insensitive Prostatic Adenocarcinomas, Dunning

[CANCER RESEARCH 38, 4353-4359,
0008-5472/78/0038-OOOOS02.00
November
1978]
Animal Models of the Hormone-sensitive
Adenocarcinomas,
and -insensitive
Prostatic
Dunning R-3327-H, R-3327-HI, and R-3327-AT
John T. Isaacs, Warren D. W. Heston,2 Robert M. Weissman, and Donald S. Coffey
Department of Pharmacology and Experimental Therapeutics [J. T. I., D. S. C.], James Buchanan Brady Urologica! Institute [R. M. W., D. S. C.], Department
of Urology ¡R.M. W., D. S. C.], and The Oncology Center [D. S. C.¡,Johns Hopkins University School of Medicine, Baltimore, Maryland 21205
Abstract
that is slow growing, well differentiated, and hormone
insensitive; and R-3327-AT, a rapidly growing anaplastic
The Dunning R-3327-H is a well-differentiated transtumor, which is also hormone insensitive. All of these
plantable rat prostatic adenocarcinoma that contains both transplantable prostatic tumors originated as sublines from
hormone-sensitive and -insensitive cells. The component the original Dunning R-3327, which was first described by
composed of hormone-insensitive cells has been permit
W. F. Dunning in 1961 as a spontaneous tumor originating
ted to grow in a castrated male, and a new slow-growing,
in the prostate of an aged, syngeneic Copenhagen rat (4).
well-differentiated hormone-insensitive subline of the tu
The tumors can also be carried s.c. in F, male hybrids of the
mor has been established and designated R-3327-HI. In
Copenhagen male and Fischer female cross. The original
addition, a rapidly growing hormone-insensitive anaplastumor is hormone sensitive and metabolizes testosterone to
tic tumor has been developed, R-3327-AT. These three dihydrotestosterone (23). The R-3327-H hormone-sensitive
tumor lines have been characterized, and their histológi tumor has been characterized in detail in relation to growth
ca! and biochemical profiles are compared.
properties, morphology, histochemistry, and therapeutic
responsiveness (18, 19).
Introduction
The treatment for disseminated human prostatic cancer
has remained essentially unchanged for the past 35 years.
Androgen ablation through castration or estrogen therapy
is still the major treatment, even though this therapy pro
vides only a questionable or limited extension of life span
and very few permanent remissions. The almost universally
observed relapse from hormonal therapy is due to the
growth of hormone-insensitive cells. At present, adequate
hormonal therapy is available to regulate the growth of
androgen-sensitive cells; however, only limited insight has
been obtained into the control of hormone-insensitive cells.
The investigation of this complex problem has been
assisted by the availability of several animal models that
appear to be appropriate for the study of prostatic cancer
(4, 18, 19, 23, 24). Selecting a single animal model for
prostatic cancer may not be realistic because the human
cancer counterpart is itself a variable and multifaceted
disease. Prostatic adenocarcinoma in humans is often char
acterized by diversity in relation to pathology, state and
variability of cellular differentiation, uniformity of growth
rate, and differences in therapeutic responsiveness to hor
monal, nonhormonal, and radiation treatment. Indeed,
many of these variations can sometimes-be observed within
one patient. Because of this variability in the types of
human prostatic cancer, it is possible that more than one
animal model may be required to correspond to these
different states of human cancer. This study provides the
comparative characterization of 3 forms of transplantable
rat prostatic adenocarcinomas: R-3327-H, a well-differen
tiated hormone-sensitive tumor; R-3327-HI, a similar tumor
' Presented at the John E. Fogarty International Center Conference on
Hormones and Cancer, March 29 to 31, 1978, Bethesda, Md. This investiga
tion was supported by National Cancer Institute Grant CA15416, Department
of Health, Education and Welfare.
2 Present address: Department of Urology, Medical School, Washington
University, St. Louis, Mo. 63110.
