Volume 4 Number 12 December 1977
Nucleic Acids Research
The nucleotide sequence of a major glycine transfer RNA from the posterior silk gland of
Bombyx mori L
Martha C. Z6nigat and Joan A. Steitz*
tBiology Department and 'Department of Molecular Biophysics and Biochemistry, Yale University,
New Haven, CT 06510, USA
Received 11 October 1977
ABSTRACT
The aucleotide sequence of tRNAj y isolated from the posterior silk
gland of Bombyx mori has been determined. This transfer RNA is present in
high amounts in the posterior silk gland during the fifth larval instar. It
has a GCC anticodon, capable of decoding a major glycine codon in the fibroin
messenger RNA, GGU. Structural features of Bombyx tRNA? y and its
homology to other eukaryotic glycine tRNAs are discussed.
INTRODUCTION
The posterior silk gland of the fifth Instar larva of' Bombyx mori is an
attractive system for studies on the regulation of protein synthesis because
its major product is a single polypeptide, silk fibroin 1 ' 2 .
Not only is the
fibroin protein well studied, but its messenger RNA has been isolated and
chemically characterized3.
The fibroin message exhibits selective codon
utilization for the major amino acids which occur in the fibroin protein
(glycine, 46%; alanine, 29%; serine, 12.9%; and tyrosine, 6%**).
The triplets
GGU and GGA appear in 78.5Z of the glycine positions, in a ratio of 1.4 GGU
to 1.0 GGA 3 .
The unusual amino acid composition of the fibroin protein4 prompted an
analysis of the transfer RNAs from the posterior silk gland of the fifth
instar larva.
Previous aminoacylation studies showed that there is an
increased amino acid acceptance activity for glycine, alanine, and serine
during the maturation of the posterior silk gland 5 ' 6 .
raised several questions.
These observations
First, is the increased amino acid acceptance
activity In the fifth instar posterior silk gland due to the synthesis of
novel isoacceptors or to the accumulation of tRNA species present in other
larval tissues?
Second, does the anticodon population of the posterior silk
gland tRNAs reflect the pattern of codon utilization in the fibroin mRNA?
Third, do the transfer RNAs which accumulate for fibroin biosynthesis possess
any unusual structural features which may enhance their biological activity?
© Information Retrieval Limited 1 Falconberg Court London W1V5FG England
4175
Nucleic Acids Research
To answer these questions,
32
P-labeled transfer RNAs from the posterior
silk gland (and several other tissues) of Bombyx larvae were fractionated on
polyacrylamide gels, identified by their aminoacylation properties, and analyzed by RNA sequencing methods.
The sequences of the two major alanine iso-
acceptors have been reported elsewhere7.
The elucidation of the primary
structure of one of the major glycine acceptors, tRNAj ', is described here.
EXPERIMENTAL PROCEDURES
Sources of enzymes:
Tj ribonuclease (B grade) and T2 ribonuclease (A
grade) were purchased from Calbiochem.
Ribonuclease A (beef pancreas), beef
spleen phosphodiesterase, snake venom phosphodiesterase, and micrococcal
nuclease were obtained from Worthington.
U2 ribonuclease was a gift of Sankyo, Ltd.
Pi nuclease was a gift of
Alan Welner (Yale University); Si nuclease was a gift of Roger Wiegand (Yale
University); and the beef spleen phosphodiesterase used for partial digestions
of primary oligonucleotides was a gift from E. G. Niles (State University of
New York at Buffalo) .
Preparation of Bombyx mori RNA:
Bombyx mori were reared as described by
Suzuki and Brown3 and Sprague1.
32
P-labeled RNA was prepared from fifth larval instar posterior silk
glands, middle silk glands, or midgut tissue as described previously
except that each larva was administered 2.0 to 2.5 mCi
,
32
POi, (carrier-free,
New England Nuclear) on the fifth and sixth days after ecdysis, for a total of
4 to 5 mCl
32
POi4 per larva.
fifth larval instar3.
Animals were dissected on the eighth day of the
Specific activities obtained were 2 to 6 x 101* cpm/ug
RNA.
Low molecular weight RNA was prepared in one of two ways: by sedimentation through sucrose gradients3 or by DEAE-cellulose column chromatography 9 ' 10 .
When the latter method was used, total cellular RNA or low molecular
weight RNA obtained by precipitation of high molecular weight RNA with 1M
NaCl 1 1 was fractionated by passage through DE-52 (Whatman) as described by
Landy, e_t al.1 0 , except that the buffer used throughout was 0.05M N£ acetate,
pH 5.0, 0.01M MgCl 2 .
Unlabeled RNA was extracted from posterior silk glands of fifth instar
larvae on the 8th day after ecdysis.
Approximately 0.26 mg of total low
molecular weight RNA was obtained from each pair of posterior silk glands.
Fractionation of low molecular weight RNAs:
Bombyx mori low molecular
weight RNA was fractionated in one-dimensional and two-dimensional gels using
slight modifications of the systems reported by Ikemura and Dahlberg 12 .
4176
Nucleic Acids Research
The one-dimensional gel (and the first dimension of the two-dimensional
gel system) contained 9% acrylamide, 0.3Z N, N'-bisacrylamide, and 7M urea, in
the 0.09M TriB-borate, pH 8.3 buffer previously described13 except that EDTA
was omitted.
RNA samples were dissolved in 7M urea and then mixed with a
small amount of dye solution (0.2% xylene cyanol FT in 40% sucrose) for a
total volume of 20 to 40 ul per 1 cm-wide gel slot.
In some cases the samples
were heated to 60° C immediately prior to loading the gel; however no differences were found between electrophoretic patterns of preheated and unheated
samples.
Nine percent gels (14 x 40 x 0.15 cm) were run on a vertical slab
apparatus14 with 0.045M Tris-borate, pH 8.3, in the upper reservoir and 0.09M
Tris-borate, pH 8.3, in the lower reservoir.
Electrophoresis was carried out
at 4°C and 500 volts until the dye was 3 cm from the bottom of the gel.
For two-dimensional gel electrophoresis, RNAs fractionated on a 9% polyacrylamide gel were detected by autoradiography or by staining with methylene
blue (0.2% in 0.2M Na acetate, pH 4.7, for 20 minutes).
A gel slice contain-
ing the 4S RNAs was excised and positioned on the short side of a 14 x 16 cm
gel plate.
The second dimension gel (20% acrylamide, 0.45% N, N'-bisacryl-
amide, 0.09M Tris-borate, pH 8.3) was then formed as described by Ikemura and
Dahlberg12.
Prior to electrophoresis, Trypan red (0.2% in 40% sucrose) was
layered onto the top of the gel.
0.09M Tris-borate, pH 8.3,
Electrophoresis was carried out at 4°C in
at 450 volts until the major red-blue doublet of
the Trypan red dye migrated 2/5 of the way down the gel.
Quantitation of
P-RNAs in gel pieces was achieved by directly measuring
Cerenkov radiation using a liquid scintillation counter.
Calculations assume
that all RNAs are equally labeled, a conclusion based on specific activity
measurements on three RNA species eluted from gels (M. Zfifiiga, unpublished
observations).
RNA was routinely eluted from gel pieces by electrophoresis15 into dialysis casing containing 0.036hl Tris-borate, pH 8.3, immersed in the same buffer
made 1H in NaCl.
