Nucleotide and aminoacyl-tRNA specificity of the mammalian mitochondrial elongation factor EF-Tu·Ts complex

i~~ i~i~!i~i~!~i~i' i~ iiii~?i!i~' iiii~i~~!~~ii~i~!,~i/~i~ii,

et Biophysica A~.ta

E~~ER

Biochimica et Biophysica Acta 1307 (1996) 66-72

Nucleotide and aminoacyl-tRNA specificity of the mammalian mitochondrial elongation factor EF-Tu • Ts complex Velinda L. Woriax ", Gertrude H. Spremulli '~'~, Linda L. Spremulli ,~.b., " Deparnm'nt ~" Chemisto'. Campus Bnx 32~0. Unirersity of North Caroliml. Chapel Hill. NC 27599-3290. USA " Lim,l~erqer, Cancer Resem'ch Centel; C.mlm,S Box 3290, Unirer.¥itv. qfNorU~ Cm'olin., Chapel Hill, NC .7599-. "~ " ¢'~.90,USA

Received 4 September 1995: revised II December 1995: accepted 21 December 1995

Abstract

The bovine mitochondrial elongation factor Tu. Ts complex (EF-Tu. Ts,,,t) promotes the binding of aminoacyl-tRNA to ribosomes. In the presence o| GTP. this complex ftmctions catalytically. Both dGTP and ddGTP can replace GTP although about 4-fold higher concentrations are required. ATP. CTP and UTP are not active. ITP can replace GTP when used at I0- to 20-fold higher concentrations. The catalytic use of EF-Tu ' T.~.. is inhibited by GDP but not by GMP. XDP also inhibits although about 20-tbld higher concentrations are required. EF-Tu, Ts,. will promote the binding of Phe-tRNA to either Est'herit'hkt colt or mitochondrial ribosomes. Unlike E. colt EF-Tu. EF=Tu. Ts,,! will promote the binding of AcPhe-tRNA to riboson|es about 25% as efficiently as Phe-tRNA. E F - T u ' T s m t is active in catalyzing the binding of E. ('oli McI-tRNA • ,.~t m to ribo.~ome.~. EF-Tu. Ts,,, has about 3()~ as much activity with E. coil MeIqRNA"/~'' but has e,,,sentially no activity with E. rn/i fMet.tRNN', '~'', Neither yeast Met-tRNA",'"' nor fMet-tRNA", '~'t is recognized by bovine EF=Tu, T.~,.~. K,'v.*m'd~: Prt~leitt s~,t|lhe~i,,; l~lo,galitm; Transfer RNA; ()rt~a,elle: GTP

I. Introduction The prokaryoti¢ translational elongation factor Tu (EF= Tu) and its c~,toplasmic counterpart EF-I mediate the binding of aminoacyi=tRNA to the A-site ot' the ribosome during protein biosynthesis [I]. E. (,off EF-Tu requires GTP tbr the tbrmation of a ternary complex (EFTu:GTP:aa-tRNA) which then binds to the ribosome. Following GTP hydrolysis. EF-Tu is released from the ribosome in the form of an EF-Tu:GDP complex. The exchange of GDP tbr GTP is catalyzed by a second elongation factor (EF-Ts) with the internlediate tbrmation of an EF-Tu. EF-Ts complex [2]. Recently. a complex containing EF-Tu and EF-Ts has been purified from mammalian mitochondria (EF-Tu. T~,,, ) [3]. The properties of this complex are somewhat unusual and suggest that a variation of the traditional elongation cycle must be occurring in animal mitochondria [4]. First. unlike the E. ('off EF-Tu. Ts complex, the

C~rresponding autht~t'. Fax: +1 Linda Spre mul li (.a' unc.edu. In memory t~t"Gertrude tt, Sprcmulli.

(919) 9622388;

e-math

1)1{~7-4781/9~/$15.(~) ('~ I~% Elsevier Science B.V. All rights reserved (~;.~'DI1)!67-478 !(95)1)1)241)- 5

