Volume 2 number 9 September 1975
Nucleic Acids Research
Utilization of an Escherichia coli mutant for carbon-13 enrichment of tRNA
for NMR studies
Paul F. Agris, F.Genichi Fujiwara, Charles F.Schmidt and Richard N. Loeppky
Division of Biological Sciences and Department of Chemistry, University of
Missouri, Columbia, MO 65201, USA
Received 4 June 1975
ABSTRACT
The enrichment of tRNA at specific sites with carbon-13 has been accomplished in^ vivo using a mutant of Escherichia coli.
A relaxed strain of E_.
coli auxotrophic for methionine was grown in a specifically defined medium
supplemented with either [
c] or [
c]-methyl labeled methionine.
Cells were
collected at the end of the log-phase of growth and tRNA was extracted.
Anal-
ysis of the radioactivity of the [ c]-labeled tRNA established an incorporation ratio of three labeled carbons per tRNA molecule.
[
Incorporation of the
c]-label in_ vivo was confined to the methylation of nucleotides as deter-
mined by thin layer chromatography of nucleotides resulting from a ribonuclease digestion of [
c]-labeled tRNA.
The carbon-13 NMR spectrum of [ c ] -
enriched tRNA indicated a similar degree of incorporation into the methylated
nucleotides by the substantial enhancement of [
c]-methyl NMR signals only.
Assignment of signals has been made for the methyl groups of ribothymidine and
N -methylguanosine in E. coli tRNA.
INTRODUCTION
Nuclear magnetic resonance spect.roscopy is one of the few tools available
for structural and functional studies of biopolymers in solution.
tRNA being
the smallest nucleic acid has been studied extensively with proton magnetic
resonance spectroscopy (1-7).
In addition the NMR spectroscopy of naturally
occurring carbon-13 in tRNA has been done (8,9).
Since carbon-13 NMR spectra
for complex molecules tend to be in general, more resolved and easier to interpret than the proton spectra (10), it could provide valuable information.
However, the major drawback in doing carbon-13 NMR of tRNA has been that some
of the most interesting spectral lines are the weakest.
These weak lines are
of interest because they are due to the minor bases which always occupy specific sites in the tRNA nucleotide sequence (11) and because they are res-
1503
Nucleic Acids Research
onance signals occurring in spectral regions free from ribose and major base
interference (9). The weak intensity of the lines is due to the unavoidable
low concentration of tRNA solutions in conjunction with the low, 1.1%, natural abundance of the carbon-13 isotope.
To improve the potential of the carbon-13 NMR technique two things can be
done.
One is to improve the sensitivity of the instrumentation (9) and the
other is to enrich the tRNA in carbon-13 at specific sites.
The work pre-
sented here established a technique for the carbon-13 enrichment of minor nucleotides within tRNA.
MATERIALS AND METHODS
Growth of cells.
E_. coli C6 rel~ met" cys" (12) a gift from Dr. D. Soil (Yale
University) was grown in a defined medium of the following composition for 1
liter of solution:
0.302g KH PO ; 0.602g Na HPO ; O.lOOg NH Cl; 0.049g
MgSO • 7H 0; 5.5mg CaCl ; 45yg FeSO • 7H 0; 20mg uracil; 20mg xanthine; 16mg
guanine; 7.3mg adenine; 50g dextrose; and l.lmg cysteine.
the medium were of analytical reagent grade.
All chemicals for
The medium was considered com-
plete with the addition of either 20.8mg methionine-methyl-[
c] (New England
Nuclear) with a specific activity of 58m Ci/mmole or 51mg methionine-methyl[
c] (Merck of Canada) specified to have 90 atom % methyl carbon-13.
This
latter specification was confirmed by mass spectroscopy and carbon-13 NMR
spectroscopy (Fig. 1) . The innoculum consisted of E_. coli cells which had
been grown in nutrient broth and had been tested for methionine and cysteine
auxotrophy prior to innoculation of the defined medium.
Cultures were grown
at 37° in a shaking water bath.
tRNA Extraction.
At the end of the log-phase of growth the cells were col-
lected by centrifugation and their nucleic acids extracted (13). tRNA was
then isolated by DEAE-cellulose column chromatography and concentrated by alcohol precipitation (13). The resulting precipitates were collected by centrifugation, dissolved in glass distilled H O and dialyzed against glass distilled H O extensively.
