[CANCER RESEARCH 36, 4044-4051 , November 1976]
Cell Surface Changes in HeLa Cells as an Indication of Cell
Cycle Events'
Erik Lundgren and GOran Roos
Department of Pathology, University of Umed, S-901 87 Umed, Sweden
SUMMARY
A HeLa cell line synchronized by double thymidine block
and mitotic shake off was shown to have a characteristic
surface
morphology
for each of the different
cell cycle
stages. Inhibitors of cell multiplication were used to arrest
cells in specific cell cycle phases, and these cells had a
surface morphology
similar to that of synchronized
cells in
the same phase. The results indicated a close association
between the cell surface topography and the cycles of DNA
synthesis in the cell nucleus of this HeLa line.
INTRODUCTION
Many reports have appeared in recent years on the cyclic
changes of membrane structure and function that occur
when cells proceed through the different cell cycle phases.
The carbohydrate content has been shown to fluctuate dun
ing the cell cycle, as well as the expression of blood group
and transplantation antigens (7, 10). Concanavalin A and
hormone receptors also display cyclic changes during cell
cycle stages (9, 17).
Considering these fluctuations of plasma membrane
components, it is not surprising to find a close association
between different plasma membrane movement patterns
and cell cycle phases (11). This has also been demonstrated
by a morphological approach to CHO cells by SEM2 (14).
We studied synchronized HeLa cells by SEM and found
variations in surface morphology when the cells progress
through the cell cycle. Through the use of various inhibitors
of cell multiplication, cells were arrested in specific cell
cycle phases. These cells had the characteristic surface
morphology of synchronized cells in that stage, which sug
gested that the structure of the cell surface was associated
with different phases of DNA processing.
MATERIALS AND METHODS
Cell Culture. The HeLa cells were cultured in 90-mm
Nunclon plastic dishes (Nunc, Roskilde, Denmark) in Ea
gle's minimum essential medium supplemented with 10%
calf serum (Flow Laboratories, Irvine, Ayrshine, Scotland),
100 units penicillin pen ml, and 50 @g
streptomycin pen ml.
I This
investigation
was
supported
by
grants
from
the
Swedish
Cancer
Society, Project 659-B74-02X. and from the Medical Faculty, University of
Umeá.
2 The
abbreviation
used
is:
SEM,
scanning
electron
Received February 18, 1975; accepted July 2, 1976
4044
microscopy.
For the experiments the cells were transferred to round
covenglasses placed in 30-mm Nunclon plastic dishes. Fun
then culture conditions have been given earlier (8).
Synchronization was performed either by a double thymi
dine block, according to the principles given by Galavazi et
a!. (5), or by the mitotic shake-offtechnique (19). The thymi
dine concentration was 5 x 10@ M, and the 2 incubation
periods lasted 16 hr each, with an interval of 8 hr. For the
mitotic shake-off procedure, cells were grown in 75-sq cm
plastic flasks (Falcon Plastics, Los Angeles, Calif.) until
confluence was nearly reached, and the medium was man
ually passed gently over the monolayer 20 times. After cen
tnifugation, the cells were resuspended in a small amount of
medium and transferred to plastic dishes.
Autoradiognaphy was used to characterize synchroniza
tion. After a 20-mm pulse with 1 pCi [methyl-3H]thymidmne
pen ml (Aadiochemical Centre, Amensham, England), the
cells were fixed in alcohol:acetic acid (3:1), washed in 70%
ethanol, and air dried. The bottoms of the dishes were cut
out, dipped in Ilford G5 emulsion (Ilfond Ltd. , Ilfond, Eng
land), and exposed for 2 on 3 weeks. After being developed
and stained with May-GnUnewald-Giemsa, cells with labeled
nuclei were counted by means of a light microscope and
immersion objective.
Mitotic index was measured by light microscopy, and the
total cell number was determined by an electronic cell
counter (Linson Instrument AB, Linson Counter 411, Stock
holm, Sweden).
Inhibitors of cell multiplication were used in the doses
shown in Table 1. The inhibitors were dissolved in the
medium or in a stock solution of ethanol (hydrocortisone),
the final concentration of which was less than 0.01%. This
concentration could be shown not to affect surface mon
phology. The cells were treated for 48 hn, and a trypan blue
dye exclusion test determined that cell viability was not
affected in the used concentrations. Vinblastine treatment
could not be performed for more than 24 hr because of
serious cell detachment.
