33.6 after digestion with HinfI (B). Arrows indicate new minisatellite bands
gained in the immortal cell lines. Size markers are indicated on the left side of
the autoradiogram.
PD
Genomic DNA isolated from cell lines at various PDs was digested with
HaeIII or HinfI, and hybridized with alkaline phosphatase-labeled multilocus
probes (33.15 and 33.6) after Southern blotting as described in Materials and
methods. DNA fingerprint profiles of the cell lines at different PDs were
compared with those of the parental MDAH 087 cells at 18 PDs.
b
Names of multilocus probes used for DNA fingerprinting.
c
Restriction enzymes used for DNA digestion.
d
Bands identical to those present at the earlier time points examined.
e
Loss of the new band present at the earlier time points examined.
f
Two bands were identical to those present at the earlier time points examined.
The other one was a new band.
g
New bands detected at the later time points examined.
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
Minisatellite instability
Minisatellite alterations in the five immortal cell lines with
different levels of PDs were determined from their DNA
fingerprint profiles obtained after digestion of their DNAs
with HaeIII or HinfI followed by hybridization with the multilocus probe 33.15 or 33.6. When compared with DNA fingerprint profiles of the parental MDAH 087 cells at 19 PDs, those
of the five immortal cell lines at 81 to 120 PDs were related but
not identical with each other (Figure 4). One or two new bands
were gained in LCS-ST cells at 109 PDs, LCS-AF.1-2 cells at
81 PDs, LCS-AF.1-3 cells at 105 PDs and LCS-AF.3-1 cells at
120 PDs (Figure 4 and Table IV), but no new bands were
observed in LCS-4 2 cells at 120 PDs (data not shown)
(Table IV). When assayed in the cell lines with the late levels
of PDs (318±499 PDs), some new bands detected at the early
levels of PDs were stably present in LCS-ST cells (Figure 5A)
and LCS-AF.1-3 cells (Figure 5A and B), but lost in LCSAF.1-2 cells (Figure 5B) and LCS-AF.3-1 cells (Figure 5A).
Meanwhile, one or two new bands not seen at the earlier time
points were gained in LCS-AF.1-3 cells (data not shown),
LCS-AF.3-1 cells (Figure 5B) and LCS-4 2 cells (data not
shown) (Table IV). The DNA fingerprints of wild HT 1080
cells and HeLa cells were markedly different from those of
these immortal cell lines (Figure 5). No minisatellite alterations were observed in the DNA fingerprint profiles of a
normal human fibroblast strain (WHE-7) with different levels
of PDs (5 PDs and 53 PDs) (data not shown).
To examine clonal heterogeneity in minisatellite alterations
in the immortal cell lines, we isolated five to 10 subclones
from each of LCS-ST cells, LCS-AF.1-2 cells and LCS-AF.1-3
cells. Ten subclones were isolated from LCS-ST cells at 318
PDs, and DNA fingerprinting was performed after digestion
with HaeIII or HinfI followed by hybridization with 33.15 or
33.6. DNA fingerprint profiles of each of 10 subclones were
identical to those of the parental LCS-ST cells. Band losses
and gains of new bands were not observed in any of the 10
subclones (data not shown). Five subclones were isolated from
LCS-AF.1-2 cells at 487 PDs. DNA fingerprint profiles of four
of the five subclones were identical to those of the parental
LCS-AF.1-2 cells when obtained after digestion with HaeIII or
HinfI followed by hybridization with 33.15 or 33.6. However,
one subclone (LCS-AF.1-2 Cl 1) gained a new minisatellite
band not present in the parental cell line when the DNA was
digested with HaeIII and subsequently hybridized with 33.15
(Figure 6A). One new band was also observed in LCS-AF.1-2
Cl 1 subclone when the DNA was digested with HinfI and then
hybridized with 33.15 or 33.6 (data not shown). Six subclones
were isolated from LCS-AF.1-3 cells at 496 PDs, and the
DNAs were digested with HaeIII or HinfI and hybridized
with 33.15 or 33.6. One (LCS-AF.1-3 Cl 4) of the six subclones gained a new band when DNA fingerprints were
obtained with the multilocus probe 33.6 after digestion with
HaeIII (Figure 6B). EPDs of the subclones after subcloning
were 22 to 24 PDs when they were analyzed for DNA fingerprinting. These results indicate that each of the parental cell
Genetic instability and telomere maintenance
lines are not heterogenous populations at the level of minisatellite sequences, but minisatellite alterations are detected
more frequently in cell lines immortalized by AFB1 than in a
cell line immortalized spontaneously.
