Mutagenesis vol.15 no.3 pp.187–193, 2000
Loss of chromosome 14 increases the radiosensitivity of
CGL1 human hybrid cells but lowers their susceptibility to
radiation-induced neoplastic transformation
Marc S.Mendonca*, Lael A.Desmond, Toni M.Temples,
Daphne L.Farrington and Brendan M.Mayhugh
Radiation and Cancer Biology Laboratory, Department of Radiation
Oncology, 975 West Walnut Street, IB-346, Indiana University School of
Medicine, Indianapolis, IN 46202, USA
Loss of active tumor suppressor alleles on fibroblast
chromosomes 11 and 14 are involved in radiation-induced
neoplastic transformation of human hybrid CGL1 cells.
Loss of either chromosome 11 or 14 alone is not sufficient
for neoplastic transformation. To gain insight into the
potential functions of these tumor suppressor loci, we have
investigated the effects of chromosome 11 or 14 loss on
radiation-induced neoplastic transformation. We recently
demonstrated that loss of chromosome 11 increases the
susceptibility to X-ray induced cell killing, neoplastic transformation and the expression of delayed death. The data
suggested that one possible function of the chromosome 11
tumor suppressor gene may be to help maintain genome
stability after radiation damage. We postulated that if the
chromosome 14 allele is functioning in a similar manner,
then the loss of chromosome 14 may also make the hybrid
cells more susceptible to radiation-induced cell killing and
neoplastic transformation. A hybrid cell line which has lost
one copy of chromosome 14 was isolated and designated
CON3(–14). CON3(–14) cells were more sensitive to X-rayinduced cell killing when compared with parental CGL1
cells. However, the susceptibility to radiation-induced neoplastic transformation was significantly reduced (by a
factor of two) compared with the parental CGL1 cells. The
expression of delayed death in the progeny of the irradiated
CON3(–14) cells, growing in transformation flasks, was
similar to CGL1 cells during the 21 day assay period.
Taken together, the data indicate that loss of chromosome
14 alone increased the X-ray sensitivity of the hybrid
cells but reduced their susceptibility to radiation-induced
neoplastic transformation. These data suggest that the
tumor suppressor alleles on chromosomes 11 and 14 may
be functionally distinct in terms of their regulation of
genomic instability and neoplastic transformation after
radiation exposure.
Introduction
The loss of several tumor suppressor genes appears to be
required for neoplastic transformation of human cells (Boyd
and Barrett, 1990; Stanbridge, 1990; Loeb, 1991; Weinberg,
1995). In HeLa⫻human skin fibroblast CGL1 hybrid cells,
chromosomes 11 and 14 of fibroblast origin contain functional
tumor suppressor loci which are responsible for the observed
tumor suppression in this system (Stanbridge et al., 1981;
Mendonca et al., 1995, 1998a). We have shown that loss of
tumor suppressor alleles on chromosomes 11 and 14 appear
to be required for the radiation-induced neoplastic transformation of CGL1 hybrid cells (Mendonca et al., 1995, 1998a).
Loss of either chromosome 11 or 14 alone is not sufficient for
neoplastic transformation (Mendonca et al., 1995, 1998a).
The individual functions of the chromosome 11 or 14 tumor
suppressor genes remain to be elucidated and will require
cloning of both genes. However, we propose that studies of
the effect of chromosome 11 or 14 loss on radiation-induced
neoplastic transformation may give us insight into how these
tumor suppressor loci may be controlling tumorigenicity. We
have recently shown that CON104(–11) hybrid cells which
have lost a complete copy of fibroblast chromosome 11 are
more sensitive to X-ray-induced cell killing and to radiationinduced neoplastic transformation when compared with the
parental cell line CGL1 (Mendonca et al., 1999a). Loss of
chromosome 11 also results in an increase in the expression
of delayed death or lethal mutations post-irradiation during
the 21 day neoplastic transformation assay period. These
observations were interpreted to be the result of increased
genomic instability in CON104(–11) cells after radiation exposure (Mendonca et al., 1999a). One possible function of the
chromosome 11 tumor suppressor gene may be to help maintain
genome stability after radiation damage.
The functional significance of genes expressed on chromosome 14 has remained unexplored. If chromosome 14 is
functioning in a similar manner to chromosome 11, then loss
of chromosome 14 might also make the hybrid cells more
susceptible to radiation-induced cell killing and neoplastic
transformation. A hybrid cell line, CON3(–14), which has lost
one copy of chromosome 14 of fibroblast origin was isolated
and investigated. These data indicate that loss of chromosome
14 increases X-ray sensitivity but reduces their radiationinduced neoplastic transformation potential compared with the
parental CGL1 hybrid cells. Therefore, the tumor suppressor
alleles on chromosomes 11 and 14 appear to be functionally
distinct.
Materials and methods
Parental human hybrid cell line CGL1
The parental cell line, CGL1, is a hybrid cell line originally isolated from a
fusion of the tumorigenic HeLa cell line D98/AH-2 and a non-tumorigenic
normal human skin fibroblast cell (GM00077) (Stanbridge et al., 1981). CGL1
is non-tumorigenic when inoculated s.c. into nude mice, negative for expression
of the HeLa tumor-associated antigen intestinal alkaline phosphatase (IAP)
and is genetically stable (Stanbridge et al., 1981; Mendonca et al., 1995,
1998a). It contains, on average, four copies of each chromosome, two of
fibroblast and two of HeLa origin. The CGL1 cell line has functional p53
even though these cells express the E6 oncoprotein (Mendonca et al., 1999b).
