[CANCER RESEARCH 34, 643-653, March 1974]
Transfer RNA Base Composition Studies in Morris Hepatomas and
Rat Liver1
Erika Randerath, Li-Li S. Y. Chia, Harold P. Morris, and Kurt Randerath2
Department of Pharmacology, Baylor College of Medicine, Texas Medical Center, Houston, Texas 77025 [E. R., L. S. Y. C., K. /?.], and Department of
Biochemistry, Howard University, School of Medicine, Washington, D. C. 20001 [H. P. M.]
SUMMARY
The major and modified base composition of cytoplasmic
tRNA from Morris hepatomas 7777 and 5123D and from
rat liver has been determined for 16constituents using a new
chemical tritium derivative method. Analysis of the total
base composition of the tumor tRNA's did not reveal a
general overmodification or overmethylation, but the total
modified and methylated base content was found to be
slightly lower than in tRNA from normal liver and host
liver. As shown by analysis of the modified base composi
tion, hepatoma 7777 tRNA contains less 5-methyluridine,
pseudouridine, 3-methylcytidine, and 5-methylcytidine and
more dihydrouridine and 1-methylguanosine than does
liver, and hepatoma 5123D contains less dihydrouridine
and 5-methylcytidine and more /V2,Ar2-dimethylguanosine
and X (an unidentified modified nucleoside) than does liver.
Differences ranged from 2 to 17%, Ar2,/V2-dimethylguanosine and X showing the greatest deviations in hepatoma
5123D. For the major nucleosides, differences of 2 to 4%
were observed. Both tumor RNA's contain more uridine
and less cytidine, and hepatoma 7777 has in addition less
adenosine and more guanosine. No unusual modified bases
were detected in these tumors by the Chromatographie
procedures used. A comparison of the two tumor tRNA's
does not provide evidence for a distinct correlation between
degrees of modification and methylation of tRNA and
growth rates and histological characteristics of the tumors.
The following possibilities exist for explaining the ob
served base composition differences: (a) changes in concen
trations of specific tRNA's; (b) presence in the tumors of
tRNA's with altered sequences; and (c) aberrant modifica
tion.
INTRODUCTION
Although the general mechanism of protein synthesis in
cancer cells appears not to differ from that in normal cells,
specific components of the protein-synthesizing system may
conceivably be altered and such alterations may be involved
1Supported by American Cancer Society Grant NP-37, USPHS
Grants CA-13591 and CA-10729 and USPHS Center Grant CA-10893.
2Recipient of American Cancer Society Faculty Research Award
PRA-108. To whom requests for reprints should be addressed.
Received September 10, 1973; accepted December 3, 1973.
MARCH
in carcinogenesis. A component of the protein-synthesizing
system particularly likely to be affected is tRNA because of
the multitude and complexity of processes leading to the
formation of a mature tRNA population. Changes in the
structure of tRNA can be expected to have a profound
influence on protein synthesis and structure, since tRNA
probably is more responsible than any other macromolecule
for the correct translation of the genetic code into the amino
acid sequence of proteins.
During the past decade a considerable amount of indirect
evidence has been accumulated in an effort to support the
assumption that tRNA may be altered in tumor cells. In
particular, the activity of tRNA methyltransferases (for
reviews, see Refs. 7, 22, and 52) has been shown in many
laboratories to be increased severalfold in extracts from a
variety of neoplastic tissues including virally transformed
cells and chemically induced tumors (for a recent review, see
Ref. 6). Elevated methylase levels have been assumed to
lead to overmethylation of RNA and to be associated with
carcinogenesis (48, 49). This hypothesis was supported by
the fact that the alkylating carcinogen dimethylnitrosamine
leads to a methylation of nucleic acids in vivo (27). Indirect
evidence for changed tRNA species in various neoplastic
cells has also been provided by comparative investigations
of the column Chromatographie profiles of aminoacylated
tRNA's (for reviews, see Refs. 6 and 53).
Although there is considerable indirect evidence for
abnormal tRNA in tumors, little is known about chemistry
and structure of tRNA in normal and neoplastic mamma
lian cells. The question of whether there is excessive
methylation of tumor tRNA in vivo has been studied to a
much lesser extent than the levels of the tRNA methylases
have been, a few reports claiming substantial overmethyla
tion of tRNA in tumor tissues (4, 31, 55-57) and others
finding no overmethylation (1, 18, 33, 37).
Direct investigation of mammalian tRNA structures has
been impeded by the multitude of relatively similar species
present in the cell and the necessity to work with large
amounts of tissue because of the lack of sufficiently sensitive
analytical methods. Most of the structural work has been
carried out with bacterial tRNA thus far, using procedures
based on in vivo 32P-labeling of nucleic acids as described
by Sanger et al. (46) as reviewed by Barrell (2). This ap
proach, however, is not feasible in mammalian cells, unless
they can be grown in culture. In vivo labeling of tRNA with
methionine-I4C has been used to analyze methylated bases
(12, 17), but this approach does not allow one to determine
1974
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
643
E. Randerath,
L. S. Y. Chia, H. P. Morris, and K. Randerath
RNA constituents other than methylated bases.
A recently developed method (34, 35, 39 42) for base
analysis of RNA, which is based solely on chemical
postlabeling of nonradioactive RNA derivatives, thus not
requiring in vivo labeling, is suitable for assaying the 4
major and most modified bases in mammalian tRNA. This
tritium derivative method has since been used for base
composition analysis of various RNA's, e.g., tRNA from
human brain and brain tumors (33, 37) and other normal
and neoplastic mammalian tissues,3 tRNA from avian liver
and avian leukemic cells and the avian myeloblastosis virus
(43), tRNA and rRNA from Mycoplasma hominis (19),
and various nuclear and nucleolar RNA's from Novikoff
hepatoma cells (44).
In this communication we report tRNA base composition
data for Morris hepatomas 5123D and 7777 and rat liver
obtained by the tritium derivative method. Part of this
work has been reported in a preliminary form (32). Since
the tRNA methyltransferase activity has been described to
be 3.7 and 5 times higher in extracts from hepatomas
5123D and 7777, respectively, than in those from normal
liver (47) and since alterations of column Chromatographie
profiles of aminoacylated tRNA's have been reported for
several hepatomas, including hepatoma 5123D (50, 58), it
was of particular interest to compare the degree of methylation of tRNA from normal liver and these tumors. It was
also of interest to look for differences in terms of individual
bases or the presence of unusual bases in the tumors, which
had to be regarded as a distinct possibility in view of the
fact that bacterial tRNA's contain modified bases that do
not occur in mammalian tRNA (14, 35).
animals and used immediately or stored frozen at -70 to
-80°.tRNA was isolated from 2 to 3 g of tissue taken from
1 liver and from 2 tumors of the same animal. Supernatant
solutions (100,000 x g) of tissue homogenates were pre
pared, from which tRNA was isolated by precipitation at
pH 5 and subsequent phenol treatment. Preparation of the
high-speed supernatant solution and precipitation at pH 5
were done by minor modifications of procedures originally
described by Keller and Zamecnik (21) and Blobel and
Potter (5). The complete procedure has been described in
detail previously (10).