To examine the minisatellite instability, 30 re-subclones
were isolated from LCS-AF.1-2 Cl 1 subclone and underwent
a DNA fingerprint analysis, 22 to 24 PDs after re-subcloning.
When compared with DNA fingerprint profiles of the parental
subclone LCS-AF.1-2 Cl 1, three of the 30 re-subclones gained
one or two new bands after re-subcloning (Figure 7 and
Table V). Thirty re-subclones isolated from a subclone (LSTST Cl 1) of LCS-ST cells at 311 PDs were also analyzed with
DNA fingerprinting performed after digestion of their DNAs
with HaeIII or HinfI followed by hybridization with 33.15 or
33.6. The DNA fingerprint profiles of all the 30 re-subclones
were identical to those of the parental subclone (data not
shown). All new minisatellite bands detected in the present
study were not due to partial digestion of cellular DNAs with
HaeIII or HinfI, because of the reproducibility of the new band
gains and the presence of two clear bands when the same
membranes used in the DNA fingerprint analyses were hybridized with the single locus probe MS51 (data not shown). The
MS51 locus is particularly sensitive to partial digestion (29).
Table V. Re-subclones which gained new minisatellite bands not present in
the parental subclone LCS-AF.1-2 Cl 1 and the number of new bandsa
Re-subclones
LCS-AF.1-2 Cl 1-2
LCS-AF.1-2 Cl 1-7
LCS-AF.1-2 Cl 1-9
33.15b
33.6b
HaeIIIc
HinfIc
HaeIIIc
HinfIc
0
0
0
0
0
1
0
1
0
1
1
0
a
Thirty re-subclones were isolated from LCS-AF.1-2 Cl 1 subclone and
analyzed for minisatellite alterations as described in Materials and methods.
Three of the 30 re-subclones listed in this table gained one or two new bands.
b
Names of multilocus probes used for DNA fingerprinting.
c
Restriction enzymes used for DNA digestion.
Lack of microsatellite instability
Microsatellite sequences in the five immortal cell lines at
318 to 496 PDs were not significantly altered when
compared with the parental MDAH 087 cells. Typical results
are shown in Figure 8. Microsatellite alterations were also not
detected in a comparison between five subclones and 30
re-subclones of LCS-AF.1-2 cells when analyzed with a
959
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
Fig. 5. DNA fingerprints of the immortal cell lines with the late levels of population doublings. DNA fingerprints were obtained with either the multilocus
probe 33.15 after digestion with HaeIII (A) or the multilocus probe 33.6 after digestion with HinfI (B). An arrow indicates a new minisatellite band present in
the immortal cell lines. Solid arrowheads represent the minisatellite bands stably present in the immortal cell lines after the early population doublings
examined. Open arrowheads denote a loss of minisatellite bands present in the immortal cell lines when assayed at the early population doublings. Size markers
are indicated on both sides of the autoradiogram.
T.Tsutsui et al.
microsatellite marker BAT25 (Figure 8). The same results
were obtained with the other immortal cell lines (data not
shown).
Chromosome instability and minisatellite instability in p53
affected tumor-derived human cell lines
As the wild-type p53 is lost in the MDAH 087-derived immortal cell lines (17,18), inactivation of p53 is considered as a
possible mechanism for the chromosomal and minisatellite
instabilities in the cell lines. To examine the possibility, chromosome aberrations and DNA fingerprint profiles of wild
HT1080 cells having two normal p53 alleles were compared
with those of mutant HT1080 cells carrying mutated p53
alleles, one of which was affected at codon 245 in exon 7
and the other of which was mutated at codon 277 in exon 8
(24). Chromosome aberrations in both types of HT1080 cells
were detected at low frequencies. However, there were no
significant differences in the frequencies between both
HT1080 cells (Table III). The frequencies of chromosome
aberrations were also compared between HeLa cells with low
and high levels of PDs. There were 302 PDs of difference
between the HeLa cells. Both HeLa cells had chromosome
960
aberrations at low frequencies, but there were no significant
differences between them (Table III).
DNA fingerprint analysis using two multilocus probes 33.15
and 33.6 after digestion with HaeIII or HinfI demonstrated that
DNA fingerprints were identical between wild HT1080 cells
and mutant HT1080 cells, as well as between HeLa cells with
low and high levels of PDs. The difference in the number of
PDs was 302. The results are shown in Figures 9 and 10. In
addition, DNA fingerprint profiles of each of five subclones
isolated from each type of HT1080 cells and both HeLa cells,
and 30 re-subclones isolated from one of the five subclones
were also identical to those of their parental cells when analyzed with the multilocus probes 33.15 and 33.6 after digestion
with HaeIII or HinfI (data not shown). Each type of HT1080
cells exhibited a distinguishable morphology (Figure 11), suggesting that the identical patterns in their DNA fingerprints are
not attributed to cell contamination with each other.
