Recognition
of Altered Patterns of Messenger
O
O
Synthesis in a Mouse Hepatoma*
RNA
CHEV KiDSONf AND K. S. KIRBY
(Chester Beatty Research Institute,
Institute of Cancer Research: Royal Cancer Hospital, Fulham Road, London, S.W.S. England)
SUMMARY
Patterns of messenger RNA synthesis in a transplanted spontaneous BR6 mouse
hepatoma have been compared with those of normal BR6 mouse liver by countercurrent distribution.
The liver patterns showed little variation, whereas in the
hepatoma there was progressive alteration in the messenger RNA profile with pro
gression of the tumor. Evidence from countercurrent distribution profiles of liver
and hepatoma DNA and from comparative rates of messenger RNA synthesis sug
gests that less of the tumor genome is being transcribed into messenger RNA. From
the messenger RNA profiles on countercurrent distribution it is evident that, of those
regions of the hepatoma genome which are undergoing transcription, many are dif
ferent from those being transcribed in the normal liver.
Genetic changes involved in the transition from normal
to neoplastic cells have been inferred generally from
changes in growth pattern, in some cases from observed
chromosome aberrations, and in many instances from
alterations in enzyme patterns. Quantitative variations
in enzyme levels, the absence of an enzyme in a given
tumor, and the failure to induce it do not distinguish
between structural and regulatory gene mutations, nor
indeed do they necessarily indicate the occurrence of
true mutation. Such changes may also be explained in
terms of altered transcription of given DNA base sequences
(15). Qualitative changes in enzyme structure or kinetics
in tumor cells frequently appear to indicate underlying
mutation in the structural gene concerned, but the change
may represent an induced isozyme produced minimally
in the original normal cell or an alteration in translation
of messenger RNA into specific protein, analogous to the
effects of suppressor mutations in bacteria (4).
What is required ideally is a means of direct "reading"
of the genome of the tumor cell and the cell of origin, at
the levels of the total genome, the effectivegenome, and the
facultative genome (5). This is especially true with respect
to regulatory loci, the action of which is far from clear in
higher organisms, in contrast to the accumulation of data
supporting the postulates of Jacob and Monod (9) for
microorganisms.
* This investigation was supported by grants to the Chester
Beatty Research Institute (Institute of Cancer Research: Royal
Cancer Hospital) from the Medical Research Council, The British
Empire Cancer Campaign, and the National Cancer Institute of
the National Institutes of Health, U. S. Public Health Service.
t Arthur A. Thomas Cancer Research Fellow of the AntiCancer Council of Victoria, on leave from the Baker Medical
Research Institute, Melbourne, Australia.
Received for publication March 27, 1964.
At present this is far from possible at the desirable level
of DNA base sequences. However, analysis of patterns
of messenger RNA (mRNA) synthesis does permit as
sessment in less precise terms of the expression of the
genome under a given set of conditions. This has been
approached in a variety of ways: in bacteria by chromatography on methylated albumin kieselguhr columns (14,
16, 17) and by base sequence complementarity through
hybrid formation with homologous or heterologous DNA
in DNA-containing columns (1, 2, 6); in mammalian
tissues by hybrid formation (8) and by countercurrent
distribution (CCD).1 This latter method allows direct
recognition of altered patterns of messenger RNA syn
thesis and has provided a means of identifying tissuespecific mRNA patterns.
In the present studies CCD has been used to identify
altered patterns of mRNA synthesis in a transplanted,
spontaneous mouse hepatoma (BR6). This tumor, re
taining certain normal metabolic features such as synthesis
of glycogen, in some respects conforms to the definition of
a "minimal deviation hepatoma."2 As such it is a useful
model for analysis of patterns of messenger RNA synthesis
in cells of established tumors.
MATERIALS AND METHODS
Materials.—The BR6 hepatoma is a spontaneous
parenchymal cell tumor of mouse liver which maintains
at least some aspects of normal liver metabolic function,
notably glycogen synthesis. In transplanted form it is
1C. Kidson and K. S. Kirby, J. Mol. Biol. (in press).