Purification of RNA by Polyacrylamide Gel
Electrophoresis. For purification, the crude tRNA was
subjected to polyacrylamide gel electrophoresis on 10%gels
at pH 7.8 using a Tris/sodium acetate/EDTA buffer similar
to that described by Loening (26). Staining of the gels was
with méthylène
blue essentially according to the procedure
of Peacock and Dingman (30). The 4 S RNA band was ex
cised from the gel and subjected to homogenization and
phenol treatment. The RNA was finally isolated from the
aqueous phase, which had been concentrated under vac
uum, by precipitation with acetonitrile/ethanol. The details
of the procedure have been described (10).
Preparation of tRNA from 15,000 x g Pellets. Pellets
(15,000 x g) were obtained in the course of preparing
100,000 x g supernatant solutions of tissue homogenates
(10). Nucleic acids including tRNA were isolated from the
low-speed pellets by phenol extraction following essentially
the same procedure as described for the extraction of tRNA
from pH 5 precipitates (10). In order to separate tRNA
from rRNA's and DNA, the nucleic acid preparation was
subjected to polyacrylamide gel electrophoresis using
stacked gels (10 and 3.5% gels) and a buffer similar to that
MATERIALS AND METHODS
described by Loening (26). Isolation of 4 S RNA from the
gels was performed as described for cytoplasmic tRNA
Materials. Morris hepatomas 7777 and 5123D were (10).
originally induced by Dr. H. P. Morris and continuously
Enzymatic Digestion of RNA and Tritium Labeling of
transplanted i.m. Female Buffalo rats (150 to 250 g) were Digests. The purified 4 S RNA was digested to nucleosides
used for transplants and as a source for normal liver. using a mixture of RNase A, snake venom phosphodiesterHepatomas 7777 and 5123D have growth rates of 5.0 and ase, and alkaline phosphatase as described previously (35,
3.7 cm/month and are histologically "poorly differenti
40). The solution of /V,Ar-bis(2-hydroxyethyl)glycine buffer
ated" and "intermediate between well differentiated and
poorly differentiated," respectively (47). Apparatus used for used for preparing the enzymatic digest (40) was freshly
prepared, since storage of this buffer in the frozen state
tissue and gel homogenization included a Heidolph "In- (-20°)
led to changes in its properties that interfered with
framo" stirrer (Will Scientific, Cambridge, Mass.) and
the subsequent tritium-labeling procedure. Incubation of the
Teflon pestle homogenizers (A. H. Thomas Co., Philadel
digests was for 6 hr at 37°.After appropriate dilution, the
phia, Pa.). Triton X-100â„¢ was from Beckman Instru
enzymatic digests were directly subjected to periodate
ments, Fullerton, Calif. Chemicals for gel electrophoresis oxidation and subsequent reduction with KBH4-3H (about 2
and a Model EC-474 standard vertical gel apparatus were
as described, which results in the formation of
from EC-Apparatus Corp., St. Petersburg, Fla. Enzymes Ci/mmole)
tritium-labeled nucleoside trialcohols (39, 40). Labeled
and chemicals used for digestion of tRNA and labeling of digests were stored at -70 to -90° and analyzed within 2
the digests, as well as materials for chromatography and weeks. All operations involving KBH4-3H were carried out
liquid scintillation counting, have been described elsewhere under a well-ventilated hood. It was important to use a
(35, 40, 42).
potassium borotritide solution not older than about 4
Isolation of tRNA from Tissues. Tissue was excised months (storage at -70 to -90°). The use of older
quickly from etherized rats immediately after bleeding the borotritide solutions or of solutions that had been subjected
to repeated freezing and thawing may result in incomplete
3K. Randerath, S. K. MacKinnon, and E. Randerath, unpublished reduction of the nucleoside dialdehydes to tritium-labeled
"monoaldehydes" (40), particularly in the case of dialdeexperiments, 1971 to 1972.
644
CANCER RESEARCH VOL. 34
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
iRNA Base Composition in Morris Hepatomas and Rat Liver
hydes derived from pseudouridine, guanosine, and adenosine. These compounds travel in close proximity to the
corresponding trialcohols on the cellulose maps under our
conditions (40).
The exact adjustment of the borohydride concentration
was found to be crucial for the labeling reaction. When
borohydride concentrations lower than that recommended
(35, 40, 42) were used, the formation of "monoaldehydes"
was observed, similarly as with aged borohydride solutions.
Too high borohydride concentrations may cause a pro
nounced loss in hU4 but also losses in m3C and m'G. With
several batches of borotritide, variations in the concentra
tion of active borotritide have been observed, leading either
to formation of li'-monoaldehyde or losses in hU (both up to
about 15%). The reducing activity of a borohydride solution
may be checked and optimum conditions established by
labeling a periodate-oxidized nucleoside digest derived from
a reference preparation of tRNA at several borohydride
concentrations and analyzing the nucleoside trialcohols
affected, particularly i£<
and hU.
Analysis of Labeled Digests. Tritium-labeled digests ( 1 to
5 /¿Ci)were applied to cellulose thin layers and subjected to
2-dimensional chromatography as described previously
(35). Tritium-labeled compounds were detected on film by
fluorography (36). Exposure usually was for 2 to 4 days.
Compounds were located on the chromatograms, cut out,
eluted, and assayed by liquid scintillation counting as
described (35, 40, 42). The base composition was calculated
from the count rates according to
/*
=
cpm
x 100%
, cpm,
where N is the number of radioactive nucleoside derivatives
separated and cpm¡is the count rate of an individual
compound (35, 40, 42). For each labeled digest, 4 to 6
analyses were carried out.
RESULTS
Requirements for Comparative Investigations on tRNA.
For accurate determination of degrees of modification and
methylation of tRNA preparations from various tissues, it
was necessary to free the tRNA preparation of any
contaminating RNA [such as 5 S RNA, 4.5 S RNA,
various low-molecular-weight nRNA's (45), and highmolecular-weight rRNA]. As described earlier (10), purifi
cation of the RNA by polyacrylamide slab gel electrophoresis, followed by extraction of the 4 S band, leads to RNA
preparations that are pure in terms of molecular weight and
completely free of contaminating RNA. It was also advan-
tageous to isolate the RNA by precipitation at pH 5 (21)
rather than by phenol extraction of the tissues (9), since the
former procedure gave cleaner preparations from the tissues
investigated as judged by analytical gel electrophoresis (10).
tRNA purified by phenol extraction contained varying
amounts of background material which overlapped the 4 S
RNA band on polyacrylamide gels and consisted of degra
dation products of high-molecular-weight RNA, whereas
tRNA isolated by pH 5 precipitation was completely free of
such background material. The 4 S RNA band of tRNA
isolated by pH 5 precipitation usually represented about 85
to 90% of the total RNA (10). The residual material
consisted of several faint bands of low-molecular-weight
RNA, e.g., 5 S RNA and 4.5 S RNA, and traces of rRNA
(10). Thus polyacrylamide gel electrophoresis proved to be a
valuable means to remove all impurities contained in the
crude tRNA preparation. As shown previously (10), polya
crylamide gel electrophoresis may be successfully combined
with subsequent base analysis by the tritium derivative
method.
Another important factor in comparative investigations is
the integrity of the RNA to be analyzed. As judged by
polyacrylamide gel electrophoresis, tRNA prepared by pH
5 precipitation from Morris hepatomas 7777 and 5123D and
liver contained <3% polynucleotide material moving ahead
of 4 S RNA. However, even under strictest precautions,
undegraded tRNA cannot be isolated from certain Morris
hepatomas, e.g., hepatoma 9121.5 tRNA derived from such
hepatomas is thus unsuitable for comparative investiga
tions. As a control, tRNA was also isolated directly from
low-speed suprnatant solutions of hepatomas 7777 and
5123D and liver by phenol treatment, precipitation of the
RNA from the aqueous phase, and gel electrophoresis5 in
order to exclude the possibility of degradation and other
artifacts during the pH 5 precipitation procedure. Identical
base compositions were obtained irrespective of the mode of
preparation.