Discussion
We examined telomerase activity and TRF length in immortal
cell lines derived from a LFS patient with a mutated p53 allele.
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
Fig. 6. DNA fingerprints of the immortal cell lines and their subclones obtained after digestion with HaeIII followed by hybridization with the multilocus
probe 33.15 (A) or 33.6 (B). Arrows indicate new bands present in the subclones. Size markers are indicated on both sides of the autoradiogram.
Genetic instability and telomere maintenance
Fig. 8. Microsatellite sequences in the immortal cell lines and their subclones and re-subclones.
961
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
Fig. 7. DNA fingerprints of LCS-AF.1-2 cells and their subclone (LCS-AF.1-2 Cl 1) and re-subclones obtained with the multilocus probe 33.15 after digestion with
HinfI (A), the multilocus probe 33.6 after digestion with HaeIII (B), and the multilocus probe 33.6 after digestion with HinfI (C and D). Arrows indicate new
minisatellite bands gained in the re-subclones. Size markers are indicated on the left side of the autoradiogram.
T.Tsutsui et al.
Telomerase activity was undetectable in all the cell lines, and
all their subclones and re-subclones. The mean TRF lengths
varied considerably between the cell lines. The variability in
TRF lengths was also observed within subclones and resubclones isolated from the cell lines. The TRF variability
in the subclones and re-subclones was observed when the
cells were examined at 22 to 24 PDs after subcloning or resubcloning. The variability in the TRF lengths was not owing
to DNA degradation, because the same DNAs used for the
TRF analysis exhibited no degradation as shown in the DNA
fingerprint profiles (Figure 4). The results indicate that TRF
variability is not owing to heterogeneity of the parental MDAH
087 cell population rather to clonal heterogeneity. In addition,
they show that the variability arises in short time periods and is a
continuous phenotype of the cell lines examined.
Recent studies on TRF lengths in in vitro immortalized cell
lines reported that telomerase-negative human cell lines have
considerably longer TRF lengths than the normal cells from
which they are derived (5,7,8,30,31). The long TRF lengths
are maintained in the clones of the cell lines (32). All of
telomerase-negative cell lines derived from human tumors and
the small proportion of human tumors that are negative for
telomerase activity also maintain very long TRF lengths (9,31).
The mechanisms of telomere maintenance might be different
between telomerase-negative cell lines with variable and
unstable TRF lengths and telomerase-negative cell lines with
962
Fig. 10. DNA fingerprints of HeLa cells obtained with the multilocus probe
33.15 or 33.6 after digestion with HaeIII or HinfI. HeLa (A) is different
from HeLa (B) in the number of population doublings (PDs). HeLa (A) was
302 PDs younger than HeLa (B).
very long and stable TRF lengths. Bryan et al. (33) reported
that mutations in human telomerase RNA (hTERC) were
not the mechanism of the telomere elongation in telomerasenegative human cell lines with very long and stable TRF
lengths. In addition, it has also been demonstrated that lack
of telomerase activity of the cell lines is not due to the absence
of hTERC gene expression (33). All the immortal Li-Fraumeni
cell lines used in the present study were positive for hTEP1 or
hTERC but negative for hTERT gene expression, which is consistent with other cell lines lacking telomerase activity (34).
Various types of chromosome aberrations were found in
the parental MDAH 087 cells and their immortal cell lines
(Table III). High frequencies of chromosome aberrations are
also observed in MDAH 087 cells reported by others (19) and
in fibroblasts derived from a LFS patient different from the
MDAH 087 kindred (35). The frequencies of cells with chromosome aberrations were higher in the cell lines immortalized
by a chemical or physical mutagen (AFB1 or X-ray) than in a
cell line immortalized spontaneously. The most remarkable
difference in the chromosome aberrations between the mutagen-induced cell lines and the spontaneously immortalized cell
line was an increase in the frequency of DMs in the mutagenimmortalized cell lines. Because viability of cells in the cell
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
Fig. 9. DNA fingerprints of both types of HT1080 cells obtained with the
multilocus probe 33.15 or 33.6 after digestion with HaeIII or HinfI. Wild
HT1080, HT1080 cells having two normal p53 alleles; mutant HT1080,
HT1080 cells having two mutant alleles of p53.