SEM. At various times after the 2nd thymidine block was
released on after mitotic shake off, the cells were fixed in
2.5% glutanaldehyde in 0.1 M phosphate buffer at pH 7.2 for
1 hr. The cells incubated with the inhibitors were fixed
immediately at the end of the incubation period. After fixa
tion, the cells were washed once in buffer and postfixed in
1% 0504 in the same buffer for 1 hr. They were then dehy
drated in increasing concentrations of ethyl alcohol up to
99% and thereafter in a graded series of amyl acetate in
alcohol up to 100% amyl acetate. The preparations were
transferred to a critical-point drying apparatus (Polanon E
CANCER RESEARCHVOL. 36
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1976 American Association for Cancer Research.
Cell Cycle and Surface Morphology
Table I
Variousinhibitors
of cell multiplication usedin the ex
andInhibitor
their modes of actionperiments
actionRef.Hydrocortisone,
10 M
1-@-D-Arabinofuranosyl-
Mode of
Prolongation of G,
Arrest at G,-S
21
23
Arrest at G,-S
Not specific
Prolongation of S
Arrest at M6,
23
1
22
20
cytosine, 5 @tg/ml
Hydroxyurea, 10@M
Mitomycin C, 1 @.tg/ml
Bleomycin, 10 @tg/ml
Vinblastine, 0.01 .@g/ml
3000, Polaron Equipment Ltd. , Watford, England) and dried
with the use of CO2. They dry specimens were coated with a
thin layer of gold, and the examinations were performed in a
Cambridge Steneoscan 54 electron microscope (Cambridge
Scientific Instruments Ltd. , Cambridge, England) operated
at 20 kV.
RESULTS
Synchronization and Surface Morphology. It can be seen
from Chart 1 that the degree of synchronization was rather
high up to 10 hr after release of the thymidine block. Nearly
75% of the cells were in the S phase after 5 hn, and a mitotic
index of 20% was achieved after 9 hr. Apparently, the de
gree of synchronization was low after 15 hn, when the next S
phase started. The mitotic shake-off procedure yielded
about 80% mitotic cells (Chart 2). From 2.5 to 10 hr, no
mitotic cells could be seen. Therefore, thymidine block was
chosen for a study of the surface morphology of S, G2, and
M cells, and mitotic shake off was used to study that of G
cells.
The different surface features were best expressed when
the cells were in contact with each other, a finding that is in
agreement with studies on CHO cells (15). When the cul
tunes were overcrowded, all cells appeared round and were
covered with microvilli (Fig. 1).
Characteristics of the Cell Cycle Phases. Cells that were
shake of
10
hours
Chart 2. Cell cycle analysis after mitotic shake off. 0, percentage of
mitotic cells; •.percentage of labeled cells.
fixed immediately after the 2nd thymidine block was re
leased showed no signs of toxic effects, compared with
control cells. The thymidine block arrested the cells at the
G,-S boundary; at this stage, the cells were rather flat and
attached to the coverglass. Cell surfaces were sparsely coy
ened with short microvilli and some cytoplasmic blebs (Fig.
2).
In the beginning of the S phase, the cells looked like G-S
cells, except for the microvilli, which seemed to accumulate
oven the central parts of the cells. This phenomenon was
more accentuated in mid-S when the microvilli were taller
and more numerous (Fig. 3). The cells were still rather flat,
and few blebs were seen. Toward the end of S, the micnovilli
increased in number and covered a greater proportion of
the cell surface.
Before mitosis, the cells began to appear rounder, the
entire surface was covered with microvilli, and blebs ap
peaned only occasionally (Fig. 4). It seems reasonable to
assume that these surface characteristics represented cells
in G2, but the possibility exists that they were cells in very
late S phase on even early G,, since, obviously, the degree of
Chart 1. Cell cycle analysis after a double thymidine
block. •,
percentage of labeled cells; i@,percentage of
mitotic cells; 0, percentage of increase in cell num
bar.
NOVEMBER 1976
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1976 American Association for Cancer Research.