Isolation of IAP-negative irradiated control cells (CON lines) and
identification of the CON3(–14) subclone
The CON cell lines were isolated from irradiated populations (7 Gy γ-rays)
after 10 days growth. They are morphologically very similar to the original
CGL1 cell population and enzymatic assays and flow cytometry studies
confirmed them to be negative for the HeLa tumor-associated antigen IAP
(Mendonca et al., 1991a). CON cell lines were used as irradiated non-
*To whom correspondence should be addressed. Tel: ⫹1 317 278 0404; Fax: ⫹1 317 278 0405; Email: [email protected]
© UK Environmental Mutagen Society/Oxford University Press 2000
187
M.S.Mendonca et al.
tumorigenic controls and designated CON1–CON5 (Mendonca et al., 1991a).
Cytogenetic and molecular analyses indicated that CON3 had undergone loss
of an entire fibroblast chromosome 14, but remained IAP-negative and nontumorigenic when injected s.c. into nude mice (Mendonca et al., 1991a, 1995,
1998a). For the studies reported here CON3 has been designated CON3(–14).
Cell culture, growth curves and plating efficiency determination
Cell lines were grown in 75 cm2 flasks (Corning) containing Eagle’s modified
minimum essential medium (Flow Laboratories) supplemented with 5% calf
serum (JRH), 2 mM glutamine (Sigma), non-essential amino acids (Sigma),
20 mM sodium bicarbonate and 100 IU/ml penicillin (Sigma) as previously
described (Mendonca et al., 1991b). Cells were incubated in a 37°C incubator
containing 5% CO2 in humidified air. Cell numbers for growth curves and
plating efficiencies were performed as follows. On the designated days, flasks
were removed and single cell suspensions prepared by trypsinization followed
by resuspension in complete medium. An aliquot was counted using a Coulter
Zm counter to determine cells/ml. Appropriate cell numbers were plated to
measure clonogenic potential after 0–7 Gy X-irradiation. In vitro plating
efficiencies, population doubling times and surviving fractions were determined
by standard methods (Mendonca et al., 1989b, 1991b, 1993, 1998b).
Survival and neoplastic transformation frequencies
CON3(–14) and the parental cell hybrid CGL1 were irradiated with 0, 2, 3,
5 and 7 Gy of 250 kvP X-rays (Siemens; 1 mm Cu filtration) at a dose rate
of 68.7 cGy/min and plated for cell survival and neoplastic transformation.
The standard doses of 2–7 Gy X-rays were utilized so that the complete dose–
response curves for neoplastic transformation of CON13(–14) and CGL1
could be compared. The cell numbers for each cell line were adjusted so that
at the time of radiation exposure there are 1–1.5⫻106 cells/25 cm2 flasks
(Corning). After irradiation, cells were incubated for 6 h at 37°C to allow for
repair of potentially lethal damage. The cells were then trypsinized, counted
and plated in six 25 cm2 flasks for survival determination. After 7–10 days
growth, colonies were fixed and stained with a solution containing 0.35%
crystal violet in 35% ethanol. Only colonies with greater than 50 cells were
scored. The survival curves for each cell line were the combined results of
three independent experiments. Data are expressed as means ⫾ SD.
To measure the neoplastic transformation frequency for each cell line after
a particular X-ray dose, 30–60 75 cm2 flasks were plated with 5000–50 000
cells/flask depending on the expected cell survival at 0, 2, 3, 5 or 7 Gy. The
cell numbers plated were adjusted to achieve a standard viable cell density in
the transformation flasks of 50 cells/cm2 (Redpath et al., 1987; Sun et al.,
1988; M.S.Mendonca, unpublished results). After 21 days, flasks were fixed
and stained for neoplastically transformed foci using the Western Blue method
(Mendonca et al., 1992). Briefly, on day 21 the medium was removed and
cultures were rinsed twice in phosphate-buffered saline (PBS), fixed with 2%
formaldehyde in PBS for 20 min and rinsed with PBS four times. Western
Blue (alkaline phosphatase detection reagent; Promega) was added to the
flasks and incubated for 7 min to visualize neoplastically transformed foci of
cells expressing the tumor-associated antigen IAP p75/150 (Mendonca et al.,
1992, 1993). The blue IAP-positive neoplastically transformed foci were
visually scored using a stereomicroscope. Neoplastic transformation frequencies were expressed as the total number of foci per surviving cells (Mendonca
et al., 1993, 1998b). The results were combined from three independent
experiments and are presented as accumulated data.
Kinetics of appearance and number of cells per neoplastically transformed
foci
To compare the kinetics of the appearance of IAP-expressing, neoplastically
transformed foci after 7 Gy X-rays in CGL1 and CON3(–14), a standard
transformation assay was performed. One hundred and fifty 75 cm2 flasks of
each cell line were plated so that 15 flasks of each could be fixed and stained
for the presence of foci every other day from day 4 to 21. Transformation
flasks of unirradiated cells for each cell line were also plated, fixed and
stained to determine the spontaneous background on designated days. Transformation flasks were screened for foci with a stereomicroscope so that small
foci containing fewer than 10 cells were not missed (Mendonca et al., 1993,
1998b, 1999a). Results of two independent experiments were accumulated
and plotted as neoplastic transformation frequency over time.