Aminoacylation studies:
Cell extracts containing tRNA-aminoacyl ligase
activity were standardly prepared from three pairs of posterior silk glands
taken from fifth instar larvae seven days after ecdysis.
The protocol used
16
was that of White et_ a^. , except that 0.20MPTU (phenylthiourea, Sigma) was
included in all buffers (including the 0.15M NaCl, 0.015M Na citrate, pH 7.0,
used for dissection) and the DEAE-cellulose chromatography step utilized
Cellex D (Biorad) as the resin.
Activity was assayed across the elution
profile, using the assay reported elsewhere7.
Each fraction was made 50%
in glycerol and stored up to two months at -20°C.
4177
Nucleic Acids Research
Amlno acid acceptance activity of larval RNAs was assayed using 3 H- or
C-amino acids as described previously
. Background values for each amlno
acid were obtained from reactions omitting either tRNA or the tRNA-aminoacylligase preparation.
at high pH
7
In early experiments, tRNA was deacylated by incubation
prior to aminoacylation.
However, values obtained with gel-frac-
tionated tRNAs subjected to this treatment did not differ from those obtained
with untreated RNAs.
For calculations of percentage aminoacylation an absor-
bance of 1.0 OD per ml at 260 run was assumed to represent 40 yg of tRNA.
Sequencing procedures:
The sequence of purified tRNAj ^ was analyzed
using the methods developed by Sanger and his colleagues18.
Unless otherwise
stated, digested products were quantitated by counting in toluene scintillation fluid (containing 0.42% PPO, New England Nuclear).
Products of complete Tj RNase or RNase A digestion were fractionated by
electrophoresis on cellulose acetate (Kalex) at pH 3.5 in the first dimension
followed either by electrophoresis on DEAE-paper (Whatman) in IX formic acid 18
or by homochromatography on PEI-cellulose thin layers (20 x 20 cm, Brinkman) ^.
Where necessary for further analysis, oligonucleotides obtained from
homochromatograms were purified of carrier RNA by chromatography in pyridineformate, pH 3.4 and 7M urea7 on DEAE-cellulose thin layer plates (MacheryNagel CEL300).
For short oligonucleotides (five nucleotides or less in
length), 1.5M pyridine-formate was used; for long oligonucleotides (six
resi-
dues or longer), 2.0M pyridine-formate.
Analysis of primary digestion products:
Minor nucleotides (and pGp) were
characterized by thin layer chromatography in two dimensions20, 3MM paper
electrophoresis18, and descending paper chromatography18'21 after digestion of
Tl ribonuclease oligonucleotides with T2 ribonuclease18, spleen phosphodlesterase7, snake venom phosphodiesterase18, Pi nuclease 22 , micrococcal nuclease 2 3 , or alkaline hydrolysis* 8 ' 24 .
Identification of modified nucleotides
was based both on the mobilities of experimental samples relative to unmodified marker nucleotides and on published mobility values 1 8 ' 2 5 ' 2 0 .
Micrococcal nuclease was employed in the analysis of the dinucleotide X
(UmC) from T\ ribonuclease oligonucleotide 11.
As described for dTpCp by
Hikulski e_t JL!. 2 3 , the sample was dissolved in 5 ul of 0.05M TrisHCl, pH 9.0,
0.0111 CaCl2 containing 5 yg micrococcal nuclease (enzyme to substrate ratio of
5:1) and incubated at 37°C for six hours.
The reaction product and marker
nucleotides were then spotted onto two thin layer chromatography plates
(Avicel), and one dimensional chromatography was run using the two solvent
systems of Nishimura 20 .
4178
Nucleic Acids Research
Pj nuclease22 was also used Co analyze the dlnucleotlde X
X
(Um C) .
Since
proved resistant to digestion by spleen phosphodiesterase and venom phos-
phodiesterase under conditions where other dinucleotides are hydrolyzed
(reference 7 and this report), harsh Pi nuclease digestion conditions were
utilized.
The sample (eluted from paper and estimated to contain about 1 to
2 yg RNA) was incubated with 2.5 gg enzyme in 0.0AM NH 4 acetate, pH6.0 for one
hour at 50°C.
The reaction product was applied to thin layer chromatography
plates along with micrococcal nuclease digestion products and analyzed as
described above.
Pj nuclease contains a 3'-phosphomonoesterase activity
,
thus the major product from the dinucleotide U m C was pC.
The isomers CUCG and CCUG (Tj ribonuclease oligonucleotide 14) were separated by one dimensional ascending chromatography at room temperature on DEAEcellulose thin layer plates (Machery-Nagel CEL300) in 7.5% triethylamine carbonate, pH 1 0 2 7 .
Partial digestion with spleen phosphodiesterase was carried out on carrier-free oligonucleotides using 0.2 to 1.0 mg/ml enzyme in 10 ul of 0.1M
Nl^-acetate, pH 5.7, 0.002M EDTA, 0.05% TWEEN 8 0 1 8 at 37°C for 1, 2 and 3
hours.
The samples, along with an untreated control and marker oligonucleo—
tides, were fractionated in one dimension either by chromatography on PEIcellulose in 3% homomixture C
formic acid
or by electrophoresis on DEAE-paper in IX
. The composition of each product was then analyzed by complete
digestion with T2 RNase or spleen phosphodiesterase .
The length of the 3'-terminal oligonucleotide of tRNA^lv was determined
by a modification of a procedure reported by Keith and Gilham2° for the stepwise degradation of a polynucleotide chain.
Oxidation of the sample in 15 ul
H2O at 0°C was initiated by adding 2 ul of freshly made 0.2H Na periodate.
After one hour, 2 ul of 0.4M glycerol was added and the reaction incubated an
additional 30 minutes at 0°C.
Then 6 ul of 2.0M cyclohexamineHCl, pH 8.0, was
added and the reaction incubated at A5°C for 30 minutes before analysis.
Partial digestion with T) ribonuclease:
Limited digestion of tRNA^ly
with Ti ribonuclease was carried out at 0°C for 10 minutes in 5 to 10 ul of
either 0.01M TrisHCl, pH 7.5, 0.01H MgCl 2 or 0.01M TrisHCl, pH 7.5, 0.001M
EDTA.
In two experiments, the enzyme to substrate ratio was 1:1000 in the
magnesium-containing buffer.
In a third experiment, one half of the sample in
the magnesium-containing buffer was digested at an enzyme to substrate ratio
of 1:750, whereas the other half in the EDTA-containing buffer waa treated at
an enzyme to substrate ratio of 1:1500.
Partial digestion products were fractionated by pH 3.5 lonophoresis on
cellogel (Kalex) in the first dimension followed by homochromatography on
4179
Nucleic Acids Research
DEAE-cellulose thin layer plates (Analtech).
Fragments ranging from 15 to 50
nucleotides were fractionated with 3% homomixture B 1 8 , while more dilute homomixture B (about 1 to 1.5%) fractionated fragments ranging from A to 20
nucleotides in length 17 .
Each partial digestion product was identified by complete digestion with
Ti ribonuclease and ribonuclease A.