*V. R. Potter, Biochemical Studies on Minimal Deviation
Hepatomas. In: Unió Intern. Contra Cancrum Symposium on
Cellular Control Mechanisms and Cancer, Amsterdam, 1963 (in
press).
1604
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1964 American Association for Cancer Research.
KiDSON ANDKIEBY—MessengerRNA Synthesis
slow-growing compared with transplanted
azo dyeinduced hepatomas and is of relatively low-grade malig
nancy. From about 6 weeks onward there is increasing
central necrosis not infrequently accompanied by hemor
rhage into the tumor. In the present investigations the
hepatoma was studied under conditions of serial trans
plantation into male or female BR6 mice 6-8 weeks old,
the tumor having been maintained by transplantation for
5 years at the time of these experiments. Tumors were
used 4-10 weeks after transplantation.
Animals were
killed by breaking their necks, and liver and hepatoma were
removed immediately from each animal and frozen on
dry ice or in liquid nitrogen. The respective tissues were
pooled from at least four mice in each experiment.
Histology.—Histological sections showing the presence
of glycogen in the hepatoma cells were prepared on tumors
fixed in Bouin's fluid. Staining was with Best's carmine,
counterstaining
with Ehrlich's hematoxylin. Control
sections were stained after treatment with saliva (Figs.
1,2).
Pulse-labeling.—Inexperiments on mRNA pulse-labeling
was carried out by intraperitoneal injection of 100-250
fie. of uridine-H3 per mouse. Pulse time was assessed
from time of injection to freezing of the tissues. The
usual pulse time was 20 minutes, when maximal incorpora
tion of label into the rapidly labeled RNA fraction has been
found to occur (11).
Extraction of mRNA and DNA.—Extraction of the
rapidly labeled RNA containing most of the mRNA with
a high turnover rate was carried out essentially as de
scribed for rat liver (11). Ribosomal and soluble RNA
were first removed by extraction with 0.5 per cent naphthalene-1,5-disulfonate/90
per cent phenol/0.1 per cent
8-hydroxyquinoline and discarded in the aqueous phase.
DNA and mRNA were then extracted from the insoluble
phenolic interface with 6 per cent 4-aminosalicylate/90
per cent phenol/0.1 per cent 8-hydroxyquinoline, pre
cipitated from the aqueous phase with ethanol, washed,
and dried. This interfacial DNA/RNA was then used
to study mRNA patterns.
DNA was extracted and purified by the 4-aminosalicylate method (12).
Sucrose density gradient centrifugation.—Analysis of
mRNA preparations (1 mg. DNA/RNA) was carried out
by centrifugation in sucrose gradients of 5-20 per cent
sucrose in 0.01 M sodium acetate, pH 5, at 24,000 r.p.m.
at 4°C. for 14 hours in a Spinco SW 25 rotor. Fractions
of 10 drops were collected from the bottom of the tube,
diluted with water to 1.5 ml., and the nucleic acids were
extracted and counted by liquid scintillation spectrometry
as described previously (11),
Countercurrent distribution of mRNA.—Solvent systems
for CCD were similar to those developed for ribosomal
RNA (13), except that trilithium citrate was substituted
for tripotassium citrate in the aqueous phase. The
proportions in the CCD system employed here for mRNA
were: organic mixture, 28 volumes; amine solution, 12
volumes; 0.033 M trilithium citrate, 13.5 volumes; and
water, 38.5 volumes (solvent system 150/13.5).
