[CANCER RESEARCH 47, 5112-5115, October 1, 1987]
Failure to Transmit Diethylnitrosamine Tumorigenicity from Transplacentally
Exposed FI Generation Syrian Hamsters to the Respiratory Tract of F2 and F3
Generations
H. Ernst,1 M. Emura, B. Bellmann, D. Seinsch, and U. Mohr
InstituÃ-fürExperimentelle Pathologie, Medizinische Hochschule Hannover, Konstanty Gutschow Str. S, 3000 Hannover 61, Federal Republic of Germany
ABSTRACT
successive generations
treated with DEN.
also occurs in Syrian hamsters when
A multigeneration study with four successive generations of Syrian
hamsters was conducted to determine whether a single s.c. injection of
MATERIALS AND METHODS
different doses of diethylnitrosamine (DEN) (1.25, 2.5,5,10, and 20 mg/
kg body weight) on day 15 of pregnancy induces respiratory tract tumors
Thirty-six female 12-week-old Syrian hamsters (TNO, Zeist, The
not only in the treated P generation mothers and their I-, progeny but
Netherlands) were mated and divided into 5 treatment groups consisting
also in 1; and I i generations. In this study, the P generation mothers
of 6-10 animals each (Fig. 1). The day of mating was counted as day
only were given a single injection of DEN during the period of gestation.
Fifty-six % of the 36 DEN-treated mothers and 52% of their !•', 0. On day 15 of gestation, all animals (P generation) received a single
s.c. injection of 0.2 ml of physiological saline in which 1.25, 2.5, 5.0,
generation offspring (total, 233 animals) developed neoplasms in the
10.0, and 20.0 mg of DEN (provided by Professor Preussmann, German
respiratory tract. A single respiratory tract tumor was found in one DENCancer Research Center, Heidelberg, Federal Republic of Germany;
unexposed !•',
generation control hamster as well as in one F2 generation
purity, 98%), respectively, had been dissolved. A group of 6 hamsters
animal (total, 209 animals) descended from DEN-exposed P generation.
treated with the solvent alone on day 15 of gestation served as the
Both tumors were considered to have arisen spontaneously. No respira
vehicle control.
tory tract tumors were observed in the I <generation (total, 160 animals)
After delivery, the FI generation was raised. Because of the limita
descended from a DEN-exposed P generation. Thus our results indicate
tions on space and the costs of keeping progenies over 3 successive
that the vertical transmission of the tumorigenic effect of DEN in Syrian
generations, the sister-brother mating of !•',
generation was carried out
hamsters is limited to one generation and does not persist in the F2 and
only for two representative DEN doses, i.e., the intermediate (5-mg)
I«generations.
and the highest (20-mg) dose. In order to check the possibility of any
INTRODUCTION
Transplacental induction of tumors by DEN2 has been re
ported in the first generation of Syrian hamsters (1, 2), Djungarian hamsters (3), mice (4-6), and rats (7, 8).
In the Syrian hamster, the respiratory tract proved to be the
main target organ for DEN tumorigenesis, regardless of the
route of administration of the carcinogen (9). However, only
animals whose mothers had received DEN on one of the last 4
days (days 12-15) of pregnancy developed neoplasms in the
respiratory tract (10). It is only at this late stage of pregnancy
that the respiratory tract mucosa has attained the capacity to
enzymatically activate the indirect acting DEN to its ultimate
carcinogen (11, 12).
A particular aspect of transplacental carcinogenesis is that in
experiments using certain animal species and certain types of
carcinogens the carcinogenic risk can be observed not only in
the FI generation but also in subsequent !•'..
and I <generations.
Such experiments include those carried out in mice and rats
with 3-methylcholanthrene (13), DMBA (14-17), o-aminoazotoluene (18), methylnitrosourethan
(19), methylnitrosourea
(20), ENU (21-23), urethan (24), isoniazid and hydrazine sul
fate (25), and diethylstilbestrol (26). However, some of these
experiments produced controversial results, giving rise to such
problems as species or strain specificity and organotropism of
carcinogens, which still remain unsolved in this particular type
of carcinogenesis.
