Download Loss of the Intrinsic Heat Resistance of Human Cells and Changes

(CANCER RESEARCH 51. 26.16-2641. May 15. 1991|
Loss of the Intrinsic Heat Resistance of Human Cells and Changes in Mt 70,000
Heat Shock Protein Expression in Human x Hamster Hybrids1
Robin L. Anderson, Kathy J. Fong, Tim Gabriele, Pablo Lavagnini, George M. Hahn, James W. Evans,
Charles A. Waldren, Thomas D. Starnato, and Amato J. Giaccia2
Department of Radiation Oncology, Stanford I'nirersily, Stanford, California 94305 [K. J. F., T. G„P. L., G. M. H., J. W. E., A. J. G.]; Radiology and Radiation
Biology, Colorado State I'nirersily, Fort Collins, Colorado 80523 ¡C.A. H'.J; Lankenau MédicalResearch Center, Philadelphia, Pennsylvania 19151 [T. D. S.]; Peter
MacCallum Cancer Institute, 4X1 Little Lonsdale St., Melbourne, l'ictoria 3000, Australia ¡R.I.. A.]
ABSTRACT
Since mammalian cells vary widely in their intrinsic thermoresistance,
we have investigated the genetic basis underlying this phenomenon in
human and rodent cell lines. Typically, human cells are considerably
more resistant to killing by heat than rodent cell lines. To determine
whether the heat-resistant phenotype is dominant or recessive and to
locate the chromosome(s) bearing determinants for heat resistance, we
have prepared hybrids of heat-resistant human HT1080 cells and heatsensitive Chinese hamster ovary (CUO) cells to test their response to
heat. For both mass hybrid cultures and individual clones, the heat
response of the hybrids was similar to that of the CHO parent. Analysis
by in situ hybridization revealed the presence of five to 20 human
chromosomes per cell in the mass hybrids and four to eight intact
chromosomes plus some fragments in individual clones isolated from the
hybrid cell population. A similar result was obtained using a different
human cell line, AC 1522. These data suggest that heat resistance is a
recessive trait. Consistent with this conclusion are the results from a
study of a fusion of HT1080 to a CHO mutant, BL-10, which was found
to be hypersensitive to heat-induced killing. These hybrids had a normal
CHO heat response and not the more heat-resistant phenotype of
HT1080 cells. Two hybrid clones, H2 and H4, from the HT1080/BL-10
fusion were studied in more detail. Both clones possess similar amounts
of M, 70,000 heat shock protein (HSP70), despite the fact that H4
contains three human chromosomes (Nos. 6, 14, and 21) which carry
HSP70 genes while 112 contains only one (chromosome 6). Both hybrid
cell lines have the same response to heat. Although we found a wide
range of sensitivities to heat, all cell lines contained a similar amount of
constitutive HSP70, suggesting that HSP70 levels per se are not the
critical determinant of intrinsic heat resistance.
INTRODUCTION
An understanding of the different genetic mechanisms un
derlying intrinsic cellular thermoresistance is essential for the
development of more effective hyperthermic treatment of tu
mors. The rationale for the use of hyperthermia in cancer
therapy has come largely from studies on the effects of heat on
rodent cell lines in tissue culture and on model tumor systems
in rodents. However, recently it has been shown by us and
others that human tumor cells grown in tissue culture tend to
be more resistant than rodent cell lines (1). The increased
resistance is displayed over the whole of the relevant tempera
ture range for clinical hyperthermia (40-46°C) and, typically,
the breakpoint in an Arrhenius curve for hyperthermic
killing is 1-1.5°Chigher in human cells (1).
cell
It is unclear why human cells should be more resistant to
heat than rodent cells. Since it is not known whether the
Received 12/17/90; accepted 3/5/91.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 17.14 solely to indicate this fact.
' Supported in pan by Grant CA 44665 from the National Cancer Institute
(G. M. H.). Grant CA 45277 to T. D. S.. and Gram CA 36447 to C. A. \V.
2 To whom requests for reprints should be addressed, at Dept. of Radiation
Oncology. CA
CBRL.