The secondary digestion products, frac-
tionated by homochromatography in 3% homomixture C in one dimension19, were
then analyzed by digestion with T2 ribonuclease, Ti ribonuclease (in the case
of ribonuclease A digestion products) or ribonuclease A (in the case of Ti
ribonuclease digestion products).
This method allowed the ordering of most of
the T 1 ribonuclease partial digestion fragments.
However, to order the
C(U,C)G isomer8, secondary Tj RNase products suspected of containing these
oligonucleotides were also analyzed by incubation with carbodiimide reagent
followed by digestion with ribonuclease A 1 8 .
Digestion with Si nuclease:
Specific cleavage of tRNAj
codon loop 29 was achieved by incubating carrier-free
32
at the anti-
P-tRNA (A to 6 ug, 2
to 5 x 10-" cpm, eluted from a two dimensional polyacrylamide gel) with Sj
nuclease in 10 ul of 0.09M NaCl, 0.03M Na acetate, pH 4.5, 0.001M ZnCl 2 , 0.01M
MgCl2, 5? glycerol at 20°C for eight to twelve hours.
ratio of 1 unit nuclease to 1 ug tRNA was used.
An enzyme to substrate
In some experiments, the tRNA
was renatured by heating to 60°C for three minutes and then cooled on ice for
five minutes before the addition of the nuclease.
The inclusion of magnesium
in the buffer minimized the cleavage by Si nuclease at tRNA loop regions other
than the anticodon, as determined by Ti ribonuclease fingerprint analysis of
cleaved tRNA (M. Zufliga, unpublished observations).
RESULTS
Purification and identification of Bombyx mori tRNAi y :
Bombyx morl low
molecular weight RHAs can be fractionated into approximately eight bands by
electrophoresis in one dimension on a 9% polyacrylamide gel in 7M urea (Figure
1, upper panel).
Fractionation in a second dimension (20X acrylamide, no
urea) generates about 20 spots as detected by autoradiography of
RNA (Figure 1, lower panel).
32
P-labeled
Note that gel bands 1, 2 and 3 of the first dimen-
sion each give rise to a single major spot in the second dimension.
Finger-
print analysis of the RNAs eluted from the two dimensional gel indicated that
spots 1, 2 and 3 indeed contain single RNA species which are greater than 80Z
pure (Figure 2 and reference 7).
Bombyx morl low molecular weight RNAs fractionated on the first dimension
9% gel were examined for their aminoacylation properties.
4180
Data for bands 1
Nucleic Acids Research
Acrylamide, 7 M Urea, pH 8 3
00
b.
T
Q.
U
X
<__>
U
O
cu
—
*—
Blu*
Posterior silk gland tRNAs
Figure 1. Two dimensional polyacrylamide gel electrophoresis of Bombyx mori
posterior silk gland tRNAs. "P-labeled low molecular weight RNA (4 x 10*
cpm) was prepared from fifth instar posterior silk glands by DEAE-cellulose
chromatography as described in Experimental Procedures. Fractionation of
RNAs in the first dimension (left to right, upper panel) was by electrophoresis on a 9% polyacrylamide gel containing 7M urea and in the second dimension
(top to bottom) by electrophoresis on a 20% polyacrylamide gel without urea.
XCFF marks the position of the xylene cyanol FF dye in the first dimension.
The positions of the major "red" and "blue" components of Trypan red in the
second dimension are indicated. Aminoacylatlon data obtained on the first
dimension bands labeled 1,2,3 and 4 are shown in Table I.
through 4 (which comprise the majority of the material in the 4S region,
Figure 1, upper panel) are given in Table I.
Bands 1 and 2 aminoacylate with
alanine and have been described previously7.
Band 3 aminoacylates primarily
with glycine (77Z of the theoretical value) and is called tRNA^ly.
The small
response to alanine (4.8Z) is probably due to contamination of band 3 by band
2.
Band 4 also responds primarily to glycine (about 76Z of the theoretical
value).
Together bands 3 and 4 account for about 70Z of the total glycine
acceptance activity in the posterior silk gland of the fifth instar larva.
The occurrence of pure RNA species in gel spots 1, 2 and 3 of the two
dimensional fractionation system (Figure 1, lower panel) permitted an examination of the relative level of these RNAs in several larval tissues.
Table II
4181
Nucleic Acids Research
1
I
f
»#
T, rlbonudeost fingerprint of ipot 3 t RNA * "
Figure 2. Tj ribonuclease fingerprint analysis of 32P-labeled t R N A ^ . 3 2 P _
labeled spot 3 tRNA (about 2.5 x 1 0 5 cpm), eluted from a two dimensional gel,
was digested with Tj ribonuclease. Products were fractionated in the first
dimension by electrophoresis on cellulose acetate at pH 3.5 and in the second
dimension by electrophoresia on DEAE-paper (DE81) in 7% formic acid. 0, B, P,
and Y represent the origin and the blue, pink, and yellow dyes, respectively.
Oligonucleotides are labeled as in Table III.
shows that the level of tRNA^ly (spot 3) relative to tRNA* 13 (spot 2) is
greater in late fifth instar posterior silk glands than it is in late fifth
instar middle silk glands or midgut tissue.
the hypothesis that tRNAi y , like tRNAi
fibroin biosynthesis.
These data are consistent with
, is selectively synthesized for
Although Table II al30 shows an increase in the rela-
tive level of tRNAj y during the maturation of the posterior silk gland
4182
Nucleic Acids Research
Aainoacylation of tRHA isolated frco posterior silk glands
of Boabyx morl fifth instar larvae
91 polyacrylanlda
gel b»nd
occurrtnce
relative to
band 2
1
2
3
4
1.3
1
2.8
2.5
aaino acid
tasted
Z aninoacylation
mean
ranRe
alanine
glycine
• •Tine
tyro*in*
leucine
proline
methionlae
75
1
0
1
0
0
0
56-95
alanin*
glyclna
serine
tyroiine
leucine
proline
methlonine
66
1
0
1
0
0
0
54-79
alanine
glycine
••rlne
tyroaine
leucine
proline
7
77
1
0
0
0
4-9
62-98
alanine
glycine
aerine
tyrofline
leucine
prollna
15
76
2
0
0
0
6-21
53-98
0-1
0-1
0-1
0
0
0
1-2
0
0-1
0-1
0-1
0
1-2
0
0
0-1
1-2
0
0
0-1
number
of trials
10
10
2
2
2
2
2
8
8
2
2
2
2
2
7
10
2
2
2
2
4
4
2
2
2
2
Table I. Low molecular weight RNA (32P-labeled in three preparations,
unlabeled in remaining preparations) was purified by DEAE-cellulose chromatography (sucrose gradients in one experiment), fractionated on a one dimensional 9% polyacrylamide gel and aminoacylated as described in Experimental
Procedures. At least two different tRNA preparations were examined for each
amino acid. Quantitation of the tRNA bands was achieved as described in
Experimental Procedures; data are expressed relative to the amount of band
2 = 1 . The observed percentage of aminoacylation was calculated by dividing
the observed cpm incorporated into TCA-precipitable
counts above background
by the expected TCA-precipitable counts if 100% aminoacylation occurred. In
all experiments the specific activity of 3 H- or ll|C-amino acid was set so
that at least 10,000 cpm were incorporated when 0.0A nM tRNA was used as
acceptor. Background values were about 100 cpm.
