[CANCER RESEARCH 26, 786-796,May 1966]
Quantitative
Analysis
of the Walker Tumor
KAREL SMETANA,
K. SHANKAR
of Ultrastructural
and Liver1
NARAYAN,
AND
Components
of Nucleoli
HARRIS BUSCH
Department of Pharmacology and The Tumor By-products Laboratory, Baylor University College of Medicine, Houston, Texas
Summary
Nucleolar components were defined as fibrillar, granular or
light areas on the basis of ultrastructural appearance. Quantita
tive measurements were made of these areas with the aid of a
planimeter and the percentage of total nucleolar area composed
of the various areas was determined. The highest content of
fibrillar component was found in the nucleoli of the livers of
actinomycin D-treated animals and the lowest was found in the
Walker tumor. The Walker tumor nucleoli contained a number of
regions that could be classified as fibrillogranular regions, but
very few pure fibrillar regions were found. The finding that the
percentage of the total nucleolar area composed of fibrillar regions
was related in part to the adenylic acid + uridylic acid/guanylic
acid + cytidylic acid ratio of nucleolar RNA supports previous
suggestions that the fibrillar regions contain an RNA fraction
richer in adenylic and guanylic acid.
Introduction
Recent studies have suggested that nucleoli may be biochemi
cally and morphologically differentiated during various stages
of growth, maturation and function of cells (5, 6, 23, 28, 30).
This differentiation is related to the amounts and distribution of
the subcomponents of the nucleoli which include granular and
fibrillar ribonucleoprotein structures as well as the light spaces of
varying size that contain DNP2 fibrils and possibly an amorphous
"nucleoloplasm." Compact nucleoli (23) have a rather uniform
distribution of the ribonucleoprotein structures and a relatively
small amount of the light spaces. These nucleoli are found in
young or growing cells such as tumor cells (3, 5), immature leuko
cytes (23), and cells of pupal stages of Drosophila melanogaster
(30). Ring-shaped nucleoli are found in mature cells such as
lymphocytes (23), plasmocytes (23), cells of mature pupae of
D. melanogaster (30), and mature oocytes (7), or in cells about
to undergo division such as Ehrlich ascites cells in preprophase
(31). In addition, they were also found in HeLa cells and cells of
livers of actinomycin D-treated animals (8, 23, 26). These nu
cleoli have ribonucleoprotein structures predominantly in their
periphery.
1These studies were supported in part by grants from the Jane
Coffin Childs Fund, the American Cancer Society, the Public
Health Service (CA 08182), and the National Science Foundation.
2Abbreviations used in this paper: A, adenylic acid; C, cyti
dylic acid; G, guanylic acid; U, uridylic acid; DNP, deoxyribonucleoprotein; RNA, ribonucleic acid.
Received for publication September 17, 1965.
786
Studies of Marinozzi and Bernhard (12, 13) have suggested
that the fibrillar component of the nucleolar ribonucleoproteins
may represent messenger RNA and the granular component
may represent the precursor of the ribosomal elements in the
cytoplasm. In view of the marked differences in the base ratios
of the nuclear and nucleolar nucleic acids found in tumors and
other tissues in recent studies in this laboratory (4, 14-16, 18,
25), it seemed possible to correlate quantitative studies on the
relative size of areas containing ribonucleoprotein components
with the composition of the nucleolar nucleic acids. To make a
quantitative analysis of the ultrastructural
nucleolar com
ponents, maps of nucleoli were prepared from which areas of
fibrillar and granular components were determined with the aid
of a planimeter. These studies showed that of the tissues ex
amined, the Walker tumor had the lowest relative areas of
fibrillar components, a finding which agrees substantially with
the high content of GC-rich RNA found in previous studies
(14).