Room GK1I5. Stanford University. School of Medicine.
Stanford".
94305-5468.
differences in intrinsic thermoresistance result from new HSP3
gene expression or differential regulation of preexisting HSPs,
we decided to take a genetic approach and measure the heat
response of hybrid cells derived from the fusion of rodent and
human cells. From such an analysis, we can determine whether
the heat-resistant phenotype of human cells is dominant or
recessive and, if dominant, determine the human chromosome(s) and subsequently the gene(s) bearing determinants for
heat resistance.
This paper describes the results of studies in which we fused
heat-sensitive CHO cells to two different heat-resistant human
cell lines and compared the heat response of the hybrids with
that of the parent cells. By in situ hybridization with biotinylated human genomic DNA of human x hamster metaphase
spreads, we could determine the number of human chromo
somes in the hybrids. In addition, we have identified a mutant
hamster cell line, BL-10, which is hypersensitive to killing by
heat. Hybrids formed between BL-10 and HT 1080 cells exhib
ited a heat response similar to that of CHO and not HT 1080
cells.
We also present data on the levels of HSP70 in the parent
and hybrid cells. HSP70 measurements were included because
of the widely held belief that the level of HSP70 is an indicator
of the level of heat resistance exhibited by cells (2). Although
two-dimensional gel electrophoresis clearly demonstrates the
induction of the human HSP70 gene, the CHO-human hybrid
H4 still possesses a heat response similar to that of rodent cells
and not human cells. These results will be discussed in relation
to what is already known about heat resistance in mammalian
cells.
MATERIALS
AND METHODS
Cell Culture. The following cell lines were used: CHO cells, subline
HA-1; BL-10/6TGR/OuR, a heat-sensitive mutant of CHO cells, ini
tially isolated for its sensitivity to killing by bleomycin (3), carries
genetic markers for resistance to 6-thioguanine and ouabain (4); CHO/
6TGR/OuR. a derivative from CHO-K1 cells, carries genetic markers
for resistance to 6-thioguanine and ouabain (4); HT 1080, human fibro
sarcoma cells, and AG 1522, normal diploid human fibroblasts. All drug
resistance markers were obtained spontaneously without the use of
mutagens. Cells were maintained in a-MEM containing 10% fetal calf
serum and grown at 37°Cin a humidified incubator in an atmosphere
of5%CO2/95%air.
Cell Fusion. Hybrids were obtained using a modification of the fusion
protocol of Davidson and Gerald (5). Approximately 1 x IO6 CHO/
6TGR/OuR or BL-10/6TGR/OuR cells were plated into a T25 flask
along with a 6-thioguanine- and ouabain-sensitive human or hamster
cell line (HT1080, AG1522, or CHO HA-1) and allowed to attach
overnight. The cells were then washed twice with serum-free a-MEM
3The abbreviations used are: HSP, heat shock protein; HSP70, M, 70,000
heat shock protein; CHO, Chinese hamster ovary; «-MEM,^-minimal essential
medium; HATO, hypoxanthine (1 x 10~4M):aminopterin (4 x 10~7M):thymidme
(1.6 x 10"* M):ouabain (1 x 10~' M); ELISA, enzyme-linked immunosorbent
assay.
2636
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HEAT RESISTANCE LOSS IN HUMAN CELLS
and exposed to fusogen-45% Polyethylene Glycol 1000 and 7.5%
dimethyl sulfoxide for 1.5 min, followed by 3 washes in a-MEM
containing 20% fetal calf serum. After 12 h, the cells were trypsinized
and replated in drug-selective medium («-MEMcontaining 10% fetal
calf serum and HATO). The medium was replenished every 3 days.
These selection conditions only permit hybrids formed between drugsensitive and drug-resistant cells to survive. Hybrid colonies appeared
after 2 to 3 wk of growth in HATO medium. Mass hybrid cultures were
composed of 100 to 1000 individual clones. Individual colonies were
subcloned using glass cloning cylinders and transferred into T25 flasks
for further analysis.