Interfacial DNA/RNA (1.5 mg.) was distributed in
this system over 100 transfers. The DNA/RNA was
1605
TABLE 1
RELATIVEINCORPORATION
OF URIDINE-H' INTO
RAPIDLYLABELEDMRNA OF BR6 LIVER
ANDHEPATOMA*
Weeks after
tumor
transplantation468TissueLiverHepatomaLiverHepatomaLiverHepatomaCounts/min/Ajeomp186445218
liver/nepatoma4.13.84.9
* Pulse: 200 nc. uridine-H3/niouse/20 min.
first dissolved in 3 ml. of each phase of a 150/2 system,
then transferred to the lower phase of the final solvent
system via a two-tube gradient of increasing lithium ion
concentration. After 100 transfers 1 ml. of ethanol was
added to each tube to give one phase, and radioactivity
was measured in 2-ml. aliquots by liquid scintillation
spectrometry (10 ml. of scintillation mixture).
Countercurrent distribution of DNA.—Native DNA from
liver or hepatoma was distributed in the same solvent
system as for mRNA, but containing 11.7 volumes of
trilithium citrate and 40.3 volumes of water (solvent
system, 150/11.7). One mg. was distributed over 80
transfers (2 ml. each phase), ethanol added as above,
and the absorption read at 260 m/i (10).
Chromosome counts on BR6 hepatoma.—-Animalswere
given injections intraperitoneally of 0.04 per cent (w/v)
colcemid (CIBA), a portion of the tumor was removed,
the cells were freed from stroma, treated for 20 minutes
with 1 per cent citrate solution, then fixed for 1 hour in
45 per cent acetic acid. After acid hydrolysis (6 min. in
NHC1 at 60°C.), the tissue was stained with Feulgen, and
chromosome counts were carried out on squash prepara
tions.
RESULTS
Rate of mRNA synthesis.—The relative rates of in
corporation of uridine-H3 into rapidly labeled mRNA
in a 20-minute pulse in mouse liver and hepatoma in a
series of experiments are listed in Table 1. These are
expressed as counts/ min/A^o m«,
the absorption represent
ing mainly DNA. The rate of synthesis of mRNA as
judged by uridine incorporation was consistently higher
in liver than ha the hepatoma, at all time periods of tumor
growth studied.
Sucrose gradients of mRNA.—Chart 1 shows the patterns
obtained by centrifugation of mRNA samples in sucrose
density gradients. The tumor patterns were less con
sistent than those for liver but mRNA's from both tissues
had a wide range of s values which, for mouse liver (12-40
s) were essentially the same as for rat liver (II).1 The
hepatoma mRNA had profiles different from those for
liver but with a similar range of s values. These findings
indicate that little or no degradation of mRNA occurred
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1964 American Association for Cancer Research.
1606
Vol. 24, October 1964
Cancer Research
1250-
1000-
LIVER
A/
WEEKS AFTER
TUMORTRANSPLD.
4
\
6
8
750
500-
250-
400-
HEPATOMA
L/1
§300
O
O
200-
100
10
15
20
25
30
TUBE No.
CHART1.—Sedimentation profiles obtained by centrifugation
in sucrose gradients of rapidly labeled mRNA from BR6 livers and
hepatomas, from the same animals at varying times after trans
plantation of the tumors.
during preparation and are consistent with considerable
heterogeneity in size of mRNA's in both tissues.
Countercurrent distribution of mRNA.-—Mouse liver
mRNA's showed relatively consistent distribution pat
terns on CCD in a series of experiments (Chart 2). DNA,
rRNA, and sRNA remain largely in the first twenty
tubes in the solvent system used.1 The radioactive pat
terns obtained for liver mRNA were seen only under the
standardized conditions of diet and methods of prepara
tion that have been used. The CCD patterns of tumor
mRNA taken at 4, 6, and 8 weeks showed considerable
heterogeneity in profile, little consistency among them
selves, and great differences from the patterns of liver
mRNA. There was a trend away from the liver pattern
the longer the time of growth after transplantation, but
the distribution at no time closely resembled that of
liver in all respects.