Plating efficiencies of irradiated and non-irradiated CON3(–14) and
parental CGL1 cells during the 21 day neoplastic transformation assay
(delayed death assays)
The plating efficiencies (PEs) of control and irradiated CGL1 and CON3
(–14) cells in the transformation flasks on days 4, 6, 8, 11, 13, 15,18, 20 and
21 were measured (Mendonca et al., 1989a, 1993, 1998b). Cells in 75 cm2
transformation flasks were trypsinized, counted and 100–500 irradiated and
non-irradiated cells were plated per 25 cm2 tissue culture flask, which
contained pre-equilibrated medium (pH 7.2), six flasks per point. The PE
188
Fig. 1. X-ray survival curves of CON3(–14) and the parental CGL1 hybrid
cell. The survival levels after 2, 3, 5 and 7 Gy are shown as means ⫾ SE.
CON3(–14) is more radiosensitive than the parental CGL1 at doses ⬎3 Gy
due to a decrease in the shoulder of the survival curve of CON3(–14).
Graphical analysis confirmed that the survival curve parameter Dq, a
measure of the survival curve shoulder, was significantly different at
2.0 ⫾ 0.2 and 3.0 ⫾ 0.2 Gy for CON3(–14) and CGL1, respectively
(P ⬍ 0.05). The data for both cell lines were average values from three
independent experiments.
flasks were incubated for 7–10 days to assess colony-forming ability. Colonies
were stained and scored as described above. The experiments were repeated
three times, using six replicates per condition. Plating efficiencies for irradiated
and control samples were calculated by dividing the average number of
colonies in the six flasks by the number of cells initially plated (Mendonca
et al., 1989a, 1993, 1998b).
Data presentation and statistical analyses
Radiation-induced neoplastic transformation data for CGL1 and CON3(–14)
were scored by counting the total number of foci within the 75 cm2 flasks, as
well as the total number of cells surviving each treatment protocol. Transformation frequency was expressed as the total number of foci/total number of
surviving cells. When necessary, a χ2 analysis was performed to compare the
accumulated data for CGL1 with CON3(–14) to test for statistical significance.
For transformation frequency, this was done using the total number of foci
and the total number of cells surviving treatment (Mendonca et al., 1990b,
1998b, 1999a). Plating efficiencies or survival fractions at various dose points
for CON104(–11) versus parental CGL1 cells were considered significantly
different if the 95% confidence intervals (2 SD) did not overlap (Mendonca
et al., 1991b, 1999a).
Results
X-ray-induced cell killing of CON3(–14) versus parental
CGL1 hybrid
Cytogenetic and molecular analyses of the non-tumorigenic
control hybrid cell lines (CON1–CON5) indicated that CON3
was missing an entire copy of human fibroblast chromosome
14, which contains a tumor suppressor locus for this system
but remained non-tumorigenic (Mendonca et al., 1998a). In
Figure 1, the X-irradiation responses of CON3(–14) versus
CGL1 are shown. CON3(–14) was more radiosensitive than
the parental CGL1 cells. The dose of X-rays which reduced
survival to 50% (D50) was 4.0 Gy for CON3(–14) versus 5.0
Gy for CGL1. The difference in radiation response was due
to a difference in the shoulder size of the survival curves,
which analyses confirmed; Dq values for CON3(–14) and
CGL1 were significantly different at 2.0 ⫾ 0.2 and 3.0 ⫾ 0.2
Gy, respectively (P ⬍ 0.05). The D0 values of the curves were
found to be equivalent at 2.8 ⫾ 0.2 Gy.
Chromosome 14 loss increases radiosensitivity
Radiation-induced neoplastic transformation of CON3(–14)
and parental CGL1 hybrid cells versus X-ray dose
Standard neoplastic transformation assays were performed with
CON3(–14) and parental CGL1 cells to investigate whether
loss of fibroblast chromosome 14 influenced the number of
neoplastically transformed foci produced 21 days after radiation
exposure. Accumulated data for neoplastic transformation
frequency versus X-ray dose for CON3(–14) and CGL1 are
shown in Figure 2. The transformation frequency of CON3
(–14) was consistently lower than the parental cell line on day
21 post-irradiation. The data from three independent neoplastic
transformation experiments for CGL1 versus CON3(–14) after
7 Gy are shown in Table I. χ2 analyses on the accumulated
data at 7 Gy confirmed that the neoplastic transformation
frequency was significantly lower for CON3(–14) versus CGL1
(2.2⫻10–4 versus 4.4⫻10–4, respectively; P ⬍ 0.05). The
radiation-induced neoplastic transformation frequency of
CON3(–14) was also significantly lower than CGL1 at the 3
and 5 Gy dose points (P ⬍ 0.05). The data indicate that
CON3(–14) is less susceptible to X-ray-induced neoplastic
Fig. 2. Accumulated neoplastic transformation frequency (foci per surviving
cell) versus X-ray dose for CON3(–14) and the parental CGL1 hybrid cell.
CON3(–14) appears to be less sensitive to X-ray-induced neoplastic
transformation than the parental CGL1 hybrid cell line when the
neoplastically transformed foci are scored 21 days post-irradiation. χ2
analysis on the accumulated data for the end-point transformation frequency
(foci per surviving cell) confirmed that the decrease in radiation-induced
neoplastic transformation frequency of CON3(–14) is statistically significant
at 3, 5 and 7 Gy (P ⬍ 0.05) (see Table I for 7 Gy accumulated data).
transformation than the parental cell line CGL1 from which it
was derived.