(fourth versus fifth instar), such results are not consistently obtained (see
Table II in reference 7 ) .
Sequence analysis of
ribonuclease:
products of complete digestion with
Figure 2 shows a two dimensional fractionation of the digestion
products generated by complete hydrolysis of spot 3 tRNA with T\ ribonuclease.
All of the eighteen Tj RNase digestion products are numbered in Figure 2, with
the exception of 19b which does not occur in all tRNAi
molecules.
Table III
4183
Nucleic Acids Research
Occurrance of tMAi
, tEHA^1*, and tlHAi
J
in larval tissues
tENAf/tENAf1'
tBHA^/tBHAf*
V inatar
(day 7)
1.3
3.2
mlddl< silk
gland (8)
V inatar
(day 7)
0.2
2.9
midgut (12)
V Inatar
(day 7)
0
1.9
posterior
silk gland (6)
IV instar
(day 5 )
0.5
2.4
tissua
stage
posterior
•ilk gland (12)
Table II. 32 P-lou molecular weight RNAs were prepared as described in Experimental Procedures and fractionated by two dimensional polyacrylamide gel electrophoresis as in Figure 1. The age of each larva was measured in days since
last molt (column two). The ratios shown for tRNAi a (spot 1), tRNA2 a
(spot 2), and tRNAi " (spot 3) are the mean values obtained from the number of
independent measurements given in parenthesis in column one. Minimum Cerenkov
radiation assayed was 101* cpm per gel spot.
summarizes the molar yields of the Ti ribonuclease digestion products of
tRNAi y .
Note that the ollgonucleotide AG does not occur in tRNAi y ; thus it
could be judged from yields of this product that tRNA preparations used for
sequence analysis were over 80% pure.
The sequences of most of the oligonucleotides from the T\ ribonuclease
fingerprint were readily established by quantitation of the products of ribonuclease A and T2 digestions, supplemented by additional analyses using ribonuclease U2 and/or RNase A after chemical modification with carbodiimide
reagent.
These data are presented in Table III.
In some cases the sequences
were confirmed by complete digestion with snake venom phosphodiesterase or by
partial hydrolysis with spleen phosphodiesterase (Table III). Oligonucleotidea which gave special problems are discussed below.
T; ribonuclease digestion product 5:
The U*p in oligonucleotide 5
(m1AU*UCCCG) has mobilities identical to those of Up in thin layer chromatography in two solvent systems20 and on 3 MM paper upon electrophoresis at
4184
Nucleic Acids Research
Cat
dltHtia a
product
5
prtxlo c c a
of tEJUj
J
MJlar yl« l d «
UU< A
T 2 DaM
otbar t r u t w
0.5O:3.OC:
1.2D:2.3C:
0.9»'AD* 1C
(b)-'A£*OC.C^. t«)1.0.'i:10.
0.6* l AD*i
0 9»'AU*DilC
6
• loSM ot Ti rlkooocl " * • di»"tioci
predact* of furtli*r dliafCloa
1.60I0.9C
1AC
• 1AD*tJCCCC
0 9
1
(•)AC,CAO,DCAC.
CDCAC
UUCAC
1 1
1
(c)pt,pC,pC
TtCC
1.1
1
AAOC
0.9
1
CAD mCC
1.3
1
AOC
1 1
1
CCOQ
CDCC
2.0
1 ot *acb
t{f)l.U*-«A:lD*
(1)CCCG,UCCCC.I lAE*OCCCC
2.2Dil.2O
O.SAilC
2-KT++) l.OC
1C
7
1.1(T+»):0 SC:
1C
10
1 1AAU 1C
2.tA 0.9O:lC
11
l.K:0.2D:
1.3Cil.0Ai
0.2Ui2.1(X •«)
l.QAX* 1C
axp*ct*d
j
0>)66c,Ai
(b)(T»)C,4.
(a)CA,IfC
1*. 1C11A
(
(b>C.Ai ,C
+ (f)D-C
+(«)pC.pCp
+ (h)D»,C
(•)n «CC,AD •CC.GAD »CC
12
l.LUftlC
14
0.8D 1.9C-1C
I.OO1I.6A 1C
0-«Ci2 SCilG
(b)C,DC,6c.C
{c)0 9pC'lpO:
l.Bppp
(e)OC,CC,(C,D)C,
C(C.01G
(k)c,c6i t dx.b
1 1
1
l.OOilC
1.0D:1.0AilC
l.tHJ 1G
OAC
17
uc
1.8
1
18
pC
PC
PC
0.8
1
19
0.90 L B ' C
0.9Oil«'c
1.10** 1C*
D-'C
u**c*
0 5
0 4
—4 t*xt
CCA» 5 C0
1.0
1
CACCA
1.2
1
16
l.LDilAC
19b
20
2.2CI1.0AJMSC
1C
21
3 KiLAC
22
2.6«sC.ia
14
l.SCilC
15
1
n
c
ICIIC
I.JCH O-SCJ
l.UilC
3.X:1A
{•XXA.B^X
(OCA.CCA^
(1)1.2CAJ2C
(c)0.9pAilpC
(j>1.0ACi2C
OB
1
3.3«*CilC
i.6CilC
1.1C 1C
•5c«sC«lCC
1.0
1
ccc
cc
1.1
1
Q
0
2.4
2
10.7
9
Table III. Products of T2 RNase digestion were fractionated by electrophoresis
on 3MM paper at pH 3.5.
Other treatments include the following:
(a) U2 ribonuclease digestion
(b) ribonuclease A digestion after chemical modification with carbodiimide
reagent
(c) digestion with snake venom phosphodiesterase
(d) total digestion with beef spleen phosphodiesterase
(e) partial digestion with beef spleen phosphodiesterase
(f) alkaline hydrolysis
(g) total digestion with Pi nuclease
(h) digestion with micrococcal nuclease
(i) periodate oxidation and 6-elimination followed by U2 ribonuclease digestion
(j) periodate oxidation and 6-elimination followed by ribonuclease A digestion
(k) treatment (b) on isolated isomers (see text for details)
Products of treatments (c), (d), (g), and (h) were fractionated by thin
layer chromatography in the two Nishimura solvent systems,
t
These digestions were performed on isolated dinucleotides.
pH 3.5. However, we believe U*p to be a modified nucleotide for the following
reasons:
1.
The 3'-5' phosphodiester linkage of m'AU* is not cleaved by T2 RNase
4185
Nucleic Acids Research
(Table III) under conditions which hydrolyze n^AU isolated from B^. mori
2.
n^AU* has a different mobility than n^AU upon electrophoresis on DEAE-
paper at pH 3.5.
3.
The 3'-5' phosphodiester linkage of n^AU* is hydrolyzed by alkali, ruling
out the possibility of a 2'-0-methyl group on the adenosine residue. Moreover, the adenosine residue displays the characteristics of n^A, including
rearrangement to N6-mA in the presence of alkali 24 and its mobility in two
different systems.
Thus, it is unlikely that the failure of T 2 RNase to
hydrolyze n^AU* is due to further modification of the A residue.
4.