NOVEMBER
Materials and Methods
The Dunning R-3327-H hormone-sensitive prostatic ad
enocarcinoma was originally obtained from the Mason
Research Institute and is carried in the F, hybrids of male
Copenhagen x female Fischer rats. This R-3327-H tumor
gave rise in our laboratory to trie R-3327-HI (hormoneinsensitive) and R-3327-AT (anaplastic tumor) lines.
The animals were housed in cages containing 3 to 5
animals. One Copenhagen breeding male serviced 3 to 4
female rats. With these methods healthy litters with a
survivorship exceeding 90% were obtained once the good
breeding mothers were selected. The F, hybrid adult males
60 to 80 days old carried the tumor lines and were given
s.c. transplants of cell suspensions of the tumors. Once the
tumor had reached 2 cm in diameter, it was excised, minced
with scissors, passed through 2 wire mesh gauze strainers
(1.5 and 0.5 mm), and resuspended in cold Roswell Park
Memorial Institute Tissue Culture Medium 1640. Cell sus
pensions were then made up to a dose of 106 cells/ml. The
resultant cell suspension (1 ml) was then injected s.c. into
the flank of the recipient animal.
After the s.c. injection of tumor cell suspension, tumors
were measured in all animals at various times with microcalipers. Three diameters of the mass were determined, and
the tumor volume was calculated according to the formula
(1 x w x h x 0.5236) descried by Janek ef al. (10). When
the tumor was excised, the actual volume and tumor weight
were measured. Total tissue DNA, RNA, and protein were
measured according to the method described by Coffey ef
al. (1). Knowledge of the tumor volume and weight and the
total DNA/mg tissue, as well as the DMA content/isolated
nucleus, made it possible to determine the total number of
cells per tumor.
The measured tumor volumes along the growth curve of
each tumor were analyzed semilogarithmically. The slope
1978
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1978 American Association for Cancer Research.
4353
J. T. Isaacs et al.
of the semilog plot, determined by statistical linear regres
sion, determined the actual growth rate of each tumor, as
described previously (19).
Tumor biopsies, all normal sex accessory tissue, and the
control tissues of liver and kidney were stained routinely
with hematoxylin and eosin, periodic acid-Schiff reagent,
and Masson trichrome stain.
All enzyme assays were performed on aliquots of whole
homogenates of tissue. Homogenization
was by means of
an all-glass conical homogenizer. Tissue was diluted with
50 mM Tris buffer, pH 7.4, to a final concentration
of 1:10
(w/v). 5«-Reductase was assayed by the method of Moore
and Wilson (14). 3«-and 3/S-hydroxysteroid dehydrogenase
was assayed according to the method of Jacobi and Wilson
(9). 7«-Hydroxylase was assayed by the method of Isaacs
and Coffey.3 Acid phosphatase was assayed by the method
of Royef al. (17). Alkaline phosphatase determinations were
according to the method of Roy (16). The method of Talalay
ef al. (21) was used for ß-glucuronidase determinations,
yGlutamyltransferase
was assayed according to the method
of Tate and Meister (22). Leucine aminopeptidase
was
assayed by the methods of Goldbarg and Rutenburg (5).
Glucose-6-phosphate
dehydrogenase was assayed accord
ing to the method of Taketaand Watanabe (20). Glutathione
reducÃ-ase was assayed by the method of Pinto and Bartley
(15). Catalase was assayed by the method of Luek (12).
LDH4 was assayed according to the method of Kornberg
(11). LDH isozymes were determined
Dietz and Lubrano (3).
by the procedure
of
Results and Discussion
Hormone-sensitive
R-3327-H. The histology of this well-
differentiated
adenocarcinoma
indicates abundant acini
with secretions, and the pathology is almost identical at the
light and electron microscopic
level with that of a welldifferentiated
human prostatic adenocarcinoma
(19). The
tumor epithelial cell plasma membranes contain microvilli
and secretory granules. The cell and nuclei shapes appear
malignant.
Cell kinetic and growth rate studies indicate that 93% of
the inoculated tumor cells grow s.c. in an adult intact male;
the doubling time of the tumor is approximately
20 days.