Chromosome Analysis. Cells were prepared for cytogenetic analysis
by incubation with 0.1 ng/ml of Colcemid for 1 h, followed by mitotic
shake-off, swelling in 0.075 M KC1 containing 10% fetal calf serum for
30 min at 37°C,and fixation in methanohglacial acetic acid (3:1).
Fluorescent in situ hybridization was performed by the procedure of
Giaccia et al. (6) using biotinylated human DNA for the identification
of human genetic material present in a rodent background. The hybrids
H2 and H4 were characterized by Giemsa-trypsin banding (7) and
Giemsa-11 staining (8) to identify their human chromosome content.
Measurement of Heat Response. Exponentially growing cells were
heated as monolayers in specially designed incubators that were tem
perature controlled to ±0.1°Cand in an atmosphere of 5% CO2/95%
20
RESULTS
Our initial experiments started with testing the heat response
of hybrids between CHO/6TGR/OuR and HT 1080 cells. The
heat response of the hybrid cells was measured within 2 wk of
fusion to minimize the loss of human chromosomes that occurs
during culturing of hybrid cells (11). These freshly prepared
hybrids had a human chromosomal content which ranged from
5 to 20 chromosomes per cell and displayed a heat response
similar to that of CHO cells (Fig. 1). The survival data shown
are for cells heated at 45°C.However, this result is not specific
to 45°Cheating, as hybrids heated to 43°Calso have the same
response as CHO cells (data not shown).
In an attempt to find a thermorésistant clone, we performed
a heat selection on the hybrids by exposing them to 60 min at
45°Cand allowing regrowth of the survivors. After 4 cycles of
selection, the heat response of the survivors was still the same
as that of CHO cells (data not shown).
60
80
Fig. I. Clonogenic heat sun ¡valof recently fused hybrids (D) of HT 1080 (O)
and CHO/6TG*/OuR (•)cells. The hybrids were tested in mass, without cloning.
For comparison, the response of tetraploid CHO cells (•)is included. In this and
in subsequent figures showing survival data, results are averaged from 2 to 6
separate experiments, and the error bars show the standard deviation of the mean.
air. The pH was maintained between 7.2 and 7.4, and the medium was
changed before heating. To measure survival, the cloning assay of Puck
and Marcus (9) was used. The anticipated number of cells to give
approximately 100 clones was plated into 6-cm Petri dishes and left to
grow for 10 days at 37°C.At this time, the colonies were fixed, stained
and counted.
Measurement of HSP70. An ELISA has been developed to quantitate
the levels of HSP70 in cell homogenates (method to be published
elsewhere). Briefly, the cells were disrupted in phosphate-buffered saline
containing 1 HIMATP and sonicated, and aliquots containing known
amounts of protein were placed in wells of a 96-well plate. A monoclo
nal antibody against HSP70 (N6 from Dr. W. Welch) was added, and
the amount of bound antibody was quantitated by the rate of enzyme
activity of horseradish peroxidase conjugated to goat anti-mouse immunoglobulin G which was subsequently added to the wells.
Gel Electrophoresis. For two-dimensional gel electrophoresis, expo
nentially growing cells were incubated at 37°Cfor 2 days in «-MEM
containing 10% fetal calf serum and 2 ¿iCiof [35S]methionine (Amersham Corporation, Arlington Heights, IL; specific activity, >1000 Ci/
mmol) per ml as previously described by Anderson et al. (10). First
dimension isoelectric focusing was performed in 3-mm-diameter tube
gels in 3.5% acrylamide prior to molecular weight separation on a 10%
sodium dodecyl sulfate-acrylamide slab gel (10). After second-dimen
sion electrophoresis, the gels were stained with Coomassie blue, destained, dried, and exposed to Kodak XAR-2 film for 7 to 12 days.
40
Time at 45 C (min)
I
10
1020
40
80
Time at 45°C(min)
Fig. 2. Clonogenic heat survival of individual clones isolated from the mass
fusion shown in Fig. 2. •clone 3; o, clone 8; A, clone 18; D, clone 25; •,
clone
29; x, clone 30; O. HT 1080; •.CHO.
