[CANCER
RESEARCH
37, 3136-3140,
September
1977]
A/,A/-Dimethylformamide-induced
Morphological
Differentiation
and Reduction of Tumorigenicity
Rhabdomyosarcoma
Cells1
in Cultured Mouse
Daniel L. Dexter
Department
02908
ol Medicine,
Roger Williams
General Hospital,
and Division
of Biological
SUMMARY
A/./V-Dimethylformamide treatment of cell cultures estab
lished from a transplantable murine rhabdomyosarcomainduced morphological differentiation and a marked reduc
tion in the tumorigenicity of the sarcoma cells. Fourteen of
17 CE/J mice receiving injections of inducer-treated cells
did not develop tumors after 6 months, whereas all 21 mice
receiving inocula of untreated sarcoma cells died of disease
between 11 and 31 days. The drug-treated cells did not grow
in soft agar; untreated tumor cells grew in the semisolid
medium. The untreated tumor cells showed a reduced se
rum requirement and had a higher saturation density com
pared to drug-treated cells. Thus the reduction in tumori
genicity of A/./V-dimethylformamide-treated cells correlates
with certain in vitro growth properties that are more charac
teristic of normal, mesenchymally derived cells than of sar
coma cells.
INTRODUCTION
Induction of differentiation of tumor cells has been inves
tigated in several in vitro systems. Paran et al. (6) have used
culture medium conditioned by human spleen cells to effect
the differentiation of human leukemic cells to granulocytes.
Silagi ef al. (11) found that mouse melanoma cells treated in
culture with 5-bromodeoxyuridine lost their pigment and
piled-up morphology, had a reduced tumorigenicity, and
possessed an increased number of C-type particles of mu
rine leukemia virus and an enhanced immunogenicity. A
variety of polar solvents have been shown to be good inducers for erythroleukemic cells (2, 10, 15). Dimethyl sulfoxide-treated cells have been characterized by the synthesis of
hemoglobin (2), accumulation of globin mRNA (9), and the
appearance of an erythrocyte membrane antigen (4). Kimhi
ef al. (5) have reported on the dimethylsulfoxide-induced
maturation of murine neuroblastoma cells in culture; the
treated cells showed pronounced morphological changes
and a high degree of electrical excitability (5). This paper
reports the in vitro induction of morphological differentia1 This work was supported by USPHS NIH Grants CA14520. CA13548, and
CA13943. Part of this study was completed at the Wisconsin Clinical Cancer
Center, University of Wisconsin. Madison, Wis. 53706.
Received September 27, 1976; accepted June 7, 1977.
3136
and Medical
Sciences,
Brown
University,
Providence,
Rhode Island
tion and reduction of tumorigenicity in mouse rhabdomyosarcoma cells by another polar solvent, DMF.2
MATERIALS
AND METHODS
Cell Cultures. Rhabdomyosarcoma cells were obtained
from a transplantable tumor (BW10139; The Jackson Labo
ratory, Bar Harbor, Maine) passaged in CE/J mice. Tumor
tissue was excised and minced, and cultures were obtained
either from cells growing out from small tumor fragments or
from tumor mince treated with 0.25% trypsin (Grand Island
Biological Co., Grand Island, N. Y.) to obtain single-cell
suspensions. Cell suspensions or fragments were cultured
in tissue culture dishes (Falcon No. 3001 or No. 3002; Fal
con Plastics, Oxnard, Calif.) that had previously been
treated with 2 ml of 0.02% gelatin (Difco, Detroit, Mich.) for
3 hr at 4°to provide a substratum. Growth medium con
sisted of Eagle's basal medium supplemented with 15%
heat-inactivated horse serum (both from Grand Island Bio
logical Co.) and containing 20 ¿¿g
gentamicin (Schering
Corp., Port Reading, N. J) and 100 units mycostatin (Grand
Island Biological Co.) per ml, respectively. Some cultures
were maintained in growth medium that contained 2%
chicken embryo extract (Grand Island Biological Co.). Cul
tures were kept at 37°in a humidified atmosphere of 5% CO-,
in air. For routine passage, cells were treated with 0.25%
trypsin for 2 min and then replated at 1 x 10" cells/35-mm
dish.