The present study was undertaken to investigate whether
similar vertical transmission of tumorigenicity over multiple
sex-associated transmission of carcinogenic risk to the I i generation,
FI animals exposed in utero to the two representative DEN doses were
cross-mated with the control FI animals of either sex. The F3 generation
was obtained by brother-sister mating of F2 litters (Fig. 1) only. The
hamsters were mated when they were 12 weeks old.
All animals of the P, I-',. I ...and F3 generations were housed individ
ually in Makrolon cages, type III (E. Becker and Co., Castrop-Rauxel,
Federal Republic of Germany), under standard laboratory conditions
(room temperature, 22 ±2'C; relative humidity, 55 ±5%; air change,
8 times/h) and were given a pelleted diet (RMH-TMB; Hope Farms,
Woerden, The Netherlands) and tap water ad libitum. They were
checked twice daily, weighed at weekly intervals, and either kept until
spontaneous death or killed when moribund. After autopsy, all organs
were fixed in 10% buffered formalin, the skulls were decalcified with
DecaÃ-(Scientific Products, Evanston, IL), and all tissues were em
bedded in Paraplast. Sections (5 ¿tmthick) were stained routinely with
hematoxylin and eosin and, if necessary, also with special stains. Step
sections were prepared from the nasal cavities, larynx, trachea with
stem bronchi, and lungs. To determine statistical significance, Fisher's
one-sided exact probability test was used. Calculations were based on
the number of hamsters histologically examined (= effective number of
animals), since some animals were lost due to cannibalism.
RESULTS
Fig. 1 shows the mating schedule and the group-based and
total effective number of males and females in both the control
and experimental hamsters over 4 generations. While each of
the 6 FI groups (total, 273 hamsters) was derived from 6-10 P
generation mothers (total, 42 females), the 7 F2 (total, 251
animals) and 7 I , groups (total, 178 animals) were delivered by
only 5-6 mothers per group of FI and F2 mothers, respectively.
Since most deliveries occurred during the night, the preweaning
Received 3/16/87; revised 6/30/87; accepted 7/9/87.
mortality through miscarriage, cannibalism, and perinatal death
The costs of publication of this article were defrayed in part by the payment
of the young could not be exactly determined but, based on data
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
provided by the breeding stock supplier on the average litter
1To whom requests for reprints should be addressed.
size, this was estimated to range between 15 and 25%. Pre
* The abbreviations used are: DEN, diethylnitrosamine; DMBA, 7,12-dimethylbenz(a)anthracene; ENU, ethylnitrosourea; TBA, tumor-bearing animals.
weaning mortality was especially high in the F3 generation
5112
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1987 American Association for Cancer Research.
MULTIGENERATION
«UMBERS OF «IIMllS
10t
SCiiection
gestation
day15^ li d
(mgfcgb.w.l
4 o-MC
OBI 12
IScrlOc.
l
l 8t
l |
DE»
?W
11o-11g
l» GROUPS
6t
| | 6t
DEI 5
7o- 12?
TUMORIGENICITY
Kl 10
Mo-15 9
13er 17ç Wer 16? -178
Fig. I. Mating schedule. Data represent effective number of animals: f,
pregnant females, b.w., body weight.
offspring where some entire litters were cannibalized by their
mothers. Fertility rates in terms of the total number of animals
at weaning showed no significant differences between controls
and treatment groups of each generation or between animal
groups of different generations.
Data on survival times and tumor incidences are summarized
in Table 1. Although the percentage of TBA, the overall tumor
incidence, and the tumor multiplicity (number of tumors per
hamster) of the DEN-treated P mothers and their F, progeny
were markedly higher than in the untreated controls, the life
span was apparently little affected. The average survival time
of DEN-treated P mothers was 77 weeks as compared to 81
weeks in the untreated controls (TBA, 72.2 versus 50%) and
those of DEN-exposed FI males and females were 90 and 76
weeks (TBA, 64.2 and 76.1%), respectively, as compared to 83
weeks in both sexes of the unexposed FI controls (TBA, 33.3%
males and 57.9% females). The life spans of F2 and F3 animals
ranged between 84 and 99 weeks (TBA, 25-50.6%). The num
ber of tumors per hamster was nearly twice as high in the DENexposed P and F] animals as in the controls. No significant
differences were observed between the percentages of TBA, the
overall tumor incidence, and the tumor multiplicity of F2 and
Fi descendants from DEN-treated mothers and from untreated
control animals.