Interestingly, chromosome ploidy has little effect on resist
ance of mammalian cells to heating. Since the chromosome
content of human x hamster hybrids is hyperdiploid, but their
parent lines are diploid, the response of these interspecies
hybrids was compared with that of near-tetraploid CHO cells
prepared by fusion of CHO/6TGR/OuR to wild-type CHO cells
(modal number of 21 chromosomes). The heat resistance of the
near-tetraploid CHO line (modal number of 41 chromosomes)
is slightly greater than that of the diploid cells, but still much
less than the HT 1080 cells (Fig. 1). This result suggests that
gene dosage itself is not the critical factor that determines heat
resistance.
A series of individual clones was isolated from the CHO/
HT 1080 fusion described above. The heat responses of the
clones varied slightly, but all were similar to that of the hybrid
line from which they were derived (Fig. 2). The human chro
mosome content of the clones was measured by fluorescence in
situ hybridization of biotinylated total human genomic DNA
to metaphase chromosomes (Fig. 3). In Fig. 3, the hamster
chromosomes appear red due to staining with propidium iodide
(gray), and only human chromosomes stain bright yellow white
due to efficient hybridization with biotinylated probe. Fluores
cent in situ hybridization of chromosomes with biotinylated
genomic DNA is specific for human chromosome material and
does not cross-hybridize to hamster chromosomes (6). The
number of human chromosomes varied from 4 to 8 intact
chromosomes, with a variable number of fragments integrated
into the hamster genome (Table 1).
To check whether the failure of the heat-resistant phenotype
to dominate was a general feature of human x hamster hybrids
2637
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HEAT RESISTANCE LOSS IN HUMAN CELLS
10
I
Ì
20
40
60
Time at 45 C (min)
Fig. 4. Clonogenic heat survival of recently fused hybrids (•)of AG 1522 (O)
and CHO/6TGR/Ou" (D) cells. The hybrids were tested in mass, without cloning.
a
"
40
60
80
Time at 45 C (min)
Fig. 5. Clonogenic heat survival after varying times at 45°Cof two characterized
hybrid clones, H2 (Ü)and H4 (x), from a fusion of human HT1080 (O) with BL10/6TG"/Ou" cells (•).For comparison, the response of CHO HA-1 cells (•)is
shown as well.
Fig. 3. Fluorescence in situ hybridization of metaphase spreads from hybrid
clones from a HT 1080 x CHO fusion (clones i.A. and 25.Ä). The human
chromosomes, in white, are tagged with biotinylated total human genomic DNA.
human chromosomes in the hybrids (Fig. 4).
We further studied the restoration of heat resistance in hy
brids of a variant CHO heat-sensitive line, BL-10 with a human
fibrosarcoma cell line, HT 1080. Originally, BL-10 cells were
isolated for their sensitivity to killing by bleomycin, but their
response to other DNA-damaging agents is normal (3). Com
plementation studies between BL-10 and HT 1080 have shown
that the gene that corrects the defect in BL-10 resides in human
chromosome 6.4 The development of two different BL-10 x
HT 1080 hybrids with known human chromosome content gave
us the opportunity to measure their heat response and compare
Table 1 Human and total chromosome content of six hybrid clones derived from
it with the parent lines. Such data are presented in Fig. 5. The
a HTI080 x CHO fusion
results represent the average of 2 to 6 experiments per curve.
There is a marked difference in response to 45 °Chyperthermia
Intact human chromosomes are seen, as well as fragments integrated into
hamster chromosomes. The number was scored by counting the chromosomes in
between BL-10 and HT 1080 cells. A heat treatment of 30 min
the metaphase spreads as shown in Fig. 3.
at 45°C,which results in a survival of only one cell in IO4 in
Chromosome content
BL-10, causes little if any killing in HT1080 cells. The two
hybrid lines, designated H2 and H4, show the same response
no.(range)32-5338-6033-5228-6426-4224-82
Clone3g18252930Modal
no.447845HumanFragments645553Total
to heat as do wild-type CHO cells. It is noteworthy that H2
contains human chromosomes 5, 6, 9, and 11, while H4 con
tains chromosomes 4, 5, 6, 9, 11, 12, 14, 15,19, 21, 22, and X.