Radiation-induced neoplastic transformation of CON3(–14)
and the parental CGL1 cells versus time after 7 Gy of
X-rays
To determine if loss of chromosome 14 influenced the kinectics
of focus formation after radiation exposure, a comparison of
neoplastic transformation frequency versus time postirradiation for CGL1 and CON3(–14) after 7 Gy was performed
(Figure 3). In CGL1 cells, the appearance of neoplastically
transformed foci was delayed. The majority of foci did not
appear until 12 or 13 days post-irradiation and all were
fully expressed by day 21. These data agree with previously
published γ- and X-ray-induced foci kinetic data for CGL1
(Sun et al., 1988; Mendonca et al., 1993, 1998b). In CON3
(–14), an increase in neoplastic transformation to 2⫻10–4 was
evident by day 6 post-irradiation and remained at about that
level until day 21. Differences in neoplastic transformation
Fig. 3. Accumulated neoplastic transformation frequency (⫻10–4) versus
time after 7 Gy of X-rays for CON3(–14) versus CGL1. Fifteen T-75 flasks
from each cell line were removed and scored on the indicated days for the
presence of neoplastically transformed foci. In the parental CGL1 cells the
majority of the foci appear after day 12 post-irradiation. In CON3(–14), an
increase in neoplastic transformation to 2⫻10–4 was evident by day 6 postirradiation and remained at that level until day 21. Therefore, in CON3(–14)
the radiation-induced foci arise early and no additional foci appear at later
times. χ2 analysis verified that the differences in neoplastic transformation
frequency versus time for CON3(–14) versus CGL1 were statistically
significant on days 15, 18 and 21. The data were accumulated from two
independent experiments.
Table I. Neoplastic transformation data for parental CGL1 and CON3(–14) cells
Dose
CGL 1
Foci
CON3(–14)
Cells at risk
Transformation
frequency
Foci
0
0
0
0
Gy
Gy
Gy
Gy total
1
0
2
3
63 000
110 000
106 000
279 000
1.6⫻10–5
1.9⫻105
1.1⫻10–5
1
1
1
3
7
7
7
7
Gy
Gy
Gy
Gy total
37
63
39
139
63 000
151 000
104 000
318 000
5.9⫻10–4
4.2⫻10–4
3.8⫻10–4
4.4⫻10–4
40
14
14
68
Cells at risk
Transformation
frequencya
147 000
72 300
63 000
282 300
6.8⫻10–6
1.4⫻10–5
1.6⫻10–5
1.06⫻10–5
188 000
66 600
51 400
306 000
2.1⫻10–4
2.1⫻10–4
2.7⫻10–4
2.2⫻10–4
frequency calculated per surviving cell. A χ2 analysis was performed on the accumulated 7 Gy data for CGL1 versus CON3(–14). Statistical
analysis on the end-point foci per surviving cell (transformation frequency) indicates a significant difference (P ⬍ 0.05) between CGL1 and CON3(–14).
aTransformation
189
M.S.Mendonca et al.
Fig. 4. Growth curves for 7 Gy X-irradiated CGL1 and CON3(–14) cells in
the 75 cm2 transformation flasks during the 21 day expression period. CGL1
and CON3(–14) cells have very similar growth characteristics during
exponential growth until day 10 and have equivalent population doubling
times (20 ⫾ 1 h). The similarity persists even after the irradiated cells reach
a steady-state plateau on days 11–20 with both cell lines averaging 13 ⫾
1.2⫻106 per flask. These data were accumulated from three independent
experiments.
Fig. 5. Plating efficiency as a function of time for CON3(–14) and CGL1
cells during a standard neoplastic transformation assay after 0 or 7 Gy. On
the designated days the cells from the 0 and 7 Gy transformation flasks of
both cell lines were trypsinized and replated to assess clonogenic potential.
The PE of progeny of the 7 Gy irradiated CGL1 and CON3(–14) cells both
initially recovered by day 8, plateaued and then declined during the next 10
days post-irradiation. The expression of delayed death is very similar for
these two cell lines. The 0 Gy control plating efficiencies for both cell lines
remained at 0.75 ⫾ 0.10 during the 21 day assay period. These data were
averaged from three independent experiments.
frequency for CON3(–14) versus CGL1 were statistically
significant on days 15, 18 and 21. The data indicate that in
CON3(–14) the radiation-induced foci arise early and no
additional foci appear at later times (Figure 3).
Growth curves of CON3(–14) and CGL1 cells after 7 Gy of
X-rays
To evaluate whether a difference in cell population doubling
times could explain the reduction in numbers of foci seen in
CON3(–14) after radiation exposure or their early appearance,
growth curves of irradiated (7 Gy) CON3(–14) and CGL1
cells in the transformation flasks were determined (Figure 4).
The growth curves overlap during the exponential growth
period until day 10, indicating similar population doubling
times for CON3(–14) and CGL1 (20 ⫾ 1 h). Even on days
10–20, when the irradiated cells reach a steady-state plateau,
the average densities of CON3(–14) and CGL1 cells were
similar with both cell lines averaging 13.0 ⫾ 1.2⫻106 per
flask. The data for CGL1 are in agreement with previous
observations (Mendonca et al., 1993, 1999a).
Plating efficiency of CON3(–14) and CGL1 cells after 7 Gy
of X-rays
The equivalent steady-state plateau observed in the growth
curves in Figure 4 for the two irradiated hybrid cell lines
suggest that expression of delayed death due to genomic
instability in the cells would be similar (Mendonca et al.,
1993, 1998b). In Figure 5, the PE of irradiated CON3(–14)
and CGL1 cells from transformation flasks are shown during
the 21 day transformation assay period. The data indicate that
the PEs of the irradiated CON3(–14) and parental CGL1 cells
were not significantly different during the 21 day assay period.
