JOURNAL OF VIROLOGY, Oct. 1968, p. 1115-1121
Copyright © 1968 American Society for Microbiology

Vol. 2, No. 10
Prit.ted in U.S.A.

Biophysical, Biological, and Cytochemical Features
of a Virus Associated with Transplantable

Hamster Tumors1
W. A. STENBACK, G. L. VAN HOOSIER, JR., AND J. J. TRENTIN

Division of Experimenztal Biology, Baylor Unziversity College of Medicinie, Houstoni, Texas 77025

Received for publication 3 June 1968

A virus resembling type C murine leukemia viruses, which is associated with
transplantable hamster tumors, was partially characterized with respect to certain
biological, biophysical, and cytochemical features. As determined by electron
microscopy, high concentrations of the virus appeared in the blood plasmas of
tumor-bearing hamsters. Hamsters inoculated with virus concentrates did not show
gross evidence of disease, and preliminary attempts to infect various hamster cells in
tissue culture were unsuccessful. A line of cells from a virus-containing tumor which
had been established in tissue culture released large numbers of the virus into the
supernatant fluids by budding. The buoyant density peak of virus concentrates in
potassium tartrate density gradients was 1.13 g/cm3 by ultraviolet absorption and
electron microscopic analysis. Acridine orange staining and nuclease digestion
methods have indicated that the virus is probably a ribonucleic acid virus.

Previously, we reported the presence of virus
particles in several transplantable hamster tumors
as revealed by electron microscopy (12). One of
the two types of particles found resembles type C
murine leukemia virus, the first report of such an
agent in hamsters. One of our transplantable
tumors (D-9), a malignant lymphoma, con-
sistently contained many budding type C particles,
as revealed by electron microscopy, and a second
type particle which appears to be morphologically
identical with the particle first described by
Bernhard and Tournier (2) and recently studied
by Thomas et al. (13). Blood plasma pellets from
animals bearing this tumor contain large numbers
of mature and immature type C particles but no
"Bernhard" type particles. The present report is
concerned primarily with studies of the type C
virus.

MATERIALS AND METHODS

In vivo virus. Groups of twenty-five 6- to 8-week-old
hamsters received transplants of about 1 mm3 of
tumor subcutaneously by trocar. This tumor, a malig-
nant lymphoma, grows rapidly and usually kills
within 12 to 14 days. At about 12 days, the bloods of
the moribund animals were withdrawn by cardiac
puncture into syringes rinsed with 1,000 units of
heparin per ml, and were pooled in a chilled container
in an ice bath. The blood cells were removed by low-
speed centrifugation, and the plasma was stored at 4 C

1 Presented in part before the American Association
for Cancer Research, Atlantic City, N.J., April 1968.

until concentration of the virus was performed (within
1 day).

Tissue-culture virus. Cells obtained by trypsinization
from a freshly excised D-9 tumor were placed in
Eagle's minimal essential medium supplemented with
10%c fetal calf serum in 32-oz (0.946 liter) bottles and
were incubated under 5%7 CO2 at 37 C. After approxi-
mately 1 month, with minimal disturbance and feeding
only once a week, a luxuriant growth of cells was evi-
dent. These cells were subsequently serially passed at
approximately 10- to 12-day intervals. Thin sections
of cell pellets and pellets of supernatant fluids showed
large numbers of mature and immature type C par-
ticles, many in the process of budding from the cyto-
plasmic membranes. The "Bernhard" type particles,
present in the original tumor, were not evident in the
cultured cells.

Virus concentrationz. Virus-rich hamster plasmas or
supernatant tissue culture fluids were centrifuged at
10,000 X g for 2 min and were then filtered through a
thin layer of Celite (10). The virus was then pelleted
from the clarified fluids at 30,000 rev/min for 1 hr in
the no. 50 rotor of the Spinco model L ultracentrifuge
or was collected on a cushion of potassium tartrate
(density, 1.25 g/cm3) in the SW 25.1 rotor at 25,000
rev/min for 1 hr; the supernatant fluids were dis-
carded.