4045
@
io@@
E. Lundgren and G. Roos
synchronization had deteriorated in this stage (Chart 1). The
mitotic cells were spherical and covered with microvilli of
varying lengths (Fig. 5). Many cells were attached to the
glass by long filopodia that often bifurcated at the ends. In
late mitosis, just before separation of the 2 daughter cells, a
general phenomenon was the appearance of many cyto
plasmic blebs (Fig. 6).
In early G1, the cells were somewhat rounded and had
begun to spread out. Microvilli of varying lengths appeared
all oven the surface. Blebs were often but not always seen
(Fig. 7). During the G1 phase, the cells successively flat
tened, microvilli became shorter, and blebs were still pres
ent. Cells fixed at the G,-S boundary appeared similar to
those fixed immediately after the thymidine block was re
leased (Figs. 2 and 8).
The described surface features represented the main
characteristics of each cell cycle phase. Chart 3 summa
rizes the percentage of cells displaying a certain surface
character after the 2 methods of synchronization. The cells
were counted in the scanning micrographs. Obviously, not
all cells have the described typical appearance, since the
degree of synchronization is never 100%. Moreover, the
morphology changed gradually during the cell cycle, and it
did not seem possible to assign all cells in an unsynchnon
ized culture to a specific stage according to surface mor
phology.
S
,62N, 61 • S
so@7N@@b
The Effect of Inhibitors of Cell Multiplication. To check
the validity of these results, we tested some inhibitors of cell
multiplication with well-characterized modes of action on
the cell cycle. According to Table 1, some inhibited the cells
in specific stages of the cell cycle, while others arrested the
cells more unspecifically.
Hydnocortisone is known to prolong the G, phase (6, 21),
and the principal part of the cells incubated with this drug
appeared as G1 cells, a similarity based on the general
shape as well as the distribution and size of microvilli (Fig.
9).
Incubation with 1-f3-D-arabinofunanosylcytosine and hy
droxyurea has been shown to arrest cells at the G,-S bound
ary (23). Treatment with these drugs induced surface char
acteristics similar to those described occurring immediately
after the 2nd thymidine block was released (Fig. 10; cf. Fig.
2),but the association between surface morphology and cell
cycle stage seemed lower, as many cells with charactenis
tics of other cell cycle phases could be seen.
Most of the cell population appeared as mitotic cells after
incubation with vinblastine (Fig. 11). As a rule, microvilli
were normal, compared with mitotic cells in the thymidmne
synchronized population.
Cells with the appearance of all cell cycle phases were
present after incubation with mitomycin C and bleomycin,
and no specific cell cycle stage seemed to be predominant
(Fig. 12). Some of the cells showed shrinkage of the cell
surface and loss of microvilli, which could be a toxic effect,
although trypan blue exclusion was not affected.
Ruffles The phenomenon of ruffling of the cell mem
bnane seen in CHO cells by SEM has been described (14). In
our HeLa cells, similar structures appeared at the free mar
gins of some cells, but not to the same extent as in the CHO
cells. The ruffles did not seem to be more numerous in any
specific cell cycle phase. After incubation with the inhibi
tors, ruffling was still visible. Fig. 13 shows ruffling in a cell
treated with mitomycin C.
DISCUSSION
Our observations indicate that it is possible to recognize
different cell cycle phases by the surface morphology of the
HeLa line. Microvilli increased in number during the S
phase and were most abundant at G2. M. Porter et a!. (14)
50
a
found that microvilli were most numerous during G2, M, and
G1, in CHO cells, and similar findings have been made in a
melanoma cell line (16). This discrepancy may be due to
different culture conditions on to the fact that the cell lines
50
f
were derived from different species.
During the S phase, microvilli accumulated over the cen
tral parts of the cell, a phenomenon reported in unsyn
I hours
chronized rat sarcoma cells (13). Studies of unsynchronized
Chart 3. Appearance otsome surface ch&acteristicsdunng the cell cycle.
HeLa S3 cells by SEM or transmission electron microscopy
Time scale (abscissa) indicates hr after removal of the 2nd thymidine block.