Treatment of oligonucleotide 5 with RNase A (0.3 mg/ml, 37° C, one hour)
produces m:AU*U in addition to n^AU*, U, C, and G (Table III). Although this
slow hydrolysis of the U*U bond could be due to other causes, it is nonetheless consistent with possible modification.
T] ribonuclease digestion product 11:
In oligonucleotide 11 (CAX G ) ,
the assignment of U m C to the alkali-resistant dinucleotide X
by the fact that X
is supported
has the same mobility as the published value for C m U in
four different systems: electrophoresis on 3MM paper at pH 3.5 18 , descending
paper chromatography in the isopropanol:HCl:H20 system 18 ' 21 , electrophoresis
on DEAE-paper in IX formic acid25, and two dimensional thin layer chromatography 25 .
Digestion of X
with Pi nuclease gave pC (Table III), confirming
that C is the 3' nucleotide in X . Micrococcal nuclease digestion of X
gave
products whose mobilities In the isobutyric acid:NH3:H2O and isopropanol:HCl:
H2O systems 20 are consistent with the assignment U m C .
However, the latter
system did not give an unambiguous result, since U m p and P
mobility.
have the same
Finally, the assignment U m C is consistent with the observation
that a small percentage of the tRNAi
molecules are undermodifled at posi-
tion 4 ( O B ) and give free U (and increase the molar amount of C) upon RNase
A or T 2 RNase digestion of oligonucleotide 11 (Table III).
Ti ribonuclease digestion product 14:
nucleotide in tRNAj
y
There are two moles of this tetra-
(Table III). Analysis of the products obtained by ribo-
nuclease A digestion with and without carbodiimide modification and of the
products of snake venom phosphodiesterase suggested that this oligonucleotide
is an equimolar mixture of two isomers, CUCG and CCUG (Table III). Chromatography of oligonucleotide 14 on a DEAE-cellulose thin layer plate in 7.5Z
triethylamlne carbonate, pH 10, yielded two components.
RNase A digestion
after carbodiimide modification of these fractionated isomers yielded unique
U- and G-containing products (Table III) . The sequence assignments CCUG and
4186
Nucleic Acids Research
CUCG were confirmed by partial digestions of oligonucleotide 14 with spleen
phosphodiesterase (Table III).
T] ribonuclease digestion products 19 and 19b: heterogeneity at positions
8 and 9 in tRNA^f y :
Most preparations of tRNAj
y
contained only Tj ribonucle-
ase product 19, I W G (Table III). However several preparations contained, in
addition to oligonucleotide 19, another product 19b in low yields (0.4 moles,
*#
40
f
m
Ribonucltox A fin«trprint of ipot 3 I R N A 6 "
Figure 3. Two dimensional fractionation of the ribonuclease A digestion
products of tRNA? v . 32P-tRNA?1>r (spot 3 tRNA, 2.5 x 10 6 cpm) was obtained
and fingerprinted as in Figure 2, except that RNase A rather than Ti RNase
was used. 0, B, P, and Y show the positions of the origin and the blue, pink
and yellow marker dyes, respectively. Oligonucleotides are numbered as in
Table IV.
4187
Nucleic Acids Research
Table III). The occurrence of product 19b did not appear to be correlated
with the developmental stage of the larva.
Partial digestions of tRNAj
with Ti ribonuclease showed that 19 and 19b occurred in the same location
(positions 8 and 9) in the tRNA molecule.
Oligonucleotide 19b appears to be U**G*, where U** and G* are both modified nucleotides (Table III). U** is not ifi, T or S"*U as determined by its
mobility on thin layer chromatography20.
of its chromatographic mobility
18
paper at pH 3.5 .
fied.
20
G* is not m 2 G or m^G on the basis
or its electrophoretic mobility on 3MM
It may be that G* is a m 1 G residue which is further modi-
A second methylation (for example at the N 2 position) could occur,
however no tRNA sequenced to date has m^G.
position 9 in tRNAi
Alternatively, heterogeneity at
could be due to degradation or undermodification of
this G residue (see for example the yields of GGU in RNase A digestions of
this RNA, Table IV). Heterogeneity at position 9 has also been reported for
tRNA\ ys isolated from haploid yeast 30 ' 31 ; in this case m 2 G occurs with 30Z
frequency and G occurs with 70% frequency in all preparations of tRNA.
Ti ribonuclease digestion product 21: This oligonucleotide contains no
Gly
G and therefore must be the 3' terminus of the tRNAi
molecule. RNase A
digestion consistently gave 3.9 C to 1.0 mole AC (Table III), and U2 RNase
digestion gave only CA and another product which yielded C upon T2 RNase
digestion.
This suggested the sequence CACCCA QH .
However, equal yields of
pA and pC from snake venom phosphodiesterase (svpde) digestion of oligonucleotide 21 (Table III) were compatible with the sequence C A C C A Q H rather than
CACCCA,,,,. The conflicting RNase A and svpde data were reconciled by the
Un
following experiment. Oligonucleotide 21 was subjected to periodate oxidation and B-elimination and the product chromatographed on PEI-cellulose thin
layers in 3Z homomixture C, along with CCG, m 5 Cm 5 Cm 5 CG. AAUG, CCAm5CG, and
uTJCAG (as size markers).
The resulting oligonucleotide comigrated with
m 5 Cm 5 Cm 5 CG and AAUG, suggesting it was a tetranucleotide.
Subsequent RNase
A and RNase U2 digestion yielded products which gave the order CACC to the
periodate oxidation-g-elimination product (Table III). The high yields of C
obtained in RNase A (and T2 RHase) digests of oligonucleotide 21 may be due
to a higher specific activity of C residues in the CCA . sequence, which is
usually added to the tRNA posttranscriptionally32.
Products of complete digestion with ribonuclease A:
G
nuclease A fingerprint of tRNA ^y.
are listed in Table IV.
Figure 3 is a ribo-
The molar yields of the digestion products
All of the RNase A oligonucleotides were sequenced
by secondary digestions with ribonucleases Ti and T2 (Table IV), with the
4188
Nucleic Acids Research
A dlgaatloo pradncta
dlgaatloo
prodact
prodacta of furtbar dltMtloa
•olar
ial
•pi MO
!,«.„
TlUMa
dwloccd
1
2.KilT
2.6C:1T
CCCT
0.8
1.0
2
Z.2CilC/9
2.0CilD/D
CCD.CCD
2.1
1 of Mch
3
0 7»1CC:10i
0.7»'C:1.0C
1.2..C.0.K.H,
•'occ
0-6
I
4
2 KtlC
l.SCilC
CCCC
0.6
1
5
l.OACiLAAD
3.2A:l.lCilD
AU4D
1.0
1
t
1 OACilD
l.OAil.OCilD
ACU
1.0
1
7
l.OCilAD
1.1C1.4A.1U
GAD
1-2
1
•
1 9C.1C
1 KilC
ccc
0 9
1
9
1.7C 1»*C
1.W.1-5C
10
1.2C.1* 1 AD*
l.OC:l»1AD*
11
0 7pCilC
13
14
a*p«t..