Density gradient centrifugation. Virus pellets were
gently resuspended in about 0.5 ml of single strength
phosphate-buffered saline (PBS) with a Potter-Elve-
hjem homogenizer (8). The material was layered over
linear(1.04 to 1.25 g/ml) potassium tartrate gradients
(6) preformed with the aid of a layering device (4),
and was centrifuged at 37,500 rev/min for 4 hr in the
SW-39 rotor of a Spinco model L ultracentrifuge. The

1115

 o
n
 A

p
ril 5

, 2
0
2
1
 a

t C
A

R
N

E
G

IE
 M

E
L
L
O

N
 U

N
IV

 L
IB

R
h
ttp

://jvi.a
sm

.o
rg

/
D

o
w

n
lo

a
d
e
d
 fro

m
 

http://jvi.asm.org/


STENBACK, VAN HOOSIER, AND TRENTIN

gradient was then sampled either by collecting four-
drop fractions after piercing the bottom of the tube or,
for ultraviolet (UV) analysis, by utilizing a top-un-
loading piercing unit (Buchler Instruments, Inc., Fort
Lee, N.J.). For further purification, the fluid above the
opalescent band near the center of the gradient was
removed into a syringe equipped with a 2-inch, 19-
gauge needle bent at right angles. The band was then
similarly removed with a fresh syringe and was sub-
mitted to a second cycle in a potassium tartrate gradi-
ent after dialysis against single strength PBS. Densities
were determined by weighing 20-/Aliter samples (I 1) or
by measurement of the refractive index (8).

Electronz microscopy. Low-speed cell pellets and
high-speed virus pellets were fixed in 3%6 phosphate-
buffered glutaraldehyde for 1 hr or, if necessary, were
stored in this solution for several days at 4 C before
processing as previously described (12). Briefly, the
pellets were postfixed in osmium tetroxide, dehy-
drated in graded alcohols, and embedded in Araldite
resin. Thin sections were double-stained with uranyl
acetate and lead citrate and were examined in the
Siemens Elmiskop IA electron microscope.

Cytochemical staininlg technique. The virus band
from the second cycle of density gradient cen-
trifugation was dialyzed against single strength PBS
and then was pelleted at 30,000 rev/min for I hr.
Small droplets of the concentrated virus were placed
on the center of small cover slips and were allowed to
dry thoroughly in air. After fixation in Carnoy's fixa-
tive, acridine orange fluorochrome staining and
nuclease digestion procedures were performed accord-
ing to the methods of Mayor (5).

Fluorescence microscopy. The Leitz Ortholux
fluorescence microscope equipped for dark-field
illumination was used. For observation of acridine
orange-stained virus materials, the BG-38 (heat) and
BG-12 (bluLe) filters and the OG-1 (orange) eye piece
barrier filter were employed.

RESULTS

Lack of serological cross-reactivity with known
murine viruses. To help rule out the possibility
that the agent might be a murine leukemia virus
inadvertently introduced into our hamster colony,
3 frozen, virus-rich tumor samples and 30 sera
from tumor-bearing hamsters and 60 sera from
normal control hamsters were sent to the labora-
tories of Janet Hartley and W. Rowe, National
Institute of Allergy and Infectious Diseases, to
whom we are indebted for testing these materials.
No specific cross-reactivity was found in the
complement-fixing (CF) test with their murine
leukemia group-reactive rat tumor antigen and
antiserum reagents (3).

Attempts to demonstrate biological activity.
High-speed pellets from the blood plasma of D-9
tumor-bearing hamsters contained large numbers
of type C virus particles as revealed by electron
microscopic examination of thin sections (Fig. 1).
These particles consist of an electron-dense

nucleoid surrounded by a loose membranous
envelope. Virus samples concentrated on a
cushion of potassium tartrate and dialyzed over-
night against PBS were inoculated into the follow-
ing hamster cells in tissue culture: hamster
embryo, hamster mixed spleen-thymus cells,
adeno-12 induced tumor cells, and BHK-21 cells.
No cytopathic effects (CPE) were seen, and thin
sections of cell pellets examined by electron
microscopy showed no evidence of type C viral
replication, although the "Bernhard" type parti-
cles were observed in the BHK-21 cells, as re-
ported by others (2, 13), as well as in the spleen-
thymus cells and adeno-12 tumor cells. Super-
natant fluid pellets from the above cells contained
type C particles only on the first medium change,
most probably representing the inoculum.
Newborn and weanling hamsters were inocu-

lated with cell-free virus concentrated from
tumor-bearing hamster plasma. Tumors were
found in two animals; because of the low in-
cidence and advanced age of the animals at the
time of autopsy, the relationship of the tumors
to the inoculum is uncertain. Membranous
glomerulonephritis was also observed in the
older animals coming to autopsy. The possible
etiological relationship between the virus and the
kidney lesions is under investigation.