The absence of androgen (obtained by castration or estro
gen or antiandrogen therapy) reduced the size of the tumor
growth by 84 to 92% (Table 1).
Although the R-3327-H tumor attains a larger size when
grown for 6 months in the presence of androgens, some
limited growth was observed even in the castrate animals.
When the growth of the tumor in the castrate was monitored
and the log of the tumor volume was plotted against time, a
straight line was obtained, which did not extrapolate to the
origin (for details see Ref. 19). These cell kinetic studies
indicated clearly that a specific fraction or clone (8 to 29%)
of the total original tumor cells inoculated were hormoneinsensitive cells that were capable
sence of androgens.
of growing
in the ab
3 J. Isaacs and D. S. Coffey. Characterization of the 3/3 Hydroxy Steroid
7«Hydroxylase Activity of Rat Ventral Prostate, submitted for publication.
' The abbreviation used is: LDH. lactic dehydrogenase.
4354
Hormone-insensitive
R-3327-HI.
When the R-3327-H tu
mor was placed in a castrate animal and grown for 180
days, the small fraction or clone of hormone-insensitive
cells within the original inoculum continued to grow at a
similar tumor-doubling
rate of approximately 25 days (Table
1). The fraction of tumor cells growing in the castrate for
180 days in the absence of hormone yielded a different
histology (compare Fig. 1). This hormone-insensitive
tumor
is composed of small acini possessing secretion, and the
epithelial
cells appear almost cuboidal.
This hormoneinsensitive tumor tissue has been removed, transplanted,
and carried in castrate animals as a new slow-growing
hormone-insensitive
prostatic adenocarcinoma
line termed
R-3327-HI. The histology and growth rate of this tumor has
remained essentially constant through multiple passages.
The importance
of this new line (R-3327-HI) is that it
represents the type of cells that grow following relapse to
hormonal therapy, the type of cells requiring nonhormonal
or special therapeutic considerations.
Like human prostatic
adenocarcinomas,
castration and estrogen therapy will not
cure animals carrying the R-3327-H tumor because of the
small clone of these hormone-insensitive
cells (R-3327-HI)
that continues to grow.
Markland and Lee (13) reported that the Dunning tumor
contained androgen and estrogen cytoplasmic
receptors.
These studies have been extended by Heston ef al. (8), who
have compared the receptors in the hormone-sensitive
(R3327-H) and -insensitive (R-3327-HI) cells. The insensitive
R-3327-HI line has a 68% reduction
in the cytoplasmic
androgen-binding
protein, but the level is still higher than
in the normal dorsal-lateral prostate tissue (8). In contrast,
the cytoplasmic estrogen receptor level does not change
and remains in both tumors almost 10-fold higher than in
the normal prostate. Small levels of the progesterone recep
tor appear in the hormone-insensitive
lines (8).
Anaplastic Tumor (R-3327-AT). A third tumor line, a
hormone-insensitive
anaplastic tumor, has arisen from the
original R-3327-H on a transfer to intact males. The reason
for the transformation
is unknown. This rapidly growing
anaplastic tumor grows equally well in female, intact male,
and castrate male rats and has remained stable through
over 50 passages.
The R-3327-AT tumor does not contain acini structures
but contains sheets of anaplastic cells, and there is no
evidence of secretory activity. The histology of all 3 tumor
lines is compared in Fig. 1.
Comparison of the 3 Types of Prostatic Tumors. The
general properties of the 3 tumor lines are compared in
Table 2. The R-3327-H tumor was thought to arise originally
from the dorsolateral prostate of the Copenhagen rat. The
enzymatic profile of 13 different enzymes is compared for
each of the 3 tumors and for the normal sex accessory
tissues. The activity is expressed on a cellular basis by
normalizing
to a unit amount of DMA (see Table 3). A
comparison of the relative rates, normalized to 1.0 for the
R-3327-H tumor, indicates the similarity of the enzymatic
profile to that of the dorsolateral prostate (Chart 1).