Materials
and Methods
Animals
Male albino rats from the Holtzman Rat Company, weighing
180-250 gm, were used in this study. Partial hepatectomy was
performed as described previously (15) and the animals were
sacrificed 15 hr later. Thioacetamide-treated animals received
i.p. injections of a 2% solution of thioacetamide in 0.15 M NaCl
in a dose of 50 mg/kg body weight daily or at specified times
prior to death (1). Rats were killed 9 days after the initial in
jection of thioacetamide. Animals treated with actinomycin D
received 1 i.v. injection of 150 ¿ug
of the drug/kg body weight 60
min before they were sacrificed. The Walker 256 carcinosarcoma
used in these studies was excised 6-8 days after transplantation.
Electron Microscopy
Thin slices of Walker tumor and rat liver from experimental
and control animals were fixed in 1% osmium tetroxide buffered
with s-collidine at pH 7.4 (2) for 1 hr at 4°C.Specimens embedded
in Epon 812 (11) were cut with an LKB ultramicrotome and
post-stained with uranyl acetate (29) and lead hydroxide (9).
Ultrathin sections were examined with a Philips Norelco 200
electron microscope.
Planimetry
Whole nucleolar areas and areas containing fibrillar and granu
lar components were measured on electronmicrographs magniCANCER RESEARCH VOL. 26
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
Nucleolar Components of Walker Tumor and Liver
fied 2-6 times with a planimeter (OTT, Burrel Corporation,
Pennsylvania), as shown in Figs. 1 and 2. The proportions of the
total nucleolar areas containing the fibrillar and granular com
ponents were calculated as well as the areas occupied by the
light spaces of the nucleolus.
By definition, a "fibrillar area" is one of greater density than
the light spaces but does not contain defined granular components
up to 30,000 magnifications directly in the electron microscope.
By definition, at the same magnification, a granular area con
tains "granules" predominantly.
TABLE 2
PERCENTAGECOMPOSITIONOF TOTAL NUCLEOLARAREAS IN
SERIALSECTIONSOF NORMALLIVER"
containing
containing
section12341234Areas
fibrillar components14.814.215.514.219.fi18.720.817.0Areas
granular components83.681.883.083.577.580.07
Nucleolus
1Nucleolus
2Serial
Results
Studies on Normal Liver Nucleoli
Fig. 1 shows the ultrastructure of normal liver nucleoli. Fig.
2 shows the method for preparation of the nucleolar map which
differentiates the fibrillar areas from the total nucleolar area.
On the basis of planimetrie analysis, approximately 19% of the
total area of nucleoli of normal liver was accounted for by the
fibrillar component. For the purposes of statistical analysis, a
number of electron micrographs of normal liver nucleoli were
analyzed in the same way. The values in Table 1 show the per
centages of the fibrillar and granular areas in nucleoli from var
ious tissues.
To determine the reliability of planimetrie studies on nucleoli,
a series of serial sections was prepared from 2 different liver
nucleoli. Consecutive sections, 800-1000 A in thickness, were cut
and mounted on separate grids. Out of each group of 20 sections,
1 was selected for planimetrie analysis. The selection was made
in such a manner that no 2 sections were closer than 1 /t to one
another in situ. The results of the planimetrie analyses are shown
in Table 2 for 2 nucleoli of normal liver. Only minor changes
TABLE 1
PERCENTAGECOMPOSITIONOF TOTAL NUCLEOLARAREAS"
(10)Regenerating
liver
afterpartial liver, 15 hr
(6)Liver hepatectomy
9 days after thioaceta
(6)Liver
mide administration
actinomycinadministration
1 hr after
(29)Walker
tumor (15)Areas
were found in the values for the areas containing the fibrillar
and granular nucleolar components.
Regenerating Liver and Liver of Thioacetamide-treated Rats
Serial Sections
TissueNormal
" Serial sections of a given nucleolus were prepared by mount
ing 10 consecutive sections, 800-1000 A in thickness, on separate
grids. The 10 consecutive sections were placed on a grid and the
next 10 sections were discarded. This process was repeated several
times. One micrograph was prepared for each of the 10 sections;
thus the micrographs taken represented sections of the nucleolua
spaced 2-^ thickness apart.
containing
fibrillar
components19
containing
granular
components77
412±
34.5 ±
584±
491±
0.348
±
250±
233
±
2293
±
±2Areas
±3
" Areas containing fibrillar and granular components were
measured with a planimeter on micrographs of ultrathin sections
and were expressed as % of the total nucleolar areas measured in
the same sections. The means and standard deviations of values
obtained are presented. The numbers of different nucleoli studied
are shown in parentheses. A representative section from each
nucleolus was analyzed. The numbers of animals used for studies
on the normal liver, regenerating liver, livers of thioacetamidetreated rats, actinomycin D-treated livers and Walker tumors
were 6, 2, 6, 2, and 4, respectively.