Growth Rates. For determination of cell-doubling times,
replicate culture dishes received inocula of 5 x 10" cells,
and cells from 2 dishes were harvested separately each day
and counted with a hemacytometer. Results were plotted on
semilogarithmic paper, and doubling times were calculated
from the curves.
Plating efficiencies on plastic were determined per inocu
lum by counting under a light microscope the colonies
present in duplicate dishes 1 week after seeding.
Growth in Agar. For growth in semisolid medium, cells
were resuspended in 0.5% agar (Difco) in complete growth
medium, and 1 ml of this suspension was layered on a 2-ml
base of 1.0% agar in medium in a 35-mm dish (Falcon No.
3001). Treated cells were seeded in agar medium containing
1%DMF.
2 The abbreviations
used are: DMF, N.N-dimethylformamide;
N8,O2'-dibutyryl adenosine 3':5'-cyclic
monophosphoric
acid.
CANCER
RESEARCH
dbcAMP.
VOL. 37
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1977 American Association for Cancer Research.
Effects of DMF on Mouse Rhabdomyosarcoma
Cloning efficiencies in agar were determined by counting
under a light microscope
colonies larger than 50 ^m in
duplicate dishes per inoculum. Plates were scored 2 weeks
after cultures were initiated.
Saturation Densities. These data were obtained from the
results of experiments done to determine doubling times.
The numbers of cells present in confluent dishes which
determined the plateau regions of the growth curves pro
vided the maximum cell number per 35-mm dish.
Induction Experiments. A 10% (v/v) solution of DMF
(Fisher Scientific Co., Fair Lawn, N. J.) in basal medium was
added to cultures to give a final concentration
of 1% DMF in
the growth medium. Other cultures received growth me
dium that contained dbcAMP (5 x 10" M) (Sigma Chemical
Co., St. Louis, Mo.).
Cytogenetic
Studies. Exponentially growing cultures
were incubated with 0.06 ^9 colcemid (Grand Island Bio
logical Co.) per ml for 2 hr. Cells were harvested, and
"squash"
preparations were made according to standard
techniques. The preparations were observed under a phase
microscope, and 25 metaphases were analyzed for each cell
type.
Autoradiography.
The method is essentially that de
scribed by Pariza et al. (7). Cells were plated on covered
tissue culture chamber slides (Lab-Tek Products, Naperville,
III.), and when cultures
were 30% confluent,
[3H]leucine (1 ¿¿Ci/ml
in the growth medium; 5 Ci/mmole;
New England Nuclear, Boston, Mass.) was added for 4 hr.
Autoradiograms
were prepared in the laboratory of Dr. Van
R. Potter, Oncology Department, University of Wisconsin,
Madison, Wis.
Mice. CE/J mice were obtained from The Jackson Labo
ratory. Athymic (nude) mice bearing thenu/nu genotype on
a BALB/c background were provided by Dr. Robert Auerbach, Zoology Department, University of Wisconsin.
Inoculation of Mice. Male CE/J (or BALB/c nude) mice, 6
to 10 weeks of age, were used as hosts for tumor cell
injections. Cells in cultures were trypsinized, and the cell
suspension was centrifuged (5 min at 1400 rpm in a Sorvall
GLC-1 centrifuge). The pellet was washed twice with 0.9%
NaCI solution and then resuspended in the NaCI solution to
give a cell density of 3 x 106 cells/ml. One million cells
(except in studies of doses lethal to 50% of mice) were
injected either s.c. or i.p.
Cells
Histology. Dr. William Ribelin of the University of Wiscon
sin Research Animal Resources Center, Madison, per
formed routine hematoxylin-eosin
staining and histology of
both tumors obtained from mice and tumor cells growing in
vitro.
RESULTS
Effect of DMF on Cell Morphology, Growth Properties,
and Viability. Untreated rhabdomyosarcoma cells grew with
a generation time of 12 hr and were always randomly ori
ented with respect to each other. Piling up and growth of
cells on top of layers of confluent cultures were always
observed (Fig. 1). The cells had an irregular morphology but
were often triangular, with short processes. Hematoxylineosin staining of cultures showed scanty cytoplasm. These
cells formed colonies in soft agar; a cloning efficiency of
48% was obtained when 3 x 102 cells were seeded in the
semisolid medium. The plating efficiency on plastic for
these cells was 40% at 100 or 300 cells seeded (Table 1). The
modal chromosome number for untreated cells was 80.