1.0
1
C'AU*
0 e
1
0 7pCilC
pec
0.7
1
AD.C
lA.o.nr-c
AOB
1.1
1
0 SCilC
1 OC:1C
cc
(total dltastlon)
13
C
Z lAil.lC
16
c.'c
17
0,*
c,.>c
0,*
4.9
5
1.8
2
9.5
13
3.6
5
Table IV. The observed molar yields are based on the results of seven experiments and are calculated relative to the average cpm/POi, in AGAAU, GGC, and
AGU. Oligonucleotide 2 contains roughly 1.3 to 1.4 moles GGU to one mole GGD,
presumably because of undermodification of D (in GGD) and undermodification of
n^G (in m'GGU,oligonucleotide 3 ) . The molar yields of oligonucleotides 9 and
10 were determined after correction for cross-contamination; these two trlnucleotides were separated by elec'trophoresis on DEAE-paper at pH 3.5 before
further analysis. Products of T2RNase digestion (and total spleen phosphodiesterase digestion, oligonucleotide 10) were fractionated by electrophoresis
on 3MM paper at pH 3.5.
exception of ribonuclease A digestion product 3, m'GGU, which was solved by
partial spleen phosphodiesterase digestions (Table IV).
Partial digestion with T; RNase: derivation of an unambiguous sequence
for tRNAt
y
: The sequence of tRNAj y was established by overlapping the
products generated by limited digestion with Tj RNase.
three experiments are summarized in Figure 4.
The data obtained in
Each partial digestion fragment
(except the two discussed below) was readily identified by analysis of its
constituent products after complete RNase Tj and A digestion.
T; RNase partial digestion product 1:
The structure of this product
must be GCGGm5Cm5Cm5CG for the following reasons.
5
5
5
Overlapping of the products
5
of total Ti RNase (m Cm Cm CG) and RNase A (GGm C) digestion gave the partial
sequence GGm5Cm5Cm5CG.
Since GGm5C (and not Gm 5 C) was observed among the
4189
Nucleic Acids Research
31
5'
l
5
5
5
l
PGCAUmCGGUn GGUUCAGUGGDAGAAUGCUCGCCUGCCAni CGCGGGC6Gm Cim CGGGTi|rCGm AUUCCCGGCCGAUGCACCAw
F i g u r e 4 . D e r i v a t i o n of the sequence of 15. mori tRNAi y . The b r a c k e t s above
and below the l i n e a r sequence i n d i c a t e t h e Ti r i b o n u c l e a s e p a r t i a l d i g e s t i o n
products used to d e r i v e the sequence. These were obtained from t h r e e d i f f e r ent experiments. All products shown were obtained more than once except for
CCGAUG.
CACCA,,,
AUG
y
ue
DAS
P
Tyco
J
o
It
al«ctropnor«sis. pM 1 5
AUO,
OCUO "CCG
T^ICO
AAUG
UUCAG *
»CAU-iCG
N*.AlA>CCCG *
—
.UUSUCCCQ
ctectraprioresift.pH 3.5
Figure 5a. Two dimensional fractionation of a Ti ribonuclease digest of the
3'-half of tRNA? ly . Tj partial digestion product 3 (Figure 4) was purified
of other surrounding partial digestion products by two dimensional polyacrylamide gel electrophoresis according to De Wachter and Fiers^. The eluted sample
was then digested to completion with Tj RNase, and the products separated by
electrophoresis on cellulose acetate (Kalex) at pH 3.5 in the first dimension
followed by homochromatography in the second dimension. 0, B, P, and Y refer
to the origin and the blue, pink, and yellow dye markers, respectively. Note
the breakdown of oligonucleotide 5 (m1AU*UCCCG), which is accelerated by the
high temperatures used in homochromatography.
Figure 5b. Two dimensional fractionation of a Tj ribonuclease digestion of
tRNAY ly . Intact tRNA^ly was digested with Ti and the products fractionated
as described in Figure 5a. Oligonucleotides were analysed by digestion with
RNase A and T 2 RNase.
4190
Nucleic Acids Research
RNase A digestion products of Ti partial digestion product 1, GGm5C cannot
occur at the 5' terminus and must be preceded by a pyrimidine in the fragment.
This placed the Ti RNase-generated dlnucleotide, CG, 5' to Gm 5 Cm 5 Cm 5 CG to give
CGGm 5 Cm 5 Cm 5 CG.
Furthermore, RNase A digestion of Ti RNase partial digestion
product 1 yielded the dinucleotide GC.
5
5
Thus, the sequence of this RNA frag-
5
ment is GCGGm Cm Cm CG.
T] RNase partial digestion product 2: Tx RNase partial digestion product
2 must be a mixture of two partial digestion products:
5
CCAm CGCG
C(U,C)G,CCAm5CG and
where the exact length of the latter product is unknown.
The
Gly
Specific cleavaje of tRHAi J with Si nuclease
molar yield tRHA<;ly
Tj ribanucleas*
digestion product
control
.'ABIICCCG
1.0
UDCAG
1.1
1.0
T»CG
0.9
0.9
AAUG
1.1
1.1
CAUsCC
1.0
0.8
AUC
1.2
1.1
CCUG + CUCG
2.1
1.6
Si
nucleafle-treated
1.1
DAG
1.0
1.0
UG
1.6
1.9
PC
0.9
0.9
Un'C
0.4
0.8
CCAB5CG
1.2
0.6
CACCA0H
1.7
—
a Ci> Cii CG
1.4
1.2
CCC
1.3
1.1
5
5
5
CG
2.6
1.9
C
8.9
11.2
Table V. 32 P-tRNAj y (2 to 5 x 101* cpm), purified by two-dimensional polyacrylamide gel electrophoresis (Figure 1 ) , was digested with Si nuclease and
the products fractionated from undigested material on a 9Z polyacrylamide gel.
Half molecules (which comigrated in this gel system) were eluted and analyzed
by Tj ribonculease fingerprint analysis using homochromatography in the second
dimension. Tj RNase digestion products were quantltated by direct Cerenkov
counting. The molar yields are average values obtained from three experiments
(expressed relative to DAG • 1). The yields of Um'G varied considerably from
preparation to preparation, presumably because this oligonucleotide migrates
near the front on PEI-cellulose and is sometimes lost. The identities of the
Ti RNase digestion products were confirmed by secondary analysis with RNase A.
4191
Nucleic Acids Research
reasons for this assignment are as follows.
We obtain two moles of CCAm5CG
for each mole of C(U,C)G in complete Ti RNase digests of partial digestion
product 2.
Since there is only one mole of CCAm5CG
and two moles of C(U,C)G
in intact tRNAj^ ^, product 2 must be a mixture of digestion products both of
which contain CCAm5CG.
In the complete RNase A digests of the mixture, the
products Am^C and GC occur in equal amounts, from which we conclude that
there must also be two moles of each in partial digestion product 2.
However,
there is only one mole of CG present in the complete T\ RNase digests.
Thus,
the two RNA fragments present in partial digestion product 2 must be
C(U,C)G,CCAm5CG and CCAm5CG,CG.•••, in which the constituent Ti RNase products
are not ordered.
The orders given in Figure A are based on the observation
that Tj partial product 3, which is the 3'-half of the molecule (Figure 5 ) ,
contains two moles of CG (as determined by visual inspection of the autoradiograph) and one mole of CCAm5CG but no C(U,C)G.
Thus, C(U,C)G must be 5' to
CCAm5CG, and by process of elimination, CG must be 31 to CCAm5CG.