Similar analyses were carried out on regenerating liver 15 hr
after hepatectomy, as shown in Figs. 3 and 4, at this same point.
The fibrillar component accounted for 12% of the total nucleolar
area. Although the fibrillar component was somewhat decreased
in the nucleoli in cells of regenerating liver, planimetrie analysis
of the percentage of fibrillar component in the nucleoli of livers
of rats treated with thioacetamide for 9 days (Table 1, Figs. 5,
6) revealed a much lower content of fibrillar components, i.e.,
4.5%.
Livers of Rats Treated with Actinomycin D
By contrast with regenerating liver and the livers of rats
treated with thioacetamide in which the percentage of nucleolar
area accounted for by the fibrillar component was decreased,
there was a marked increase in the fibrillar component following
administration of actinomycin D. Although a series of changes
occurred within 60 min after the i.v. injection of actinomycin D,
the nucleolar pattern was relatively stabilized in treated livers
at 60 min (Figs. 7, 8); at this time, the fibrillar component com
prised 48% of the total nucleolar area. There was variability in
the appearance of nucleoli of liver cells of actinomycin D-treated
animals; some nucleoli were ring-shaped and some were compact
with segregation of granular and fibrillar components (8, 17, 20,
26, 27).
Walker Tumor
Figs. 9 and 10 present the planimetrie maps for the Walker
tumor. Unlike the liver nucleoli of all types studied in which
fibrillar areas were virtually completely devoid of granules,
granular components in the Walker tumor were found in almost
all areas. The appearance of the nucleoli in electron micrographs
MAY 1966
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
787
K. Smetana, K. S. Narayan, and Harris Busch
suggests that 2 predominant types of areas exist in the Walker
tumor nucleolus, i.e., granular areas and fibrillogranular areas.
By comparison with the liver in which structures in some nucleolonemata were virtually completely fibrillar, there were few, if
any, such nucleolonemata in the Walker tumor. As a result, the
percentage of total nuclcolar area in the Walker tumor composed
of fibrillar components was extremely low, i.e., 2.8%.
Granular Component
Table 1 also presents the content of the granular components
of the nucleolus which ranged in area from 52% to 93% of the
total nucleolar area. Inasmuch as the fibrillar component is
such a large part of the total nucleolar area in nucleoli treated
with actinomycin D, the granular component only accounted
for about 0.5 the total nucleolar area. As shown in Figs. 7 and 8,
the line of demarcation of the granular and fibrillar component
is frequently very sharp in nucleoli of livers of actinomycin Dtreated animals. Although the granular component was largest
in area in nucleoli of the Walker tumor, the differences in the
percentages of the total nucleolar areas accounted for by the
granular component were not significantly different from the
nucleoli of thioacetamide-treated livers. In addition, the nucleoli
from regenerating livers had a higher content of granular com
ponent than those of normal liver. On the other hand, the per
centages of areas occupied by the nucleolar light spaces were not
markedly different in the various nucleoli analyzed.