Identification of In_ Vivo Methylated Nualeotides in tRNA.
The [
c]-labeled
tRNA was subjected to enzymic digestion according to Agris, et al. (13). The
resulting mononucleotides and dinucleoside-diphosphates were subjected to thin
layer chromatography (250y thick, microgranular cellulose, Analtech) in two
dimensions:
(I) isobutyric acid-concentrated ammonia-H^O (198:3:99, V/V/V)
and (II) isopropanol-concentrated HC1-H2O (68:17.6:14.4, V/V/V).
Autoradio-
graphy of the developed chromatograms was used to detect the migrations of the
1504
Nucleic Acids Research
B
NH,
12
2'5
50
75
100
3
4
5
150
M/E
I
200
150
100
50
PPM
13.
Mass and NMR Spectra of Methionine-Methyl- [ c ] .
1.
The mass spectrum of the methionine-methyl-[
c] that was used in the
experiments reported here is shown in section A of the figure.
A CEC 21-110
mass spectrometer (70ev, 250°C source temperature) coupled to a JEOL-6 computer
was used to record the spectrum.
This spectrum was compared to the spectrum of
methionine-methyl [
c] (not shown) which had prominent signals at 61 and 149
12
12
M/E corresponding to the ion - CH S- CH and the molecular ion, respectively.
13
Section B of the figure is the proton-decoupled [ c]-NMR spectrum of methionine
methyl-[
c ] . The resonance signal number 5 obtained from the carbon-13 en-
riched methyl group (C ) and the signals obtained from the other carbons (C 13
C ) , containing [ c] in natural abundance only were assigned according to
Horsley, et^ al^.
[Horsley, W., Sternlicht, H. and Cohen, J.S.
(1970)
J. Am.
Chem. Soc. 92, 680-686].
rl4
L
CJ-labeled nucleotides, which were compared to the migrations of standards
for the purposes of identification (13). Each radioactive area of the cellulose
plates was scraped from the plate, the scrapings packed into a Pasteur pipette
1505
Nucleic Acids Research
and the radioactive nucleotide eluted with glass distilled water.
The radio-
activity of each nucleotide was measured by scintillation counting.
NMR Samples and Instrumentation.
Besides the [
c]-enriched tRNA obtained
from cells grown in medium with methionine-methyl-[
parations were used for NMR studies.
c] two other tRNA pre-
tRNA from E_. coli B (Schwarz-Mann) was
a gift from Dr. B. Ortwerth (University of Missouri-Columbia). tRNA deficient
in methylated nucleotides was extracted from E_. coli C6 rel
met
had been grown under conditions of methionine-starvation (12,14).
cys
that
The two
latter tRNA samples which contained carbon-13 in natural abundance only were
dissolved in glass distilled H O and dialyzed against glass distilled H O
extensively.
Proton-decoupled carbon-13 NMR spectra were recorded at 22.63 MHz on a
Brucker HFX-90 Fourier transform NMR spectrometer.
ning sample tube with a 10mm o.d.
The probe uses a spin-
A Nicolet 1080 computer with a limit of
resolution of 0.1 ppm was used for signal averaging and transformation.
The
strongest resonance observed was assigned to the ribose 2' and 3' carbons and
was used as an internal reference (A
=73.7 ppm; ref. 9 ) . A repetition time
of 0.5 sec between spectra was used.
RESULTS
Cultures of E. coli C6 rel
met
cys
were grown in the completely de-
fined medium that was described in the Materials and Methods section.
Bases
(adenine, guanine, uracil and xanthine) were added to the medium to insure
that methyl groups from the labeled methionine compounds were not utilized
through the "one-carbon" pool for the biosynthesis ^n_ vivo of the major nucleotides.
The growth curve of a culture grown in medium with methionine-
methyl-[
c] (20.8 mg/1) and that of a culture grown in medium with methionine-
methyl-[
c] (51mg/l) are shown in Figure 2.
were comparable considering that the [
much methionine as did the [
The growth of the two cultures
c]-labeled culture contained twice as
c]-labeled culture.
without methionine (results not shown).
No growth occurred in medium
The cells of these cultures were col-
lected at the end of the log-phase of growth.