mitochondrial complex is very tightly associated and cannot he dissociated even in the presence of high concentrations of guanine ,ucleotides [4]. A similar observation has been made with the (EF-Tu. Ts)., complex IYom Thermus thermoldlilus [5]. Secondly. EF-Tu. Ts,,. does not bind guanine nucleotides detectably in the absence of aminoacyl-tRNA. Fin.'dly. in contrast to other translational systems. EF-Ts,. t carl be detected along with EF-Tumt on ribo~otncs (V. Woriax. unpublished data). Despite these differences, the catalytic use of the organellar factor requires GTP hydrolysis [4]. In addition, sequence analysis of the cDNAs encoding bovine and human EF-Tu,,,t indicates that the mitochondrial factors contain the 4 consensus sequences observed in the other guanine nucleotide binding proteins [6.7]. These observations have led us to examine the nucleotide specificity of EF-Tu .Ts,, in order to determine whether it has the characteristics tbund in other GTPases [8]. in bacterial systems, EF-Tu discriminates against the initiator fMet-tRNA both through the formylation of the a-amino function of the methionine and through the tRNA itself [9]. Unlike other genetic systems, bovine mitochondrial DNA has only one gene encoding tRNAm*t [10]. There is currently no evidence tot' the mitochondrial im-

V.L, Woriax et a l . / Biochimica et Biophysica Acta 1307 ¢1996) 66-,72

port of a tRNAm~t species from the cytoplasm, and the single mitochondrial gene is thought to give rise to both the initiator and elongator tRNAm~t [! I]. The initiation of protein synthesis in mitochondria requires formylation of the initiator tRNA [12], and the initiation and elongation factors must be able to discriminate between mitochondriai Met-tRNA''~ and tMet-tRNA"I~L In the present work, we have examined the ability of the bovine EF-Tu'Tsmt complex to distinguish elongator and initiator tRNAs having free or blocked a-amino functions.

2. Experimental procedures

67

mixtures (100 btl) prepared as described [4] and containing 0.5 to 1 pmol of EF-Tu • Ts,, and the indicated concentrations of the nucleotides to be tested. All dilutions of EF-Tu. Ts,, were made using Buffer A minus GDP but containing 2 mg/ml BSA. The reaction mixtures were incubated at 37°C for 15 min and then filtered through nitrocellulose membranes (Millipore Type HA). Filters were washed with Buffer B, dried and counted. Experiments to determine the ability of various nucleotides to inhibit the GTP-dependent recycling of EF-Tu - Tsmt in the ribosome binding assay were carried out as described above. The reaction mixtures contained ! or 2 p.M GTP and varying amounts of the nucleotides to be tested as indicated.

2. !. Materials 2.5. Aminoacyl-tRNA spec(lh'it3' Poly(U) and E. coli tRNA were purchased from Boehringer Mannheim. Phosphoenolpyruvate, pyruvate kinase, bovine serum albumin (BSA), XDP and glycerol were from Sigma. Poly(A,U,G), GMP, GDP, GTP, ITP, ATP, UTP and CTP were from P-L Biochemicals. Nitrocellulose membrane filter paper, Type HA (0.45 /zm pore size), was purchased from the Millipore Corporation. [J4C]Phe-tRNA was prepared from E. coil tRNA using the method of Ravel and Shorey [13] and synthetases were prepared by the method of Muench and Berg [ 14]. [ 3.~S]Met (1094 Ci/mmol) was obtained from DuPont-New England Nuclear. Yeast and !:.'. coli [3"~S]tMet-tRNA'li'~'~ and [~5S]Met-tRNA"~'~. and bovine mitochondrial ribosomes were generously provided by Dr. Hua-Xin Liao (Department of Chemistry, University of North Carolina) and had been prepared as described previously [15,16]. E. coil tRNA",~~ was kindly provided by Dr. Takuya LJeda and Dr, Kimitsuna Watanabe (University of Tokyo. Japan) :rod 't IIKq ['~'~S]Mct-tRNA,,I was aminoacylated as described [17], [l'~C]N.AcetyI-Phe-tRNA was provided by Qiong Lin (Department of Chemistry. University of North Carolina).

Ribosomal A-site binding assays with various aminoacyl-tRNAs were carried out in reaction mixtures (50 /.tl) basically as described [4] using E. coil EF-Tu. Ts or EF-Tu • Ts n, (100 units), E. coli ribosomes ( 1.44 A.,~,t~). the indicated amount of aminoacyi-tRNA and either 10/.tg poly(A.U,G) or 12.5 /.tg poly(U) as appropriate. When necessary, EF-Tu • TSmt was diluted into Buffer A containing 50 mg/ml BSA before use. Following incubation at 37°C tbr 10 min. samples were filtered through nitrocellulose mernbranes, washed with Buffer B. dried and counted as described [4]. Incubations using mitochondrial ribosomes (0.54 A2~,~) were performed as indicated above, except that the KCI concentration was reduced to 45 mM and the MgC! 2 concentration was increased to 12 raM. The filters from these reactions were washed with Buffer C. Nonenzymatic binding representing 0.1~0.4 pmol in assays on E. coil ribosomes and 0.5=2 pmol in incubao tions using mitochondrial ribosomes has been subtracted I'rom each value,