The PE of progeny of irradiated CGL1 and CON3(–14) cells
both initially recover by day 8, plateau and then decline during
the next 10 days post-irradiation. This characteristic pattern
has been attributed to the expression of delayed death in
irradiated CGL1 cells (Mendonca et al., 1989a, 1993, 1998b,
1999a). Therefore, the expression of delayed death in irradiated
CON3(–14) cells appeared to be similar to CGL1 cells. The
replated PEs of the unirradiated controls for both cell lines
were in the range 0.75 ⫾ 0.10 during the 21 day assay period.
Discussion
Radiation-induced neoplastic transformation of the human
hybrid cells requires loss of fibroblast chromosomes 11 and
14 containing functional tumor suppressor loci (Mendonca
et al., 1995, 1998a). Since loss of either chromosome 11 or
14 alone was insufficient for neoplastic transformation of the
hybrid cells (Mendonca et al., 1998a), we began experiments
to investigate whether chromosome 11 or 14 loss altered
radiation-induced cell killing and neoplastic transformation.
We recently demonstrated that CON104(–11) hybrid cells
which have lost a complete copy of fibroblast chromosome 11
are more sensitive to X-ray-induced cell killing and to radiationinduced neoplastic transformation when compared with the
parental cell line CGL1 (Mendonca et al., 1999a). To determine
whether the chromosome 14 allele is functioning in a similar
manner, CON3(–14) hybrid cells missing a copy of chromosome 14 were investigated.
Loss of chromosome 14 increased X-ray-induced cell killing
of the hybrid cells
CON3(–14) hybrid cells were found to have increased sensitivity to X-ray-induced cell killing when compared with the
parental CGL1 cells (Figure 1). The difference is due to a
decrease in the shoulder of the CON3(–14) survival curve.
This has been interpreted to be a reflection of decreased repair
capacity in many mammalian cells in vitro and in vivo (Elkind,
1984; Iliakis, 1988; Mendonca et al., 1989b, 1990a; Hall,
1994). A number of the genes involved in repair of X-ray
induced DNA damage, which may be important in controlling
X-ray sensitivity of mammalian cells, have been identified and
a partial list includes p53, ATM, Ku70/80, hMRE-11, DNAPK, c-ABL, HRAD50, HRAD51, APE (apurinic endonuclease)
and polymerase ε (reviewed in Taylor and Lehmann, 1998;
Rosen et al., 1999; Yu et al., 1999). Some of these human
DNA repair genes are located on chromosome 14 (Yu et al.,
1999). For example, HRAD51 is located on chromosome
14q23–q24 and appears to be involved in recombinational
repair and homologous recombination. APE/HAP1/REF1 is
located on 14q11.2–q12 and is an apurinic endonuclease
involved in base excision repair. Polymerase ε, a DNA
190
Chromosome 14 loss increases radiosensitivity
polymerase involved in nucleotide excision repair, is located
on 14q13–q21 (for an overview of these genes see Taylor and
Lehmann, 1998; Yu et al., 1999). It is tempting to speculate
that loss of one of these genes on chromosome 14 in CON3
(–14) cells could be responsible for their increased radiosensitivity.
The HeLa/cervical cancer tumor suppressor locus on chromosome 14 has not yet been clearly defined. Whether loss of
one copy of the tumor suppressor gene on chromosome 14 or
changes in other genes involved in DNA repair on this
chromosome listed above can explain the increased radiosensitivity of CON3(–14) will require additional studies. However,
our data do suggest that loss of chromosome 14 alleles
increases the radiation sensitivity of the hybrid cells. Since
loss of a complete copy of fibroblast chromosome 11 containing
the other tumor suppressor locus for these hybrid cells also
increases radiosensitivity, it suggests that genes on both chromosomes may be involved in controlling X-ray sensitivity in
these cells (Mendonca et al., 1999a).
Loss of chromosome 14 decreases the radiation-induced
neoplastic transformation potential of the hybrid cells
Standard neoplastic transformation assays with CON3(–14)
and CGL1 demonstrated that loss of chromosome 14 decreased
the sensitivity of the hybrid cells to X-ray-induced neoplastic
transformation. Significantly less foci were produced at 3, 5
and 7 Gy X-rays (Figure 2 and Table I; P ⬍ 0.05). In addition,
a study of neoplastic transformation frequency versus time
post-irradiation indicates that all of the foci in CON3(–14)
appear 6 days post-irradiation, which is significantly earlier
than in CGL1 cells (Figure 3). Growth curves indicate that
large differences in cell population doubling times for the two
cell lines are not evident and therefore cannot explain the
difference in the reduced number or timing of the appearance
of foci post-irradiation in CON3(–14) versus the parental
CGL1 cells (Figure 4). This is in contrast to our recent data
which indicate that loss of chromosome 11 increases the
X-ray induced neoplastic transformation frequency of these
CGL1 hybrid cells (Mendonca et al., 1999a).
The above data suggest that the tumor suppressor alleles on
chromosomes 11 and 14 appear to be functionally distinct.