As the hormone-sensitive
tumor (R-3327-H) becomes hor
mone insensitive
(R-3327-HI) or anaplastic
(R-3327-AT),
there is a marked shift in the enzymatic profile, with a
progressive decrease in 5a-reductase, 7a-hydroxylase,
and
CANCER
RESEARCH
VOL. 38
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1978 American Association for Cancer Research.
Rat Prostatic Adenocarcinomas
Table 1
Growth parameters of the R-3327-H tumor
All animals were given identical s.c. inoculations containing 1.5 x 10s viable R-3327-H tumor cells.
statusAnimalIntact
Hormonal
weightAbsolute
treat
ment"None
(9)3.47
±1.48
4.25 ±1.10
0.41 ±0.06
±0.190.31
0.60
cellAbsolute
(10")6.48
tive100
tive100
in
crease in
initial cell
inoculum432
in
oculated
cells
growing93
±2.7
6.40 ±1.7
122
0.65 ±0.21
12
1 .20 ±
168151Total0.320.68
Intact
Androgen (TP)
427
74
99
Intact
Estrogen (DES)
10
43
8
IntactCastrate Antiandrogen
162983Cell
1810134Fold 8045
(flutamide)
±0.07
±0.16
None
8.73 ±3.0no.Rela
Castrate6-mo. Androgen (TP)Tumor 5.26 ±1.69Rela
582%of
" Animals were treated daily as indicated with testosterone
propionate (TP), 20 mg/day; diethylstilbestrol
100 yug/kg/day;
or flutamide,
Table 2
of the transplantable
Dunning
R-3327-H
19.8
(DES),
tumors in rats
R-3327-AT
Well-differentiated
prostatic
adenocarcinoma
Well-differentiated
prostatic
adenocarcinoma
Origin
Spontaneous in 1961
in Copenhagen rat
Histology
Large, well-developed
acini filled with
secretions
Hormone sensitive;
maximal growth in
intact adult males;
heterogeneous,
80% of cells are
hormone sensitive
and 20% are
hormone
insensitive
Slow
15-20 days
Subline developed in
1977 from carrying
R-3327-H in
castrate males
Smaller acini with less
secretions
Biochemical
profile
Enzyme profile of
tumor similar to
dorsal lateral
prostate tissue;
moderate activity of
5a-reductase;
androgen and
estrogen
cytoplasmic
receptors present;
large tumor mass
reduces size of sex
accessory tissue
and elevates serum
acid phosphatase
levels
Anaplastic prostatic
tumor; not a
squamous cell
carcinoma
Subline developed
spontaneously in
1976 from R-3327-H
No acini; sheets of
anaplastic cells
Hormone insensitive;
growth rate equal in
both intact and
castrate males
Hormone insensitive;
growth rate equal in
both intact and
castrate males
Slow
15-20 days
Fast
2 days
Probably <1%
Slow local invasion to
lymph nodes, but
no distant
metastasis detected
Low biochemical
correlation to
normal lobes of rat
prostate; low in 5areductase activity;
androgen receptor
absent; no effect of
tumor mass on sex
accessory tissue;
no increase in
serum acid
phosphatase levels
Moderate levels of 5areductase;
decrease in level of
androgen
cytoplasmic
receptor, but no
decrease in
estrogen receptor;
appearance of
progesterone
receptor; elevated
serum acid
phosphatase levels
alkaline phosphatase activity. In contrast, other enzymes
increase, such as 3a-hydroxysteroid dehydrogenase and
LDH(Chart2).
The increase in LDH was resolved into the 5 isozyme
NOVEMBER
prostatic
R-3327-HI
Type
Growth rate
Tumor
doubling
time
Metastatic
rate
18.7
20.2
20.324.8
50 mg/kg/day.
Properties
Androgen
sensitivity
dou
bling
time
(days)20.9
patterns (I to V). This was of interest because Denise?al. (2)
reported a relative increase in the LDH V isozyme of pros
tatic carcinoma but not for benign prostatic hyperplasia.
