Nucleotide sequence of Scenedesmus obliquus cytoplasmic initiator tRNA

Nucleic Acids Research

Volume 8 Number 4 1980

Nucleotide sequence of Scenedesmus obliquus cytoplasmic initiator tRNA

P.O.Olins* and D.S.Jones Department of Biochemistry, University of Liverpool, P.O. Box 147, Liverpool L69 3BX, UK

Received 25 December 1979 ABSTRACT The cytoplasmic initiator tRNA from the green alga Scenedesmus

obliquus has been purified and its sequence shown to be p A G C U G A GUmC G G C G C A G D G G A A G C m2G G G C U C A U t6A A C C C A U A G m7G D m5C A C A G G A U C G m A A A C C U dm U C U C A G C U A C C A-O H. The sequence has been deduced and confirmed using several different 32P-post labelling techniques. The sequence is similar to those of other eukaryotic cytoplasmic initiator tRNAs and it has the sequence G A U C in place of the usual G TIfC. Although it resembles lower eukaryotic species in having a U preceding the anticodon and a modified G in the T, C stem, in overall homology it is closer to the higher eukaryotic than to the fungal initiator tRNAs.

GA

A4'G

INTRODUCTION The sequences of the cytoplasmic initiator tRNAs from several sources Comparison of these sequences shows that those isolated

are now known.

from mammalian sources are homologous [l] and that there is considerable

homology between those isolated from mammals, yeast, Neurospora crassa and wheat germ [2]. Scenedesmus obliquus is a unicellular green alga that contains a large

chloroplast.

[3]

We have reported the isolation of tRNA from this organism

and partial purification of the cytoplasmic initiator tRNA

[4].

Here

we describe the sequence of this tRNA and compare it with the sequences of those obtained from other sources. The sequence has been deduced from the results of three sets of

experiments. 1) Complete digestion of the tRNA with RNase A and RNase T1, separation of the fragments by fingerprinting [5] and determination of their sequences. This gave a good indication of the positions of modified nucleosides. 2) Sequencing of [5'-'32p-labelled tRNA. The sequence of fifteen nucleotides at the 5'-end of the molecule were deduced by partial nuclease P1 digestion followed by separation of the products on two

©) IRL Press Limited, 1 Falconberg Court, London W1V 5FG, U.K.

715

Nucleic Acids Research [5'-32P]P-labelled tRNA was sequenced using the rapid read-off gel technique introduced by DonisKeller et al. [8]. In addition to the use of the extra-cellular enzyme dimensional homochromatography [6,7].

from Bacillus cereus [9] which is pyrimidine specific and the Physarum

polycephalum RNase, phyI, which cleaves after all residues except C [10], U's and C's were distinguished by measurement of the distance between successive bands in the formamide ladder. 3) Limited hydrolysis sequence method described by Stanley and Vassilenko [11].

fast and accurate.

This method proved to be

It gave most of the sequence except for 5 nucleotides

at the 3' end and allowed modified nucleosides to be assigned. The sequence of S. obliquus tRNAMet has 87% homology with wheat germ tRNAMet [2,12] and 65% homology with all other eukaryotic cytoplasmic

initiator tRNAs whose sequences have been published [2,13,,14]. Although many initiator tRNA sequences are known this represents only the second published sequence of an initiator tRNA from the green plant kingdom and the first sequence of a green algal tRNA.

MATERIALS AND METHODS

Analytical grade chemicals were used where available. Uniformly labelled [1XC] methionine (specific activity 50-200 Ci/mol), sodium boro ['H] hydride (specific activity 7.2 Ci/mmol) and [y_32p] ATP (specific activity 2000-4000 Ci/mmol) were obtained from the Radiochemical Centre, Amersham, U.K. Enzymes. Calf intestinal alkaline phosphatase (CAP) and yeast hexokinase were obtained from Boehringer Mannheim (BCL, London). RNases T1, T2 and U2 (Sankyo) were purchased from Calbiochem. RNase A and snake venom phosphodiesterase were supplied by Worthington Enzymes. Nuclease P1 from Penicillum citrinum and T4 polynucleotide kinase were obtained from PL Biochemicals. The extra-cellular RNase from Bacillus cereus and Phy 1 from Physarum polycephalum were a generous gift from Dr. U.L. RajBhandary. Reagents.