Nucleic acids were extracted
from the differentially labeled cells and tRNA isolated (see Materials and
Methods) . Approximately 2g wet weight of packed cells obtained from a liter
culture yielded 5mg of either [14c] or [
c]-labeled tRNA.
Analysis of the radioactivity of the [
c]-labeled tRNA isolated from
various cultures indicated that 2-4 methyl groups had been incorporated in_ vivo
into each molecule of tRNA.
1506
This incorporation was equivalent to the number
Nucleic Acids Research
6
12
18
24
INCUBATION TIME
(hr)
2.
Growth of E_. coli Cultures.
E. coli C6 rel
Materials and Methods.
figure.
met
cys
was grown in the minimal medium described in
The growth curves of two cultures are depicted in the
One culture was grown in medium containing methionine-methyl-[
c]
at a concentration of 51mg/l (o). The other culture was grown in medium containing methionine-methyl-[
c] at a concentration of 20.8mg/l (•).
Both cul-
tures had an initial period of no growth due to the inoculum being maintained
in nutrient broth.
Growth of the cultures was determined by analyzing the
change in optical density at 590nm.
expected from consideration of the known E^. coli tRNA nucleotide sequences (15).
In order to determine that the incorporation of methyl groups from methionine
into tRNA had been limited to post-transcriptional methylation, the radioactive
nucleotides within [
c]-labeled tPNA were identified and quantitated (see
Materials and Methods and ref. 13 for procedures).
Analysis of the radioactive
nucleotides by their migration within a two-dimensional, thin layer chromatography system and detection by autoradiography showed that there was no radioactivity associated with the major nucleotides, Ap, Cp, Gp and Up (16). Radioactivity was associated with nucleotides that had chromatographic migrations
comparable to riboTp, m Gp and a methylated Ap tentatively identified as m Ap.
1507
Nucleic Acids Research
Quantification of the radioactivity in these nucleotides indicated that they
occurred in each molecule of tRNA in the following approximate amounts:
1.0
riboTp, 0.5 m Gp and 0.1 m Ap (Table I ) .
Table I
Methylated Nucleotides Within [14c]-labeled tRNA
Nucleotide
Associated
Radioactivity
(CPM)
Approximate moles of [ c ] methylated nucleotides per
mole of tRNA
riboTp
674
1.0
m Gp
337
0.5
m Ap
120
0.1
NpNp
Others
"J
These values are equivalent to the amounts of riboTp, m Gp and m Ap in tRNA
from other strains of E_. coli (11,17,18).
The identity of riboTp was confirmed
by the mass number of the nucleoside, 258 daltons, as determined by mass
spectroscopy of the nucleotide that had been eluted from cellulose scrapings of
thin layer chromatography plates.
Various radioactively labeled, ribose
methylated, nucleoside-diphosphates (NpNp) were identified by their migration
in the two-dimensional chromatography system (13) but were not quantitated.
The presence of the identically methylated, but [ c]-labeled nucleotides
within the tRNA from cells grown in medium with methionine-methylated-[
c] was
[-14 T
assumed because of the comparable growth characteristics of the L Cj and the
[
c]-methionine cultures (Fig. 2 ) . The presence of [
tides was confirmed by carbon-13 NMR spectroscopy.
c]-methylated nucleo-
The spectrum of [ c ] -
enriched tRNA was compared to the spectra of two other tRNA preparations that
contained carbon-13 in natural abundance only.
NMR spectra of the three tRNA samples.
Figure 3 shows the carbon-13
A moderately concentrated solution of
tRNA (3mM) from E. coli B gave the spectrum shown in Figure 3A. This natural
abundance spectrum is similar to that of yeast tRNA (8/9).
Komoroski and
Allerhand (8,9) have assigned, in general, low field signals between 63.2 and
191.6 ppm in Figure. 3A to the ring carbons of the bases and to the carbons of
ribose; whereas they have assigned the high field signals between 12.0 and 36.6
ppm to carbons of the methyl substituents of minor nucleotides.
Figure 3B is
the natural abundance spectrum of a dilute solution of tRNA (0.2mM) that was
deficient in methylated nucleotides.
1508
The only detectable signals in the
Nucleic Acids Research
3.
Carbon-13 NMR Spectra of 12. coli tRNA.