The overall incidences of respiratory tract tumors induced by
OF DEN
DEN at 1.25 to 20 mg/kg body weight were 55.5% in the
treated P mothers and 47.5% in their male and 56.6% in their
female FI descendants. A single respiratory tract tumor was
found in one F, control animal (papilloma of the larynx) and
in one F2 descendant of a DEN-treated mother (adenocarcinoma of the nasal mucosa). No respiratory tract tumors could
be detected in any of the F3 animals.
Table 2 shows the incidence, type, and organ distribution of
respiratory tract tumors in the DEN-exposed P and F, genera
tions. In both generations, the main target organ for DEN
tumorigenesis was the trachea. This was followed in decreasing
order of incidence by the larynx, lungs, and nasal cavities in the
Fi generation and by the larynx, nasal cavities, and lungs in the
P generation. A positive dose-response relationship was evident
except in the highest dose group of the F, generation (Fig. 2).
At as little as 10 and 20 mg/kg body weight fetal respiratory
tissues appeared to be less responsive to DEN tumorigenicity
than the adult tissues. At a much higher dose (45 mg/kg body
weight) this difference still remained consistent (10).
Histologically, the observed benign tumors were papillomas
and adenomas of the nasal cavities, papillomas of the larynx
and trachea, and adenomas of the lung. At ages ranging from
3 to 8 months these benign tumors became clinically apparent
due to sonorous rales caused by obstruction of the airways.
Additionally, 1 adenocarcinoma of the nasal cavities, 2 adenocarcinomas, and 1 primary fibrosarcoma of the lungs were
observed in FI generation.
The incidence of various tumors of other locations (Table 1)
showed large fluctuations between the treatment and control
groups of one and of different generations (range, 25-63.2%).
However, these differences were statistically insignificant and
none of the tumors could be causally related to DEN treatment.
Furthermore, there was no obvious correlation between the
occurrence of tumors at any sites in the P, I,, and 1 •
generations
and their respective progeny either in the control animals or in
the treatment groups.
DISCUSSION
Our results confirm previous studies (1,4) that the adminis
tration of DEN to pregnant Syrian hamsters at the late stage
Table 1 Animal numbers, survival times, tumor multiplicity, and tumor incidences in DEN-exposed and control Syrian hamsters of 4 generations
Animals with tumors of
no.V
effective
survival
no. of
time
animals36/36121/120115/113104/104IOS/
(wk
SD)77
±
GroupDENPF,F,<F,'ControlPF,F,F,SexFMFMFMFFMFMFMFTotal
siteNo.13424052434027371212715%36.135.0
of
tractNo.
no. of
tumors/
tumors*44(3)121hamster1.221.011.180.650.520.570.370.670.480.790.680.470.250.36Respiratory
%20
1190±
±
1576
1597
±
±2184
1599
±
10580/7986/816/621/2119/1925/2517/175/414/14Av.
1884
±
1881
±
55.557
(13)132
47.564
(22)68
56.61
(27)55
(22)45(11)30(7)4(1)10(0)15(6)17(11)8(4)KD5(3)No.
1.01
±1483
±2783
4.8Other
1497
±
1193
±
1299
±
1990
±
±13TBANo.26778651434026371112715%72.264.276.149.041.050.632.150.033.357.948.041.225.035.7Total
'Attinie of weaning.
b Malignant tumors in parentheses.
' Animals of which both parents were descendants of DEN-exposed mothers and animals of which only one parent (male or female) was descendant of a DENexposed mother were grouped together.
5113
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1987 American Association for Cancer Research.