Chromatographic and electrophoretic materials. DEAE-cellulose (DE-32) for column chromatography, DEAE paper (DE-81) (old type) and 3MM paper were from Whatman. Machery-Nagel Polygram Cel 300 cellulose thin layers were obtained from Camlab, U.K. Cellulose acetate strips were purchased from Schleicher and SchUll. DEAE-cellulose thin layers (500 pLM) were prepared by briefly homogenising microcrystalline cellulose (14 g) (DC Pulver 144, Schleicher and Schtll), Fibrous cellulose (14 g) (MN 300 HR, Machery Nagel) and DEAE-cellulose

716

(4 9) (MN 300 DE) inl6o ml of 4,0 and spreading on glass

Nucleic Acids Research plates).

Benzolylated DEAE-cellulose (BD-cellulose) was prepared according

to the method of Gillam et al. [15].

Arginine-agarose was prepared as

described by Porath and Fornstedt [16].

RPC-5 resin was prepared according

to method 'C' of Pearson et al. [17].

Assays. Methods for isolation of crude aminoacyl-tRNA synthetase from E. coli [18] and S. obliquus [3] and assays for aminoacylation [19] and formylation [4] have been described previously. Isolation of S. obliquus tRNA and purification of the cytoplasmic initiator tRNAMet. S. obliquus tRNA was isolated and 4000 A260 units were fractionated on

BD-cellulose as described previously [3].

The fractions were assayed for

methionine acceptance and formylation. Fractions (54-60) containing the first major peak of methionine accepting material were pooled. 73 A260 units of this partially purified material were added to a column of arginine-agarose (2 x 80 cm) and eluted with a linear gradient (1.5 1) of 0.35 - 0.7M NaCl in 20 mM Tris HC1 (pH 7.5), 10 mM MgCl2. Fractions (116-124) containing the single sharp peak of methionine acceptor activity were pooled. 11.9 A260 units of this material were added to an RPC-5 260~~~~ column (0.65 x 55 cm) and eluted at 37 and under a pressure of 160-180 p.s.i. with a linear gradient (150 ml) of 0.5 - o.8 M NaCl in the above Met buffer to yield 5.8 A260 units of highly purified tRNA. Nucleoside analysis was performed according to the tritium derivative method described by Randerath et al. [20,213 except that CAP was used in place of bacterial phosphatase. Total digestion of tRNA and analysis of fragments. These experiments were carried out using methods described in detail in ref. [22]. The digestions with RNase T1 or RNase A were carried out on a small scale [23], the resulting fragments were 51-labelled with 32P [24] and separated by fingerprinting [5]. Fragments were partially digested with either snake venom phosphodiesterase

[23] or nuclease P1 [7].

The digestion products

from RNase A and RNase T1 fragments were separated on DEAE-cellulose paper

electrophoresis at pH 1.9 or pH 3.5 respectively and their sequences deduced by M-value determination [25]. The sequences of some fragments were determined or confirmed by mobility shift analysis on two-dimensional homochromatography [6,23]. Met

3

5'-End group labelling of tRNA4 with P32p was carried out by a modification of the method described by Silberklang et al. [7,26].

717

Nucleic Acids Research Partial digestion of shift analysis

using

a

was

[5'-_32P]-labelled

carried out

as

tRNA with nuclease P1 and mobility

described by Silberklang

et

al. [7]

carrier tRNA-P1 ratio of 100:3 (w/w).

Rapid read-off sequencing of [5 '_32P]-labelled tRNAMet was based on the method described by Donis-Keller et al. [83 and Simonesits et al. [27]. The ladder

was

digestions

were

produced by hot formamide hydrolysis [27] and partial performed with RNases T1, RNase tA, Phy 1 RNase and the

extra cellular RNase from B. was

carried out

on

Polyacrylamide gel electrophoresis

cereus.

10% gels.

Limited hydrolysis sequencing of tRNA

was

carried out largely according to

Stanley and Vassilenko [11_. [ 1 1 ] . 11 jig of S. obliquus tRNAi was hydrolysed at 1000 for 10 mins. in 5 Ill formamide. After [5-32P]-labelling, the oligonucleotide products (0.35

mm

the gel

thickness).

The

separated

on a

10% polyacrylamide gel

[5'-32P]-oligonucleotides

contained in bands

on

isolated, and hydrolysed to the [5!-I)2P]-mononucleotide with

were

nuclease P1

were

as

described by Kuchino et al. [13] except that with the thin

gel the oligonucleotides eluted well from the gel pieces without homo-

genisation.1[5'-'32P]-mononucleotides standard nucleotide markers

which did not comigrate with

on paper

electrophoresis at pH 3.5

identified by two-dimensional chromatography

isobutyric acid

-

conc.