Countercurrent distribution of DNA.—Native DNA from
mouse liver and hepatoma (6 weeks) gave CCD patterns
illustrated in Chart 3, in solvent system 150/11.7. That
for mouse liver, with a narrow initial peak and a large
number of subsequent peaks, follows the same general
pattern observed in other mammalian tissues (10). The
initial peak represents DNA present mainly as firmly
hydrogen-bonded double helices, whereas the remainder
represent DNA with partly separated strands, denaturation being greatest in those molecules traveling furthest in
the organic phase.3 In the hepatoma the same initial
peak was observed, but there was less spreading of the
remainder, indicating the presence of fewer molecules,
3 C. Kidson and K. S. Kiiby,Biochim. Biophys. Acta (in press).
10
20
30
40
50
60
70
80
100
90
TRANSFER No.
CHART2.—Profilesobtained by countercurrent distribution of
rapidly labeled mRNA from BR6 livers and hepatomas from the
same animals at varying times after transplantation of the tumors.
Solvent system: 150/13.5, 100 transfers.
LIVER
i
--
HEPATOMA
CD .
CO
CM
UJ
O
z
<
m
ce .2
*
o
to
m
10
20
30
TRANSFER
50
60
70
80
No.
CHART3.—Profilesobtained by countercurrent distribution of
DNA from BR6 livers and hepatomas from the same animals, 6
weeks after transplantation of the tumors. Solvent system:
150/11.7, 80 transfers.
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1964 American Association for Cancer Research.
KiDSON AND KIRBY—MessengerRNA Synthesis
30-
LJ
O 20-
LJ
i
O
n
20
40
60
¡i
n
CHROMOSOME
80
100
No.
CHART 4.—Histogram of chromosome numbers in squash
preparations of a BR6 hepatoma 6 weeks after transplantation.
with a high degree of strand separation. The degree of
spreading of the partly denatured DNA from the hepatoma
varied from preparation to preparation but appeared
always to be less than for liver DNA.
Chromosome counts on BR6 hepatoma.—Chart4 shows a
histogram of chromosome numbers on a 6-week-old
hepatoma. The majority of cells had 40 or 80 chromo
somes, but considerable heteroploidy was evident, the
range being from 7 to 96 per cell.
DISCUSSION
Evidence for markedly differing patterns of messenger
RNA synthesis in a BR6 mouse hepatoma compared with
mouse liver has been derived from sucrose gradient centrifugation and from countercurrent distribution. The
mRNA in these DNA/RNA preparations consisted mostly
of free mRNA with small amounts present as ribonucleaseresistant RNA/DNA hybrid or associated with DNAprotein complex: the comparisons are as much between
patterns of mRNA synthesis as between mRNA fractionation profiles.1 The use of the term "messenger"
for the whole of the rapidly labeled RNA present in the
interfacial DNA/RNA is for convenience. In addition
to rapid turnover rate and wide range of s values, about
70 per cent of an analogous fraction from rat liver attaches
to rat liver ribosomes in the presence of 0.01 Mmagnesium,
and its synthesis is inhibited by actinomycin D.4 Al
though these DNA/RNA preparations do appear to con
tain the great majority of the rapidly labeled mRNA, it
is not yet known whether all groups of rapidly labeled
RNA thus fractionated are equally active in directing
specific polypeptide synthesis.
In our hands reproducibility of minor peaks of rapidly
labeled RNA on sucrose gradients has not been sufficiently
good to warrant precise interpretation of small differences.
The sedimentation characteristics of hepatoma and liver
mRNA in the present studies, however, did reveal differ
ences in general profile, although there was similarity in
the range of s values. This suggests that both mouse
4 B. R. Wilkinson,
to be published.
1607
liver and hepatoma mRNA's are heterogeneous in size
(7,11).
Fractionation by CCD, in which nucleotide base com
position plays a major role,1 has been found to give a high
degree of reproducibility for a given preparation of mRNA
and for different preparations made from a given tissue
type under standardized conditions.1 CCD has revealed
striking differences between the patterns of mRNA syn
thesis in liver and hepatoma. In contrast to the very
similar liver patterns, the CCD profile for the hepatoma
mRNA varied with different stages of tumor growth.