The 4 S RNA isolated from pH 5 precipitates was found
to represent about 50% of total cellular 4 S RNA, the rest
being present in the low-speed (15,000 x g) pellets and
consisting of membrane and ribosome-bound tRNA, as well
as mitochondrial and nuclear tRNA. We have also carried
out base composition studies on tRNA isolated from such
low-speed pellets of the same tissues as well as rat liver
mitochondria5 and shall refer to these results in connection
with the base composition data for tRNA's from pH 5
precipitates.
During the course of our work it was found that
experiments must be conducted in parallel if the 2 RNA
preparations to be compared exhibit only small differences.
In particular, the labeling reactions must be carried out
simultaneously using the same borotritide preparation, and
conditions leading to the formation of monoaldehydes must
be avoided (see "Materials and Methods"). It is also
essential to cut the nucleoside trialcohol spots from the
4The abbreviations used are: hU, dihydrouridine; m3C, 3-methylcytidine; m'G, 7-methylguanosine; \L, pseudouridine; tA, ¿V-[9-(|3-D-ribofu- chromatograms in an identical manner. This could best be
ranosyl)purin-6-yl carbamoyljthreonine; m5U, 5-methyluridine; m'C, 5methylcytidine; m'A, 1-methyladenosine; m22G, A^-W-dimethylguano5K. Randerath, L. S. Y. Chia, and E. Randerath. unpublished
sine; m'G, 1-methylguanosine; m'A, jV-methyladenosine.
experiments, 1972 to 1973.
MARCH 1974
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
645
E. Randerath,
L. S. Y. ChÃ-a,H. P. Morris, and K. Randerath
achieved if the same person did the particular experiment.
Chromatographyof Tritium-labeled tRNA Digests. Fig. 1
shows fluorograms (36) of 2-dimensional thin-layer chromatograms of tritium-labeled nucleoside trialcohols (35, 40)
obtained by digestion and labeling of tRNA of Morris
hepatoma 5123D (lop) and host liver (boitom). The RNA
rr^A'
m'A1
m'G1
B
hU1
gly
m'A1
m'G'
Fig. 1. Fluorograms of tritium-labeled digests of hepatoma 5123D
tRNA (lop) and host liver tRNA (bottom) chromatographed on cellulose
thin layers. To each chromatogram were applied 4.8 jiCi of labeled digest.
Development with Solvents A and B of Ref. 35. Fluorography according to
Ref. 36. Exposure of films for 3 days. B, background, not derived from
tRNA; X, unidentified modified nucleoside originating from tRNA; gly,
glycerol (35); m'A', trialcohol of /V'-methyladenosine formed from m'A
under labeling conditions (35); \i-D, decomposition product derived from
pseudouridine; U', A', C', G', msV, etc., nucleoside trialcohols of uridine,
adenosine, cytidine, guanosine, m5U, etc; m2G, 2-methylguanosine.
646
was prepared by extracting the 4 S band from polyacrylamide gels following isolation by pH 5 precipitation (21) (see
"Materials and Methods"), Both maps (Fig. 1) show the
same spots, spot intensities closely resembling each other,
thus indicating an overall similarity of the base composi
tions. There is no evidence for the presence in the tumor of
any additional, unusual bases. Very similar patterns had
been observed previously for tRNA from human brain and
brain tumors (33, 37), chicken liver and avian myeloblasts
(43), and various mammalian tissues,3- 5 whereas labeled
digests of tRNA from the avian myeloblastosis virus (43)
and from bacteria (35) showed considerably different spot
patterns and quantitative results. Fluorograms obtained
from hepatoma 7777 and host liver tRNA were also very
similar to each other and to those depieted in Fig. 1, again
indicating an overall similarity of the base compositions.
Comparison between tRNA Base Composition of Hepa
toma 7777 andLiver. In Table 1 the total base composition
of tRNA from Morris hepatoma 7777 is compared with
that of host liver. Sixteen compounds were analyzed as
trialcohol derivatives by the tritium derivative method
(35, 40) including most base-modified and all base-methyl
ated nucleosides known to occur in mammalian tRNA.
2'-0-Methylated nucleosides, thio derivatives, and N6(A2-isopentenyl)adenosine have not been assayed for
reasons discussed elsewhere (35). tA (16, 29) (nucleoside
trialcohol tA' on Fig. 1) was not analyzed, since only after
completion of the work reported here was the nucleoside
trialcohol of this compound prepared by derivatization of
an authentic sample of tA.5 We have, however, analyzed
the radioactive derivative of an unidentified modified nu
cleoside X (Fig. 1), which appears to be identical with a
nucleoside occurring next to m'G in several Escherichia
coli tRNA species of known sequence (3, 11, 28, 59, 60).
This was shown by subjecting several purified E. coli
tRNA's known to contain X (tRNAmMet, tRNAPhe,
tRNA! Arg) to tritium base composition analysis.5 Since
some other tRNA species from E. coli as well as from yeast
contain uridine or hU instead of X in the same position, X
is assumed to be a derivative of uridine (60).
As can be seen from Table 1, the base composition of
hepatoma 7777 tRNA is similar to that of host liver tRNA.
There are, however, small but statistically significant
differences in the 4 major and 5 modified constituents
ranging from 2 to 10%, the greatest differences being in the
content of m5U (10%), m3C (8%), and m5C (7%). All
modified constituents significantly altered in the tumor
tRNA, i.e., m5U, ^, m'A, m3C, and m5C, were found to be
lower than in liver tRNA. A slight undermodification of the
tumor tRNA is reflected by the ratios 2modifiednucieosides/
'major
nucleosides
and 2 modified
nucleosides
/
al
nucleosides
which are about 3% lower for the tumor tRNA. The hepa
toma tRNA is also slightly undermethylated, the corre
sponding ratios being again about 3% lower. Regarding the
major constituents, the tumor tRNA contains more cyti
dine and guanosine and less adenosine and uridine than
does liver tRNA, the differences being 2 to 4%.
We have also investigated the total base composition of
tRNA from livers of normal Buffalo rats.5 This analysis
CANCER RESEARCH VOL. 34
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
iRNA Base Composition in Morris Hepatomas and Rat Liver
Table 1
Total base composition
of tRNA
from Morris hepatoma
7777 and host liver
tRNA was prepared from pH 5 precipitates of high-speed supernatant solutions of tissue homogenates. Base analysis was carried out by chemical
tritium derivatization of nucleosides in en/ymatic digests (40). Base composition data were calculated from count rates of tritium-labeled nucleoside
trialcohols.
7777
compared with
7777/
host
liver"©©©4$©t©©Î;îÎi-^hepa
•¿Xhost
hver0.9830.9561.0311.0220.9000.9
NucleosidesU
ridineAdenosineCytidineGuanosinem5UhUìtm'A'Inosinem3C'm5Cm/Gm'G«m'Gm'G"XHepatomaX13.98316.61826
56828.4600.5152.8683.4551.2600.3300.3151.8030.5821.2030.8420.8030.3887777V0.07200.06370.13800.08370.01640.04020
.