However, it is difficult to understand why loss of one copy of
chromosome 14 containing a tumor suppressor locus resulted
in a reduction in radiation-induced neoplastic transformation
while loss of chromosome 11 containing the other tumor
supressor locus led to an increase. Logically you might expect
that loss of either tumor suppressor locus should lead to an
increase in radiation-induced neoplastic transformation because
one of the required events has already occurred. It is possible
that loss of other linked genes on chromosome 14 are the
reason for this observation. Alternatively, loss of one copy of
the chromosome 14 tumor suppressor allele may result in a
reduction in the amount of suppressor protein produced. A
reduction in the amount of this particular protein may alter its
function so significantly that, as a protective mechanism, it
leads to the death of some of the cells undergoing neoplastic
transformation. Further analysis and the eventual cloning and
characterization of the chromosome 11 and 14 tumor suppressor
genes will allow us to better address this paradox.
We have previously shown that the delayed appearance of
radiation-induced neoplastically transformed foci in CGL1
correlates with a lower steady-state plateau and a persistently
reduced PE which we and others attribute to the expression
of delayed death or lethal mutations (Seymour et al., 1986;
Gorgojo and Little, 1989; Mendonca et al., 1989a, 1993,
1998b; Chang and Little, 1991). Loss of chromosome 11
increased both the radiation-induced neoplastic transformation
frequency and the amount of delayed death post-irradiation,
suggesting that these processes are linked (Mendonca et al.,
1999a). However, we have shown here that loss of a copy of
fibroblast chromosome 14 decreased the probability of radiation-induced neoplastic transformation. We therefore investigated whether chromosome 14 loss altered the expression of
delayed death in CON3(–14) versus CGL1.
Loss of chromosome 14 does not increase the expression of
delayed death post-irradiation in CON3(–14) cells
The nearly identical steady-state plateau seen in the growth
curves for irradiated CON3(–14) and CGL1 cells (Figure 4)
indicated that the expression of delayed death in CON3(–14)
should be similar to that observed in CGL1 (Mendonca et al.,
1998b, 1999a). In Figure 5, a PE versus time post-irradiation
study with CON3(–14) and CGL1 indicated that after 7 Gy
X-rays the PE of CON3(–14) was quite similar to that observed
for CGL1. This is in contrast to what we observed with
CON104(–11) cells in which the PE was significantly lower
during the 21 day assay period (Mendonca et al., 1999a). This
would suggest that loss of chromosome 14 does not increase
the expression of delayed death in CGL1 cells, as did loss of
chromosome 11. It is possible that the mode of delayed
cell death in CON3(–14) versus CON104(–11) cells may be
different (necrosis versus apoptosis for example) or different
apoptotic pathways may be being activated in these cell lines.
Future experiments will investigate these possibilities.
Are the loci on chromosomes 11 and 14 functionally distinct
in terms of their regulation of radiation-induced genomic
instability and neoplastic transformation?
The observation that neoplastic transformation of human cells
requires a number of genetic changes in a wide variety of
genes has led to the concept of a mutator phenotype and
associated genomic instability being involved in neoplastic
progression (Loeb, 1991; Cheng and Loeb, 1993; Lengauer
et al., 1997). We and others have proposed that genomic
instability may also lead to the expression of delayed death
by induction of apoptosis (Mothersill et al., 1996; Limoli
et al., 1998; Mendonca et al., 1999b).
We have recently shown a correlation between the expression
of delayed death and the onset of delayed apoptosis in CGL1
cells during the 21 day neoplastic transformation assay period
(Mendonca et al., 1999b). We proposed that radiation-induced
instability in CGL1 cells has two relevant outcomes: (i) delayed
cell death by p53-dependent post-mitotic apoptosis in cells
which have developed large-scale genomic damage or loss
through misrepair or translesion DNA synthesis; (ii) neoplastic
transformation of a subset of survivors which have lost
fibroblast chromosomes 11 and 14 (tumor suppressor loci),
but have not acquired enough genetic damage to induce
apoptosis or have found pathways for avoidance of apoptosis,
perhaps by mutation of genes involved in its regulation
(Mendonca et al., 1999b).
In CGL1 cells the mechanisms for either escaping death or
controlling genomic instability may lie on chromosomes 11
and/or 14, since maintaining these chromosomes allows suppression of neoplastic transformation (Mendonca et al., 1995,
1998a,b, 1999b). Previous loss of chromosome 11 increases
the sensitivity of CGL1 hybrid cells to radiation-induced
191
M.S.Mendonca et al.
neoplastic transformation and increases the expression of
delayed death, both due to an increase in genomic instability
(Mendonca et al., 1999a). We have shown here, however, that
loss of chromosome 14 decreases the sensitivity of CGL1 cells
to radiation-induced neoplastic transformation and does not
alter the expression of delayed death post-irradiation.
The above observations suggest that the alleles on chromosomes 11 and 14 may be functionally different in terms of their
regulation of genomic instability and neoplastic transformation.
The chromosome 11 allele may help control genomic instability
post-irradiation and its loss is necessary for the onset of
genomic instability and neoplastic transformation. Therefore,
previous loss of chromosome 11 should increase the amount
of radiation-induced instability post-irradiation and increase
the number of neoplastically transformed foci produced, which
is what we have previously reported (Mendonca et al., 1999a).
The loss of one copy of chromosome 14, however, decreases
radiation-induced neoplastic transformation. As discussed
above, the loss of one copy of the chromosome 14 allele may
lead to a reduction in protein level that alters its suppressor
function and results in the death of cells which could potentially
neoplastically transform. Therefore, the loss of the second
chromosome 14 allele may be necessary for neoplastic transformants to survive. Analysis of radiation-induced neoplastically transformed cell lines isolated from irradiated
CON3(–14) cells will allow us to investigate this further.