When subconfluent
(1 x 105 cells/35-mm dish) cultures
treated with DMF (final concentration of 1%) were examined
with a light microscope 1 day after drug addition, the cells
were much more elongated than the untreated tumor cells.
All cells remained attached to the dish surface, and after a
lag time of several days, the treated cells began to prolifer
ate. Autoradiograms
prepared from subconfluent
cultures
of DMF-treated
and untreated
cells incubated
with
[3H]leucine showed that virtually every cell in both induced
and control cultures contained label. As cultures
with DMF approached confluency, the elongated,
treated
needle-
shaped cells oriented themselves with a definite polarity,
producing swirl patterns visible under light microscopy.
Piling up in confluent DMF-treated cultures was never ob
served. These cells often contained 2 or 3 nuclei compared
to the mononucleated
tumor cells (Fig. 2).
In vitro passages of treated cells were never accompanied
by any noticeable DMF-induced toxicity; resuspensions of
trypsinized
confluent DMF-treated cell cultures routinely
contained 95% viable cells as shown by trypan blue dye
exclusion.
Resuspensions
of trypsinized,
untreated cells
from confluent dishes usually contained 90% viable cells.
The in- vitro growth properties of treated cells are summa-
Table 1
In vitro growth properties of DMF-treated and untreated cells
Cell suspensions in growth medium on agar-containing growth medium were plated in
35-mm tissue culture dishes. The various parameters were determined as described in
"Materials and Methods." Duplicate dishes were analyzed in all experiments. In the
determination of doubling times and saturation densities, the cell numbers determined on
a given day for each of 2 duplicates never differed by more than 5%. The values for plating
or cloning efficiencies agreed within 3% for all duplicates seeded at a given density. Each
value reported is the average of duplicate determinations.
density (cells/
typeSarcomaDMF-treated
Cell
time(hr)1222Saturation
cm)1.1
sq
10s2.9
x
efficiency
agar48%;
in
3 x 102 cells
seededNo
x 10"Cloning
sarcomaDoubling
SEPTEMBER
1977
colony growth
in agarPlating
efficiency
dishes40%
on culture
at 3 x I02and 1
plated8%
x 102 cells
at 3 x 102 cells
plated
3137
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1977 American Association for Cancer Research.
D. L. Dexter
rized in Table 1. The doubling time of DMF-treated cells in
their 10th passage in DMF-containing medium was 22 hr,
considerably longer than that of untreated cells (12 hr). The
saturation density of untreated cells was almost 4 times that
of the drug-treated cells. DMF-treated cells did not form
colonies in soft agar; their plating efficiency on plastic was
8% with an inoculum of 3 x 102 cells. The modal chromo
some number of the treated cells was 80, identical to that of
untreated cells.
The following experiment was done to determine whether
the morphological changes effected by DMF were reversi
ble. DMF-containing medium was removed from cell cul
tures that were growing in the presence of the drug for 4
passages (30 days). Fresh medium without DMF was added
to the cultures, and over a period of 4 days the cells gradu
ally became less elongated, lost their polarity, and finally
resembled in every way untreated cells. Thus the continued
presence of the drug is required for DMF-effected morpho
logical differentiation.
Tumorigenicity Experiments. DMF-treated and untreated
cells were injected i.p. into CE/J mice to test for tumorigenicity. Treated cells used in injections were always pas
saged at least twice in culture in the presence of DMF and
were maintained in the inducer for at least 10 days. Cells
from noninduced cultures were injected into control mice at
the same time to provide controls for each injection.