Genetic instability and telomere maintenance
lines was ~99%, and the cell lines showed neither senescence
nor changes in growth rates, most cells with aberrant chromosomes maintain viability. High frequencies of chromosome
aberrations were also detected in the subclones and resubclones of the immortal cell lines. Moreover, the frequencies
of chromosome aberrations were maintained at similar levels
between the subclones and re-subclones as well as between the
same cell lines with different levels of PDs. The results indicate that chromosome instability is a stable characteristic of
these immortal cell lines.
The loss of p53 tumor suppressor function results in genetic
instability, characteristically associated with changes in chromosome ploidy and gene amplification (13,14,36,37). Abnormal amplification of centrosomes, which become spindle poles
during mitosis and participate in chromosomal segregation, is
frequently observed in embryonic fibroblasts derived from
p53-nullizygous mice (p53 ÿ= ÿ mice). The cells from various
organs of the p53 ÿ= ÿ mice display changes in chromosome
ploidy (aneuploidy) (37). We demonstrated previously that the
number of chromosomes of MDAH 087 cells (PD 19) are
mainly distributed in the hypodiploid range (33±38) with a
modal number of 40 (17). In contrast, the chromosome number
of LCS-ST cells (PD 108), LCS-AF.1-2 cells (PD 100), LCSAF.1-3 cells (PD 141), LCS-AF.3-1 cells (PD 75) and LCS4 2 cells (PD 123) are mainly distributed in the hypotriploid
and hypotetraploid ranges (61±80) (17,18). Changes in the
chromosome ploidy observed in these cell lines may be attributed to the loss of p53 functions, because the cell lines, which
were derived from the MDAH 087 cell strain carrying one
963
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
Fig. 11. Phase contrast photomicrographs of both types of HT1080 cells. W,
wild HT1080 cells; M, mutant HT1080 cells. Bars denote 100 mm.
wild-type p53 allele and one allele with a mutation at codon
248 in exon 7, lose the wild-type p53 allele during immortalization (14,17).
Dicentric chromosomes were detected at similar or lower
frequencies in the immortal cell lines than in the parental
MDAH 087 cells at 19 PDs. This is consistent with the findings
that the frequency of dicentric chromosomes in human fibroblasts increases as the cells enter senescence (38) but decreases
upon immortalization (39). Dicentric chromosomes may arise
from telomere association and fusion as a direct result of loss
of telomeric sequences (40). The high frequencies of dicentric
chromosomes observed in the immortal cell lines suggest that
telomere lengths of the cell lines are not stably maintained, and
that the telomere shortening is a usual event in the cell populations. It is worthy to note that resulting dicentric chromosomes
can enter bridge-breakage±fusion cycles that lead to gene amplification (41). Smith et al. (41) found that in N-(phosphonoacetyl)-L-aspartate-resistant Syrian hamster BHK cells, about
one-third of the newly formed chromosomes carrying amplified CAD genes are dicentric chromosomes. Gene amplification, a type of genomic rearrangement, leads to increased gene
expression through alteration of gene copy number, and is
often accompanied by the generation of DMs (42). Livingstone
et al. (13) demonstrated that the frequency of CAD gene
amplification was at undetectable levels in early passage
MDAH 087 cells at 15 or 27 PDs, which carry the wild-type
p53 allele, but increased in immortal cell lines after loss of
wild-type p53 allele. The wild-type p53 allele is also retained
in MDAH 087 cells at 18 PDs used in the present study (17),
predicting that gene amplification may not occur in the cells.
However, as shown in Table III, 14% of the MDAH 087 cells
at 19 PDs are affected with DMs, a structure associated with
gene amplification (42). Furthermore, the frequencies of DMs
observed in AFB1 - or X-ray-induced MDAH 087 immortal
cell lines (LCS-AF.1-2, LCS-AF.1-3, LCS-AF.3-1 and LCS4 2) are much higher than those observed in the spontaneously immortalized LCS-ST cells (Table III). When analyzed at 4350 PDs, the immortal cell lines including LCS-ST
cells and the mutagen-induced immortal cells lose the wildtype p53 allele, and no additional mutations are induced in
exons 2±10 of the other allele (17,18). In addition to the loss of
the wild-type p53 allele, some other mutational event might be
necessary to allow the cells to undergo gene amplification.
Substantial levels of dicentric chromosomes and DMs were
detected in the immortal cell lines, suggesting that the cell
lines undergo gene amplification.
Minisatellite alterations defined as gain of new minisatellite
bands were observed in all immortal cell lines examined.