12040.18650.01170.02500.01970.01510.03210.00520.06110.01720.03930.0498
99.993
Total
2l
100.003
nucleosides
o©
v
/ —¿major
1678
0.08560.1437 00 00104001
V—
11
modified nucleosides // —¿total
n
V methylated nucleosides/ / —¿totalnucleosidesucleosldes
V
0.07330
—¿
000800
0.00151
nucleosides0.
modified
/VV
V
00 08861477
00.0014600124
0.010.001
0©
07550.
0.1733
001140.001
0.00169
0.01(D
©
966.973
0
0969
971
" x = mean of 6 Chromatographie analyses: for calculation, consult "Materials and Methods."
" Standard deviation s\ = \/[Z(x - x)']/(N - 1).
[NiN¿Nt + N, - 2)1
' Probability values estimated on the basis of Student's / test: t
Ni + N,
a Value for hepatome > value for liver represented by J. Value for hepatoma < value for liver represented by ¡.Statistically significant differ
ences (p < 0.05) are circled.
' The value for m'A was obtained from the sum of the count rates of the trialcohols-3H of m'A and m'A (35. 40).
Corrected for 85% recovery (35).
"m2G, 2-methylguanosine.
" Corrected for 65% recovery (35).
showed statistically significant differences for the same
major and modified nucleosides as the comparison between
hepatoma 7777 and host liver tRNA. A comparison be
tween tRNA from normal liver and host liver (hepatoma
7777) showed great similarity in terms of base composition,
m5U, m5C, m22G, and X being 5 to 8% lower in host
liver tRNA than in normal liver tRNA. No significant dif
ferences were found for any other nucleoside.
By excluding the major constituents from the calcula
tions, one is able to compare the modified base composi
tion of tRNA from various samples. Table 2 shows such a
comparison for hepatoma 7777 and host liver tRNA. The
values for 4 modified pyrimidines, including \L and the 3
base-methylated nucleosides m5U, m3C, and m5C, are
slightly but significantly lower in the tumor tRNA, whereas
m'G and hU are slightly elevated. Differences are again
relatively small, the greatest difference being 11% for m'G.
Similar results with regard to the modified base composi
tion were obtained with normal liver tRNA.
Comparison between tRNA Base Compositions of Hepa
toma 5123D and Liver. Table 3 compares the total base
composition of hepatoma 5123D and host liver tRNA.
Again, there is great overall similarity in the 2 base com
positions with differences ranging from 1 to 16%, being for
the most part < 10%. Two major nucleosides (uridine and
cytidine) and 6 modified nucleosides (hU, m'A, m3C,
m5C, m22G, and X) show statistically significant differ
ences. Four modified constituents (m5U, m'A, m3C, and
m5C) are lower and 2 modified constituents (m22G and X)
are higher in the tumor tRNA, the greatest differences
observed being 13 and 16% in m22G and X, respectively. In
spite of the relatively high values for these 2 nucleosides in
the tumor tRNA, the ratios 2modlflednucieosides/2majl,rnu.
cleosides
and
"modified
nucleosides
/ "total
nucleosides
are
1OWCT
(about 3 and 2%, respectively) for the tumor tRNA. The
ratiOS
¿methylated
nucleosides
/ "major
nucleosides
IS alSO
aoOUt
3% lower. The tumor tRNA contains more cytidine and
less adenosine than does liver tRNA, these differences be
ing about I to 2%.
In Table 4 the modified base compositions of Morris
MARCH 1974
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
647
E. Randerath. L. S. Y. ChÃ-a.H. P. Morris, and K. Rande rath
Modified
For explanatory
base composition
of lK.\'A
Table 2
from Morris hepaloma
7777 and host liver
text, consult Table 1.
*7"7~7~THepatoma
1
l'itnm
i
////
/•*ho»l7777
compared
withhost
liver"©©©iTT4©îi•^hepaloma
Nucleosidesm5UhU<Lm'A'Inosinem'C'msCm22Gm'G"m'Gm'G"XHepatomaXa(%)3.58719.97024.0178.7672.3022.20212.5484.0558.3975.8555.6032.700miI*
liver0.9271.0220.9840.9931
.0440.9410.9511
17830.15860.31850.32870.1521HostX<%)3.86819.53324.4108.8302.2052.34013.1933.8708.4335.2755.3282
.0480.9961.1101.0520.998
. 705liverf|0.07940.10760.33250.11210.20610.03580.34510.09200.20910.29780.30810.2112P'0.0010.010
100.003
Total
99.990
"'" See Table 1, Footnotes a to h.
Table 3
For explanatory
Total hase composition
text, consult Table 1.
of tRNA
from Morris hepaloma
5123D and host liver
5I23Dcompared
withp'
5123D-*ho8t
liver"0.05 host
NucleosidesUndineAdenosineCytidineGuanosinem5UhU<l>m'A'Inosinem3C'm5Cm22Gm'G8m'Gm'G"XHepatomajt"(%)14.02516.92826.72027.9080.5852.5433.5251.2
hver0.9880.9901.0171.0080.9510.9510
Q0.3
i0.05
©0.2
T0.2
i0.01
©0.1
i0.05
Q0.6
02990.01260.05480.03320.01000.03680.02380.0096Host—(%)14.20017.09326.28027.6750.6152.6733.5951.2
.
i0.05
©0.02
|0.01
©>0.90.4
1291.0000.9721.0001.157
17670.15680.07140.03110.04270.05070.00960
i>0.90.001
©•^hepaloma
Total
100.005
99.999
16850.08870.14420.07590.000001480012000111000930.17300.09120.14750.07780,000018000117001290008800,0,002010510§
nu¿methylatedy
"modified
inucleoside«
/ ¿major
¿majocleosides/
nu¿methylatedcleosidea
*modified
ninucleoaides/¿total
nucleoidesje
idesnucleoaides0.
leus
0.976
/ ¿totallut-leosidesr
"x = mean of 4 Chromatographie analyses: for calculation, consult "Materials and Methods.'
"" See Table I, Footnotes b to h.
hepatoma 5I23D and host liver tRNA are compared. Four
of the modified constituents investigated are significantly
different in the tumor tRNA, hU and m5C being lower and
m22G and X being higher than in liver tRNA. Differences
are relatively small, m22G and X show the greatest devia
tions (15 and 17%, respectively).
A comparison of total and modified base compositions
648
of
normal
Buffalo
rat liver tRNA5
with hepatoma
5123D
tRNA gave essentially similar results, again indicating
great similarity between normal and host liver tRNA. Only
X was found to be slightly (6%) but significantly lower in
host liver when host liver and normal liver tRNA were
compared.
In terms of base composition
normal liver
tRNA resembles more closely host liver tRNA from ani-
CANCER RESEARCH VOL. 34
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
tRNA Base Composition in Morris H epatamos and Rat Liver
Table 4
Modified base composition of ¡RNAfrom Morris hepatoma 5123D and host liver
For explanatory text, consult Table 1.
5123DNucleosidesm5UhU$m'A'Inosinem3C'm5Cm22Gm'G8m'Gm'G"XTotal1?(%)4.05317.63324.4838.6181.9202.13313.9454.6758.7735.3605.1033.31010
Hepatoma
ÇI'ÎIPÕriepdiomd
JIZJLI
withpc
compared
5I23D*hont
liver00.4
host
liver0.9730.9741.0041.0020.9820.9610.9451.1541.0230.99
-L0.05
Q0.8
Õ0.8
T0.8
10.2
40.05
©0.01
©0.1
t0.9
i0.5
"}0.001
©^hepatoma
"x = mean of 4 Chromatographie analyses; for calculation, consult "Materials and Methods."