The indication from CCD profiles is that in the hepatoma
there is both diminished synthesis or absence of certain
groups of mRNA's produced by liver, and synthesis of
mRNA's absent from or produced to a much less extent
by liver. Any effects due to variation in quantity of
protein bound would be most noticeable in the first ten
to twenty tubes, since DNA-protein and RNA-protein
complexes tend to remain in the aqueous phase.1 De
tailed chemical and biological characterization is required
to delineate the extent of identity and difference between
the mRNA fractions from liver and hepatoma.
The presence in the BR6 hepatoma of considerable
heteroploidy makes interpretation of the pattern of
messenger RNA synthesis in the tumor somewhat difficult.
The altered patterns of mRNA may have resulted from
variations in the activities of the different chromosomes
in the cells, and the progressive change in the mRNA
pattern may be an indication of the variation of mRNA
synthesis with respect to time. These questions cannot
be clarified at the present time. It is relevant, however,
to note that, although the over-all rate of mRNA syn
thesis in the tumor is apparently greatly diminished com
pared with that in the liver, there appears to be consider
able emphasis on the synthesis of groups of mRNA's
which are present in very small amounts or are absent
in the normal liver. This is strongly suggestive of in
duction of mRNA's not normally produced in liver; a
situation which is unlikely to derive from heteroploidy
alone.
CCD behavior of the hepatoma DNA presents features
which are helpful in understanding the functioning of the
tumor genome. The initial peak (about tube 10) repre
sents DNA in which the strands are almost completely
linked by hydrogen bonds.3 The DNA molecules present
in tubes 20-80 are characterized by an increasing degree
of strand separation which is related to the distance
traveled in the organic phase, the strand separation being
particularly marked in tubes 60-80. There is clearly
much more of this latter material in the normal liver than
in the tumor. In the present solvent system natural
DNA/RNA hybrids, judged by that portion of the mRNA
which is resistant to ribonuclease, were associated with
those molecules of DNA which exhibit a degree of strand
separation and not with the initial DNA peak in which
strand separation is minimal.3 Thus, the DNA pattern
on CCD, together with the observed decrease in rate of
mRNA synthesis, is consistent with the transcription of
a reduced amount of the genome in the hepatomas com
pared with that in the liver. However, uridine incorpora-
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1964 American Association for Cancer Research.
1608
Cancer Research
tion alone may not be an adequate criterion for mRNA
synthesis, since this may be affected by the rate of uptake,
utilization, and pool size of RNA precursors. Polyploidy
could also influence the rate of synthesis of mRNA per
unit DNA. As already discussed, the appearance of
"new" groups of mRNA's suggests that some of those
regions of the tumor genome which are being used for
transcription are different from those of the liver. Further
studies on the regulation of synthesis of individual groups
of mRNA's thus fractionated and their chemical and
biological characterization may be expected to reveal
some of the mechanisms involved in the progressive altera
tion of expression of the liver cell genome in the transition
to hepatoma cell.
ACKNOWLEDGMENTS
Gratitude is expressed to Prof. A. Haddow, F. R. S., for his
interest in this work, to Mr. S. R. Scarfe for histological sections,
to Miss W. House for chromosome counts, and to Mr. E. A. Cook
for excellent technical assistance.
REFERENCES
1. BAUTZ, E. K. F., ANDHALL, B. D. The Isolation of T4-Specific
RNA on a DNA-Cellulose Column. Proc. Nati. Acad. Sci.,
48:400-408, 1962.
2. BOLTON, E. T., ANDMCCARTHY,B. J. A General Method for
Isolation of RNA Complementary to DNA. Proc. Nati. Acad.
Sci., 48:1390-97, 1962.
3. BOTTOMLEY,R. H.; PITOT, H. C.; AND MORRIS, H. P. Meta
bolic Adaptations
in Rat Hepatomas.
IV. Regulation
of
Threonine and Serine Dehydrase.