Goldman ef al. (6) have also shown a preferential increase
1978
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1978 American Association for Cancer Research.
4355
J. T. Isaacs et al.
o
coÕJLU1-
'S"51c5
n^
r> n .o .Q jD
n r¡o
CM6
CDco m
r-.CM
't in tco+1
co in CMCMo>
ocoocoo
co
t Tco •¿*
co o
»-
+1 +1 +1 +1 +1
+1 +1 -H
+1 +1 +1
—¿^+1o>co
m"pp
roCE
jjX
inT^
^ h»rCDco o
co o r^ CMco
K co o
incocMcoCO
CMin
•¿*COCM
co r- LOin
T-
O)T—
OÃCD
cooj^i"coolem
t—
in o>co
co
cococomCMCMCO
T- CMo co co
iî
ÄCM
r^- o
COin**E3Hoa<o15
°>coE
»-»-CO+1
T-
+1h-.
+1+1+1
T-
+1+1+1+1+1
cco£
o 0)
so3O).gCCtissue
CMI
^j—
—¿
ino
¿-.ER-3327-H
+1+1+1
cqi
p u> hh»^ ^Ã-"o co
^
co
co
inT-o) o
*"? 'S T
T—
T—
r^
co ^ CM
CMO)
coinooo
in co
cooicoo)CO
ÓCOÒCMÓ
00
/xg
sensitive)coCD
00*~CD CD
M
T(hormone
Du^o%CDOoCO«"¡5O.g1•CSa1^§"o|C
and Dti."coo~•
£
-H +1 +1 +1
Np
B
+1 +1 +1 +1 II
N1
e o> 10 in o>
inocMcoT-T-O)CO
r^cocNO
* CDi^-coco01!
CDCD
T-
T-
O
'.C
t^_^>
O5
T-
r^^t
w
CD
E
II
c
*
O
Ä
»—°
"~
II
II
II
m.
CD CD ^
o
'S'S.
•¿j1
5T-
•¿*T-
-*±+1
h^ in o
—¿1CD
co
i— T-
3*ra
«o
IICM
+1 II II
II II II II II
T-c\icoino
O)
cocMcom
OCOCMP^C\l
oS
CNJCM
r^cooicoi^
m
in
CM
°-0«3~
3
CO
COh-CD3
II II II
^
ocoOiTI^o*o"QO)0OCi0ECxiZQO)a.8»—f-"5Eo.coCO'E3
'S03
E0zCDE
O
>Q.«
in oo co
oocoo»m^coo CM
mf-T-inco
mincM^r^>
or^T—
oo ~
^ o
o
T-IIin
in
IIOti II II
II II II II II
COCNJ
cocococoo
o co o
r*-coc\j^tO
O)CM
i- 0
inSh~
cocom^tmr^-co
in
n8.EoO^*•^ate/1
is'E
°E
ioCD
•¿*CMII
co i- m
II II II
(D tCM
co
W >CyLUcocococo
co^ T—m o
^ CD o
inCO
O^~C01co
CO CM
O)
00 CO O)
om^^fco
T—
T—
T-
CD
" L
T}-
II II II
in T—
m
CO inO
•¿<Õ^Õ;T^
OO>T—mo
in
T—
t—T—
r*T- T-
II II IIin
II II II
II II II
in
co
co T—co co in
O)
rCM
CO
CMT—CM
*~ r—CM
T-
*Oco
co
^*e-00*-co
t33m
T3
CDSS«
TJ CO
-S1"•0-Dx1
0.3ili!0 '6
0
$
(D
CO
CO
coco^
CD'S
» S"
CO^CD^g^
I!^^^2
Stili
»rr
O^jzinnnr^
^ ^**ri
4356
T3
•¿â€¢
â„¢¿0
CO>,NCLUo
•¿^coâ„¢c
S le
< < cb. ?. ®
0 <5O 3ZQ|C
—¿3 "3 '^
CANCER
RESEARCH
VOL. 38
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1978 American Association for Cancer Research.
flaf Prostatic Adenocarcinomas
30ProstateTiO)-Lateral
VesicleVen
-i!