NH,0H

-

on

H60 (66:1:3 v/v/v)

and 0.1 M sodium phosphate, pH 6.8

-

were

cellulose plates using in the first dimension

ammonium sulphate

-

n-propanol

(100:60:2 v/v/v) in the second dimension [10,22].

RESULTS

Purification of

tRNAMet

Crude transfer RN'A from S. obliquus

isolated

was

previously [3] and the cytoplasmic tRNA.Met

as

species was

described

purified in three

consecutive column chromatographic steps (see Materials and Methods) on benzoylated DEAE (BD)-cellulose, arginine-agarose and reversed phase (RPC-5)

respectively

as

shown in Figs. 1A, B and C.

The peak RPC fractions (34-37)

had

a methionine acceptance of about 2.2 nmole/A260 unit, and formylatability of about 12%.

an average

Nucleoside composition

Spots on the fluorogram of nucleoside analysis were identified by comparison with published fluorograms [20, 21, 283. As well as the four major triols A', U', C' and G' several minor

718

ones were

found in relative

Nucleic Acids Research molar yields as follows:

m5C',

o.8;

t6A',

m1G',

0.8; m2G'

1.6; m7G', 0.8; D', 1.8;v

",

0.9;

0.9; m A', 1.4 (this figure includes m A', a trace

rearrangement product [29] ).

These figures were largely reproducible except for,t' which gave a value of 2 (the predicted figure) in another

analy6is.

Analysis of nuclease digestion fragments Figs. 2A and B show the autoradiograms of the fingerprints obtained Met with RNase T1 and RNa§e A respectively. The after total digestion of tRNAi fingerprint patterns show considerable similarities to those of other eukaryotic cytoplasmic initiator tRNAs. Table 1 summarises the sequences of the fragments. Although all the sequences have been included some of the modified nucleosides were only tentatively assigned from this data. Also spot T15 was found to be composed of two fragments neither of whose sequence was deduced at this stage. On all fingerprints the sum of T6a and T14 gave an experimental molar ratio of 1. T6 is therefore likely to arise from further degradation of T14. TX did not appear on all fingerprints but when it did the molar yields of spots Tll and T15 were lowered. TX arises as a result of incomplete RNase T1 cleavage in the anticodon arm.

Sequencing of [5'-32P]-labelled tRNA Under conditions described elsewhere [26], which in our hands efficiently labelled E. coli tRNAMet, incorporation of 32P-label into f S. obliquus tRNAMet was very poor. In order to increase the incorporation of label, carrier free [Y-"2P]-labelled ATP (2000-3000 Ci/mole) was used at a much lower ATP concentration (0.5 ILM) than previously described. Although under these conditions the kinase reaction is very inefficient, sufficient 32 was incorporated into S. obliquus tRNAM.let for use in the following 1 experiments. (i) Partial digestion with nuclease P1 2P].aeld______ uAMet was digested partially with S. obliquus tRNA1 [5[513 _PI-labelled nuclease P1 under two different sets of conditions.

The products of

digestion obtained under the more vigorous conditions were separated by electrophoresis on DEAE-paper at pH 1.9 alongside a nuclease P1 digest of 32 P-pAGC. Both digests gave a similar pattern of fast moving spots and the mobilities of the digestion products indicated that the 5' terminal trinucleotide in the tRNA is AGC. The products of nuclease P1 digestion obtained under milder conditions were separated by two-dimensional homochromatography [6,7] using the "1OmM`

719

Nucleic Acids Research

-5 -4 E

w 0 z3

zC

gIC 0

0

'i

0 0 w

I-

0 ,

o

FRACTION

720

NUMBER

I

Nucleic Acids Research

0.s

1-

2 o

w z

-C

m

cr

0 U)

(a

O1

1

26

T

-.

32

36

LO

_

44

L8

FRACTION NO

Fig. 1. Chromatographic purification of S. obliquus tRNAi A-) BD-cellulose; B) Arginine-agarpse; C) RPC-5.

homomix.