Fibrillogranular Areas
Frequently in nucleoli of the Walker tumor and occasionally
in the nucleoli of thioacetamide-treated rats (Figs. 11, 12),
nucleolonemal areas were composed of masses which appeared
to contain a mixture of fibrillar component and granular com
ponent. In the Walker tumor, only minute areas of "pure"
fibrillar component were found as contrasted to the nucleoli of
any of the liver samples.3
Discussion
Previous studies on the ultrastructural components of the
nucleolus have been directed at analysis of the types of compo
nents present (3, 12, 13, 20, 27, 28). In the present study, an
effort has been made to determine the approximate quantitative
composition of the nucleolus in terms of the areas occupied by
components visible at magnifications providing optimal resolu
tion of these various components. Some agreement with bio
chemical analysis has been obtained in that, on the basis of
calculations of the base composition, there is an agreement be
tween the areas composed of fibrillar components and the com
parative A + U/G + C content of the nucleolar UNA (5, 14,
15, 18, 25). A strict comparison cannot be made because, in the
biochemical analyses, tens of billions of nucleoli are analyzed for
individual components as compared to the few nucleoli analyzed
with the planimetrie procedures. In addition, the analysis of
*The same result was found when nucleoli were analyzed from
immature cells of a plasmacytoma, lymphatic leukemia, and monocytic leukemia. The predominance of the mixed granular and
fibrillar components in nucleolonemata of these tumor cells was
such that it was not possible to differentiate fibrillar regions.
788
the nucleoli by biochemical methods does not take into con
sideration the marked variations from the periportal to the cen
tral vein in liver which are important in the effects of various
drugs, as was pointed out by Khoulish and Kleinfield (10).
The greatest changes found in the compositions of the nucleoli
of liver cells on the basis of planimetry were those in the thio
acetamide-treated rats as compared to the livers of rats treated
with actinomycin D.4 These quantitative results compare favor
ably with the previous over-all impressions from observations
of the electron micrographs (6, 8, 17, 20, 24, 26, 27) and also
with the changes in base composition found for isolated nucleoli
and nuclei of rats treated with these 2 agents (4, 14, 18, 25).
However, only minor changes were found in the base composition
of nucleoli of regenerating livers (.15),and these results correspond
to the relatively small changes in the amounts of the areas oc
cupied by the granular and fibrillar ribonucleoprotein compo
nents.
As noted previously by a number of workers (21, 22), there
were apparently no specific morphologically differentiated nu
cleolar structures characteristic of the cells of the Walker tumor
with the possible exception of dense granules 200-400 A in size
(19). However, it would appear that the nucleolonemata of the
Walker tumor nucleoli contain very few areas that are pure
fibrillar areas. In many areas, the fibrillar components contain
many granules in the nucleolonemata of the Walker tumor. Since
similar areas were noted in the nucleolonemata of the nucleoli
of livers of thioacetamide-treated rats, it is not clear whether
these "fibrillogranular" areas are representative of changes in
neoplastic tissues or preneoplastic tissues or simply represent
an excessive production of the granular nucleolar component.
Acknowledgments
The authors are extremely appreciative to Dr. William J. Steele
for supplying samples of animals treated with actinomycin D and
thioacetamide, to Dr. Tae S. Ro for supplying samples of livers of
animals treated with actinomycin D, to Dr. Masami Muramatsu
for supplying samples of regenerating livers, and to Mr. Charles
Taylor for supplying samples of Walker tumor. The authors are
also appreciative to Dr. Steele for his many helpful suggestions
and comments.
References
1. Adams, H. R., and Busch, H. Effects of Thioacetamide on
Incorporation of Orotic Acid-C14into RNA Fractions in Liver.
Cancer Res., 23: 576-82, 1903.
2. Bennet, H. S., and Luft, J. H. s-Collidine as a Basis for Buf
fering Fixatives. J. Biophys. Biochem. Cytol., 6: 113-14, 1959.
3. Bernhard, W., and Granboulan, N. The Fine Structure of the
Cancer Cell Nucleus. Exptl. Cell Res., 9 (Suppl.).- 19-53,
19C3.
4. Busch, H., and Adams, H. A. Effects of Thioacetamide on
4When representative sections containing all the nucleolar
elements were analyzed the means of the values for the areas con
taining fibrillar and granular components were very similar to
the means obtained when all sections were included. However,
the range of the deviations were actually lower when many of the
representative sections were analyzed from livers of actinomycin
D-treated animals.
CANCER RESEARCH VOL. 26
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
Nucleolar Components
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Nuclear and Nucleolar UNA of the Rat Liver. Texas Kept.
Biol. Med., S3: 38-44, 1965.