?o
ACTIVITY^
T
3l~J
o-iK.2.li!
i
^
J
2000-2O
fro)
Pros
teDorraito
1000-oC
0O-
o-1 >
«¡ELATIVE
1 -
Prostate¿^\C¿f
Ventral
277-1T31010Klarise
1T
0O
R-3327'H
ompared
tunl\3
to 4W>C0
ENZYME
ACTIV
^2000-N
1000T3'xo00
2000327-M
Ad^n1•o0Co.0
IVITY 1
1000-*Uisal-la i
-Chart
~
n
¡
DinOÃ-o:b2omp
•¿oaVa
r^-p"Tumor
R-3327-H
Hormone
SensitiveTumor
<Tumor
2000-1—
K
Lateral
Prostotea
1 T
R-3327-H
Hormone
Insensitivel—
1000-13000-Q_)
Q.0.C"5mati
0m.:
0XO
oaCenz>
oC
R-3327- AT
0aiC«f
Q,i0udor
AnaplastiaTi
T3üns
jofie
2000-teral
1. <
and
o thè
prò
c thè
ventral pros ate t 3 the R-33 27-H umo r.AII activ ties aree)¡près
sedi s relative
-H tu-L.iSo
values comf lared toll- e act ivitie softînastoDC.
he R 3327iransferose0
mor,R-3
thelQevelsO3.
of whichSemino/
are set at 1.(Dorsal
l.detL*l¡yd.,T0)•o•D
dehy-ÃŽ-B-I111I1T2000-1000-R-3327-H
drogfiiDSa ».io•DC
3oi- hydroxy sferoid den/cj
30r
1000
n
mET
V
LDH ISOZYMES
Chart 3. LDH isozyme patterns. The relative migration rate is in decreas
ing order from LDH I to LDH V.
in LDH V ¡sozymesfor many different malignant tissues in
comparison to benign tumors. Grayhack ef a/. (7) have
presented evidence that the prostatic cancer causes an
increase in the LDH V:LDH I ratio above 2 in expressed
prostatic fluid but not from benign prostatic hyperplasia. It
was of interest to compare the LDH isozyme pattern. There
was a marked increase in the LDH V isozyme patterns in all
of the tumors (Chart 3), and isozyme V increased with
increasing levels of total LDH. The meaning of this shift in
isozyme pattern has not been resolved.
In summary, a spectrum of animal models for prostatic
cancer are now available and have been characterized. The
Dunning R-3327 series mimics many of the properties of
human prostatic cancer including initial response to hor
monal therapy followed by relapse to a hormone-insensitive
state. It is anticipated that these animal models will provide
new insight into the understanding of prostatic cancer. The
ideal animal models for prostatic cancer do not exist; they
would need to fulfill the 16 criteria listed in Table 4. The
Chart 2. Comparisons of the enzymatic activities in the various Dunning
tumor lines. Values are relative to the activities of the R-3327-H tumor, the
levels of which are assigned a value of 1.0. dehyd., dehydrogenase.
NOVEMBER
1978
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1978 American Association for Cancer Research.
4357
J. T. Isaacs et al.
Table 4
Idealized properties of animal models of prostatic cancer
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Spontaneous origin
Development in aged animals
Proven origin from prostatic tissue
Adenocarcinoma
Slow-growing tumor
Histological similarity to human prostatic cancer, with diversity
of form
Biochemical profile similar to normal prostate and human
prostatic cancer
Malignant and metastatic patterns to bone and lymph nodes
Elevation of serum prostatic acid phosphatase levels
Hormone-sensitive, requiring presence of androgen for maxi
mal growth
Capability to respond to castration and estrogen therapy fol
lowed by relapse to hormone-insensitive state
Immunoloical parameters
A. Developed in syngeneic animals
B. Transplantable or easily induced
C. Containing tumor-specific antigens
Availability of large numbers of animals for statistical consid
erations
Similar therapeutic response to human prostatic cancer; cor
relates with past clinical experience with:
A. Hormonal therapy
B. Nonhormonal therapy
C. Radiation sensitivity
Wide diversity of tumor types (differentiation, hormonal re
sponse, etc.) corresponding to human tumor variability
Ability to predict accurately response to future modalities of
human therapy
Dunning R-3327 models approach most of the points in
Table 4 (Points 1 through 13), with the exception of Point 8.