Twenty-one mice received injections of untreated rhabdomyosarcoma cells from various passages. All mice died of
tumors in an average time of 18 days (Table 2). Seventeen
mice received cells from cultures that had initially received
DMF at various passage levels and were then grown in the
presence of the drug for several additional passages. Three
mice died of tumors; the other mice were surviving 6
months later with no sign of disease (Table 2). Two of the
3 mice that developed tumors had significantly increased
survival times compared to those of mice that received
inocula of tumor cells from untreated cultures harvested
at approximately the same passage number. These 2 mice
died on the 47th and 49th days after injection of treated
cells from Passage 21. Two mice receiving untreated cells
from Passage 20 died after 13 and 25 days, whereas 2
others receiving untreated cells from Passage 22 survived
only for 11 and 13 days. That the method of injection does
not affect the results was shown when 8 of 8 CE/J mice
given s.c. injections of control tumor cells developed solid
tumors, whereas 4 of 4 CE/J mice given DMF-treated cells
s.c. remained tumor free.
Experiments in which varying doses of untreated cells
were injected i.p. into mice established that 1 to 2 x 105
cells is the minimum inoculum capable of producing tu
mors. This finding, together with the fact that only 3 of 17
mice given 1 x 106 treated cells developed tumors, sug
gested that addition of DMF might result in a significantly
altered antigenicity, responsible for the rejection of the
treated tumor cells. This possibility was further studied by
injection of cells into sublethally irradiated CE/J mice or
into nude mice (nu x nu BALB/c). All 3 irradiated mice
given treated cells i.p. did succumb to disease, but at a
significantly later time than did irradiated mice given un
treated cells (Table 3). Treated cells from the same inocu
lum preparation were injected into 4 unirradiated mice at
3138
Tumorigenicity
Table 2
experiments
with untreated and DMF-treated
rhabdomyosarcoma
cells
Cells from confluent dishes were removed by trypsinization,
washed twice with 0.9% NaCI solution, and then were resuspended
in the NaCI solution. Each male CE/J mouse, 6 to 10 weeks old,
received an i.p. injection of 1 million trypan blue-excluding cells.
All animals reported as surviving showed no sign of disease 6
months after challenge.
no. of
no. of
cells when
No. of mice cells in
DMF was
inoculatedTotalTotal
jected13151720222732372121293436413638Passage
addedNo (days)"13, time
drugNo
1413,
drugNo
2113,
drugNo
1713,
16,
drugNo
2511,
drugNo
1313,
drugNo
312718,
15, 16, 18, 19,
drugNo
drug1719263232323434Survival
31122,
30daysBoth
alive, 1 dead after
49daysBoth
dead after 47 and
aliveBoth
aliveBoth
aliveBoth
aliveAliveAll
alive
17223226132122222214Passage
" Average survival, 18 days with a range of 11 to 31 days.
Table 3
Tumorigenicity studies with irradiated hosts
Cells from confluent dishes were removed by trypsinization,
washed twice with 0.9% NaCI solution, and then were resuspended
in the NaCI solution. Each male CE/J mouse, 6 to 10 weeks old,
received an i.p. injection of 1 million trypan blue-excluding cells.
Irradiation dose was 600 R, administrated 2 hr before injection. All
animals reported as surviving showed no sign of disease 6 months
after challenge.
No. of mice
inoculated2
no. of
no. of cells when DMF
cells injected37
was
addedNo
time
(days)21,
irradiated
drug
23
No drug
17
1 irradiated
39
2irradiated1
384438Not
343434Survival
35,3747All
irradiated4
nonirradiated2
aliveBoth
injectedPassage
alive 6
irradiatedPassage
monthslater
the same time, and all these mice were surviving 6 months
later with no sign of tumor. Also, 1 x 106 DMF-treated cells
injected s.c. into nude mice did produce small solid tumors
in 3 animals.
Experiments with dbcAMP. Since dbcAMP has been ob
served in other systems to induce morphological differen
tiation similar to what was observed here with DMF, it was
of interest to test dbcAMP in cultures of these sarcoma cells
(3, 8). Cells in cultures treated with the cyclic nucleotide
analog showed morphological changes similar to those obCANCER
RESEARCH
VOL. 37
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1977 American Association for Cancer Research.
Effects of DMF on Mouse Rhabdomyosarcoma Cells
served with DMF-treated cells. However, the swirling pat
tern seen in dishes confluent with DMF-treated cells was not
nearly as pronounced in the dbcAMP-treated cultures.