The proton-decoupled carbon-13 NMR spectra of three different samples
of E_. coli tRNA are shown.
Methods.
Spectra were taken as described in Materials and
Section A is the spectrum from 534,000 scans of E_. coli B tRNA at
a concentration of 3mM.
Section B is the spectrum from 650,000 scans of a
dilute solution (0.2mM) of methyl-deficient tRNA obtained from E_. coli C6 rel
met
cys
grown under conditions of methionine starvation.
Section C is the
spectrum from 910,000 scans of a dilute solution (0.2mM) of tRNA that had been
obtained from E_. coli C6 rel
[
met
cys
grown in medium with methionine-methyl-
c ] . Resonance signal 1 in the figure has been assigned to the naturally
abundant carbon-13 present in ribose carbons 2' and 3' (9). Signals numbered
3-5 in spectrum C have been assigned to the [
nucleotides in [
c]-enriched tRNA.
c]-enriched carbons of methylated
These assignments are described in the text.
spectrum of this dilute solution could be assigned to carbons of the ribose
moiety (64.8 ppm, 71.6 ppm and peak number 1 at 73.7 ppm).
In contrast to
these natural abundance spectra, the spectrum of a dilute solution (0.2mM) of
1509
Nucleic Acids Research
[
c]-enriched tRNA (Fig. 3C) contained substantially enhanced high field sig-
nals (12.0, 29.1, 36.6 ppm; peak numbered 5, 4 and 3, respectively) that could
be assigned, according to Komoroski and Allerhand (8,9), to methyl carbons of
the minor nucleotides in tRNA.
The assignments of these signals are described
in the Discussion section.
DISCUSSION
A relaxed strain of E_. coli auxotrophic for methionine has been used for
the production of tRNA that is enriched in carbon-13 at specific sites.
En-
richment was achieved through the transfer in_ vivo of the methyl group of
methionine to tRNA.
Incorporations of methionine-methyl-[
parallel with the incorporation of methionine-methyl-[
c] were done in
C] because of the ease
of following the radioactive label into the minor nucleotides of tRNA.
The
comparable growth characteristics of cultures grown in medium with either [
or [
for methyl carbons of the minor nucleotides in [
established that [
[
c]
c]-labeled methionine (Fig. 2) and the enhanced carbon-13 NMR signals
c]-enriched tRNA (Fig. 3)
c]-methyl incorporation into tRNA was similar to that for
c]-methyl incorporation (3 moles of methyl per mole of tRNA).
NMR signals obtained from the minor nucleotides of [
Since the
c]-enriched tRNA were
comparable to the resonance of ribose carbons 2' and 3" (73.7 ppm) as shown
in Figure 3, an
80-fold enrichment of carbon-13 must have occurred in these
minor nucleotides.
Ribose carbons 2' and 31 occur some 80 times in each tRNA
molecule; whereas the methyl carbon of riboTp occurs only once.
Thus an 80-
fold enrichment of carbon-13 in the methyl carbon of riboTp would produce an
NMR signal comparable to that of the ribose carbons which contain carbon-13
in natural abundance only.
We believe that this 80-fold enrichment, achieved
in vivo, is sufficient for further NMR studies especially when more concentrated solutions of [
c]-enriched tRNA are used.
The enhanced high field signals in the spectrum of [
c]-enriched tRNA
(Fig. 3C) could be tentatively assigned utilizing the assignments made by
Komoroski and Allerhand (8,9) and our nucleotide analysis of the [ c]-methylated tRNA.
Nucleotide analysis showed that [
c]-methyl groups of methionine
were used in the biosynthesis of riboTp, m Gp, m Ap and NpNp.
hanced NMR signals from [
Thus the en-
c]-enriched tRNA were probably those for the methyl
groups of riboTp (12.0 ppm), m Gp (36.6 ppm) and m Ap (29.1 ppm).
advantage from the NMR analysis of this [
An added
c]-enriched £. coli tRNA is the
ability to assign and study signals from methyl groups, such as m Ap, that
occur in only 10% of the tRNA molecules (Table I ) .
1510
Nucleic Acids Research
The locations of the minor nucleotides within the structures of tRNA
molecules are one of their most common features (11,17).
For instance, riboTp
is consistently found in the "GTfC loop" of the cloverleaf structure of E_. coli
tRNA and m Gp in the "extra arm" of this structure.