When compared with DNA fingerprints of the parental
MDAH 087 cells, most cell lines gained new bands when
assayed at the early levels of PDs. In LCS-ST cells, the new
bands were stable for 4200 PDs and no additional alterations
were observed in the cells with the late levels of PDs. On the
other hand, in each of the other four cell lines, which were
mutagenized by AFB1 or X-ray (17,18), minisatellite alterations were observed both at the early and late levels of PDs. In
LCS-AF.1-2 cells, new bands present at the early passage were
lost at the late passage. In LCS-AF.1-3 cells, not only all new
bands detected at the early passage but also an additional new
band was observed at the late passage. In LCS-AF.3-1 cell
lines, all new bands observed at the early passage were lost but
three new bands were present at the late passage. In LCS-4 2
cells, no new bands were detected at the early passage but a
T.Tsutsui et al.
964
observed TRF variability in the cell lines might be coordinately
controlled by a mechanism(s) associated genomic instability
that is involved in gene amplification and/or recombination,
although the data presented support this hypothesis only indirectly. Minisatellite-binding proteins (Msbps) are postulated to
play a role in DNA recombination process (47). Msbps bind to
a repetitive guanine-rich strand of the minisatellite duplex
(47,48), and the binding properties of Msbp-4 are affected by
dephosphorylation (48). Because minisatellite alterations are
induced in mice by DNA damaging agents (44,45) or in mouse
cells by okadaic acid, a potent inhibitor of protein phosphatase
(49), some structural or functional alterations in Msbps could
lead cells to undergo recombination of the G-rich strand of
telomeric DNA. The TRF variability observed in the present
study might be due to the Msbp alterations that cause recognition errors at recombination sites.
Acknowledgements
We appreciate the help and advice of Drs E.Hiyama and K.Hiyama (Hiroshima
University, School of Medicine, Japan) and the critical comments of
Drs Hitoshi Nakagama (National Cancer Center Research Institute, Japan)
and Pat Volt (National Institute of Environmental Health Sciences). We
thank Dr J.W.Shay (The University of Texas) for providing the internal
TRAP assay standard probe through Dr E.Hiyama. This study was supported
in part by grants-in-aid from the Ministry of Education, Sciences and Culture
in Japan.
References
1. Hayflick,L. and Moorhead,P.S. (1961) The serial cultivation of human
diploid cell strains. Exp. Cell Res., 25, 585±621.
2. Goldstein,S. (1990) Replicative senescence: the human fibroblast comes of
age. Science, 249, 1129±1133.
3. Harley,C.B., Futcher,A.B. and Greider,C.W. (1990) Telomeres shorten
during aging of human fibroblasts. Nature, 345, 458±460.
4. Greider,C.W. and Blackburn,E.H. (1985) Identification of a specific
telomere terminal transferase activity in Tetrahymena extracts. Cell, 43,
405±413.
5. Kim,N.W., Piatyszek,M.A., Prowse,K.R., Harley,C.B., West,M.D.,
Ho,P.L.C., Coviello,G.M., Wright,W.E., Weinrich,S.L. and Shay,J.W.
(1994) Specific association of human telomerase activity with immortal
cells and cancer. Science, 266, 2011±2015.
6. Murnane,J.P., Sabatier,L., Marder,B.A. and Morgan,W.F. (1994)
Telomere dynamics in an immortal human cell line. EMBO J., 13,
4953±4962.
7. Bryan,T.M., Engelzou,A., Gupta,J., Bacchetti,S. and Reddel,R.R. (1995)
Telomere elongation in immortal human cells without detectable
telomerase activity. EMBO J., 14, 4240±4248.
8. Sugihara,S., Mihara,K., Marunouchi,T., Inoue,H. and Namba,M. (1996)
Telomere elongation observed in immortalized human fibroblasts by
treatment with 60 Co gamma rays or 4-nitroquinolone 1-oxide. Hum.
Genet., 97, 1±6.
9. Morii,K., Tanaka,R., Onda,K., Tsumanuma,I. and Yoshimura,J. (1997)
Expression of telomerase RNA, telomerase activity and telomere length in
human gliomas. Biochem. Biophys. Res. Commun., 239, 830±834.
10. New,J.H., Sugiyama,T., Zaitseva,E. and Kowalczykowski,S.C. (1998)
Rad52 protein stimulates DNA strand exchange by Rad51 and replication
protein A. Nature, 391, 407±410.