Busch, H., Byvoet, P., and Smetana, K. The Nucleolus of the
Cancer Cell: A Review. Cancer Res., 23: 313-30, 1963.
Davis, J. M. G. The Ultrastructure
of Nucleoli and Chromo
somes during the Early Ktages of Regeneration
and the
Changes Produced in These Structures by X-Kudiation. Phil.
Trans. Roy. Soc. London, Ser. B, 246: 291-303, 19(ì3.
Esper, H. Studies on the Nucleolar Vacuoles in the Oogénesis
of Aratoria punctulata. Exptl. Cell Res., 38: 85-96, 191)5.
Journey, L. J., and Goldstein, M. N. Electron Microscopic
Studies on HeLa Cell Lines Sensitive and Resistant to Actinomycin D. Cancer Res., 21: 929-32, 1961.
Karnovsky, M. J. Simple Methods for Staining with Lead at
High pH for Electron Microscopy. J. Cell Biol., 11: 729-32,
1961.
Khoulish, S., and Kleinfeld, R. G. The Role of the Nucleolus.
I. Tritiated Cytidine Activity in Liver Parenchymal Cells of
Thioacetamide Treated Rats. Ibid., 23: 39-51, 1964.
Luft, J. H. Improvements
in Epoxy Resin Embedding
Methods. J. Biophys. Biochem. Cytol., 9: 409-14, 1961.
Marinozzi, V. Cytochimie Ultrastructurale
du Nucleole-RNA
et ProtéinesIntraniicleolaires.
J. Ultrastruct. Res., 10: 433-56,
1964.
Marinozzi, V., and Bernhard, W. Presence dans le Nucléolede
Deux Types de Ribonucleoproteines
Morphologiguement
Distinctes. Exptl. Cell Res., S2: 595-98, 1963.
Muramatsu, M., and Busch, H. Studies on Nucleolar RNA of
the Walker 256 Carcinosarcoma and the Liver of the Rat.
Cancer Res., 24: 1028-34, 1964.
. Studies on Nucleolar and Nucleolar RNA of Regener
ating Liver. J. Biol. Chem., 240: 3960-66, 1965.
16. Okamura, N., and Busch, H. Base Composition of High Mo
lecular Weight Nuclear RNA of Walker Tumor and Liver of
Rat. Cancer Res., 25: 693-97, 1965.
17. Reynolds, R. C., Montgomery, P. O'B., and Hughes, B. Nu
cleolar "Caps" Produced by Actinomycin D. Ibid., 24: 126972, 1964.
18. Ro, T. S., and Busch, H. In Vitro Labeling of RXA in Isolated
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
of Walker Tumor and Liver
Nucleoli of the Walker Tumor and Liver. Ibid., 24: 1630-33,
1964.
Sankaranarayanan,
K., and Busch, H. Dense Granules in
Nucleoli of Walker 256 Carcinosarcoma Cells. Exptl. Cell Res.,
38: 434-37, 1965.
Schoefl, G. I. The Effect of Actinomycin D on the Fine Struc
ture of the Nucleolus. J. Ultrastruct. Res., 10: 224-43, 1964.
Smetana, K., and Busch, H. On the Ultrastructure
of the
Walker 256 Carcinosarcoma.
Cancer Res., 23: 1600-03, 1963.
. Studies on the Ultrastructure
of the Nucleoli of the
Walker Tumor and the Rat Liver. Ibid., 24: 537-57, 1964.
Smetana, K., Lane, M., and Busch, H. Studies on Nucleoli of
Leukemic Agranulocytes and Plasmocytes in Multiple Mye
loma. Exptl. Mol. Pathol., in press.
Solomon, J. C. Modification des Cellules du Parenchyme
Hépatique du Rat sous l'Effet de la Thioacetamide. J. Ultra
struct. Res., 7: 293-307, 1962.
Steele, W. J., Okamura, N., and Busch, H. Effects of Thio
acetamide on théComposition and Biosynthesis of Nucleolar
and Nuclear Ribonucleic Acid in Rat Liver. J. Biol. Chem.,
240: 1742-49, 1965.