Thus the enrichment of the
methylated nucleotides in carbon-13 provided sets of various tRNA molecules all
containing [
structures.
c]-enrichment at the same locations within their individual
Therefore, it will not be necessary, at least for initial studies,
to purify individual species of tRNA when the tRNA has been enriched in carbon13 in the manner described here.
Carbon-13 NMR studies of tRNA structure and
function that can be conducted include measurements of chemical shifts during
the "melting" of the tRNA structure, the increase of organic solvent in the
aqueous tRNA solution and the interaction of specific complimentary oligonucleotides with the tRNA.
respect to [
The latter study is of particular interest with
c]-riboTp because of the finding that an oligonucleotide com-
plimentary to GTfC exists in the nucleotide sequence of 5S rRNA (19) and may
be important to the binding of tRNA to ribosomes (20) . Some of these NMR
studies are now underway in our laboratories.
ACKNOWLEDGMENTS
This work was supported by grants to Paul F. Agris from the University of
Missouri Research Council (N.S.F. Grant 1472) and the National Institutes of
Health (1-RO1-CA16327-01).
REFERENCES
1.
Kearns, D.R., Lightfoot, D.R., Wong, K.L., Wong, Y.P., Reid, B.R., Cary,
L. and Schulman, R.G. (1973) Annals N.Y. Aoad. Sci. 222, 324-336.
2.
Kearns, D.R., Wong, K.L. and Wong, Y.P.
U.S. 70, 3843-3846.
3.
Kan, L.S., Ts'o, P.O.P., Haar, V.D., Sprinzl, M. and Coramer, F.
Biochem. Biophys. Res. Corrmun. 59, 22-29.
4.
Kearns, D.R., Wong, Y.P., Chang, S.H. and Hawkins, E.
chemistry 13, 4736-4746.
5.
Kreischman, G.P., Miller, J.P., Dea, P., Hussain, Z., Wilson, L.A. and
Schweizer, M.P. (1974) Biochem. Biophys. Res. Corrmun. 58, 27-34.
(1974)
(1973)
Proa. Natl. Acad. Sci.
(1974)
(1974)
Bio-
6.
Ladner, J.E. and Schweizer, M.P.
7.
Wong, K.L. and Kearns, D.R.
8.
Komoroski, R.A. and Allerhand, A.
1804-1808.
(1972)
Proc. Natl. Acad. Sci. U.S. 69,
9.
Komoroski, R.A. and Allerhand, A.
(1974)
Biochemistry 13, 369-372.
(1974)
Nucleic Acids Res. 1, 183-192.
Biopolymers 13, 371-380.
1511
Nucleic Acids Research
10.
Allerhand, A., Doddrell, D. and Komoroski, R.
189-198.
11.
Nishumura, S.
49-85.
12.
Harris, C.L., Titchener, E.B. and Cline, A.L.
1322-1327.
13.
Agris, P.F., Koh, H. and s8ll, D.
277-282.
(1972)
(1971)
J. Chem. Phys. 55,
in Prog. Nucleic Acid Res. Mol. Biol. Vol. 12, pp.
(1973)
J. Baeteriol. 100,
Arch. Bioohem. Biophys. 154,
14.
Agris, P.F., Soil, D. and Seno, T.
15.
Barrell, B.G. and Clark, B.F.C. (1974) "Handbook of Nucleic Acid Sequences," Joynson-Bruwers, Ltd., Oxford, England.
16.
Nucleotides are abbreviated according to the recommendations of the
IUPAC-IUB Commission on Biochemical Nomenclature [Biochemistry 9, 4022
(1970)]
Hall, R. (1971) "The Modified Nucleosides in Nucleic Acids," Columbia
University Press, New York, New York.
17.
(1973)
(1969)
Biochemistry 12, 4331-4337.
18.
Munns, T.W., Podratz, K.C. and Katzman, P.A.
4409-4416.
(1974)
Biochemistry 13,
19.
Brownlee, G.G., Sanger, F. and Barrell, B.G.
736.
(1967)
Nature 215, 735-
20.
Erdmann, V.A., Springzl, M. and Pongs, 0.
Commun. 54, 942-948.
1512
(1973)
Bioohem. Biophys. Res.