The first 15 nucleotides from the 5'-end could be read off as

(pAG)C U G A G U G G C G C A G. (ii) Rapid read-off gel sequencing The rapid sequencing gel method was used to determine the sequence of nucleotides from positions 5 to 71. It also provided the means of placing the RNase T1 and RNase A fragments in order. Figs. 3A and B show the patterns obtained after electrophoresis for 2 hr. and 7 hr. respectively on 10% polyacrylamide gels. Examination of the formamide ladders reveals heterogeneity in the spacing

The formamide ladders were enlarged photographically and distances between the bands measured. There are three classes of spacing: large, corresponding to G residues, medium corresponding to A and U; small

of the bands.

corresponding to C. Although the formamide bands form a convergent series with increasing chain length, comparison of the bands within a small region permits tentative assignment of most of the bands into these three categories. 721

Nucleic Acids Research

A

..~~~~T o,t .

t

X

15

Rlft A F g1

*l

2

13

*

AFUypNI

s

| ~~~10

*v16.:

eio

13.

4

10

294 3~

~

~

~

9

Fig. 2,. Autoradiograms of fingerprints of nuclease digests of S. obliquus A) RNase Tj; B) RNase A. (B) surrounded by dots, xylene cyanol

tRNA.Met,

blueLdye. Limited hydrolysis sequence method Figure 4 shows the autoradiogram obtained after separation of the formamide hydrolysis products on a 10% polyacrylamide gel. The 5'-nucleotide of each fragment represented by a band on the gel is given in the figure. For most bands the assignments were straightforward. A band corresponding to position 2 was missing. nfGlO and Cl were found together in one strong band. The bands between 58 & 59 and 63 & 66 on analysis gave a mixture of mononucleotides similar to background contaminants found in most of the bands. The band between 58 & 59 from other evidence on the structure of the molecule would appear to be an artifact. The sequence between 63 and 66 cannot be established from this method. The faint bands in these positions appear to be artifacts. Therefore, either the phosphodiester bonds between nucleotides 63 & 64 and 64 & 65 are resistant to hydrolysis or the fragments with residues 64 & 65 at

722

Nucleic Acids Research TABLE 1.

Fragments frau complete IRNaze digestions REdase A Fragments Sequence

Molar Yield

Experimental

RNasie T1 Fragments

pAC pm Cp

Al

2.2

2

A2

1

pCp pGC pAU

TS

2.0

2

pCG

T2 T3

1.2 1.0

2

T4

1.0

pAG pCA& pAG

T5

0.8 1.1

1 1pIUA

A3

8

A4 A5 A6

1.1 2.2

1 8 1 2

1.1

2

pAGC

A7

0.9 0.8 0.9 o.6 0.8

1

pGm

1 1

pmG12 GC

T7

ptAAC

TS

A8 A9 AIO All Al2a+b A13

A14 A15

A16 A17

x

Sequencex

Molar Yield

EKp-imual Theoretical

Theoretical

IAAAC

T6 T6a

1

1

p1mG

0.2

-

pUIG

1

1 1 2

pG#

T9

0.4 o.o6 2.4

pAGmhD pm GA* + pAGD

TIO

1.1

0.4

1 1

5.3 0.7 o.6

5

pUp

Tll T12

1.1

1

pAUIG

1 1

pGGGC pGGAAC

T13

1 1

PGDI 5CM

T14

0.9 o.8

o.8 0.7

1 1

pGAGU pAGGAU

T15a+b

o.6

2

TX

0.5

-

2.0

0

4

pDG pUG pGp pCUG

pkAG

PU1(2G Utls XA (pmAAAC p

t6AACCCAtG

pAj_GCUCAUt6AACCAAG

Although the total squence has ben included, for some of the fragments, the caplete squence could not be obtained from these experiments.

their 5'end are not 32P-labelled in the kinase reaction. In the rapid read-off gel sequencing method a band corresponding to residue 64 is missing from the formamide ladder, but a band is present corresponding to residue 65 and the method indicates this residue to be U. Since 2'-0methylated nucleoside residues are resistant to formamide cleavage and the base at position 64 would be expected to base-pair with C, the residue at this position is likely to be 2{)O-methyl-G. This would explain the missing band 64 in the read-off gel method and the missing band 65 in this method. However, the gap between bands 63 and 65 in the read-off gel and the gap between band 63 and 66 in this gel are unusually large. It is probable therefore that there is a further modification on Gm64

causing fragments containing this residue to be retarded on the gel. The missing band 64 in this method could then be due to the fragment with a hypermodified Gm at its 5' end being an unsuitable substrate for the kinase. Fig. 5 shows the sequence deduced from the total of these results.

723

I

.

Nucleic Acids Research

XJI

CI

LL.

.:,

r

6

-SK

-

-S

~. ....... ~~~~~~~~~~~~~~~~..... ....

I I

724