MULTIGENERATION
group"(sex)P(F)F,
TUMORIGENICITY
OF DEN
Table 2 Incidence, type, and organ distribution of respiratory tract tumors in the DEN-exposed P and F\ generations
Treatment
1No.»1(10)1(10)5(72)15(72)"3(72)1(72).25%10.010.06.920.84.21.4DEN
2.5No.*
%PapillomaAdenomaPapilloma(ta)Papilloma(ta)AdenomaAdenomaAdenocarcinomaPapilloma(ta)
No.4
%1
12.51
(8)
12.510(51)'
(8)
(M, F)OrganNasalcavitiesLarynxTracheaLungsNasalcavitiesLarynxTracheaLungsControlTumor
19.622(51)'
2.SPapilloma(ta)Adenoma(ta)AdenocarcinomaFibrosarcomaDEN
I (40)
43.15
(51)
(35)'17(35)'3(35)5%16.766.72.920.048.68.6DENNo.'4(6)'6(6)"8
(34)'20
(34)'5
9.8DENNo.»1(6)4(6)'1(35)7(34)'1(34)10%66.7100.023.558.814.72.9DENNo.'1(6)1(
* Dose of DEN given in mg/kg body weight.
' Effective number of animals in parentheses.
"/>< 0.001.
F2 or even F3 generations (13-15,17,19-22,
24-26, 29). There
are contradictory results for the carcinogens ENU and DMBA.
Whereas Tomatis et al. (21) reported an increased incidence of
neurogenic tumors in the F, and F2 offspring of ENU-treated
pregnant rats, Ogawa et al. (23) observed no difference in the
incidence of preneoplastic hepatic lesions in the F2 and I ,
progeny derived from Ft rats exposed to ENU in utero. In
contrast to the reports of Tomatis (14), Tomatis and Goodall
(15), and Rao (17) who observed an increased tumor incidence
in FI and 1 : descendants of mice treated with DMBA during
pregnancy, Schneider (16) could confirm these data for the 1-",
100;in_oeCoCVuk.oa90BOJ7060ÌI50403020100¿L__
O
1.25
dose
2.5
of
5
DEN (mg/kg
10
20
b.w.)
Fig. 2. Relationship between DEN doses and percentage ratios of the number
of animals with at least one respiratory tract tumor to the total effective number
of animals. The difference between P and 1 , is not statistically significant, b.w.,
body weight.
of pregnancy results in a dose-related high incidence of tumors
of the respiratory tract, the latter being the main target for
DEN tumorigenesis in both mothers and their F, offspring.
This organotropism of DEN in Syrian hamsters has been long
established (1.4. 9). It is evident from the present study that
DEN treatment of pregnant P hamsters did not produce respi
ratory tract tumors in their F2 and F3 descendants. The adenocarcinoma of the nasal cavities, which was found in one F2
hamster, and the laryngeal papilloma in one F, control animal
are most likely to be spontaneous tumors, although, as has been
reported by other authors (27), the spontaneous incidence of
these tumors should usually be very low (<2%). The incidence
and spectrum of tumors of other locations, including liver and
kidney tumors which might have been initiated by DEN treat
ments (28), showed neither statistically significant differences
between control and treatment groups, each including P, I-',.
1 •.
and F) generations, nor evidence for a causal relationship
to DEN treatment. Thus the respiratory tract tumorigenesis in
the Syrian hamsters is the direct result of DEN carcinogenicity
and not merely the enhancement of spontaneous tumor inci
dence.
Our results do not conform to those of the majority of
multigeneration carcinogenicity studies which have shown ele
vated incidences of neoplasms not only in the l;. but also in the
but not for the I .•
generation. Although the exact mechanisms
of vertical tumor induction in F2 and F3 generations having no
direct contact with carcinogens are still unclear, various hy
potheses have been put forward to explain the phenomenon of
multigeneration carcinogenicity (Refs. 15, 17, 19, 26 and 30;
see also Ref. 31).
In the present study, no DEN-induced respiratory tract tu
mors of I - and I-1Syrian hamsters were found, suggesting that
there are additional factors (or mechanisms) to consider when
interpreting the multigeneration transfer of carcinogenicity.