Arrow, time of mitotic shake off; Points, mean value of 200 counted cells. •, have not shown any unevenness in the distribution of mi
crovilli (3, 12). Our HeLa cell line, continuously cultured
synchronization by a double thymidine block; 0, synchronization by mitotic
shake off; a, percentage of cells with few microvilli; b. percentage of cells
separately from other HeLa cell lines since 1956, is ob
with the whole surface covered with microvilli; c, percentage of cells with
central accumulationof microvilli; d, percentageof flat cells; e, percentage viously different. Accordingly, it was possible to ascribe
of spherical cells; f, percentage of cells with at least 2 cytoplasmic blebs.
certain surface details to different cell cycle phases of 1
4046
CANCER RESEARCHVOL. 36
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1976 American Association for Cancer Research.
Cell Cycle and Surface Morphology
specific cell line, but obviously this was not a general phe
nomenon.
The fact that the changing distribution of microvilli is
expressed when cells grow in a monolayer and not in
overcrowded cultures (3, 12) makes it difficult to explain in
functional terms the nature of the changes in size and
distribution of microvilli. The transitional nature of micro
villi has been shown by trypsinization of HeLa cells (3).
Cytoplasmic blebs were predominantly seen during late
mitosis and G, phase, which agrees with observations made
on HeLa cells by transmission electron microscopy (2). We
are inclined to consider these morphological changes as
possible markers of nuclear activity, while their functional
meaning must await other investigational approaches.
Cells arrested at a specific cell cycle stage by incubation
with synchronizing inhibitors appeared much like synchro
nized cells in that particular phase. Some of the inhibitors,
i.e. , 1-f3-D-anabinofunanosylcytosine,
hydnoxyunea,mitomy
cm C, and bleomycin, are known to inhibit DNA synthesis,
but not RNA and protein synthesis (18, 22, 23). This means
that, although cytoplasmic and cell membrane growth con
tinued, the surface morphology did not change, suggesting
that the surface morphology was directed by the cycles of
DNA synthesis.
It is known
that thymidine
also has effects
other
than
those on DNA synthesis, which may be the reason for the
observed changes in surface morphology. However, mitotic
shake off, without use of any inhibitors, is a gentle method
of synchronization, and no principal differences could be
seen between cells synchronized with the 2 methods in
comparable cell cycle phases, although the degree of syn
chronization could be variable. Thus, the changes seemed
to be due to the cell cycle phase and not the method of
synchronization.
Our observation that ruffling of the cell membrane oc
curned throughout the cell cycle in the synchronized cells as
well as after treatment with inhibitors of DNA synthesis is in
agreement with the results of Gail et a!. (4). They found
locomotion in 3T3 cells that were cultured in medium lack
ing growth factor. Apparently, there was a dissociation be
tween the rough molecular structure, indicated by surface
morphology, and certain cell membrane movements like
ruffling. The membrane components directing surface mon
phology seemed to be governed by nuclear activity, while
ruffling was independent of it, which may indicate that in
this HeLa cell line the attaching sites for the forces causing
membrane movements were more constant than the cell
membrane structures that maintained surface morphology.
ACKNOWLEDGMENTS
The authors are appreciative of the technical assistance of C. Rooth, P.
HOrstedt, and B. Carfors.
NOVEMBER1976
REFERENCES
1. Djordjevic, B., and Kim, J. H., Different Lethal Effects of Mitomycin C
and Actinomycin D during the DivisionCycle of HeLa Cells. J. Cell Biol.,
38: 477-482, 1968.
2. Erlandson, R. A., and de Harven, E. The Ultrastructure of Synchronized
HeLa Cells. J. Cell Sci.; 8: 353-397, 1971.
3. Fischer, H. W. , and Cooper, T. W. Electron Microscope Studies of the
Microvilli of HeLa Cells. J. Cell Biol., 34: 569-576, 1967.
4. Gail, M. H., Scher, C. D., and Boone, C. W. Dissociation of Cell Motility
from Cell Proliferation in BALB/c-3T3 Fibroblasts. Exptl. Cell Res., 70:
439-443, 1972.
5. Galavazi, G., Schenk, H., and Bootsma, D. Synchronization of Mamma
han Cells in Vitro by Inhibition of the DNA Synthesis. I. Optimal Condi
tions. Exptl. Cell Res., 41: 428-437, 1966.