2.2, Bi!tfer~'

3. Results and discussion

Buffer A consisted of 20 mM Hepes-KOH, pH 7.0, 40 mM KCI, I mM MgCI.,. 0. I mM EDTA, I /xM GDP. and 10% glycerol, Buffer B consisted ot' 20 mM Tris-HCI, pH 7.6, 60 mM KCI, and 6.5 mM MgCI 2. Buffer C was 20 mM Tris-HCI, pH 7.6, 45 mM KCI and 12 rnM MgCI,.

3. I. Nm'leotide sl)eci/h'ity oI' EF.Tu' 7:%,

2.3. Factors and ribosomes Purified bovine EF-Tu. Ts,,, was prepared as described [3]. E. coil ribosomes were prepared as outlined [18]. E. coli EF-Tu. Ts was isolated ,'rod purified as described [16].

2.4. Nucleotide spec!licio' The ability of various nucleotides to promote the binding of [14C]Phe-tRNA to ribosomes was tested in reaction

Previous results from this laboratory have shown that the elongation cycle in bovine mitochondria tbllows a sequence ot' events that is different from those observed in the classical E. coli translational system [3.4]. These apparent differences led us to examine the nucleotide specificity of the organellar factor in order to determine whether it exhibite,l features cornmon to other proteins that bind GTP. In contrast to E. coli EF-Tu. mammalian EF-Tu. Ts,,, does not bind guanine nucleotides directly in detectable amounts. However. the presence of GTP allows this I'actor to recycle, and, thus. to promote the binding of aminoacyl..tRNA to ribosomes catalytically. In order to test the nucleotide specificity of EF-Tu. Tsmt. ribosome binding assays were carried out in which GTP was replaced

;

7

(

7

7

......

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

I<:L. Worhl.t el tll. / lii,chimic, ct i#iollhvxica Ac/~l 131)7 t 1996J 66-72

~8

with several oilier nucleotide triphosphates. The ability of various nucleotide analogs to stimulate the amount of Phe-tRNA bound was measured providing an indication of the interaction of the mitochondrial factor with the nucleotide in question. The relative importance of the 2' and 3' h)'droxyl groups of the ribose moiety for interaction with EF-Tu. Ts n, was assessed by comparing the amount of Phe-tRNA ~ u n d to ribosomes in the presence of GTP. dGTP and ddGTP (Fig. I). GTP was the most effective nucleotide in stimulating the catalytic use of EF-Tu'Tsmt. However. dGTP and ddGTP were only about 4-fold less effective than GTP. These experiments were carried out at limiting concentralions of the respective guanine nucleotide. Hence. the various activities observed are most likely a reflection of the relative binding constant of EF-Tu. Ts,,, for the nucleotides tested. The ability of both dGTP and ddGTP to support a significant amount of activity with EF-Tu • Ts,,, suggest that neither of the ribose hydroxyl groups plays a critical role in the interaction between this factor and the guanine nucleotides. A similar conclusion has been reached in studies of E. colt EF-Tu [2]. in the structures of the GDP bound fi)rms of E. colt EF-Tu and of ras p21 determined by X-ray crystallography, the hydroxyl groups on the ribose extend into solution and arc not believed to fore) critical interactions with the protein [19=22]. A simihlr conclusion can be reached by an examination of the structure of EF.Tu complexcd with a non-hydrolyzable GTP analog [I.23]. Molecular modeling of chloroplast EF:Tu Stlggest,~ that Lys=176 is close to the 2'-hydroxyl .~roup and ciluld potentially torm a hydrogen bond with it [24], However, dGTP binds quite well to chloropl:lst EF:Tu ~tigge~ling that there is little interaciiotl l'tetWeell this group and tile protein [='4],

t

~

I

I

'

t

1

f

6

5

~

3

~ ~

2

GTP

. °o° -°°°

°°°°"

°

~

1

oO-" -"°°i -

ATP. UTP, CTP

!