The limitation of the Dunning tumors is the low incidence
of metastatis, and this may be due to the s.c. sites of the
tumor. At present, it is being determined whether these
models predict past clinical experiences to existing thera
peutic modalities.
The ultimate test will concern the predictability and ac
curate testing of new therapeutic approaches. This study
contributes to the final point of providing a wide spectrum
of characterized animal models for the study of prostatic
cancer.
In conclusion, animal models are not sufficient within
themselves to assess the full clinical importance of a future
therapeutic approach; however, well-characterized animal
models do provide a unique opportunity to develop new
concepts that are most difficult to obtain through available
clinical studies and within reasonable cost, time, and ethi
cal restraints. Appropriate animal studies and careful clini
cal trials must be done in concert to assist the search for
new and more effective management of prostatic cancer.
Acknowledgments
The authors wish to express their gratitude for the excellent professional
4358
help given by Bobby J. Trotter, William Isaacs, and Diane Gomolka, without
whose help this project could not have been accomplished.
References
1. Coffey, D.S., Shimazaki. J., and Williams-Ashman, H.G. Polymerization
of Deoxyribonucleotides in Relation to Androgen-induced Prostatic
Growth. Arch. Biochem., 724: 184-198, 1968.
2. Denis, L., Prout, G., Van Camp, K., and VanSande, M. Electrophoretic
Characterization of the Prostate: 1. Proteins and Lactic Dehydrogenase
in Benign Hyperplasia and Carcinoma. J. Urol., 88: 77-85, 1962.
3. Dietz, A., and Lubrano, T. Separation and Quantitäten of Lactic Dehy
drogenase Isoenzymes by Disc Electrophoresis. Anal. Biochem., 20:
246-257, 1967.
4. Dunning, W. F. Prostatic Cancer in the Rat. Nati. Cancer Inst. Mono
graph, 72: 351-369, 1963
5. Goldbarg, G., and Rutenburg, A. The Colorimetrie Determination of
Leucine Aminopeptidase in Urine and Serum of Normal Subjects and
Patients with Cancer and Other Diseases. Cancer, 77: 283-294, 1958.
6. Goldman. R., Kaplan, N., and Hall, T. C. Lactic Dehydrogenase in
Human Neoplastic Tissues. Cancer Res., 24: 389-399, 1964.
7. Grayhack, J., Wendel, E., Lee, C., and Oliver, L. Analysis of Prostatic
Fluid In Prostatic Disease. Cancer Treat. Rept., 67: 205-210, 1977.
8. Heston. W. D. W., Menon, M., Tananis, C., and Walsh, P. Androgen,
Estrogen and Progesterone Receptors of the R-3327-H Copenhagen Rat
Prostatic Tumor. Cancer Letters, in press, 1978.
9. Jacobi, G., and Wilson. J. D. The Formation of 5u-androstane-3u,17/3diol by Dog Prostate. Endocrinology. 99: 602-610, 1976.
10. Janek, P., Briand, P., and Hartmann, N. R. The Effect of EstroneProgesterone Treatment on Cell Proliferation Kinetics of Hormone-de
pendent GR Mouse Mammary Tumors. Cancer Res., 35: 3698-3704,
1975.
11. Kornberg, A. Lactic Dehydrogenase of Muscle. Methods Enzymol., 7:
441^443.
12. Luek, H. Catalase. In: H. Bergmeyer (ed.), Methods of Enzymatic
Analysis, pp. 885-894. Weinheim, Germany: Verlag Chemie, GmbH,
1963.