'•*
See Table 1, Footnotes ft to h.
well reproducible if the guidelines described under "Re
quirements for Comparative Investigations on tRNA"
are followed.
The observed differences in base composition between
the tumor and host liver tRNA appear to be stable charac
teristics of each tumor cell line. For hepatoma 7777, the
alterations were observed for transplant generations 62
vidual nucleosides, there are differences. Thus, in hepa and 88, which were about 2.5 years apart.3- 5 The charac
toma 7777 tRNA, uridine, adenosine, m5U, \b, m'A, m3C, teristic base pattern for hepatoma 5123D was also found in
and m5C are lower and cytidine and guanosine are higher different generations of the tumor.
tRNA from 15,000 x g Pellets of Liver and Hepatomas
when compared with host liver tRNA (Table 1); and in
hepatoma 5123D, uridine, hU, m'A, m3C, and m5C are 7777 and 5123D. The data presented (Tables 1 to 4) per
lower and cytidine, m22G, and X are higher (Table 3). Both tain to cytoplasmic tRNA obtained from pH 5 precipi
tumor tRNA's are slightly undermodified and undermeth- tates only, which represents about 50% of the total cellular
ylated to about the same degree. Differences between the 2 tRNA. Since the possibility existed that the residual tRNA
tumor tRNA's are also apparent when the data for the of the tumors might be overmodified or overmethylated,
modified base compositions are compared (Tables 2 and 4). we have also investigated the base constituency of tRNA
Hepatoma 7777 tRNA contains less m5U, x£,
m3C, and preparations from low-speed (15,000 x g) pellets,5 which
m5C and more hU and m'G when compared with host liver contain most of the residual cellular tRNA (about 45%).
tRNA, whereas hepatoma 5123D tRNA is poorer in hU
A comparison of tumor tRNA with liver tRNA isolated
and m5C and richer in m22G and X. The comparison of the from the respective low-speed pellets revealed the fol
data for the 2 tumors does not provide evidence for a dis lowing characteristics.
tinct correlation between degrees of modification and meth1. The pellet tRNA of the tumors is considerably more
ylation of tRNA and growth rates (and histological charac
undermodified and undermethylated than the correspond
teristics) of the tumors. tRNA of the fast-growing hepa ing cytoplasmic tRNA. The degrees of modification and
toma 7777 is not more methylated or modified than tRNA methylation for hepatoma 7777 and 5123D pellet tRNA
of hepatoma 5123D, which has a much slower growth rate. are both about 11 and 14% lower, respectively, than those
Reproducibility of Base Composition Data. tRNA for pellet tRNA from liver, whereas the cytoplasmic tRNA
from hepatomas 7777 and 5123D as well as from livers from the tumors is only about 2 to 3% undermodified and
has been prepared and subjected repeatedly to enzymatic undermethylated when compared with liver cytoplasmic
digestion and tritium labeling. In each instance significant tRNA (Tables 1 and 3).
differences between hepatoma and liver tRNA were ob
2. In terms of the total base composition, differences
tained for the same nucleosides, and the ratios of for individual modified nucleosides between the pellet
tRNA's of the tumors and liver are generally somewhat
liv
•¿^hepatoma
7777/
and x hepatoma 5 123D/*host
Tables 1 to 4) were found to be essentially the same. No greater than those between the respective cytoplasmic
differences were found between tRNA preparations ob tRNA's. Some differences amount to about 20 to 25%;
tained from frozen and fresh tissues. In general, data are for example, in pellet tRNA of hepatoma 7777, m3C was
mais bearing the slowly growing hepatoma 5123D than it
does host liver tRNA from those bearing the rapidly grow
ing hepatoma 7777.
Comparison of Hepatoma 7777 and 51231) Data. Base
composition of the 2 tumor tRNA's are quite similar. A
comparison of the tumor tRNA's with the respective host
liver tRNA's shows, however, that, with respect to indi
MARCH
1974
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
649
E. Randerath,
L. S. Y. ChÃ-a,H. P. Morris, and K. Randerath
found to be 24% lower than in the corresponding liver prep
aration.
3. Pellet tRNA's and cytoplasmic tRNA's from the tu
mors show the same trends. m22G and X, which were
found to be elevated in hepatoma 5123D cytoplasmic
tRNA (Table 4), are also elevated in hepatoma 5123D
pellet tRNA. m5U and m3C, which are lower in hepatoma
7777 cytoplasmic tRNA (Table 2), are also lower in pellet
tRNA from this tumor.
Comparison
of pellet tRNA from any of the tissues
investigated with the corresponding
cytoplasmic
tRNA
shows the degrees of modification and methylation to be
generally lower for pellet tRNA.5 This difference was
considerably
more pronounced in the tumors. Whereas
the degrees of modification
and methylation
of pellet
tRNA of liver are about 3 to 6% lower than those of liver
cytoplasmic
tRNA, the pellet tRNA's
of both tumors
contain about 13 to 15% less modified and methylated
nucleosides
when compared
with tumor cytoplasmic
tRNA.
DISCUSSION
In view of the conflicting reports in the literature on
tRNA methylation in tumor tissues, it appeared desirable
to investigate the chemistry of tRNA in cancer cells in a
systematic
fashion. Since the usual spectrophotometric
techniques are not sensitive enough for the analysis of the
small amounts of material available in many instances
from mammalian sources and in vivo labeling of mamma
lian tRNA does not yield the high specific radioactivity
necessary for structural studies (8), the development of new
methods for base composition (35, 39-42) and sequential
(38) analysis was required. These methods are based on the
chemical introduction of tritium label into derivatives of
RNA. The results presented in this communication
were
obtained by applying such a method to a study of the base
constituency of tRNA in liver and 2 Morris hepatomas.
These tumors are well suited for comparative studies be
cause the cell of origin (hepatocyte) is known and availa
ble in adequate amounts for control purposes. We have
chosen the particular tumors (hepatomas 7777 and 5123D)
for 2 main reasons: (a) of several Morris hepatomas in
vestigated, these tumors exhibited the highest activities of
tRNA methylases (47); (b) in contrast to other tumors,
including certain Morris hepatomas,
undegraded tRNA
could be readily obtained.
In first attempts to study the overmethylation
question,
our laboratory concentrated on the reinvestigation of tRNA
from human brain and brain tumors (33, 37), in an effort
to reproduce and extend data published by Viale et al. (31,
55, 57). Our results, however, did not confirm the substan
tial elevation of various methylated nucleosides in tumor
tRNA [e.g., an 80-fold increase in m'G for glioblastoma
multiforme tRNA (56)] reported by these authors. Instead,
our data revealed only slight differences between the over
all modified base composition of human brain tRNA and
brain tumor tRNA, and no evidence was obtained for a
substantial increase in the ratios 2modined nucleosides/2major
650
nucleosides
and
'-'methylated
nucleosides
/ ^major
nucleusides
•¿
base composition data were, however, similar to the data
reported by Iwanami and Brown (18) and Baguley and
Staehelin (1) for tRNA from other malignant sources. The
results reported in the present communication on cytoplas
mic tRNA from Morris hepatomas and liver are very sim
ilar to our earlier data obtained for cytoplasmic tRNA
from human brain and brain tumors.