20

I

4O

Nucleotide

~

t

,

6O Concentration,

80

I

IO0 IJM

Fig. 2. Activity (if EF-Tu.Tsm~ with purine and pyrimidine nucleotides. Reaction mixtures were prepared as described in Section 2 and contained about I pmol of EF-TtI.Ts,,, and Ihe indicated concentratio GTP (OI. ATP. UTP or CTP (I"1) or ITP ( 0 ) . The anlount of Phe-IRNA retained by the ribosome in the presence of EF-Tu.Ts,,, but in the absence of guanine nucleolides has been subtracted I'rom each value.

In order to determine whether the mitochondrial factor possesses the same specificity tbr the guanine base observed with traditional guanine nucleotide-binding proteins, we tested its ability to use ATP. UTP and CTP. As indicated in Fig. 2. none of these nucleotides could replace GTP in promoting the catalytic use of EF-Tu. Ts,,,. In UTP and CTP tile purine base has been replaced with a pyrimidine base .'rod these nucleotides would cot be expected to interact with EF-Tu.Ts,,,. ATP differs I'rom GTP :it rlositions 2 and 6 on the base, both of which are thought to I'orm critical hydrogen-bond interactions with bacterial EFoTu and the guanine ring ol' GDP or GTP

I The iml:~ortance ol' tile 2-amino I'tniction ol' the guanine

ring was tested by exanlining the ability of ITP. which

iill

dGTP ,,," ,,.41

5

lit

,p ,,,,

="[/ " lk I

2

3

/ 1 4

5

Nu¢leotide Concentration, IiM Fig I The el lvci ~l (ITP. dGTI~ aiid ddGTt~ i~ii EF.Tu,I',,,,. ~:atal)ted bilitlilig oi" Ph¢-IRNA hi !~'. i'ldi ribii,~tlllles. I~ea¢lioil lilixlttrq2~ were I'tr~;pared ~is d¢,,crilt, d ill Sc'cih~ll 2 aild trolllaillcd appi
lacks the 2-alnino group, to replace GTP. ITP was able to pronlole the c a t a l y t i c use o f E F - T u , Tsmt but 10- to 20-t'old higher concentrations of this nucleotide were required (Fig. 2). These results suggest that EF-Tu. Ts,, interacts with the 2-alnino function of GTP. although this interaction probably does not contribute a large amount of energy to the binding interaction. The loss ol'a weak hydrogen bond ( ..IH = 2 kcal/mol) would be expected to result in about a 30-1bid reduction in the alTinity of the factor for the analog, a value comparable to the 20-tbld reduction seen (Fig. 2). F.. twit EF-Tu shows about a 100-1"old lower affinity for inosine nucleotides compared with guanine nucleotides while Eilgh'litl gracilis chloroplast EF-Tu does not have any detectable interaction with inosine containing nucleotides [2.24]. in order to examine the importanue of the ,8 and "y phosphates on the interaction of EF-Tu. Ts,,,t with guanine nucleotides, a variation of the assay used above was required. In this approach, we have takeq advantage of the observation that hydrolysis of the /3-7 phosphodiester bond is essential for the catalytic use of EF-Tu • Ts,,, and

V.L. Wori~tx el ++I./ Biochimit'~t el l~i+q,hx'.+it'tt At'ta 13117 +It,++61 6t~-72 T

"~ .........

l

__ +r,l,~

l

II

,,1

l"+ '

l

l

~

.

.

.

.

.

.

.

;

.

• Pn~-tigNA o.

.Q

I

g

./.'+

1

¢

.a_ E

~ . + ~ + + _ _ . + . ~ + oo, +.,,++ 2

4

6

(Ac}Phe-tRNA

8

10

added, pmol

Fig. 4, Binding ~l' Phe4RNA and AcPhe-tRNA to E. coil ril'~s~m~e.,, with EF+Tu.Ts,,, and i f, roll EF-Tu.T.,,. Reactitm mixture,, were prepared as de,,cril~d in Section 2 at'ld ctqllaincd EF-Tu.Ts,m ;.tlld the indicated ¢oncentratitm ~f Phe-IRNA ( - ~ - ) or AcPhc-tRNA (-O-). Altcnmtivcl~. rca¢litm mixtm¢,, ctmlaiued E. coil I-F-Tu.'P~ and either Phe-tRNA 1--O o °) t~r AcPhe+IRNA 1+-O--). Blanks (0. I t~ ().3 pmtfl) repre.,,enting the ~tmtmtit ~1' n|ditmctixily retained on the filter ~tt c;.tch (Ac)Phe-IRNA