11. Yeager,T.R., Neumann,A.A., Englezou,A., Huschtscha,L.I., Noble,J.R.
and Reddel,R.R. (1999) Telomerase-negative immortalized human cells
contain a novel type of promyelocytic leukemia (PML) body. Cancer Res.,
59, 4175±4179.
12. Dunham,M.A., Neumann,A.A., Fasching,C.L. and Reddel,R.R. (2000)
Telomere maintenance by recombination in human cells. Nature Genet.,
26, 447±450.
13. Livingstone,L.R., White,A., Sprouse,J., Livanos,E., Jacks,T. and
Tlsty,T.D. (1992) Altered cell cycle arrest and gene amplification
potential accompany loss of wild-type p53. Cell, 70, 923±935.
14. Yin,Y., Tainsky,M.A., Bischoff,F.Z., Strong,L.C. and Wahl,G.M. (1992)
Wild-type p53 restores cell cycle control and inhibits gene amplification
in cell with mutant p53 allele. Cell, 70, 937±948.
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
new band was detected at the late passage. Furthermore, minisatellite alterations were observed in subclones and re-subclones isolated from LCS-AF.1-2 cells but not in those done
from LCS-ST cells. The results suggest that minisatellite
instability occurs preferentially in the mutagen-treated cell
lines. AFB1 is a powerful inducer of mitotic recombination
events (43). Minisatellite rearrangements are also increased in
liver tumor induced by transplacental AFB1 treatment of hepatitis B virus transgenic mice, but not in spontaneously arising
tumor (44). Moreover, ionizing radiation induces minisatellite
alterations in mice (45). Therefore, the ongoing instability
observed in the immortal cell lines could be due to some
genetic change induced by treatment with AFB1 or X-ray.
This is the first report that minisatellite alterations were
observed in human cells. The mechanism by which minisatellite alterations were also observed in the spontaneously
immortalized LCS-ST cells is not clear. Additional genetic
event(s) along with the p53 mutation may participate in the
minisatellite alteration, because the donor of the MDAH 087
cell strain had received therapeutic treatment (chemotherapy
and/or radiation) 2±13 years before the skin biopsy (19).
DNA recombination is induced by inactivation of p53 (16).
Loss of the wild-type p53 enhances gene amplification rates
(13). The mechanism underlying gene amplification is not
elucidated but likely requires DNA strand exchange, and therefore utilizes the cell's enzymatic machinery for DNA recombination (16). WTK1 and TK6, both of which are human
lymphoblast cell lines from one donor, exert different capacities to catalyze recombination. WTK1 having a homozygous
mutation in p53 at codon 237 in exon 7 has more capacity for
catalyzing recombination than TK6 containing a wild-type p53
allele (46). However, there was no difference in the DNA
fingerprint profiles between HT1080 cells with and without
functional p53 as well as between HeLa cells with low and
high levels of PDs. Although we cannot exclude the possibility
of specific mutations of p53 having the ability to catalyze
recombination, these results suggest that inactivation of p53
is not sufficient to create genomic instability in minisatellite
sequences.
In contrast to minisatellite instability, microsatellite instability was not observed in any of the immortal Li-Fraumeni cell
lines examined here. Microsatellite instability is frequently
observed in a variety of tumors as a consequence of mutations
in mismatch repair genes (15). As impairment of the mismatch
repair system is considered mainly to result in point mutations,
the TRF heterogeneity and chromosomal and minisatellite
instabilities observed in the present study might be based on
the molecular mechanism distinct from the mismatch repair
system.
In the present study, we demonstrate that several, independently established telomerase-negative Li-Fraumeni fibroblast
cell lines that have lost the wild-type p53 allele show variability in TRF lengths between the cell lines. The clonal variability in TRF lengths occurs in each cell line. Each cell line
exhibits high levels of chromosome aberrations; the principal
types of which are dicentric chromosomes and DMs, structures
associated with gene amplification, and genomic rearrangements. In addition, minisatellite alteration is found in the cell
lines as well. In comparison, human immortal cell lines
HT1080 cells and HeLa cells exhibit few dicentric chromosomes and DMs, and little minisatellite instabilities. This indicates that chromosomal and minisatellite instabilities observed
in these cell lines may not result from loss of p53 function. The
Genetic instability and telomere maintenance
33. Bryan,T.M., MarusÏic,L., Bacchetti,S., Namba,M. and Reddel,R.R. (1997)
The telomere lengthening mechanism in telomerase-negative immortal
human cells does not involve the telomerase RNA subunit. Hum. Mol.
Genet., 6, 921±926.