During the course of these earlier investigations, it be
came apparent that it was of great importance to isolate
tRNA in a pure and undegraded form in order to exclude
artifacts caused by contamination
of the preparations with
other RNA species and degradation
products of highmolecular-weight
RNA or by tRNA loss as a result of
nuclease degradation.
These points were carefully con
sidered in the present investigation. The results of investi
gations on the chemistry of tRNA in which purity and
integrity of the preparation were not carefully evaluated
appear to be of doubtful value. The substantial overmethyl
ation of tumor tRNA reported for mouse tumors (4) and
human brain tumors (31, 55, 57) may thus conceivably be
artifacts of preparation.
This appears to be particularly
likely in view of the difficulties in obtaining undegraded
RNA from many malignant tissues.
Our data regarding the overall similarity of hepatoma
and liver tRNA are consistent with results reported by
Sheid et al. (47) showing a general similarity of in vitro
methylation patterns obtained with E. coli B tRNA sub
strate and tRNA methylase preparations
from rat liver
and 4 Morris hepatomas (7777, 5123D, 7316B, and 8995).
Klagsbrun (23), in a recent report on the methylation of
tRNA in normal and SV40-transformed
3T3 cells, also
finds a contrast between the in vivo and in vitro situation
insofar as tRNA methylase activity was found to be greater
in extracts from transformed cells than in extracts from
control cells, but the degree of methylation and methylated
base composition
of normal and transformed
3T3 cell
tRNA (determined by methylation in vivo) appeared not to
differ measurably.
Interestingly,
as pointed out under "Results,"
tRNA
isolated from 15,000 x g pellets of the hepatomas was
found to be undermodified
and undermethylated
to an
even greater extent than cytoplasmic
tRNA from the
tumors. The lack of modified and methylated nucleosides
in pellet tRNA of liver when compared with cytoplasmic
tRNA of liver can for the most part be accounted for by
the presence in the pellet tRNA preparations of mitochondrial tRNA. The base constituency of mitochondrial tRNA
of rat liver, which represents 20 to 30% of the pellet tRNA,
was found to be quite different from that of cytoplasmic
tRNA.5 For example, the degree of modification of mito
chondrial tRNA is 0.10, and that of cytoplasmic tRNA is
0.17. The corresponding values for the degree of methyl
ation are 0.052 and 0.088, respectively. At the present time
we are, however, unable to explain why the pellet tRNA
of the tumors is considerable
more undermodified
and
undermethylated
than its cytoplasmic
counterpart.
The
question of whether this may be caused by a higher content
or specific characteristics of mitochondrial tRNA in these
CANCER RESEARCH VOL. 34
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
tRNA Base Composition in Morris Hepatomas and Rat Liver
hepatomas or by some other factors is currently being in
vestigated.
The relatively small differences between tumor and liver
tRNA observed by us are consistent with differences in
column Chromatographie profiles from various Morris
hepatomas including hepatoma 5123D and rat liver (50,
58). tRNA from hepatoma 5123D was shown to contain 2
new species of tRNA^r and 1 new species of tRNAPhe,
whereas 2 of the 3 tRNAHis species found in Buffalo rat
liver could not be detected in the tumor (58). Since the
structures of only 2 mammalian tRNA's in normal tissue,
i.e., of tRNA s«from rat liver (51) and of tRNAPhe from
rabbit liver (20), are known and no sequence of a tRNA in
tumor tissue has been elucidated, it is not possible at this
time to correlate differences in base composition directly
with the changes observed in the tRNA column Chromato
graphie profiles.
If one assumes that there are about 60 tRNA species
(15) in approximately equal amounts (chain length, 80
nucleotides), a calculation shows that the insertion of 1
additional modified base into 1 chain of the total tRNA
population would increase the base composition value of
this particular base by about 0.02% in terms of the total
base composition (see Tables 1 and 3) or by about 0.15% in
terms of the modified base composition (see Tables 2 and
4). For example, the increase in X and m22G observed in
hepatoma 5123D tRNA (Table 3) would correspond to the
addition of about 3 and 4 nucleosides, respectively, and the
decreases in hU and m5C to a deletion of about 6 and 8
nucleosides, respectively. The values for hepatoma 7777
tRNA (Table 1) indicate similar changes. It thus appears
possible that the differences in individual modified nucleo
sides observed by us may be the result of the presence or
absence of several tRNA species in hepatomas.
The presence of new tRNA species in tumor tRNA may
be explained on the basis of either one or a combination of
the following mechanisms: the presence of new modifying
enzymes with specificities different from the enzymes
present in normal tissue may cause modification of tRNA
precursors in unusual positions (Mechanism 1); the absence
or decreased activity of some modifying enzymes may
lead to undermodified species (Mechanism 2); an altera
tion of the specificities of the modifying enzymes may
result in overmodified species if the altered enzymes recog
nize additional sites in the tRNA (Mechanism 3); new
precursor tRNA species may be transcribed from DNA
(Mechanism 4); partially altered tRNA sequences may be
transcribed from DNA (Mechanism 5).
Mechanisms 2, 3, and 5 may also lead to the absence of
certain tRNA species in tumor tissues. At the present time
none of these mechanisms can be ruled out completely on
the basis of experimental evidence. Mechanism 3 was
shown not to be operative in certain instances in which
purified enzymes from tumor and normal tissues were com
pared. Thus, Kuchino et al. (25) showed for instance 2
guanylate residue-specific tRNA methylases to incorporate
methyl groups only into guanosine residues in certain posi
tions of purified E. coli tRNAf Metand tRNA,Val, regard
less of whether the enzymes originated from tumors, i.e..
ascites hepatoma and 3'-methyl-4-dimethylamino-azobenzene-induced hepatoma, or from normal rat liver,
although the extent of incorporation was higher for the
tumor enzymes. In our view, it would be premature to
generalize these observations to mean that no aberrant
modification of tumor tRNA occurs as a result of changes
at the enzyme level.
The presence of new modifying enzymes with different
specificities (Mechanism 1), in addition to those present in
normal tissue, would be reflected by an increased activity
of modifying enzymes in in vitro assays. Alteration of the
specificities of the normal modifying enzymes (Mech
anism 3) will lead to an overall increase in enzyme activity
when assayed in vitro. As shown in Tables 1 and 3, sta
tistically significant elevations of modified nucleosides in
the tRNA's of the 2 hepatomas have been observed for 2
nucleosides only, namely m22G and X in hepatoma 5123D,
whereas a significant decrease was found for 9 modified
nucleosides. The overmodification in terms of these 2
nucleosides may possibly be the result of aberrant modi
fication. In this connection the report by Craddock (13) on
increased activity of tRNA /V2-guanine dimethylase in
certain chemically induced rat liver and kidney tumors
should be noted. It is difficult, however, to explain the
observed undermodification on the basis of Mechanisms
1 and 3 unless the tumor tRNA population contains se
quences that are poor substrates for the modifying enzymes
(cf. Mechanisms 4 and 5). If this were the case the in
creased methylase activities in extracts from tumor tissues
may be the consequence of some regulatory response to the
presence of new precursor tRNA species and/or new se
quences in precursor tRNA. Other possibilities of explain
ing the increased activities of the tRNA methylases are
higher synthesis rates and higher turnover of tRNA in
rapidly growing tumors as evidenced by the elevated uri
nary excretion of methylated tRNA constituents by tumorbearing hosts (reviewed in Ref. 6).