Thus, we could complement the specific defect of the BL-10
cells, but only to the level of heat resistance of CHO cells. Since
wild-type CHO cells (HA-1 strain) have the same heat response
as do H2 and H4 (Fig. 5), this further strengthens our conclu
sion that the fusion of human cells to hamster cells, whether
or a specific feature of the hybrids derived from HT1080 cells, CHO, HAI or BL-10, results in a cell possessing the heat
another fusion to CHO cells was performed, this time using a
sensitivity of the hamster cell, rather than the human cell.
normal human fibroblast line, AG 1522. Once again, the result
As some investigators consider the concentration of HSP70
ing hybrids had a heat response similar to that of the CHO
to be important in the intrinsic heat response of cells (2), the
cells. The hybrids were tested within 2 wk of performing the
' A. J. Giaccia et al., unpublished data.
fusion and were not cloned, thus maximizing the number of
2638
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HEAT RESISTANCE LOSS IN HUMAN CELLS
Table 2 Relative amount of HSP70 present in the various cell lines as measured
by an enzyme-linked immunosorbent assay using a monoclonal
antibody against HSP70
Both constitutive levels and the amounts present 17 h after a heat shock of 10
min at 45'C are presented. The data are pooled from 3 separate experiments and
plasmic reticulum and mitochondria, and removing damaged
or denatured proteins from the cell via lysosomal degradation
(18, 19).
After heat shock or other stresses, including exposure to
are standardized against the relative amount of HSP70 found in CHO cells. It is
arsenite, transition metal ions, ethanol, and release from hypreferable to present the data this way as the rate of color development in the
ELISA varied between experiments, but within one experiment, consistent results
poxia and mutagens (20), the levels of these proteins are in
were obtained.
creased. In parallel, and with a good temporal correlation, cells
Relative amount of HSP70
develop a transient resistance to heat, known as thermotolerH/C°ratio
ance (21-23). The reason why proteins with the properties
Cell line
Control
1.2±0.5"
described above should be stress inducible and whether under
HT1080
5.0
H4
0.9 ±0.4
5.5
stress the induced proteins perform the same function is un
H2
1.2±0.4
6.7
clear,
although the increase in denatured proteins caused by the
BL-10
1.0
1.4 ±0.5
CHO-HA1
stress would necessitate increased levels of HSP70 to facilitate
1.0
2.3
" H, amount present 17 h after a heat shock of IO min at 45°C;C, constitutive
removal. Indeed, microinjection of denatured protein into frog
level.
oocytes has been shown to induce HSP70 (24).
* Mean ±SD.
However, a simple correlation between levels of HSP70 and
resistance to heat is not evident when different rodent and
levels of this protein were measured by immunoassay. The human cell lines are compared (1). In fact, the relatively heatconstitutive levels of the protein and the amounts present 17 h sensitive murine RIF-1 cells contain as much total HSP70 as
after a heat shock of 10 min at 45°Care shown in Table 2 for the far more resistant human HT 1080 cells (1). Similarly, we
have shown that heat-sensitive hamster cells have as much
HT1080, CHO, BL-10, H2, and H4. Several interesting results
HSP70 as HT 1080 cells (Table 2). It must be pointed out that
were obtained, (a) H4 cells, which contain 3 of the human
these
measurements are based on total amounts of HSP70.
chromosomes that bear HSP70 genes (chromosomes 6, 14, and
Individual isoforms may prove to be more important in protec
21), have nearly the same constitutive level of HSP70 as H2
tion from stress. Further, it appears that if only levels of HSP70
with only chromosome 6(12, 13). Interestingly, chromosome
6 has been shown to possess three HSP70 genes (14, 15). (¿>) are manipulated (e.g., by appropriate transfection of cells with
HSP70 genes), then increased levels of this protein do indeed
All five cell lines have a similar constitutive level of HSP70,
but vary widely in their sensitivity to heat. HT 1080 cells are far reflect increased heat resistance. These studies suggest that the
more resistant than the others, while BL-10 cells, which seem levels of HSP70 are only one of several parameters that affect
heat resistance.
to contain slightly more HSP70, are much more heat sensitive.