aminoacyl-tRNA regardless of whether it is acting on bactelial or mitochondrial ribosomes. It has been proposed that N-blocked aminoacyl-tRNAs t'ail to bind to EF-Tu because of the loss of the positive charge on the free a-amino function that is believed to be important lbr this interaction. Alternatively. N-blocked mninoacyl-tRNAs may be stericaily prohibited from interaction with the aminoacyi-tRNA binding site on EF-Tu [31.33]. The significant interaction between EF-Tu'Tsmt and AcPhe-tRNA suggests that this factor is able to accommodate modifications of the amino function of the amino acid during the ribosome binding step. Interestingly, yeast EF-i ce also uses AcPhe-tRNA about 30% as well as Phe-tRNA [34]. In this respect, the mitochondrial factor resembles the cytoplasmic l'actor more closely than its prokaryotic cotmterpart.

3.3. Interaction o.I" EF-Tu" Ts,,,, with .IMet-tRNA and MettRNA

¢Ol1¢¢lltr~ltitql in the ab.~¢nc¢ of factor llav¢ been subtracted l'ronl each

were affected by tile ribosomal environnlent present. As indicated in Fig. 5, E. <.o[i EF-Tu. Ts still discriminates ag~dnst AcPhe-tRNA when ~sayed tm tile organellar ribosomes. EF+Tu. Ts,,,, again protlmtes the bit~ding of about 25~;~ as much AcPhe°IRNA ~ts Phe-tRNA on its hotuologott~ t'iboson~es. The lower aMotlnt o1' total binding ¢~h. ~,erv~d hi these experiments is simply ~ reflectiotl o1' tile atrmtmt and quality ~i' the t'ibo~,onl~ rweparatiot~ from the mit~clmndt'ia. The~,~ re,,ults ,,,uggest that EF:l'u. T,,,,, doe., trot di~tir~guish ,,It'ot~gly hctweetl a I~hwked aml unhhwked

it has been shown that formylation of the initiator IRNA is required l'or the formation of an initiation complex in mitochondria [12]. Since formylmethionine cannot be incorporated into internal positions in a polypeptide. EF-Tu. Ts,,, must he able to distinguish between the Ibrmylated and unformyl:tted Met-IRNA'"c' species. Studies on the itlteraction of this elongation factor with its homologous (I')Met-tRNA have been hindered by the low abundance of tile mitochondrial tRNA and by the inability to charge .'rod fm'mylale the animal mitochondrial tRNA ''''~ ill vitro, in order to gain some insight into the possible effect o1' I'ot'tuyl~tlitql on the interaction o1' EF-Tu.'I's,,, with MetIRNA. we Imve examined tile ability o1' this I'actor to promote tile I~inding of yeast and E. c'oli (I')Met*tRNA",'ct mid Mct*tRNA'~,',~'t to ribosomes. As indicated in Fig. 6,

0,g0 I

o NAcPhe- tC~NA ,''

E coh TuT~

¢~ 0.45

o J~

!

<

I

I

1.0

1.5

2.0

2.5

o.4

< 0.3 7¢, i o.2

E¢+Tu Tsm~

¢D

0.15

0.0o

r

E ,- 0.5

g o.~o

0.30

I

0.6

P COl, +u TS •

+

)

2

_--

+I. +

i

I

4

6

8

=

0.1

.t +

lo

(AcIPh@-IRNA addeO, pmol

Ig~.,, 5. Binding ut' Phe+tRNA and Acl~he°IRNA t~ mih~ctmudrial t'il~t~'~I+IIICs. Reaclit~ll IlliMtlr¢,, ~,.l~|'~t~prepared ils tlcscribed in Sectitm 2 alld

ctmtained EF+Tu.Ts,,, al|d tile indicated ct~ncentratiotl t~l' Phe-IRNA I-Q+ )~+r AcPIleqRNA ~-L)-k Altcruati~ely, rea~;litm mixtures ctmtaiucd l:~, ~'oli +~F+Tu.+i3,and either Phc4RNA ( . - ~ - -)t~r AcPhe+tRNA (+-O--), I]kmk,, +:d'~+tlt1),3 pnlt~l) representing the illllOtlltt ~t~t'raditmctivity retained on the h i~er at each (Ac)Phc.IRNA ctmcentraliot~ in the absence of EF-Tu lm~e been ,q|htracled [i'Oltl each value.