As reported elsewhere (42), the pattern of modifica
tion of vertebrate tRNA has been preserved over a period
of considerable evolutionary change. Using different meth
ods, Taylor et al. (54) and Klagsbrun (24) have reached a
similar conclusion. Against this background, even slight
alterations of tRNA modification and structure in tumors
will have to be regarded as highly significant deviations
from the normal situation. The data reported here demon
strate that there are characteristic changes in tRNA modi
fication patterns for 2 Morris hepatomas. It is not yet
possible to decide to which extent each of the mechanisms
discussed above is involved in establishing the observed
alterations. We intend to carry out comparative chemical
studies on purified tRNA's from normal and tumor tissues
in order to provide answers to these questions.
ACKNOWLEDGMENTS
We are indebted to Dr. S. Nishimura who kindly provided tA. nucleoside X, and several purified E. coli tRNA's. We also thank Dr. J. A.
McCloskey for helping to establish the location of the derivative of X on
the trialcohol map.
MARCH 1974
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
651
E. Randeraih, L. S. Y. ChÃ-a,H. P. Morris, and K. Randerath
Acid in Vivo and in Vitro by Normal and SV40-transformed 3T3
Cells. J. Biol. Chem., 247: 7443 7451, 1972.
1. Baguley, B. C., and Staehelin, M. Substrate Specificity of Adenine24. Klagsbrun, M. An Evolutionary Study of the Methylation of Trans
specific Transfer RNA Methylase in Normal and Leukemic Tissues.
fer and Ribosomal Ribonucleic Acid in Procaryote and Eucaryote
Organisms. J. Biol. Chem., 248: 2612-2620, 1973.
European J. Biochem., 6: I 7, 1968.
2. Barrel!, B. G. Fractionation and Sequence Analysis of Radioactive
25. Kuchino, Y., Endo, H., and Nishimura, S. Comparison of the Speci
ficity and Extent of in Vitro Methylation by Guanylate ResidueNucleotides. In: G. L. Cantoni and E. R. Davies (eds.). Procédures
in Nucleic Acid Research, Vol. 2, pp. 751 779. New York: Harper
specific Transfer RNA Methylases Isolated from Ascites Hepatoma,
3'-Methyl-4-dimethylaminoazobenzene-induced
Hepatoma,
and
and Row, 1971.
Normal Rat Liver. Cancer Res., 32: 1243-1250, 1972.
3. Barrell, B. G., and Sanger, F. The Sequence of Phenylalanine tRNA
from E. coli. Federation European Biochem. Soc. Letters, 3: 27526. Loening, U. E. The Fractionation of High-molecular Weight Ribonu
cleic Acid by Polyacrylamide Gel Electrophoresis. Biochem. J., 702:
278, 1969.
4. Bergquist, P. L., and Matthews, R. E. F. Occurrence and Distribu
251 257, 1967.
27. Magee, P. N., and Farber, E. Toxic Liver Injury and Carcinogenesis.
tion of Methylated Purines in the Ribonucleic Acids of Subcellular
Methylation of Rat Liver Nucleic Acids by Dimethylnitrosamine in
Fractions. Biochem. J., 85: 305 313, 1962.
5. Blobel, T., and Potter, V. R. Studies on Free and Membrane-bound
Vivo. Biochem. J., 83: 114 124, 1962.
Ribosomes in Rat Liver. I. Distribution as Related to Total Cellular
28. Murao, K., Tanabe, T., Ishii, F., Namiki, M., and Nishimura, S.
RNA. J. Mol. Biol., 26: 279-292, 1967.
Primary Sequence of Arginine Transfer RNA from Escherichia coli.
6. Borek, E., and Kerr, S. J. Atypical Transfer RNAs and Their Origin
Biochem. Biophys. Res. Commun., 47: 1332 1337, 1972.
in Neoplastic Cells. Advan. Cancer Res., 15: 163-190, 1972.
29. Nishimura, S. Minor Components in Transfer RNA: Their Charac
7. Borek, E., and Srinivasan, P. R. The Methylation of Nucleic Acids.
terization, Location and Function. Progr. Nucleic Acid Res. Mol.
Biol., 12: 49-85, 1972.
Ann. Rev. Biochem., 35: 275-298, 1966.
30. Peacock, A. C., and Dingman, C. W. Resolution of Multiple Ribo
8. Brownlee, G. G. Determination of Sequences in RNA. New York:
nucleic Acid Species by Polyacrylamide Gel Electrophoresis. Bio
American Elsevier Publishing Co., Inc., 1972.
9. Brunngraber, E. F. A Simplified Procedure for Preparation of "Solu
chemistry, o: 1818-1827, 1967.
ble" RNA from Rat Liver. Biochem. Biophys. Res. Commun., 8: 1-3,
31. Perria, L., and Viale, G. L. The RNA-Overmethylation in Brain Tu
mours. Data and Hypothesis. Acta Neurochir., 21: 257-264, 1969.
1962.
10. Chia, L. S. Y., Randerath, K., and Randerath, E. Base Analysis of 32. Randerath, E., Chia, L. S. Y., Williams, J. B., Morris, H. P., and
Ribopolynucleotides by Tritium Incorporation following Analytical
Randerath, K. Comparative Studies on Liver and Hepatoma tRNA
Polyacrylamide Gel Electrophoresis. Anal. Biochem., 55: 102-113,
Base Composition. Federation Proc., 32: 653, 1973.
1973.
33. Randerath, K., MacKinnon, S. K., and Randerath. E. An Investiga
11. Cory, S., and Marcker, K. A. The Nucleotide Sequence of Methiotion of the Minor Base Composition of Transfer RNA in Normal
nine Transfer RNA M. European J. Biochem., 12: 177 194, 1970.
Human Brain and Malignant Brain Tumors. Federation European
12. Craddock, V. M. Methylation of Transfer RNA and of Ribosomal
Biochem. Soc. Letters. /5: 81 84, 1971.
RNA in Rat Liver in the Intact Animal and the Effect of Carcinogens.
34. Randerath, E., Ten Broeke, J. W., and Randerath, K. Analysis of
Biochim. Biophys. Acta, 195: 351-369, 1969.
Nucleic Acid Derivatives at Subnanomole Level. (IV). Analysis of
13. Craddock, V. M. Increased Activity of Transfer RNA /V2-Guanine
Polyribonucleotides by Conversion to Tritiated Nucleoside Deriva
tives. Federation European Biochem. Soc. Letters, 2: 10- 12, 1968.
Dimethylase in Tumors of Liver and Kidney. Biochim. Biophys. Acta,
35. Randerath, E., Yu, C.-T.. and Randerath, K. Base Analysis of Ribo
272: 288-296, 1972.
14. Dunn, D. B., Smith. J. D., and Spahr. P. F. Nucleotide Composition
polynucleotides by Chemical Tritium Labeling: A Methodological
sition of Soluble Ribonucleic Acid from Escherichia coli. J. Mol.
Study with Model Nucleosides and Purified tRNA Species. Anal.
Biochem., 48: 172 198. 1972.
Biol., 2: 113 117, 1960.
15. Gallo, R. C., and Pestka, S. Transfer RNA Species in Normal and
36. Randerath, K. An Evaluation of Film Detection Methods for Weak
/3-Emitters, Particularly Tritium. Anal. Biochem., 34: 188-205, 1970.