Virus production in tissue culture. Immature
type C virus particles were seen budding from
cytoplasmic membranes of the D-9 tumor cells in
tissue culture (Fig. 2). These immature particles
were observed at a frequency of about 3 to 8 per
standard 200 mesh electron microscope grid
square, and the frequency remained the same
between 2 and 8 days in culture. When 105 tissue-
cultured cells were inoculated into weanling
hamsters, tumors were produced in four of four
hamsters within 50 days; these tumors also con-
tained budding type C virus particles in thin sec-
tions at a frequency of 0 to 6 per 200 mesh grid
square. The "Bernhard" type particles were not
evident in the D-9 cells in tissue culture nor in the
tumors that we obtained from these cells. Pellets
of supernatant fluids from cell cultures contained
many type C particles in thin sections (Fig. 3).

Equilibrium density gradient centrifugation. In
potassium tartrate gradient centrifugation of
virus-containing materials, two light-scattering
bands were produced (Fig. 4A). The upper band
was a diffuse, opalescent band with a 260-nm
absorption peak at a density of 1.13 g/cm3,
whereas the lower band was flocculent with a UV
absorption peak at a density of 1.19 g/cm3 (Fig.
5). Control materials contained only the lower
band (Fig. 4B). When the two bands were col-
lected and pelleted for thin sectioning after

1116 J. VIROL.

 o
n
 A

p
ril 5

, 2
0
2
1
 a

t C
A

R
N

E
G

IE
 M

E
L
L
O

N
 U

N
IV

 L
IB

R
h
ttp

://jvi.a
sm

.o
rg

/
D

o
w

n
lo

a
d
e
d
 fro

m
 

http://jvi.asm.org/


FIG. 1. Tlhin section throughl a high-speed pellet of pooled blood plasmas from 25 tumor-bearing hamsters.
Many type C virus particles may be seen throughlout the section. The bar in all micrographs represents 100 nm.

** s !4 *j * .e

X' AS*' .: szr er
A

..r. ~~~ .
_i . t * ._ t w w1'
. :4 4* w~~~~~~~

s \ X, #ts 4'

s eS . .A

I ei

FIG. 2. Immature type C virus particle budding from the cytoplasmic membrane of a D-9 tumor cell in tissue
culture.

1117

.''l'

;
" . .
4

jf!6.7V.:+ . : ...

*L .: :

-,, J

 o
n
 A

p
ril 5

, 2
0
2
1
 a

t C
A

R
N

E
G

IE
 M

E
L
L
O

N
 U

N
IV

 L
IB

R
h
ttp

://jvi.a
sm

.o
rg

/
D

o
w

n
lo

a
d
e
d
 fro

m
 

http://jvi.asm.org/


STENBACK, VAN HOOSIER, AND TRENTIN

FIG. 3. Thini section through a high-speed pellet of 6-day supernataiit fluids from D-9 tumor cells growing in
tissue culture.

1118 {l. VIROL.

 o
n
 A

p
ril 5

, 2
0
2
1
 a

t C
A

R
N

E
G

IE
 M

E
L
L
O

N
 U

N
IV

 L
IB

R
h
ttp

://jvi.a
sm

.o
rg

/
D

o
w

n
lo

a
d
e
d
 fro

m
 

http://jvi.asm.org/


VOL. 2, 1968 VIRUS ASSOCIATED WITH TRANSPLANTABLE HAMSTER TUMORS

dialysis against single strength PBS, the upper
band (density, 1.13 g'cm3) was found to contain
type C particles in large numbers (Fig. 6). No
virus particles were associated with the lower l6
band, and the nature of its amorphous con- 6
stituents is unknown. When the upper band was
collected and submitted to a second cycle of
density gradient centrifugation, a lower flocculent
band was still produced but to a lesser degree.