6. Kollmorgen. G. M. . and Griffin, M. J. The Effect of Hydrocortisone on
HeLa Cell Growth. Cell Tissue Kinet., 2: 111-122, 1969.
7. Kuhns, W. J., and Bramson, S. Variable Behaviour of Blood Group H on
HeLa Cell Populations Synchronized with Thymidine. Nature, 219: 938939, 1968.
8. Lundgren, E. IsozymeVariations in Human Cells Grown in Vitro. VI.
Further Data on the Amino Acid Naphthylamidase Isozymes. Human
Heredity,21: 1-11, 1971.
9. Makman,M. H., and Klein, M. I. Expressionof AdenylateCyclase,Cate
cholamine Receptor, and Cyclic Adenosine Monophosphate-dependent
Protein Kinase in Synchronized Culture of Chang's Liver Cells. Proc.
NatI. Acad. Sci. U. S., 69: 458-458, 1972.
10. Pasternak, C. A., Sumner, M. C. B., and Collin, R. C. L. S. Surface
Changes during the Cell Cycle. In: G. M. Padilla, J. L. Cameron, and A.
Zimmerman (ads.), Cell Cycle Controls, pp. 117-124. New York: Aca
demic Press, Inc., 1974.
11. Pontén,J. , Westermark, B. , and Hugosson, A. Re9ulation of Prolifera
tion and Movement of Human Glia-like Cells in Culture. Exptl. Cell. Res.,
58: 393-400, 1969.
12. Porter,K. A., Fonte,V., andWeiss,G. A. ScanningMicroscopeStudyof
the Topography of HeLa Cells. Cancer Rae., 34: 1385-1394, 1974.
13. Porter, K. R., Kelley, D., and Andrews, P. M. The Preparation of Cultured
Cells and Soft Tissues for Scanning Electron Microscopy. In: 0. Johari
and I. Corvin (ads.), Scanning Electron Microscopy, pp. 1-19. Chi
cago: lIT Research Institute, 1972.
14. Porter, K. A., Prescott, D., and Frye, J. Changes in Surface Morphology
of Chinese Hamster Ovary Cells during the Cell Cycle. J. Cell Biol., 57:
815-836, 1973.
15. Rubin, A. W., and Everhart, L. P. The Effect of Cell-to-Cell Contact on the
Surface Morphology of Chinese Hamster Ovary Cells. J. Cell Biol., 57:
837-844, 1973.
16. Sank, J. J., White, J. G., and Witkop, C. J. Alteration in the Surface
Morphology of Synchronized B-16 Melanoma Cell during the Cell Cycle.
Acta Dermato-Venereol., 55: 331-336, 1975.
17. Shoham, J., and Sachs, L. Different Cyclic Changes in the Surface
Membrane of Normal and Malignant Transformed Cells. Exptl. Cell Res.,
85: 8-14, 1974.
18. Studzinsky, G. P., and Cohen, L. 5. Mitomycin C-induced Increases in
the Activities of the Desoxyribonucleasesof HeLa Cells. Biochem. Bio
phys. Res. Commun., 23: 506-512, 1966.
19. Terasima, T., and Tolmach, L. J. Growth and Nucleic Acid Synthesis in
SynchronouslyDividing Populations of HeLa Cells. Exptl. Cell Res., 30:
344-362, 1963.
20. Tobey,R. A., Peterson,D. F., and Anderson,E. C. Biochemistryof G,
and Mitosis. In: R. Baserga (ed), The Cell Cycle and Cancer, pp. 309353. New York: Marcel Dekker, Inc., 1971.
21. Tu, S-H., Nordquist, R. E.. and Griffin, M. J. Membrane Changes in HeLa
Cells Grown with Cortisol. Biochim. Biophys. Acta, 290: 92-109, 1972.
22. Watanabe, M., Takabe, Y., Katsumata, T., and Teraèima,T. Effects of
Bleomycin on Progression Through the Cell Cycle of Mouse L-CeIls.
Cancer Res., 34: 878-881 , 1974.
23. Wheeler, G. P., Bowdon, B. J., Adamson, 0. J., and Vail, M. H. Compari
son of the Effect of Several Inhibitors of the Synthesis of Nucleic Acids
upon viability and Progression through the Cell Cycle of Cultured H.