Three CE/J mice were each inoculated i.p. with 1 x 106
dbcAMP-treated cells; the analog had been added at the
29th passage, and the cells were then cultured in the pres
ence of the drug for 3 additional passages. All 3 mice
developed tumors, but with significantly increased latency
periods (30, 40, and 53 days) compared to mice that re
ceived untreated tumor cells (see Table 2). Thus the reduc
tion in tumorigenicity was much less than that induced by
DMF.
ACKNOWLEDGMENTS
The author acknowledges Dr. Paul Calabresi, Chairman, Department of
Medicine, Brown University, who provided support for this study. He also
acknowledges Dr. Robert Auerbach, Zoology Department, Madison campus,
who during his association with the Wisconsin Clinical Cancer Center pro
vided space and allocated center grant resources for this work.
REFERENCES
DISCUSSION
It has been shown that a number of in vitro growth param
eters sometimes correlate with the transformation of mesenchymally derived cells to sarcoma cell types. Neoplastic
transformation can be accompanied by the acquisition of
the ability of these cells to grow in semisolid medium (13,
14), or with a reduced serum requirement (1, 12), or with an
increased saturation density (16) compared to untransformed control cells. In this study, various in vitro growth
parameters of untreated and DMF-treated sarcoma cells
were determined to see whether any of these parameters
correlated with the marked reduction in tumorigenicity ob
served with drug-treated cells. The reduction correlated
with: (a) inability of treated cells to grow in soft agar, (b)
their increased serum requirement, and (c) their decreased
saturation density. Thus DMF-treated tumor cells acquired
several in vitro properties that are usually considered more
consistent with normal, mesenchymally derived cell growth
than with sarcoma cell growth. The agreement in modal
chromosome number between treated and untreated cells
suggests that the changes effected by DMF are manifested
at the phenotype level in treated cells. However, until more
careful cytogenetic studies are done, including determina
tion of banding patterns, the possibility of drug-induced
chromosomal changes cannot be ruled out.
The reduction in tumorigenicity seen in this study can be
interpreted in several ways. It may be due to an enhanced
immunogenicity in the total population of treated rhabdomyosarcoma cells. The appearance of tumors in nude and
irradiated mice inoculated with treated cells, albeit at a later
time than in controls, is consistent with this hypothesis.
Experiments are in progress to determine the relative immunogenicities of control and DMF-treated tumor cells. Other
possibilities, such as the inability of treated cells to alicit the
vascularization necessary for tumor growth in inoculated
normal CE/J hosts, also exist. Regardless of the mecha
nism, these results show that polar solvent-induced loss of
tumorigenicity and morphological differentiation have been
achieved with cells derived from a soft tissue sarcoma. This
SEPTEMBER
extension of methods applied previously to erythroleukemic
and neuroblastoma cells indicates that chemical induction
of a reduced tumorigenicity in neoplastic cells may be pos
sible with a variety of tumor cell types.
1. Chen, T. T., and Heidelberger, C. Quantitative Studies on the Malignant
Transformation of Mouse Prostate Cells by Carcinogenic Hydrocarbons
in Vitro. Intern. J. Cancer, 4: 166-178, 1969.
2. Friend, C. W., Scher, W., Holland, J. G., and Sato, T. Hemoglobin
Synthesis in Murine Virus-induced Leukemic Cells In Vitro: Stimulation
of Erythroid Differentiation by Dimethyl Sulfoxide. Proc. Nati. Acad. Sci.
U. S., 68: 378-382, 1971.
3. Furmanski, P., Silverman, D. J., and Lubin, M. Expression of Differen
tiated Functions in Mouse Neuroblastoma Mediated by Dibutyryl-Cyclic
Adenosine Monophosphate. Nature, 233: 413-415, 1971.
4. Ikawa, Y., Furasawa, M., and Sugano, H. Erythrocyte Membrane-Spe
cific Antigens in Friend Virus-Induced Leukemia Cells. Bibliotheca Haematol., 39: 955-967, 1973.
5. Kimhi, Y., Palfrey, C., Spector, I., Barak, Y., and Littauer, U. Z. Matura
tion of Neuroblastoma Cells in the Presence of Dimethylsulfoxide. Proc.