There are also differences in the expression of the heatIn the study reported here, we used a novel approach to
inducible isoforms of HSP70. After a heat shock of 10 min at explore the basis of heat resistance by using hybrid cells formed
45°C,the CHO parent shows a 2- to 3-fold increase in HSP70,
by the fusion of heat-resistant human cells with heat-sensitive
a result similar to that shown by others (2). Surprisingly, BL- hamster cells. By measuring the heat response of the hybrids,
10 cells were unable to show any increase in HSP70 levels we hoped to learn more about the genetic basis of the heat(Table 2). HT1080 cells, in contrast, increase HSP70 levels 5- resistant phenotype. If this phenotype was dominant, we could
fold and, after more severe heat treatments, increases of 15- to determine the chromosomal location of the determinant(s) and,
30-fold are seen (1). Furthermore, hybrid cells H2 and H4 from there, isolate the gene(s) involved. With time after fusion,
display a HSP70 increase similar to that of the human parent,
it has been shown that the hybrids selectively lose human
despite their greater heat sensitivity. We have verified this chromosomes (11), thus permitting the isolation of a family of
increase in the inducible HSP70 isoform by two-dimensional
hybrid lines each containing only a few human chromosomes.
gel electrophoresis (Fig. 6).
Surprisingly, we found that the phenotype of human x ham
ster hybrids is more similar to that of the thermosensitive
hamster cell than that of the human cell lines. This conclusion
DISCUSSION
is based on three independent findings, (a) When we fused a
heat-sensitive CHO line (BL-10) to HT 1080 cells, we were able
The molecular determinants of heat resistance in mammalian
to restore the normal CHO heat response, but not confer extra
cells are not known. Much interest has centered around a group
resistance to heat, (b) Hybrids of wild-type CHO cells and
of inducible proteins known as the heat shock, or stress, pro
either HT1080 or AG1522 cells, both of human origin, dis
teins. These proteins can be divided into several subgroups
played the same phenotype as the CHO parent, (c) Finally,
based on their molecular weight on reducing gels. One or more
further heat selection of CHO x HT 1080 hybrids failed to
members of each group are present as constitutive proteins.
reveal any clones with increased heat resistance. The possibility
Some of these are further inducible by heat and other stresses,
remains that the human chromosome necessary for conferring
while others are synthesized de novo (16). Of the major families
thermoresistance to CHO cells segregated before testing. This
with molecular weights of 110,000, 90,000, 70,000, 60,000,
and 28,000, most knowledge has been obtained about the M, is unlikely, as mass hybrid cultures, which were analyzed by in
situ hybridization, possessed a human chromosome content
90,000 and M, 70,000 groups, as these are major cellular
that ranged from 5 to 20 human chromosomes. The interpre
proteins in the absence of stress, and recently, some of their
functions have been defined (16). The functions of the M, tation of these results is complex. It is possible that several
proteins located on different chromosomes must work in con
90,000 group center on binding to and maintaining steroid
cert to increase the heat resistance of hamster cells. This is not
receptors and various protein kinases in an inactive state until
the hormone ligand or appropriate substrate is present (17). likely, as selection of mass hybrids that possess almost every
The A/r 70,000 group function as housekeeping proteins, bind
combination of human chromosomes did not result in any
ing to nascent proteins, translocating proteins to the endothermorésistant clones. It is also possible that the lack of
2639
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HEAT RESISTANCE LOSS IN HUMAN CELLS
pH5
pH7
B
A
94-
67-
43-
C
9467-
43 —
Fig. 6. Two-dimensional gel analysis of ("S]methionine-labeled proteins from HAI and hybrid H4 cells before and after heating at 45°C.A ', actin; •
-, constitutive
HSP70, <. inducible. A, HAI, control; A, HAI, 45"C for 20 min; C, hybrid H4, control; D, hybrid H4, 45°Cfor 20 min.
thermoresistance in the hybrids can be attributed to suppression
of critical genes or production of nonfunctional proteins caused,
for example, by a lack of correct post-translational modifica
tion. This possibility is currently being investigated by protein
gel electrophoresis for the major heat shock proteins. Alterna
tively, instead of focusing on the isolation of the human gene
that confers resistance, we will aim to isolate, by DNA transfection and replica plating, the hamster gene that confers ther
mal sensitivity.