34. Nakamura,T.M.,
Morin,G.B.,
Chapman,K.B.,
Weinrich,S.L.,
Andrews,W.H., Lingner,J., Harley,C.B. and Cech,T.R. (1997)
Telomerase catalytic subunit homologs from fission yeast and human.
Science, 277, 955±959.
35. Rogan,E.M., Bryan,T.M., Hukku,B., Maclean,K., Chang,A.C.-M.,
Moy,E.L.,
Englezou,A.,
Warneford,S.G.,
Dalla-Pozza,L.
and
Reddel,R.R. (1995) Alterations in p53 and p16INK4a expression and
telomere length during spontaneous immortalization of Li-Fraumeni
syndrome fibroblasts. Mol. Cell Biol., 15, 4745±4753.
36. Cross,S.M., Sanchez,C.A., Morgan,C.A., Schimke,M.K., Ramel,S.,
Idzerda,R.L., Raskind,W.H. and Reid,B.J. (1995) A p53-dependent
mouse spindle checkpoint. Science, 267, 1353±1356.
37. Fukasawa,K., Wiener,F., Woude,G.F.V. and Mai,S. (1997) Genomic
instability and apoptosis are frequent in p53 deficient young mice.
Oncogene, 15, 1295±1302.
38. Benn,P.A. (1976) Specific chromosome aberrations in senescent fibroblast
cell lines derived from human embryos. Am. J. Hum. Genet., 28, 465±473.
39. Counter,C.M., Avilion,A.A., LeFeuvre,C.E., Stewart,N.G., Greider,C.W.,
Harley,C.B. and Bacchetti,S. (1992) Telomere shortening associated with
chromosome instability is arrested in immortal cells which express
telomerase activity. EMBO J., 11, 1921±1929.
40. Counter,C.M. (1996) The roles of telomeres and telomerase in cell life
span. Mutat. Res., 366, 45±63.
41. Smith,K.A., Stark,M.B., Gorman,P.A. and Stark,G.R. (1992) Fusions near
telomeres occur very early in the amplification of CAD genes in Syrian
hamster cells. Proc. Natl Acad. Sci. USA, 89, 5427±5431.
42. Wahl,G.M. (1989) The importance of circular DNA in mammalian gene
amplification. Cancer Res., 49, 1333±1340.
43. Sengstag,C. and W
urgler,F.E. (1994) DNA recombination induced by
aflatoxin B1 activated by cytochrome P450 1A enzymes. Mol.
Carcinogen., 11, 227±235.
44. Kaplanski,C., Chisari,F.V. and Wild,C.P. (1997) Minisatellite rearrangements are increased in liver tumors induced by transplacental aflatoxin B1
treatment of hepatitis B virus transgenic mice, but not in spontaneously
arising tumors. Carcinogenesis, 18, 633±639.
45. Dubrova,Y.E., Jeffreys,A.J. and Malashenko,A.M. (1993) Mouse minisatellite mutations induced by ionizing radiation. Nature Genet., 5, 92±94.
46. Xia,F., Amundson,S.A., Nickoloff,J.A. and Liber,H.L. (1994) Different
capacities for recombination in closely related human lymphoblastoid cell
lines with different mutational responses to x-irradiation. Mol. Cell Biol.,
14, 5850±5857.
47. Collick,A., Dunn,M.G. and Jeffreys,A.J. (1991) Minisatellite binding
protein Msbp-1 is a sequence-specific single-stranded DNA-binding
protein. Nucleic Acids Res., 19, 6399±6404.
48. Yamazaki,H., Nomoto,S., Mishima,Y. and Kominami,R. (1992) A 35-kDa
protein binding to a cytosine-rich strand of hypervariable minisatellite
DNA. J. Biol. Chem., 267, 12311±12316.
49. Nakagama,H.,
Kaneko,S.,
Shima,H.,
Inamori,H.,
Fukuda,H.,
Kominami,R., Sugimura,T. and Nagao,M. (1997) Induction of minisatellite mutation in NIH3T3 cells by treatment with the tumor promoter
okadaic acid. Proc. Natl Acad. Sci. USA, 94, 10813±10816.
Received July 16, 2002; revised January 7, 2003;
accepted February 11, 2003
965
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on April 26, 2014
15. Liu,B., Farrington,S.M., Petersen,G.M. et al. (1995) Genetic instability
occurs in the majority of young patients with colorectal cancer. Nature
Med., 1, 348±352.
16. Mekeel,K.L., Tang,W., Kachnic,L.A., Luo,C.-M., DeFrank,J.S. and
Powell,S.N. (1997) Inactivation of p53 results in high rates of homologous
recombination. Oncogene, 14, 1847±1857.