Ep. No. 2 Cells. Cancer Res., 32: 2661-2669, 1972.
4047
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1976 American Association for Cancer Research.
E. Lundgren
and
G. Roos
@: ;‘@j@
Fig. 1. Rounded cells covered with microvilli
inan overcrowdedculture.
x 3000.
Fig. 2. Cells near the G-S boundary fixed
immediately after the release of the 2nd thymi
dine block. Note short microvilli evenly distrib
uted over flattened cells. x 1100.
Fig. 3. Cells with accumulation of microvilli
over the central parts of the cells. Fixation was
performed 4 hr after thymidine release, i.e.,
during S phase. x 1100.
Fig. 4. Somewhat rounded cells with the
surface covered with microvilli, probably repre
senting 0, cells. Fixed 9 hr 40 mm after thymi
dinerelease.
x 1100.
4048
CANCER RESEARCH VOL. 36
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1976 American Association for Cancer Research.
@
@
/
.@
. .@
Cell Cycle and Surface Morphology
:@.
@
@
@
@
@
Fig. 5. A cluster of round, mitotic cells fixed 9
hr 40 mm after thymidine release. x 1100.
Fig. 6. Late mitosis just before cytoplasmic
separation. Note many cytoplasmic blebs. Fixed
1 hr after mitotic shake off. x 1100.
Fig. 7. Early G, cells spread out on the surface. Fixed3hraftermitoticshakeoff.
x 1100.
Fig. 8. Cells fixed at late G, 7 hr after mitotic
shakeoff.x1200.
5
‘
,,
:
•
•
@.e1
@
.‘--,@——
.
...
.@—
L.
@
@
@
6
\
L•
@
‘@@‘:@
•
.
‘—@
‘I
;
.@)
,@-,
@:.
:
.
.@,;
S @.
‘
.
@s@it@
;
‘4
I
WI
NOVEMBER1976
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1976 American Association for Cancer Research.
4049
@
@
@
@:#@:r
:
:@:
:
r
.
:@:
E. Lundgren and G. Roos
.@
“ •:‘
@
@.,.4-
,t.
@
@.pall,,,
I-
@
@
@
@
.
..-
.
.
v:
,
• .
.
. .
@
, . A@-i@@.
@
,@
..
.-;
@
@
.
--.-_@.
@
@
..
—,“ ..-@__@&.___:___._._—*@‘-@,
:.
.,
@
.4
,f
,@
.
@*
.
..
-
. “,4@:.,
.A.@
,.
.
.@
,
@—
ir@@
.
a'•4•-@@,@,
@.
@
-@ t@'
A@
..
9
10
r':@
‘+,,;
@. ‘4t@,
@
•
‘
I
@
.@
1@
Ã-
.
@
.@q..
fl.,.
@
hI.
I...,
@..
[email protected]+.. ‘..
,@
.
4
+
)
@
-
.@..
•@:@-. ,
, +-
+
*,f
.@.
@.
•1_
,.
I
.@.
@
@
.
.
‘@@i1/
@++
@.
11
Fig.
Fig.
Fig.
Fig.
4050
12
9. Cells fixed after incubation with 10 â€M̃ hydrocortisone for 48 hr. x 2400.
10. Cells fixed after incubation with 1O@PAhydroxyurea for 48 hr. x 1200.
11. Mitotic cells after treatment with 0.01 @.tg
vinblastine per ml for 24 hr. x 2400.
12. Cells with the appearance of cells in all cell cycle phases can be seen after incubation with 1 @g
mitomycin C per ml for 48 hr. x 1200.
CANCER
RESEARCH
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1976 American Association for Cancer Research.
VOL.
36
Ce!! Cycle and Surface Morphology
N
(+
@
/
‘_4*/@
,@‘
,.
.
.,@‘+‘
Fig. 13. Ruffling of the cell membrane (arrows)
in a cell treated with mitomycin C for 48 hr. x 3250.
A9
/
13
NOVEMBER 1976
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1976 American Association for Cancer Research.
4051
Cell Surface Changes in HeLa Cells as an Indication of Cell
Cycle Events
Erik Lundgren and Göran Roos
Cancer Res 1976;36:4044-4051.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/36/11_Part_1/4044
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 16, 2017. © 1976 American Association for Cancer Research.