The chromosomal location of several of the human HSP70
genes has been established. So far, genes have been found on
chromosomes 6, 14, and 21, but the location of genes of other
members of the family, of which there are at least 8, has yet to
be established (12-15). In any case, as shown in Table 2, there
is no correlation between constitutive levels of HSP70 and heat
response of the cells listed.
In the absence of heat shock, hamster, hybrid, and human
cell lines have nearly the same constitutive levels of HSP70.
After a heat shock of 10 min at 45°C,the HSP70 levels of the
hybrid cells respond in a similar way to that of the human
parent, showing a 5- to 7-fold increase in HSP70 levels. CHO
cells typically show a 2- to 3-fold increase in HSP70 after this
heat dose (2). Certain isoforms of HSP70, most notably HSP68
(10, 25), are strongly heat-inducible and thus the molar ratio of
the isoforms will be different after heat shock. The immunologie
quantitation of HSP70 does not allow us to distinguish the
constitutive form from inducible forms. Therefore, we used
two-dimensional gel electrophoresis to help distinguish the
multiple isoforms. We found that the hybrid cell lines do indeed
acquire a large increase in the HSP68 gene after heating,
whereas HAI cells did not. These results suggest either that the
inducible form is not involved in protection against heat, or
more likely, that the amount of new HSP70 produced in re
sponse to stress is not correlated with the cytotoxicity of the
inducing heat dose.
A surprising and as yet unexplained observation was the
failure of BL-10 cells to increase their levels of HSP70 after a
heat shock of 10 min at 45°C.Further experiments using
different heat treatments will determine whether this lack of
induction of HSP70 is a general response to heat shock and
whether these cells can develop thermotolerance.
ACKNOWLEDGMENTS
The authors wish to thank Dr. W. Welch for the gift of the antiHSP70 antibody.
2640
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HEAT RESISTANCE LOSS IN HUMAN CELLS
REFERENCES
1. Hahn, G. M., Ning, S. C., Elizaga, M., Kapp, D. S., and Anderson, R. L. A
comparison of thermal responses of human and rodent cells. Int. J. Radial.
Biol., 56:817-825, 1989.
2. Li, G. C. Elevated levels of 70,000 dalton heat shock protein in transiently
thermotolerant Chinese hamster fibroblasts and their stable heat resistant
variants. Int. J. Radiât.Oncol. Biol. Phys., 11:165-177, 1985.
3. Starnato, T. D., Peters, B.. Patii. P., Denko. N., Weinstein, R., and Giaccia,
A. Isolation and characterization of bleomycin-sensitive Chinese hamster
ovary cells. Cancer Res., 47: 1588-1592, 1987.
4. Giaccia, A. J., Richardson, E., Denko, N., and Starnato, T. A. Genetic
analysis of the XR-1 mutation in hamster and human hybrids. Somatic Cell
Mol. Genet., 75: 71-79, 1989.
5. Davidson, R. I... and Gerald, P. S. Improved techniques for the induction of
mammalian cell hybridization by polyethylene glycol. Somatic Cell Genet.,
2: 165-176, 1976.
6. Giaccia, A. J., Evans, J. W., and Brown, J. M. Use of fluorescent in situ
hybridization to detect chromosomal rearrangements in somatic cell hybrids.
Genes Chromosomes Cancer, 2: 248-251, 1990.
7. Seabright, M. A rapid banding technique for human chromosomes. Lancet,
2:971-972, 1971.
8. Friend, K. K.. Chen, S., and Ruddle, F. H. Differential staining of interspe
cific chromosomes in somatic cell hybrids by alkaline Giemsa stain. Somatic
Cell Genet., 2: 183-188, 1976.
9. Puck, T. T., and Marcus, P. I. Action of X-rays on mammalian cells. J. Exp.
Med., 103: 653-666, 1956.