Studies of other tumor suppressor genes such as p53 indicate
that they are involved in the regulation of DNA damage
checkpoints and in the control of genomic instability (Kastan
et al., 1995; Smith and Fornace, 1995; Weinberg, 1995; Tlsty,
1996; Baylin, 1997; O’Connor, 1997; Cahill et al., 1998). Loss
of function of these genes may allow genomically unstable
cells to acquire additional mutations required for neoplastic
transformation by not signaling removal of these cells by
apoptosis (Lane, 1992; Boukamp et al., 1995; Kastan et al.,
1995; Smith and Fornace, 1995; Anthoney et al., 1996; Wahl
et al., 1997). Our studies indicate that the HeLa/cervical cancer
tumor suppressor loci on chromosomes 11 and 14 may both be
involved in the control of genomic instability post-irradiation.
However, the data suggest that the tumor suppressor alleles
on chromosomes 11 and 14 may be functionally distinct in
terms of their regulation of genomic instability and neoplastic
transformation after radiation exposure.
Acknowledgements
This work was supported by start up funds awarded to M.S.M. from the
Department of Radiation Oncology, Indiana University School of Medicine
and in part by grant IRG-84-002-14 from the American Cancer Society
awarded to M.S.M.
References
Anthoney,D.A., Mcilwrath,A.J., Gallagher,W.M., Edlin,A.R.M. and Brown,R.
(1996) Microsatellite instability, apoptosis and loss of p53 function in drugresistant tumor cells. Cancer Res., 56, 1374–1381.
Baylin,S.B. (1997) Tying it all together: epigenetics, genetics, cell cycle and
cancer. Science, 277, 1948–1949.
Boukamp,P., Peter,W., Pascheberg,U., Altmeier,S., Stanbridge,E.J. and
Fusenig,N.E. (1995) Step-wise progression in human skin carcinogenesis
in vitro involves mutational inactivation of p53, ras H oncogene activation
and additional chromosomes loss. Oncogene, 11, 961–969.
Boyd,J.A. and Barrett,J.C. (1990) Genetic and cellular basis of multistep
carcinogenesis. Pharmacol. Ther., 46, 469–486.
Cahill,D.P., Lengauer,C., Yu,J., Riggins,G.J., Willson,J.K.V., Markowitz,S.D.,
Kinzler,K.W. and Vogelstein,B. (1998) Mutations of mitotic checkpoint
genes in human cancers. Nature, 392, 211–314.
192
Chang,W.P. and Little,J.B. (1991) Delayed reproductive death in X-irradiated
Chinese hamster ovary cells. Int. J. Radiat. Biol., 60, 483–496.
Cheng,K.C. and Loeb,L.A. (1993) Genomic instability and tumor progression:
mechanistic considerations. Adv. Cancer Res., 60, 121–156.
Elkind,M.M. (1984) Failla memorial lecture: repair processes in radiation
biology. Radiat. Res., 100, 425–499.
Gorgojo,L. and Little,J.B. (1989) Expression of lethal mutations in progeny
of irradiated mammalian cells. Int. J. Radiat. Biol., 55, 619–630.
Hall,E.J. (1994) In Ryan,J.D. (ed.) Radiobiology for the Radiologist, 4th Edn.
J.B. Lippincott, Philadelphia, PA.
Iliakis,G. (1988) Radiation-induced potentially lethal damage: DNA lesions
susceptible to fixation. Int. J. Radiat. Biol., 53, 541–584.
Kastan,M.B., Canman,C.E. and Leonard,C.J. (1995) p53, cell cycle control
and apoptosis: implications for cancer. Cancer Metastasis Rev., 14, 3–15.
Lane,D.P. (1992) Cancer, p53, guardian of the genome. Nature, 358, 15–16.
Lengauer,C., Kinzler,K.W. and Vogelstein,B. (1997) Genetic instability in
colorectal cancers. Nature, 386, 623–627.
Limoli,C.L., Hartmann,A., Shephard,L., Yang,C.-R., Boothman,D.A.,
Bartholomew,J. and Morgan,W.F. (1998) Apoptosis, reproductive failure
and oxidative stress in chinese hamster ovary cells with compromised
genomic integrity. Cancer Res., 58, 3712–3718.
Loeb,L.A. (1991) Mutator phenotype may be required for multistage
carcinogenesis. Cancer Res., 51, 3075–3079.
Mendonca,M.S., Kurohara,W., Antoniono,R. and Redpath,J.L. (1989a) Plating
efficiency as a function of time post-irradiation: evidence for the delayed
expression of lethal mutations. Radiat. Res., 119, 389–393.
Mendonca,M.S., Rodriguez,A. and Alpen,E.L. (1989b) Quiescence in 9L cells
and correlation with radiosensitivity and PLD repair. Radiat. Res., 117,
433–447.
Mendonca,M.S., Rodriguez,A. and Alpen,E.L. (1990a) Differential repair of
potentially lethal damage in exponentially growing and quiescent 9L cells.
Radiat. Res., 122, 38–43.
Mendonca,M.S., Sun,C. and Redpath,J.L. (1990b) Suppression of radiationinduced neoplastic transformation of human cell hybrids by long term
incubation at low extracellular pH. Cancer Res., 50, 2123–2127.
Mendonca,M.S., Antoniono,R.J., Latham,K.M., Stanbridge,E.J. and
Redpath,J.L. (1991a) Characterization of intestinal alkaline phosphatase
expression and the tumorigenic potential of γ-irradiated HeLa⫻fibroblast
cell hybrids. Cancer Res., 51, 4455–4462.