One of the above-mentioned hypotheses, the enhancement of
an inherited susceptibility (15), could explain the induction of
lung tumors in F2 and F3 Swiss or ICR mice or of lymphomas
in F2 and I-, MA mice; these strains of mice are all genetically
predisposed to develop the respective tumors (15, 24). This
genetic disposition to specific tumor types implies a ready
response to exogenously induced light perturbation of heritable
materials. By contrast, as regards the induction of hyperplastic
(preneoplastic) hepatic nodules by ENU in FI generation F-344
rats (23) and, in this study, papillary respiratory tract tumors
by DEN in F, Syrian hamsters, the neoplastic initiation was
caused by the organotropism of the administered carcinogens
and was not due to the acceleration of processes responsible for
spontaneous tumorigenesis. Therefore, neither tumors nor pre
neoplastic changes could be induced in the target organs of F2
and l;i animals.
In order to verify these assumptions, additional experiments
seem necessary with the use of various species and strains of
animals with low and high spontaneous tumor incidences and
diverse carcinogens.
ACKNOWLEDGMENTS
The authors would like to thank Barbara Bartels and Ann Bordiert
for their assistance in preparing the manuscript.
5114
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1987 American Association for Cancer Research.
MULTIGENERATION
TUMORIGENICITY
REFERENCES
1. Mohr, U., and Althoff, J. Möglichediaplacentar-carcinogene Wirkung von
Diäthylnitrosamin
beim
1964.
2. Mohr,
U., Allheit'!',J.,
andGoldhamster.
Authaler, A. Naturwissenschaften,
Diaplacental effect of 51:
the 515,
carcinogen
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
ilid liyIni]rosimi ine in the golden hamster. Cancer Res., 26:2349-23S2,1966.
Thust, R., and Warzok, R. Transplacental carcinogenic action of diethylnitrosamine in the Djungarian hamster. Zentralbl. Allg. Pathol. Palimi. Anal.,
121:82-86, 1977.
Mnlir. U., and Althoff, J. Die diaplazentare Wirkung des Carcinogens
Diäthylnitrosamin bei der Maus. Z. Krebsforsch., 67:152-155, 1965.
Likhachev, A. J. Transplazentare carcinogene Wirkung von N-Nitrosodiäthylaminbei der Maus. Vopr. Onkol. (Leningr.), 17:45-50,1971.
Diwan, B., and Meier, H. Transplacental carcinogenic effects of diethylnitrosamine in mice. Naturwissenschaften, 63:487-488, 1976.
Pielsticker, K., Wieser, O., Mohr, U., and Wrba, H. Diaplazemar induzierte
Nierentumoren bei der Ratte. Z. Krebsforsch., 69: 345-350, 1967.
Thomas, ( '.. and Bollmann, R. Untersuchungen zur diaplazentaren krebser
zeugenden Wirkung des Diäthylnitrosamins an Ratten. Z. Krebsforsch., 71:
129-134, 1968.
Herrold, K. M. Effect of route of administration on the carcinogenic action
of diethylnitrosamine (¿V-nitrosodiethylamine). Br. J. Cancer, 18: 763-767,
1964.
Mohr, U., Kc/n ik Scimi 1er. H., Reznik, G., and H illrich, J. Transplacental
effects of diethylnitrosamine in Syrian hamsters as related to different days
of administration during pregnancy. J. Nati. Cancer Inst., 55:681-683,1975.
Mohr, U., Reznik-Schuller, II.. and Emura, M. Tissue differentiation as a
prerequisite for transplacental carcinogenesis in hamster respiratory system,
with specific respect to the trachea. Nati. Cancer Inst. Monogr., 51: 117122, 1979.
Reznik-Schüller, H. M., and Hague, B. F., Jr. Autoradiography in fetal
Syrian golden hamsters treated with tritiated diethylnitrosamine. J. Nati.
Cancer Inst., 66:773-777, 1981.
Shay, H., Gruenstein, M., and Weinberger, M. Tumor incidence in F1 and
!•'_•
generations derived from female rats fed methylcholanthrene by stomach
tube prior to conception. Cancer Res., /-.': 296, 1952.
Tomatis, I.. Increased incidence of tumors in F, and F2 generations from
pregnant mice injected with a polycyclic hydrocarbon. Proc. Soc. Exp. Biol.
Med., 119: 743-747, 1965.
Tomatis, L., and ( .nodali. C. M. The occurrence of tumours in F,, F2 and F3
descendants of pregnant mice injected with 7,12-dimethylbenz(a)anthracene.