Stenram, U. Electron Microscopic Study on Liver Cells
Treated with Actinomycin
D. Z. Zellforsch. Mikroskop.
Anat., 65: 211-19, 1965.
Stevens, B. J. The Effect of Actinomycin D on Nucleolar and
Nuclear Fine Structure in the Salivary Gland Cell of Chironomus thummi. J. Ultrastruct. Res., //: 329-53, 1964.
. The Fine Structure of the Nucleolus during Mitosis
in the Grasshopper Neuroblast Cell. J. Cell Biol., 24: 349-68,
1965.
Watson, M. L. Staining of Tissue Sections for Electron Micros
copy with Heavy Metals. J. Biophys. Biochem. Cytol., 4475-78, 1958.
Wessing, A. Der Nucleolus und seine Besiehungen zu den des
Cytoplasmas.
Eine Untersuchung
an den malphigischen
Gefassen von Drosophila melanogaster. Z. Zellforsch. Mikros
kop. Anat., 65: 445-80, 1965.
Yasuzumi, G., and Sugihara, R. The Fine Structure of Nuclei
as Revealed by Electron Microscopy. Exptl. Cell Res., 37:
207-29, 1965.
MAY 1966
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
789
K. Smetana, K. S. Narayan, and Harris Busch
The measured lines in Figs. 1-12 each represent 1 it.
FIG. 1.—Nucleolus of a liver cell from a control rat.
790
X 48,000.
CANCER RESEARCH VOL. 26
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
Nucleolar Components of Walker Tumor and Liver
FIG. 2.—Map of the nucleolus of Fig. 1. The whole nucleolar area and the areas containing
X 48,000.
the fibrillar components
are outlined.
MAY 1966
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
791
K. Smetana, K. S. Xarayan, and Harris Busch
FIG. 3.—Nucleolusof a liver cell from regenerating liver, 15 hr after partial hepatectomy.
X 37,000.
FIG. 4.—Map of the nucleolus of Fig. 3. The whole nucleolar area and the areas containing
X 37,000.
792
the fibrillar components
are outlined.
CANCER RESEARCH
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
VOL. 26
Nucleolar Components of Walker Tumor and Liver
ñ
6
FIG. 5.—Nucleolus of a liver cell after administration of thioacetamide in a dose of 50 mg/kg body weight for 9 days. X 10,000.
FIG. 6.—Map of the nucleolus in Fig. 5. The whole nucleolar area and the areas containing the fibrillar components are outlined.
X 16,000.
MAY 1966
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
793
K. Smelano,, K. S. Narayan, and Harris Busch
'*>
~
'•.»
8
FIG. 7.—Nucleolus of a liver cell l hr after treatment of the rat with a dose of 150 /ig of actinomycin D per kg body weight.
X 45,500.
FIG. 8.—Nucleolar map of the nucleolus in Fig. 7. The nucleolar areas and the areas with the fibrillar components are outlined.
X 45,500.
794
CANCER
RESEARCH
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
VOL. 26
Nucleolar Components of Walker Tumor and Liver
iom
FIG. 9.—Nucleolus of a Walker tumor cell. X 29,700.
FIG. 10.—Map of the nucleolus of Fig. 9. The nucleolar area and the areas containing the fibrillar components are outlined.
MAY 1966
X 29,700.
795
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
K. Smetana, K. S. Narayan,
and Harris Busch
FIG. 11.—Portion of a nucleolus in a liver cell after administration
of thioacetamide showing areas in nucleolonemata
both fibrillar and granular components.
X 50,000.
FIG. 12.—Portion of a nucleolus in a Walker tumor cell with areas composed of both fibrillar and granular components.
796
containing
X 59,000.
CAXOKR HKSKARCH VOL. 26
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1966 American Association for Cancer Research.
Quantitative Analysis of Ultrastructural Components of Nucleoli
of the Walker Tumor and Liver
Karel Smetana, K. Shankar Narayan and Harris Busch
Cancer Res 1966;26:786-796.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/26/5/786
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 17, 2017. © 1966 American Association for Cancer Research.