Mendonca,M.S., Boukamp,P., Stanbridge,E.J. and Redpath,J.L. (1991b) The
radiosensitivity of human keratinocytes: influence of activated c-H-ras
oncogene expression and tumorigenicity. Int. J. Radiat. Biol., 59, 1195–1206.
Mendonca,M.S., Antoniono,R.J., Sun,C. and Redpath,J.L. (1992) A simplified
and rapid staining method for the HeLa⫻skin fibroblast human hybrid cell
neoplastic transformation assay. Radiat. Res., 131, 345–350.
Mendonca,M.S., Antoniono,R.J. and Redpath,J.L. (1993) Delayed heritable
damage and epigenetics in radiation-induced neoplastic transformation of
human hybrid cells. Radiat. Res., 134, 209–219.
Mendonca,M.S., Fasching,C.L., Srivatsan,E.S., Stanbridge,E.J. and
Redpath,J.L. (1995) Loss of a putative chromosome 11 tumor suppressor
locus after gamma ray-induced neoplastic transformation of
HeLa⫻fibroblast human cell hybrids. Radiat. Res., 143, 34–44.
Mendonca,M.S., Howard,K., Fasching,C.L., Farrington,D.L., Desmond,L.A.,
Stanbridge,E.J. and Redpath,J.L. (1998a) Loss of suppressor loci on
chromosomes 11 and 14 may be required for radiation-induced neoplastic
transformation of HeLa⫻fibroblast human cell hybrids. Radiat. Res., 149,
246–255.
Mendonca,M.S., Temples,T.M., Farrington,D.L. and Bloch,C. (1998b)
Evidence for a role of delayed death and genomic instability in radiationinduced neoplastic transformation of human hybrid cells. Int. J. Radiat.
Biol., 74, 755–764.
Mendonca,M.S., Howard,K., Desmond,L.A. and Weissman Derrow,C. (1999a)
Previous loss of chromosome 11 containing a suppressor locus increases
radiosensitivity, neoplastic transformation frequency and delayed death in
HeLa⫻fibroblast human hybrid cells. Mutagenesis, 14, 483–489.
Mendonca,M.S., Howard,K., Farrington,D.L., Desmond,L.A., Temples,T.M.,
Mayhugh,B.M., Pink,J.J. and Boothman,D.A. (1999b) Delayed apoptotic
responses associated with radiation-induced neoplastic transformation of
human hybrid cells. Cancer Res., 59, 3972–3979.
Mothersill,C., Lyng,F., O’Reilly,S., Harney,J. and Seymour,C.B. (1996)
Expression of lethal mutations is suppressed in neoplastically transformed
cells and after treatment of normal cells with carcinogens. Radiat. Res.,
145, 714–721.
O’Connor,P.M. (1997) Mammalian G1 and G2 phase checkpoints. Cancer
Surv., 29, 151–182.
Redpath,J.L., Sun,C., Colman,M. and Stanbridge,E.J. (1987) Neoplastic
Chromosome 14 loss increases radiosensitivity
transformation of human hybrid cells by irradiation: a quantitative assay.
Radiat. Res., 110, 468–472.
Rosen,E.M., Fan,S., Rockwell,S. and Goldberg,I.D. (1999) The molecular and
cellular basis of radiosensitivity: implications for understanding how normal
tissues and tumors respond to therapeutic radiation. Cancer Invest., 17,
56–72.
Seymour,C.B., Mothersill,C. and Alper,T. (1986) High yields of lethal
mutations in somatic mammalian cells that survive ionizing radiation. Int.
J. Radiat. Biol. Relat. Stud. Phys. Chem. Med., 50, 167–179.
Smith,M.L. and Fornace,A.J.Jr (1995) Genomic instability and the role of p53
mutations in cancer cells. Curr. Opin. Oncol., 7, 69–75.
Stanbridge,E.J. (1990) Human tumor suppressor genes. Annu. Rev. Genet.,
24, 615–657.
Stanbridge,E.J., Flandemeyer,R.R., Daniels,D.W. and Nelson-Rees,W.A.
(1981) Specific chromosome loss associated with expression of
tumorigenicity in human cell hybrids. Somat. Cell Genet., 7, 699–712.
Sun,C., Redpath,J.L., Colman,M. and Stanbridge,E.J. (1988) Further studies
on the radiation-induced expression of a tumor-specific antigen in human
cell hybrids. Radiat. Res., 114, 84–93.
Taylor,E.M. and Lehmann,A.R. (1998) Conservation of eukaryotic DNA repair
mechanisms [review]. Int. J. Radiat. Biol., 74, 277–286.
Tlsty,T.D. (1996) Regulation of genomic instability in preneoplastic cells.
Cancer Surv., 28, 217–224.
Wahl,G.M., Linke,S.P., Paulson,T.G. and Huang,L.-C. (1997) Maintaining
genetic stability through TP53 mediated checkpoint control. Cancer Surv.,
29, 183–219.
Weinberg,R.A. (1995) The molecular basis of oncogenes and tumor suppressor
genes. Ann. N. Y. Acad. Sci., 758, 331.
Yu,Z., Chen,J., Ford,B.N., Brackley,M.E. and Glickman,B.W. (1999) Human
DNA repair systems: an overview. Environ. Mol. Mutagen., 33, 3–20.
Received on August 1, 1999; accepted on January 4, 2000
193