Nati. Acad. Sei. U. S., 73: 462-466, 1976.
6. Paran, M.. Sachs, L., Barak, Y., and Resnitzky, P. In Vitro Induction of
Granulocyte Differentiation in Hematopoietic Cells from Leukemic and
Nonleukemic Patients. Proc. Nati. Acad. Sei. U. S., 67: 1542-1549, 1970.
7. Pariza, M. W., Yager, J. D., Goldfarb, S., Gurr, J. A., Yangi, S., Gross
man, S. H., Becker, J. E., Barber, T. A., and Potter, V. R. Biochemical,
Autoradiographic, and Electron Microscopic Studies of Adult Rat Liver
Parenchymal Cells in Primary Culture. In L. E. Gerschenson and E. B.
Thompson (eds.), Gene Expression and Carcinogenesis in Cultured
Liver, pp. 137-167. New York: Academic Press, Inc., 1975.
8. Prasad, K. N., and Hsie, A. W. Morphological Differentiation of Mouse
Neuroblastoma Cells Induced in Vitro by Dibutyryl Adenosine 3' :5'-Cyclic
Monophosphate. Nature, 233: 141-142, 1971.
9. Ross, J., Ikawa, Y., and Leder, P. Globin Messenger-RNA Induction
during Erythroid Differentiation of Cultured Leukemia Cells. Proc. Nati.
Acad. Sei. U. S., 69: 3620-3623, 1972.
10. Scher, W., Preisler, H. D., and Friend. C. Hemoglobin Synthesis in
Murine Virus-induced Leukemic Cells in Vitro. III. Effects of 5-Bromo-2'Deoxyuridine, Dimethylformamide and Dimethylsulfoxide. J. Cellular
Physiol., 81: 63-70, 1973.
11. Silagi, S., Beju, D., Wrathall, J., and DeHarven, E. Tumorigenicity,
Immunogenicity and Virus Production in Mouse Melanoma Cells Treated
with 5-Bromodeoxyuridine. Proc. Nati. Acad. Sei. U. S., 69: 3443-3447,
1972.
12. Smith, S. H., Scher, C. D., and Todaro, G. J. Induction of Cell Division in
Medium Lacking Serum Growth Factor by SV40. Virology, 44: 359-370,
1971.
13. Stoker, M. Abortive Transformation by Polyoma Virus. Nature, 278: 234238, 1968.
14. Stoker, M. Regulation of Growth and Orientation in Hamster Cells Trans
formed by Polyoma Virus. Virology, 23: 165-174, 1969.
15. Tanaka, M., Levy, J., Terada, M., Breslow, R., Rifkind, R. A., and Marks,
P. A. Induction of Erythroid Differentiation in Murine Virus Infected
Erythroleukemia Cells by Highly Polar Compounds. Proc. Nati. Acad.
Sei. U. S., 72: 1003-1006, 1975.
16. Todaro, G. J., Green, H., and Goldberg, B. Transformation of Properties
of an Established Cell Line by SV40 and Polyoma Virus. Proc. Nati. Acad.
Sei. U. S., 57: 66-73, 1964.
1977
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1977 American Association for Cancer Research.
3139
D. L. Dexter
Fig. 1. Rhabdomyosarcoma cells in vitro. Photomicrograph of living cells in a confluent culture. Control cells are small, not elongated, and are not
oriented with respect to one another. A number of cells are seen growing on top of the monolayer. Phase optics, x 80.
Fig. 2. DMF-treated rhabdomyosarcoma cells in vitro. Photomicrograph of living cells in a confluent culture. DMF-treated cells are elongated, oriented
with a definite polarity, and do not grow on top of the monolayer. Some treated cells contain more than 1 nucleus. Phase optics. •
80.
3140
CANCER
RESEARCH
VOL. 37
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1977 American Association for Cancer Research.
N,N-Dimethylformamide-induced Morphological Differentiation
and Reduction of Tumorigenicity in Cultured Mouse
Rhabdomyosarcoma Cells
Daniel L. Dexter
Cancer Res 1977;37:3136-3140.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/37/9/3136
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 18, 2017. © 1977 American Association for Cancer Research.