The penta-
5
nucleotide CCAm CG is placed in the anticodon loop on the basis of data
obtained in experiments using S\ nuclease, which cleaves intact tRNAj
y
spe-
cifically within this oligonucleotide (Table V ) .
DISCUSSION
tRNA; y is a major decoding species in the mature posterior silk gland:
Bombyx mori tRNAj y , as isolated from our two dimensional gel system, was
identified as a pure species which represents 10X to 20* of the total tRNA
and about 402 of the glycine acceptance activity in the mature posterior silk
gland.
Although this tRNA is not unique to the posterior silk gland, we conclude
that tRNA^ly accumulates to meet the needs of rapid fibroin synthesis for the
following reasons:
(a)
The relative amount of tRNAj" v is greater in the mature posterior silk
gland than in other larval tissues.
(b)
The derivation of the complete sequence of tRNA^lv shows that the mole-
cule has a GCC anticodon, which is theoretically capable of decoding the major
glycine codon in the fibroin mRNA, GGU.
Moreover, comparison of our data with
codon-response studies on glycine-accepting tRNAs fractionated by countercurrent distribution31* suggests that tRNA^lv is the major GGU-decoding species
in the posterior silk gland of J5. mori.
The occurrence of GCC rather than an ICC anticodon (which could also pair
with GGU) in glycine tRNAs was predicted by Ninio 35 on the basis of kinetic
and fidelity arguments.
4192
ACC would not be expected to appear since A is never
Nucleic Acids Research
found in the first position of the anticodon 36 ' 35 .
Analysis of the frequency
of codon usage in MS2 bacteriophage RNA 3 7 ' 3 8 ' 3 9 has led to the suggestion
(H. Grosjean, R.J. Cedergren, D. Sankoff, W. Min-Jou, H. Fiers, personal communication) that there is an optimal half-life for codon-anticodon interaction.
The occurrence of a GCC anticodon in a major posterior silk gland gly-
cine tRNA and of a GGU codon as the major glycine codon in the fibroin mRNA
would appear to support this hypothesis.
Also consistent is the fact that
the glycine codon GGC occurs in only 10.4% of the possible glycine positions
in the fibroin mRNA 3 .
Another interesting feature of the tRNAj
a U residue 5
1
v
sequence is the appearance of
and an unmodified A residue 3' to the anticodon triplet.
While
these nucleotides are not unexpected in these positions, their potential contribution to tRNA-mRNA interaction can now be evaluated since the crystalline
portion (greater than 60%) of the silk fibroin protein is known to have the
repeating sequence:
Gly-Ala-Gly-Ala-Gly-[Ser-Gly-(Ala-Gly)^2]8-Ser-Gly-Ala-Ala-Gly-Tyr1*.
Thus,
Glv
tRNAi } could potentially form five base pairs with the fibroin mRNA when
glycine resides between two alanines, for which the major codon is GCU , four
base pairs could be formed when glycine occurs between an alanine and a serine
residue, for which the major codon is UCA3.
These observations are compatible
with ideas concerning the current nature1*" and the origin
of the protein
synthetic apparatus.
Glv
Structural homology of B. mori tRNAi ' to other glycine tRNAs:
1 2 1 3
11
The
1 5 1 6 1 7
sequence of eight prokaryotic * ' ' '' *' ' ' * ' * and three eukaryotic (yeast49,
wheat germ 50 , and Bombyx mori tRHAG(ly ) are now known.
On the basis of these
sequences one can generalize that glycine tRNAs tend to be small (73 to 76
nucleotides in length), to have small variable loops (three to five nucleotides) and to have few modified residues.
Otherwise, there are no clear
sequence homologies among all the various prokaryotic and eukaryotic glycine
tRNAs.
Among the three eukaryotic glycine tRNAs (all of which have a GCC anticodon) distinct similarities do exist.
Excluding posttranscriptional modifi-
cations (modified bases and the CCA-u end), B. mori tRNA G l v has 66% homology
uti
—
*
with yeast tRNA Cly and 78% homology with wheat germ tRNAGly, suggestive of an
evolutionary relationship48.
Interestingly, the homologies among these three
tRNAs are localized in certain areas (Figure 6): loop I
IV (TifiC loop) and the T^C stem.
(the D loop), loop
The acceptor stems also exhibit some common
sequences, but the major feature of interest here is a posttranscriptional
4193
Nucleic Acids Research
Gly
Figure 6. Cloverleaf structure of J3. morl tRNAi . Regions of homology
among the three eukaryotlc glycine tRNAs sequenced to date are Indicated by
boxed areas. Arrows denote differences in modification where there is
Aj
sequence homology; in each case only one of three is different. Yeast tRNAj
(73 nucleotides in length) has a three-membered extra loop (loop III), whereas
.B. morl tRNA^y and wheat germ tRNA^ly (74 nucleotides in length) each have a
four—membered extra loop.
modification at position 4.
All three molecules have a pyrimidine with a
2'-0-methyl moiety on the ribose group at this position.
A modification of
this type in the acceptor stem is thus far unique to these tRNAs.
It is
tempting to suggest that the above mentioned homologous features may be important to recognition by the glycyl tRNA aminoacyl-ligase.
In addition, jJ. mori tRNA] y shares two further sequence homologies with
wheat germ tRNAj y :
identical anticodon loops with m"C at position 37 and
three additional adjacent 5-methyl cytidine residues in positions 46 to 48
(Figure 6) of their sequences (corresponding to positlono 48 to 50 in the
standard tRNA structure 51 ' 52 ). Five-methyl-cytldine has been found only in
eukaryotic tRNAs.
In 29 of 35 known sequences it occurs at position 48 (where
it presumably participates in the Levitt base pair 51 ) or at position 49 (in
the TtyC stem) of the standard tRNA structure.
Of the six eukaryotic tRNAs
5
lacking m C, five are from the yeasts (Saccharomyces cerevisiae and Torulopsis
utilis).
Thus m 5 C at positions 48 and/or 49 or the standard tRNA structure
appears to be a general feature of tRNAs in "higher" eukaryotes, though it has
yet to be assigned a function.
4194
Nucleic Acids Research
Although the occurrence of one or two m 5 C residues is common in eukaryotic tRNAs, the four m 5 C residues in tRNAj ' of Bombyx mori and wheat germ
are novel.
It has been suggested previously that the high content of m 5 C in
wheat germ tRNAj y , coupled with its lack of ribothymidine at U52, allows it
to function more efficiently in protein synthesis 53 ' 50 .
tRNAj
y
The fact that J3. mori
5
possesses both ribothymidine and a high m C content (and yet is pre-
sumably highly active on the ribosome5"*) suggests that the functional relationship of these two methylated residues may be more complex. Studies of
Glv
the protein synthetic capabilities of 15. mori tRNAj * are needed to evaluate
the role of these modified residues in tRNA function.
Recently, Garel and Keith 55 , using a different approach, deduced a comparable sequence for 15. mori tRNAj
ACKNOWLEDGEMENTS
We thank N. Farber, P.M.M. Rae, D. Soil and K.U. Sprague for advice, and
M. Krikeles and C. Stocking for assistance.
This work was supported by
National Science Foundation Grant PCM74-01136 and forms part of a dissertation
submitted by M.C.Z. to the Graduate School of Yale University in partial fulfillment of the requirements for the PhD degree.