Leukemic Human Lymphoblasts. J. Mol. Biol., 52: 195 219, 1970.
37. Randerath, K. Application of a Tritium Derivative Method to Hu
16. Hall, R. H. The Modified Nucleosides in Nucleic Acids, Chap. 7, pp.
339-344. New York: Columbia University Press, 1971.
man Brain Tumor Transfer RNA Analysis. Cancer Res., 31: 658
17. Inose, M., Miyata, S., and Iwanami, Y. Methylation Patterns of
661, 1971.
38. Randerath, K. Continuous Directional Degradation—A Novel
Transfer RNA from Ascites Hepatomas. Biochim. Biophys. Acta,
259: 96 103, 1972.
Method for Sequence Analysis of Polyribonucleotides. Federation
18. Iwanami. Y., and Brown, G. M. Methylated Bases of Transfer Ribo
European Biochem. Soc. Letters, 33: 143 146, 1973.
nucleic Acid from HeLa and L Cells. Arch. Biochem. Biophys., 124: 39. Randerath, K., and Randerath, E. Analysis of Nucleic Acid Deriva
472-482, 1968.
tives at the Subnanomole Level. III. A Tritium Labeling Procedure
19. Johnson, J. D., and Horowitz, J. Characterization of Ribosomes and
for Quantitative Analysis of Ribose Derivatives. Anal. Biochem., 28:
RNAs from Mycoplasma hominis. Biochim. Biophys. Acta, 247:
110 118, 1969.
262-279, 1971.
40. Randerath, K., and Randerath, E. A Tritium Derivative Method for
20. Keith, G., Picaud, F., Weissenbach, J., Ebel, J. P., Pétrissant,G.,
Base Analysis of Ribonucleotides and RNA. In: G. L. Cantoni and
and Dirheimer, G. The Primary Structure of Rabbit Liver tRNAPhe
D. R. Davies (eds.). Procedures in Nucleic Acid Research, Vol. 2,
and Its Comparison with Known tRNAphe Sequences. Federation
pp. 796-812. New York: Harper and Row, 1971.
European Biochem. Soc. Letters, 31: 345-347, 1973.
41. Randerath, K., and Randerath, E. Chemical Post-Labeling Methods
21. Keller, E. B., and Zamecnik, P. C. The Effect of Guanosine Diphosfor the Base Composition and Sequence Analysis of RNA. J. Chromatog., 82: 59-74, 1973.
phate and Triphosphate on the Incorporation of Labeled Amino
Acids into Proteins. J. Biol. Chem., 221: 45-49, 1956.
42. Randerath, K., and Randerath, E. Chemical Characterization of Un22. Kerr, S. J., and Borek, E. The tRNA Methyltransferases. Advan.
labeled RNA and RNA Derivatives by Isotope Derivative Methods.
Enzymol., 36: 1-27, 1972.
In: H. Busch (ed.), Methods in Cancer Research, Vol. 9, pp. 3-69.
23. Klagsbrun, M. The Contrast between the Methylation of Ribonucleic
New York: Academic Press, Inc., 1973.
REFERENCES
652
CANCER RESEARCH VOL. 34
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
tRNA Base Composition in Morris Hepatomas and Rat Liver
43. Randerath, K., Rosenthal, L. J., and Zamecnik, P. C. Base Compo
sition Differences between Avian Myeloblastosis Virus Transfer RNA
and Transfer RNA Isolated from Host Cells. Proc. Nati. Acad. Sci.
U. S., 68: 3233-3237, 1971.
44. Reddy, R., Ro-Choi, T.-S., Henning, D., Shibata, H., Choi, Y. C.,
and Busch, H. Modified Nucleosides of Nuclear and Nucleolar Low
Molecular Weight Ribonucleic Acid. J. Biol. Chem., 247: 72457250, 1972.
45. Ro-Choi, T. S., Moriyama, Y., Choi, Y. C., and Busch, H. Isolation
and Purification of a Nuclear 4.5 S Ribonucleic Acid of the Novikoff
Hepatoma. J. Biol. Chem., 245: 1970-1977, 1970.
46. Sanger, F., Brownlee, G. G., and Barrel!, B. G. A Two-dimensional
Fractionation Procedure for Radioactive Nucleotides. J. Mol. Biol.,
13: 373-398, 1965.
47. Sheid, B., Wilson, S. M., and Morris, H. P. Transfer RNA Methylase Activity in Normal Rat Liver and Some Morris Hepatomas.
Cancer Res., 31: 774-777, 1971.
48. Srinivasan, P. R., and Borek, E. The Species Variation of RNA
Methylase. Proc. Nati. Acad. Sei. U. S., 49: 529 533, 1963.
49. Srinivasan, P. R., and Borek, E. Enzymatic Alteration of Nucleic
Acid Structure. Science, 145: 548 553, 1964.
50. Srinivasan. D., Srinivasan, P. R., Grunberger, D., Weinstein, I. B.,
and Morris, H. P. Alterations in Specific Transfer Ribonucleic Acids
in a Spectrum of Hepatomas. Biochemistry, 10: 1966 1973, 1971.
51. Staehelin, M., Rogg, H., Baguley, B. C., Ginsberg, T., and Wehrli,
52.
53.
54.
55.
56.
57.
58.
59.
60.
W. Structure of a Mammalian Serine tRNA. Nature, 219: 13631365, 1968.
Starr, J. L., and Sells, B. H. Methylated Ribonucleic Acids. Physiol.
Rev., 49: 623 669, 1969.
Sueoka, N., and Sueoka, T. Transfer RNA and Cell Differentiation.
Progr. Nucleic Acid Res. Mol. Biol., 10: 23 55, 1970.
Taylor, M. W., Buck, C. A., Granger, G. A., and Holland, J. J. Chro
matographie Alterations in Transfer RNAs Accompanying Speciation. Differentiation and Tumor Formation. J. Mol. Biol., 33: 809
828, 1968.
Viale, G. L. Biochemical Patterns in Brain Tumours. III. Transfer
Ribonucleic Acid. Acta Neurochir., 21: 123 137, 1969.
Viale, G. L. Transfer RNA and Transfer RNA Methylase in Human
Brain Tumors. Cancer Res., 31: 605 608, 1971.
Viale, G. L., Restelli, A. F., and Viale, E. Methylated Bases in tRNAs
of Brain Tumors. Tumori, 53: 533-539, 1967.
Volkers, S. A. S., and Taylor. W. Chromatographie Comparison of
the Transfer Ribonucleic Acids of Rat Livers and Morris Hepatomas.
Biochemistry, 10: 488-497, 1971.
Yaniv, M., and Barrel!, B. G. Sequence Relationship of Three Valine
Acceptor tRNAs from Escherichia coli. Nature New Biol., 233: 113
114, 1971.
Yarus, M., and Barrell. B. G. The Sequence of Nucleotides in
tRNA"e from E. coli B. Biochem. Biophys. Res. Commun., 43: 729
734, 1971.
MARCH 1974
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1974 American Association for Cancer Research.
653
Transfer RNA Base Composition Studies in Morris Hepatomas
and Rat Liver
Erika Randerath, Li-Li S. Y. Chia, Harold P. Morris, et al.
Cancer Res 1974;34:643-653.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/34/3/643
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 15, 2017. © 1974 American Association for Cancer Research.