Acridine orange staining. When fixed and
stained with acridine orange, the partially purified
virus concentrate appeared brilliant orange-red .4
under ihe fluorescence microscope. Prior treat-
ment with ribonuclease prevented the red fluo-
rescence, whereas treatment with deoxyribonu- ECD
clease or buffer alone had no effect. These results
are consistent with a single-stranded ribonucleic
acid (RNA)-containing virus (5). a

DISCUSSION 2

Lack of a system for assaying biological activity
has limited the extent to which the characteristics
of this virus, which is associated with trans-
plantable hamster tumors, can be studied.

0 I I
Density 1.13 1.19

FIG. 5. Photographic reproductionz ofa tracing made
duiring UV absorptionz anialysis of a potassium tartrate
dentsity gradient runz similar to that shIown2 in Fig. 4A.
The initial virus pellet was derived from tumor-bearing
hamster plasma. The peak at the 1.13 g/cm3 denisity
level conitainied the viruis. The 1.19 g/cm3 density peak
consisted ofamorphouis material devoid of virus.

Nevertheless, some important characteristics
have been determined by physical and cytochem-
ical means. These characteristics include: (i)
possession of a loose sac-like envelope, (ii)
possession of high water content as evidenced by
morphological alteration in concentrated potas-
sium tartrate, (iii) particle formation by budding
from the plasma membrane with a lack of CPE,
and (iv) cytochemical evidence that the nucleic
acid of this virus is RNA. All of these charac-
teristics are common to oncogenic RNA viruses

A B_ (1). The buoyant density of 1.13 g/cm3 may be
slightly less than that of known leukemia viruses
(8); however, this density was determined by

FIG. 4. Distributiont of light-scatterinzg zontes of physical means rather than by assay of biological
virus-conttaininlg anid conitrol materials int linlear potas- activity. Possible destruction of biological ac-
sium tartrate density graidientts. Resuispended viruis pel- tivity by the potassium tartrate has not been
lets were layered over preformed gradientts anid were
centtrifuiged at 150,000 X g for 4 hr in the SW-39 ruled out; however, there iS no reason to assume
rotoofa Beckman Spinco ultracentrifue. () Vu. that brief exposure of this virus to the salt shouldrotor of a Beckmaii Spitico lultraceittrifltge. (A) VirUtsbeaymrdtienlthnscexoues

band, derived from tissute ctultuire fluiids of D-9 tutmor be any more detrimental than such exposure IS
cells, is visible nzear the middle of the tutbe. (B) Conitrol to other similar viruses (6, 8).
fluids from hamster embryo cells were devoid of virius. There is no evidence that murine agents

1119

J. .. .__ .1 .- J _.,

 o
n
 A

p
ril 5

, 2
0
2
1
 a

t C
A

R
N

E
G

IE
 M

E
L
L
O

N
 U

N
IV

 L
IB

R
h
ttp

://jvi.a
sm

.o
rg

/
D

o
w

n
lo

a
d
e
d
 fro

m
 

http://jvi.asm.org/


STIENBACK, VAN HOOSIER, AND TRENTIN

.j: .! t$.* * . *e ii
,. 4t41 ^,

.9gY' se* :' i L
.. S ,... SF. ve ..... ,

';.e, t' +:.

's-

.~~~ ~ ~

frF ...

tiv= i4

_S|-F-

FIG. 6. Thlini sectioln through a high-speed pellet ofa 1.13 g/cm9 density band isolated from a potassiuim tartrate
denisity gradielt. The pellet conisisted almost exclusively of type C virus particles. Most of the particles were dis-
torted as a resultt of osmotic effects of the contcenztrated potassium tartrate.

replicate in hamsters. It has been reported that
inoculation of the Moloney agent into hamsters
produces a reticulum cell sarcoma (7); however,
to the best of our knowledge, the agent was never
recovered from the tumor or from subsequent
transplants. The lack of CF cross-reactivity with
murine agents in the present study suggests that
the virus is a new hamster virus. The fact that the
hamster harbors a virus which resembles murine
leukemia agents is important in light of the
extensive role played by the hamster in cancer
research. The possibility has not been ruled out
that the agent is merely a passenger in these
tumors, but if the virus should prove to be
causally related it would provide a useful new
laboratory model for study.