17. Tsutsui,T., Fujino,T., Kodama,S., Tainsky,M.A., Boyd,J. and Barrett,J.C.
(1995) Aflatoxin B1 -induced immortalization of cultured skin fibroblasts
from a patient with Li-Fraumeni syndrome. Carcinogenesis, 16, 25±34.
18. Tsutsui,T., Tanaka,Y., Matsudo,Y., Hasegawa,K., Fujino,T., Kodama,S.
and Barrett,J.C. (1997) Extended lifespan and immortalization of human
fibroblasts induced by X-ray irradiation. Mol. Carcinogen., 18, 7±18.
19. Bischoff,F.Z.,
Yim,S.O.,
Pathak,S.,
Grant,G.,
Siciliano,M.J.,
Giovanella,B.C., Strong,L.C. and Tainsky,M.A. (1990) Spontaneous
abnormalities in normal fibroblasts from patients with Li-Fraumeni cancer
syndrome: aneuploidy and immortalization. Cancer Res., 50, 7979±7984.
20. Gollahon,L.S., Kraus,E., Wu,T.A., Yim,S.O., Strong,L.C., Shay,J.W.
and Tainsky,M.A. (1998) Telomerase activity during spontaneous
immortalization of Li-Fraumeni syndrome skin fibroblasts. Oncogene,
17, 709±717.
21. Jeffreys,A.J., Wilson,V. and Thein,S.L. (1985) Hypervariable minisatellite
regions in human DNA. Nature, 314, 67±73.
22. Steinmetz,M., Stephan,D. and Lindahl,K.F. (1986) Gene organization and
recombinational hotspots in the murine major histocompatibility complex.
Cell, 44, 895±904.
23. Wahls,W.P., Wallace,L.J. and Moore,P.D. (1990) Hypervariable minisatellite DNA is a hotspot for homologous recombination in human cells.
Cell, 60, 95±103.
24. Anderson,M.J., Casey,G., Fasching,C.L. and Stanbridge,E.J. (1994)
Evidence that wild-type TP53 and not genes on either chromosome 1 or
11, controls the tumorigenic phenotype of the human fibrosarcoma
HT1080. Gene Chromosome Cancers, 9, 266±281.
25. Kessis,T.D., Slebos,R.J., Nelson,W.G., Kastan,M.B., Plunkett,B.S.,
Han,S.M., Lorincz,A.T., Hedrick,L. and Cho,K.R. (1993) Human
papillomavirus 16 E6 expression disrupts the p53-mediated cellular
response to DNA damage. Proc. Natl Acad. Sci. USA, 90, 3988±3992.
26. Piatyszek,M.A., Kim,N.W., Weinrich,S.L., Hiyama,K., Hiyama,E.,
Wright,W.E. and Shay,J.W. (1995) Detection of telomerase activity in
human cells and tumors by a telomeric repeat amplification protocol
(TRAP). Methods Cell Sci., 17, 1±15.
27. Wright,W.E., Shay,J.W. and Piatyszek,M.A. (1995) Modifications of a
telomeric repeat amplification protocol (TRAP) result in increased
reliability, linearity and sensitivity. Nucleic Acids Res., 23, 3794±3795.
28. Wyllie,A.H., Kerr,J.F.R. and Currie,A.R. (1980) Cell death: the significance of apoptosis. Int. Rev. Cytol., 68, 251±306.
29. Armour,J.A., Wong,Z., Wilson,V., Royle,N.J. and Jeffreys A.J. (1989)
Sequences flanking the repeat arrays of human minisatellites: association with tandem and dispersed repeat elements. Nucleic Acids Res., 17,
4925±4935.
30. Strahl,C. and Blackburn,E. (1996) Effects of reverse transcriptase
inhibitors on telomere length and telomerase activity in two immortalized
human cell lines. Mol. Cell Biol., 16, 53±65.
31. Bryan,T.M., Englezou,A., Dalla-Pozza,L., Dunham,M.A. and Reddel,R.R.
(1997) Evidence for an alternative mechanism for maintaining telomere
length in human tumors and tumor-derived cell lines. Nature Med., 3,
1271±1274.
32. Tahara,H., Maeda,S., Watanabe,T., Tokutake,Y., Furuichi,Y., Tahara,E. Jr,
Tahara,E. and Ide,T. (1997) Regulation of telomere in telomerase negative
immortal cell lines. Proc. Am. Assoc. Cancer Res., 38, 503 (abstract).