M.C.Z. was a recipient of the
Ford Foundation Fellowship for Mexican Americans.
REFERENCES
1
2
3
4
5
Sprague, K.U. (1975) Biochem. 1A, 925-31
Suzuka, I., Tanaka, S. and Shimura, K. (1962) J. Biochem. 52, 54-58
Suzuki, Y. and Brown, D.D. (1972) J. Mol. Biol. 63, 409-429
Lucas, F. and Rudall, K.M. (1968) Comp. Biochem. 26, 475-558
Garel, J.P., Mandel, P., Chavancy, G. and Daillie, J. (1970) FEBS Lett. 7,
327-329
6 Matsuzaki, K. (1966) Biochim. Biophys. Acta 114, 222-226
7 Sprague, K.U., Hagenbuchle, 0. and Zufiiga, M.C. (1977) Cell 11, 561-570
8 Sprague, K.U., Kramer, R.A. and Jackson, M.B. (1975) Nucl. Acids Res. 2,
2111-2118
9 Holley, R.W., Apgar, J., Doctor, B.P., Farrow, J., Karini, M.A. and Merrill, S.H. (1961) J. Biol. Chem. 236, 200-202 2
10 Landy, A., Abelson, J., Goodman, H.M. and Smith, J.D. (196/) J. Mol. Biol.
29, 457-471
11 Zubay, G. (1962) J. Mol. Biol. 4, 347-356
12 Ikemura, T. and Dahlberg, J. (1973) J. Biol. Chem. 248, 5024-5032
13 Peacock, A.C. and Dingman, C.W. (1967) Biochem. 6, 1818-1827
14 Studier, F.W. (1973) J. Mol. Biol. 79, 237-248
15 Gallbert, F., Sedat, J., and Ziff, E. (1974) J. Mol. Biol. 87, 377-407
16 White, B.N., Tener, G.N., Holden, J. and Suzuki, D.T. (1973) J. Mol. Biol.
74, 635-651
17 Zufiiga, M.C. (1977) PhD Thesis, Department of Biology, Yale University,
New Haven
18 Brownlee, G.G. (1972) Determination of Sequences in RNA. Laboratory Techniques in Biochem. and Mol. Biol., Work, T.S., and Work, E., Eds. North
Holland/American Elsevier. Amsterdam
4195
Nucleic Acids Research
Squires, C , Lee, F., Bertrand, K., Squires, C.L., Bronson, M.J. and
Yanofsky, C. (1976) J. Mol. Biol. 103, 351-381
20 Nishimura, S. (1972) Prog. Nucleic Acid Res. Mol. Biol. 12, 49-85
21 Wyatt, G.R. (1951) Biochem. J. 48, 581-584
22 Yamada, Y. and Ishikura, H. (1975) Biochim. Biophys. Acta 402, 285-287
23 Mikulski, A.J., Sulkowski, E. and Stasiuk, L. (1969) J. Biol. Chem. 244,
6559-6565
24 Hall, R.H. (1971) The Modified Nucleosides in Nucleic Acids. Columbia
Univ. Press. New York, London
25 McClain, W.H., Barrell, B.G. and Seidman, J.G. (1975a) J. Mol. Biol. 99,
717-732
26 Fujimoto, M., Kuninaka, A., and Yoshino, H. (1974) Agrlc. Biol. Chem. 38,
1555-1561
27 S a n g e r , F . and B r o w n l e e , G.G. (1967) i n G e n e t i c E l e m e n t s , S h u g a r , D . , Ed.
FEBS Symp. Warsaw, 1 9 6 6 . Academic P r e s s , New York
28 K e i t h , G. and Gilham, P . T . (1974) Biochem. 1 3 , 3601-3606
29 H a r a d a , F . and D a h l b e r g , J . E . (1975) N u c l . A c i d s R e s . 2 , 865-871
30 S m i t h , C . J . , L e y , A . N . , D'Obrenan, P . and H i t r a , S.K. (1971) J . B i o l .
Chem. 2 4 6 , 7817-7819
31 S m i t h , C . J . , Teh, H . - S . , Ley, A . N . , D ' O b r e n a n , P . (1973) J . B i o l . Chem.
248, 4475-4485
32 Deutscher, M.P. (1973) Progr. Nucleic Acid Res. Mol. Biol. 13, 51-92
33 De Uachter, R. and Fiers, W. (1972) Anal. Biochem. 49, 184-197
34 Garel, J.P., Hentzan, D. and Daillie, J. (1974) FEBS Lett. 39, 359-363
35 Ninio, J. (1973) Progr. Nucleic Acid Res. 13, 301-307
36 Jukes, T.H. (1973) Nature 246, 22-26
37 Fiers, W., Contreras, R., Duerinck, F., Haegeman, G., Merregaert, J., Min
Jou, W., Raeymaekers, A., Volckaert, G., Ysebaert, M., Van de Kerckhove,
J., Nolf, F. and Van Montagu, M. (1975) Nature 256, 273-278
38 Fiers, W., Contreras, R., Duerinck, F., Haegeman, G., Iserentant, D.,
Merregaert, J., Min Jou, W., Molemans, F., Raeymaekers, A., Van den
Berghe, A., Volckaert, G. and Ysebaert, M. (1976) Nature 260, 500-507
39 Min Jou, W., Haegeman, G.. Ysebaert, M. and Fiers, W. (1972) Nature, 237,
82-88
40 P i e c z e n i k , G. (1972) PhD Thesi3, New York U n i v e r s i t y , New York
41 Crick, F.H.C., Brenner, S., Klug, A. and P i e c z e n i k , G. (1976) Origins of
Life 7, 389-397
42 Barrell, B.G., Coulson, A.R., and McClain, W.H. (1973) FEBS Lett. 37,
64-69
43 H i l l , C.W., Combriato, G., S t e i n h a r t , W., Riddle, D.L. and Carbon, J .
(1973) J . B i o l . Chem. 248, 4252-4262
44 Riddle, D.L. and Carbon, J. (1973) Nature New Biol . 242, 230-234
45 Roberts, J.W. and Carbon, J. (1975) J . B i o l . Chem. 250, 5530-5541
46 S q u i r e s , C. and Carbon, J . (1971) Nature New B i o l . 233, 274-277
47 S t a h l , S . , Paddock, G. and Abelson, J . (1973) Biochem. Biophys. Res.
Commun. 54, 567-569
48 Holmquist, R., Jukes, T.H. and Pangborn, S. (1973) J . Mol. B i o l . 78, 91116
49 Yoshida, M. (1973) Biochem. Biophys. Res. Commun. 50, 779-784
50 Marcu, K.B., Mignery, R.E. and Dudock, B.S. (1977) Biochem. 16, 797-806
51 Levitt, M. (1969) Nature 224, 759-763
52 Slgler, P.G. (1975) Ann. Rev. Biophys. Bioeng. 4, 477-527
53 Marcu, K.B. and Dudock, B.S. (1976) Nature 261, 159-162
54 Suzuki, Y., Gage, L.P. and Brown, D.D. (1972) J. Mol. Biol. 70, 637-649
55 Garel, J.P. and Keith, G. (1977) Nature 269. 350-352.
19
4196