13. Markland, F.. and Lee, L. Estrogen Receptor Characterization of the R3327 Transplantable Prostatic Adenocarcinoma. Federation Proc.. 36:
913, 1977.
14. Moore, R. J., and Wilson, J. D. Reduced Nicotinamide Adenine, Dinucleotide Phosphate: V-3-ketosteroid 5u Oxidoreductase (Rat Ventral Pros
tate). Methods Enzymol., 36: 466-474, 1975.
15. Pinto. R., and Bartley, W. The Effect of Age and Sex on Glutathione
ReducÃ-aseand Glutathione Peroxidase Activities and on Aerobic Gluta
thione Oxidation in Rat Liver Homogenates. Biochem. J., 772: 109-115,
1969.
16. Roy, A. Rapid Method for Determining Alkaline Phosphatase Activity in
Serum with Thymolphthalein Monophosphate. Clin. Chem., 76: 431436, 1970.
17. Roy, A., Brower, M., and Hayden, J. E. Sodium Thymolphthalein Monophosphate: A New Acid Phosphatase Substrate with Greater Specificity
for Prostatic Enzyme in Serum. Clin. Chem., 77: 1093-1102,1971.
18. Smolev, J., Coffey. D.. and Scott. W. Experimental Models for the Study
of Prostatic Adenocarcinoma. J. Urol., 778: 216-220, 1977.
19. Smolev, J., Heston, W., Scott, W., and Coffey, D. Characterization of the
Dunning R-3327-H Prostatic Adenocarcinoma: An Appropriate Animal
Model for Prostatic Cancer. Cancer Treat. Rept., 67: 273-287, 1977.
20. Taketa, K., and Watanabe, A. Interconvertible Microheterogeneity of
Glucose 6-phosphate Dehydrogenase in Rat Liver. Biochim. Biophys.
Acta,235:19-26, 1971.
21. Talalay, P., Fishman, W., and Huggins, C. Chromogenic Substances. II.
Phenolphthalein Glucuronic Acid as Substrate for the Assay of Glucuronidase Activity. J. Biol. Chem., 766: 757-772, 1946.
22. Täte,S., and Meister. A. Interaction of Gammaglutamyl Amino Acids,
Dipeptides, and Derivatives and Analogs of Glutathione. J. Biol. Chem.,
249:7593-7602. 1974.
23. Voigt, W., and Dunning, W. F.ln Vivo Metabolism of Testosterone-3H in
R-3327, an Androgen-sensitive Rat Prostatic Adenocarcinoma. Cancer
Res..34: 1447-1450, 1974.
24. Voigt, W., Feldman, M., and Dunning, W. F. 5a-Dihydrotestosteronebinding Proteins and Androgen Sensitivity in Prostatic Cancers of
Copenhagen Rats. Cancer Res., 35. 1840-1846. 1975.
CANCER
RESEARCH
VOL. 38
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1978 American Association for Cancer Research.
Rat Prostatic
Adenocarcinomas
«*Wf
"*>
>'%t
-â„¢
*3f
v
I ^A; %
'•%^*-*
V^****>
*¿v«w¥
•¿â€¢
" ^SfW
-J*L.1+
vOWflA"^
•¿'•¿
'
•¿
-
, -. .,4';,?-
|f-?^
:v..
wl!
-AV?,
v„
i•¿
b^*&$$ii
' ' •¿â€
.^•v^tî,:,V«EL
S£S& --. V
'
\
Fig. 1. Comparative histology. A, normal dorsal-lateral prostate; 6, R-3327-H;C, R-3327-HI;D. R-3327-AT. x 410.
NOVEMBER 1978
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1978 American Association for Cancer Research.
4359
Animal Models of the Hormone-sensitive and -insensitive
Prostatic Adenocarcinomas, Dunning R-3327-H, R-3327-HI, and
R-3327-AT
John T. Isaacs, Warren D. W. Heston, Robert M. Weissman, et al.
Cancer Res 1978;38:4353-4359.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/38/11_Part_2/4353
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. © 1978 American Association for Cancer Research.