Cancer Res., 23:392-99,
1963.
4. BRODY, S., ANDYANOFSKY,C. Suppressor Gene Alteration of
Protein Primary Structure. Proc. Nati. Acad. Sci., 60:9-16,
1963.
Vol. 24, October 1964
5. FOULDS, L. Some Problems of Differentiation and Integration
in Neoplasia. In: R. J. C. HARRIS (ed.), Biological Organiza
tion at the Cellular and Supracellular
Level, pp. 229-244.
New York: Academic Press, 1963.
6. HALL, B. D., ANDSPIEQELMAN,S. Sequence Complementarity
of T2-DNA and T2-Specific RNA. Proc. Nati. Acad. Sci.,
47:137-63, 1961.
7. HIATT, H. A Rapidly Labeled RNA in Rat Liver Nuclei. J.
Mol. Biol., 5:217-29, 1962.
8. HOYEH, B. H.; MCCARTHY,B. J.; ANDBOLTON,E. T. Comple
mentary RNA in Nucleus and Cytoplasm of Mouse Liver Cells.
Science, 140:1408-12, 1963.
9. JACOB, F., AND MONOD, J. Genetic Regulatory Mechanisms
in the Synthesis of Proteins. J. Mol. Biol., 3:318-56, 1961.
10. KIDSON, C., AND KIRBY, K. S. Fractionation
of DNA. I.
Countercurrent
Distribution of Native and Degraded DNA.
Biochim. Biophys. Acta, 76:624-27, 1963.
11. KIDSON, C.; KIRBY, K. S.; ANDRALPH, R. K. Isolation Charac
teristics of Rapidly Labelled RNA from Normal Rat Liver. J.
Mol. Biol., 7:312-15, 1963.
12. KIRBY, K. S. The Preparation of Deoxyribonucleic Acids by
the p-Aminosalicylate-Phenol
Method. Biochim. Biophys.
Acta, 36:117-24, 1959.
13. KIRBY, K. S.; HASTINGS, J. R. B.; AND O'SULLIVAN, M. A.
Countercurrent
Distribution
of Ribonucleic
Acids. IV.
Biochim. Biophys. Acta, 61:978-79,. 1962.
14. MONIER, R.; NAONO,S.; HAYES, D.; HAYES, F.; ANDGROS, F.
Studies on the Heterogeneity of Messenger RNA from E. coli.
J. Mol. Biol., 5:311-24, 1962.
15. MONOD, J., AND JACOB, F. General Conclusions: Teleonomic
Mechanisms in Cellular Metabolism, Growth, and Differentia
tion. Cold Spring Harbor Symp. Quant. Biol., 26:389-101, 1961.
16. OTAKA, E.; MITSUI, H.; AND OSAWA, S. On the Ribonucleic
Acid Synthesized in a Cell-free System from Escherichia coli.
Proc. Nati. Acad. Sci., 48:425-30, 1962.
17. SAGIK, B. P.; GREEN, M. H.; HAYASHI, M.; ANDSPIEGELMAN,
S. Size Distribution
of "Informational"
RNA. Biophys. J.
2:409-21, 1962.
FIG. 1.—BR6 hepatoma, showing the presence of glycogen
stained with Best's carmine.
X 475.
FIG. 2.—BR6 hepatoma, after saliva treatment, showing loss of
glycogen staining.
X 475.
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1964 American Association for Cancer Research.
fe *•""
*"
'•**
' .^••V
•'
L^H»m
>*
rf » »»
4t
»u ^
-•
?"• •*
*x
f--r>^£
' "^••Jf«l^.^
¿?
1009
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1964 American Association for Cancer Research.
Recognition of Altered Patterns of Messenger RNA Synthesis in
a Mouse Hepatoma
Chev Kidson and K. S. Kirby
Cancer Res 1964;24:1604-1609.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/24/9/1604
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 16, 2017. © 1964 American Association for Cancer Research.