Int. J. Cancer, 4: 219-225, 1969.
Schneider, P. Multigenerationsstudie zur transplazentaren Wirksamkeit von
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
OF DEN
7,12-Dimethylbenz(a)anthracen
bei C57BL/6-Mäusen. Thesis, University of
Hannover, 1981.
Rao, A. R. Inhibitory action of BHA on carcinogenesis in FI and I •
descend
ants of mice exposed to DMBA during pregnancy. Int. J. Cancer, 30: 121124, 1982.
Gel'shtein, V.J. The incidence of tumours among offspring of mice exposed
to orthoaminoazotoluene. Vopr. Onkol. (Leningr.), 7: 58-64, 1961.
Tanaka, T. Transplacental induction of tumours and malformations in rats
treated with some chemical carcinogens. IARC Sci. Pub., 4: 100-111, 1973.
Tomatis, L., I Ullrich. J., and Turusov, V. The occurrence of tumours in FI,
I , and F.I descendants of BD-rats exposed to JV-nitrosomethylurea during
pregnancy. Int. J. Cancer, 75: 385-390, 1975.
Tomatis, L., Ponomarkov, V., and Turusov, V. Effects of ethylnitrosourea
administration during pregnancy on three subsequent generations of BD VIrats. Int. J. Cancer, 19:240-248, 1977.
Napalkov, N. P., Tomatis, L., Likhachev, A. J., and Kolodin, V. J. Tumors
in mice BALB/c after transplacental exposure to /V-nitrosoethylurea in
progenesis. Vopr. Onkol. (Leningr.), 23:41, 1977.
Ogawa, K., Yokokawa, K., I onnivori. T., and Narasaki, M. Initiation of
focal hyperplastic hepatic lesions by transplacental administration of ethyl
nitrosourea in rats of I \ generation, and no transmission of the effect to F2
and Fj generation. Int. J. Cancer, 31: 775-777, 1983.
Nomura, T. Transmission of tumors and malformations to the next genera
tion of mice subsequent to urethan treatment. Cancer Res., 35: 264-266,
1975.
Menon, M. M., and Bhide, S. V. Perinatal carcinogenicity of isoniazid (INH)
in Swiss mice. J. Cancer Res. Clin. Oncol., 705: 258-261, 1983.
Walker, B. E. Tumors of female offspring of mice exposed prenatally to
diethylstilbestrol. J. Nati. Cancer Inst., 73: 133-140, 1984.
Pour, P., Mohr, U., Cardesa, A., Althoff, J., and Kmoch, N. Spontaneous
tumors and common diseases in two colonies of Syrian hamsters. II. Respi
ratory tract and digestive system. J. Nati. Cancer Inst., 56:937-948, 1976.
Herrold, K. M., and Dunham, L. J. Induction of tumors in the Syrian
hamster with diethylnitrosamine (JV-nitrosodiethylamine). Cancer Res., 23:
773-777,1963.
Miller, E. W., and Pybus, F. C. The incidence of spontaneous and induced
forestomach tumours in mice of two inbred strains and their reciprocal
hybrids. Br. J. Cancer, 9: 142-162, 1955.
Tomatis, L. Prenatal exposure to chemical carcinogens and its effect on
subsequent generations. Nati. Cancer Inst. Monogr., 51: 159-184, 1979.
Mohr, U., Emura, M., Aufderheide, M., Riebe, M., and Ernst, H. Transpla
cental carcinogenesis: mouse, rat, hamster. In: T. C. Jones, U. Mohr, and R.
D. Hunt («Is.),Monographs on Pathology of Laboratory Animals, Repro
ductive System, pp. 148-157. Berlin: Springer-Verlag, 1987.
5115
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1987 American Association for Cancer Research.
Failure to Transmit Diethylnitrosamine Tumorigenicity from
Transplacentally Exposed F 1 Generation Syrian Hamsters to
the Respiratory Tract of F 2 and F3 Generations
H. Ernst, M. Emura, B. Bellmann, et al.
Cancer Res 1987;47:5112-5115.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/47/19/5112
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. © 1987 American Association for Cancer Research.