In studies currently in progress, we are attempt-
ing to produce specific antisera for further
serological studies, to apply radioisotope tech-
niques to confirm the nucleic acid type, and to
disclose biological activity which might be em-
ployed for viral assay.

ACKNOWLEDGMENTS

We thank Judith Marston, Department of Micro-
biology, Baylor University College of Medicine, for

her time and the use of her equipment in our ultra-
violet analyses. We also thank Judy Adams, Carolyn
Gist, Bernie Powell, Karen Rhew, Irene S. Cox, and
H. Koo for excellent technical assistance.

This investigation was supported by American
Cancer Society Grant IN 27-H-Project 2, and by
Public Health Service research grants CA-06941,
CA-05021, and K6 CA-14,219 from the National
Cancer Institute.

LITERATURE CITED

1. Anonymous. 1966. Suggestions for the classifica-
tion of oncogenic RNA viruses. J. Natl. Cancer
Inst. 37:395-397.

2. Bernhard, W., and P. Tournier. 1964. Infection
virale inapparente de cellules de hamsters
d6cel6e par la microscopie electronique. Ann.
Inst. Pasteur 107:447-452.

3. Hartley, J. W., W. P. Rowe, W. I. Capps, and
R. J. Huebner. 1965. Complement fixation and
tissue culture assays for mouse leukemia viruses.
Proc. Natl. Acad. Sci. U.S. 53:931-938.

4. Martin, R. G., and B. N. Ames. 1961. A method
for determining the sedimentation behavior of
enzymes: application to protein mixtures. J.
Biol. Chem. 236:1372-1379.

5. Mayor, H. D. 1962. Biophysical studies on viruses

e.
4R*.

1120 J. VIROL.

"L.A:

4'o,
ll. IP.- .:,i

1%, A

W.

., I..'.
11 N."...4.4.11111111111k ..:

 o
n
 A

p
ril 5

, 2
0
2
1
 a

t C
A

R
N

E
G

IE
 M

E
L
L
O

N
 U

N
IV

 L
IB

R
h
ttp

://jvi.a
sm

.o
rg

/
D

o
w

n
lo

a
d
e
d
 fro

m
 

http://jvi.asm.org/


VOL. 2,1968 VIRUS ASSOCIATED WITH TRANSPLANTABLE HAMSTER TUMORS

using the fluorochrome acridine orange. Progr.
Med. Virol. 4:70-86.

6. McCrea. J. F., R. S. Epstein, and W. H. Barry.
1961. Use of potassium tartrate for equilibrium
density-gradient centrifugation of animal vi-
ruses. Nature 189:220-221.

7. Moloney, J. B. 1962. The murine leukemias. Fed-
eration Proc. 21:19-31.

8. O'Connor, T. E., F. J. Rauscher, and R. F. Zeigel.
1964. Density gradient centrifugation of a
murine leukemia virus. Science 144:1 144-1147.

9. Potter, V., and C. A. EIvehjen. 1936. A modified
method for the study of tissue oxidations. J.
Biol. Chem. 114:495-504.

10. Smith, K. O., and J. L. Melnick. 1964. Adeno-
virus-like particles from cancers induced by
adenovirus-12 but free of infectious virus.
Science 145:1190-1192.

11. Stenback, W. A., and D. P. Durand. 1963. Host
influence on the density of Newcastle disease
virus (NDV). Virology 20:545-551.

12. Stenback, W. A., G. L. Van Hoosier, Jr., and
J. J. Trentin. 1966. Virus particles in hamster
tumors as revealed by e'ectron microscopy.
Proc. Soc. Exptl. Biol. Med. 122:1219-1223.

13. Thomas, J. A., E. Delain, and R. Hollande. 1967.
Morphogenese d'un virus du hamster associe a
la souche BHK ou a des tumeurs. Compt. Rend.
264:785-788.

1121

 o
n
 A

p
ril 5

, 2
0
2
1
 a

t C
A

R
N

E
G

IE
 M

E
L
L
O

N
 U

N
IV

 L
IB

R
h
ttp

://jvi.a
sm

.o
rg

/
D

o
w

n
lo

a
d
e
d
 fro

m
 

http://jvi.asm.org/