10. Anderson, R. L., van Kersen, I., Kraft, P. E., and Hahn, G. M. Biochemical
analysis of heat-resistant mouse tumor cell strains: a new member of the
hsp70 family. Mol. Cell. Biol., 9: 3509-3516. 1989.
11. Weiss, M. C, and Green, H. Human-mouse hybrid cell lines containing
partial complements of human chromosomes and functioning human genes.
Proc. Nati. Acad. Sci. USA, SS: 1104-1111, 1967.
12. Goate, A. M., Cooper, D. N., Hall, C., Leung, T. K., Solomon, E., and Lim,
L. Localization of a human heat-shock HSP70 gene sequence to chromosome
6 and detection of two other loci by somatic-cell hybrid and restriction
fragment length polymorphism analysis. Hum. Genet., 75: 123-128, 1987.
13. Harrison, G. S., Drabkin, H. A., Kao, F. T., Hartz, J., Hart, I. M., Chu, E.
H., Wu, B. J., and Morimoto, R. I. Chromosomal location of the human
genes encoding major heat-shock protein HSP70. Somatic Cell Mol. Genet.,
13: 119-130, 1987.
14. Sargent, C. A., Dunham, I., Trowsdale, J., and Campbell, R. D. Human
major histocompatibility complex contains genes for the major heat shock
protein HSP70. Proc. Nati. Acad. Sci. USA, 86: 1968-1972, 1989.
15. Milner, C. M., and Campbell, R. D. Structure and expression of the three
MHC-linked HSP70 genes. Immunogenetics, 32: 242-251, 1990.
16. Lindquist, S., and Craig, E. A. The heat shock proteins. Annu. Rev. Genet.,
22:631-677, 1988.
17. Hardesty, B., and Kramer, G. The 90,000 heat shock protein, a lot of smoke
but no function as yet. Biochem. Cell Biol., 67: 749-750, 1989.
18. Pelham, H. R. B. HSP 70 accelerates the recovery of nucleolar morphology
after heat shock. EMBO (Eur. Mol. Biol. Organ.) J., 8: 3171-3176, 1989.
19. Chiang, H. L., Terlecky, S. R., Plant, C. P., and Dice, J. F. A role for a 70kilodalton heat shock protein in lysomal degradation of intracellular proteins.
Science (Washington DC). 246: 382-385. 1989.
20. Lindquist, S. The heat shock response. Annu. Rev. Biochem., 55: 11511191, 1986.
21. Landry, J., Bernier, D., Chretien, P., Nicole, L. M., Tanguay, R. M., and
Marceau, N. Synthesis and degradation of heat shock proteins during devel
opment and decay of thermotolerance. Cancer Res., 42: 2457-2461, 1982.
22. Li, G. C., and Werb, Z. Correlation between synthesis of heat shock proteins
and development of thermotolerance in Chinese hamster ovary fibroblasts.
Proc. Nati. Acad. Sci. USA, 79: 3218-3222, 1982.
23. Subjeck, J. R., Sciandra, J. J., and Johnson, R. J. Heat shock proteins and
thermotolerance; a comparison of induction kinetics. Br. J. Radiol., 55: 579584, 1982.
24. Anathan, J., Goldberg, A. L., and Voellmy, R. Abnormal proteins serve as
eukaryotic stress signals and trigger the activation of heat shock genes.
Science (Washington DC). 232: 522-524, 1986.
25. Welch, W. J., and Mizzen, L. A. Characterization of the thermotolerant cell.
2. Effects on the intracellular distribution of heat shock protein 80, inter
mediate filaments, and small nuclear ribonucleoprotein complexes. J. Cell
Biol., 106: 1117-1130, 1988.
2641
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
Loss of the Intrinsic Heat Resistance of Human Cells and
Changes in Mr 70,000 Heat Shock Protein Expression in Human
× Hamster Hybrids
Robin L. Anderson, Kathy J. Fong, Tim Gabriele, et al.
Cancer Res 1991;51:2636-2641.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/51/10/2636
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.