0.5

If)Met-tRNA added, pmol Fig, 6, Binding of I.. coli ai|d x,casl (I')MeI-IRNAs to ril~osomes hy F+F.Tu.Ts... Reacliun mixtures were ptvpared as described ill Secliml 2 and contained I'.'F-Tu.Ts.. and the indicaled amounts t~l" I-. coil MelIRNA",',',' (-1-). MeI-tRNA",'"' (-Q.). E. roll fMet-tRNA", ''~ ( . . A . . ) o r x eas! MeI-IRNA", "q and I'MeI-IRNA", '~'~ (--O--). Blank.,, representing the ilnloulll of lal~cl retained on the filter in the ahsence of factor have been sublracted from each value (O.OI to 0.2 pmol).

V.L. Woria.v et aL / Biorhimira t't Biophysira Arta 1307 ¢1996166-72

EF-Tu • Ts,,,, is active in promoting the binding of tile E. coil elongator tRNA (Met-tRNA",~,~"t ) to ribosomes. As expected, no binding can be observed when the E. coil

initiator tRNA (fMet-tRNA~'~'t) is used in this assay. However, when the E. roll initiator tRNA is present in an unformylated form (Met-tRNA"i'¢|), EF-Tu • Ts,,, promotes the binding of the aminoacyl-tRNA to the ribosome about 30-50% as well as the elongator tRNA. This result is in stark contrast to that observed with E. coil EF-Tu which will not readily bind E. coil Met-tRNA"~'~ even when it is not formylated [33]. Similar binding assays were conducted using mitochondriai ribosomes (data not shown). The results demonstrate that E. coil Met-tRNA"i'~'~ is recognized by EF-Tu • Tsmt to a greater extent than by the prokaryotic factor and that EF-Tu.Ts,,, will promote its binding to mitochondrial Ilk' t ribosomes about 50% as well as Met-tRNA,,,. The results described above indicate that EF-Tu. T s , . t strongly discriminates against the presence of a formyl group on the prokaryotic initiator tRNA. However. unlike E. coil EF-Tu, EF-Tu • Tsmt does not discriminate strongly against the bacterial tRNA",'~" species itself. Previous work [35] has indicated that E. roll EF-Tu distinguishes the bacterial initiator tRNA from other tRNAs primarily by the lack of the normal base pair involving the 5' nucleotide in this tRNA species. Clearly, EF-Tu.Ts,, does not effectively differentiate between elongator and initiator tRNA by tills means. Tills idea is consistent with the fact that while ninny mitochondrial tRNAs have seven base pairs in tile acceptor stem, others have only six base pairs in this stem [361. in contrast to the results obtained with the E. ~'oli initiator tRNA, EF-Tu. Ts,,,, is ulmblc to promote tl~e billding o1' either yeast I'Mct-tRNA",''~ or Met-IRNA",'"' to E. ('oil t'ibosolneS (Fig. 6) or mitochoildrial i'ibo.~omcs (results not shown). This obsel'vatioll suggests that the yeast initiator tRNA has features that prevent its recognition by the t11itocholldt'ial factor, Eukaryotic cytophismic initiator tRNAs ate distinguished by two unique features. First, an unusual contbrmation formed by hydrogen-bonding between residue 20 of the D loop to three purinc residues in tim T loop constructs a substructure tirol strengthens the bridge between these loops [37]. Secondly, the yeast initiator tRNA has a 5' phosphorylribosyl modification on the ribose of residue 64 attached by a 2'-O-glycosidic bond [38]. This modification is tllought to interfere with the binding of EF-I c~ to the initiator tRNA. at least in lower eukaryotes [39.40]. Tllese features may also interfere with the interaction of EF-Tu ' T s , . t with yeasl (f)MetIRNA,,i~'. Only one gene for t R N A "'~''is encoded in the mitochondriai genome. This tRNA has a fully paired acceptor stem but does not contain the modification observed with the yeast initiator tRNA. It may participate in both initiation and elongation, interacting with both IF-2,,, and EF-Tu. Tsml in an unknown way.

71

Acknowledgements Tills work was supported in part by funds provided by the National Institutes of Health (Grant GM32734).

References [I] Sprinzi. M. (1994)Trends Biochem. Sci. It). 245-250. [2] Miller. D. and Weissbach. H. 11977) in Molecular Mechanism of Protein Biosynthesis (Weissbach. H. and Pestka. S.. eds.), pp. 323-373. Academic Press. New York. [3] Schwartzbach. C. and Spremulli. L.L. (1989) J. Biol. Chem. 264. It)125-11)131. [4] Schwartzbach. C, and Spremulli. El.. (1990) J. Biol. Chem. 266. 16324-16330. [5] Arai. K-I.. Arai. N.. Nakamura. S.. Oshima. T. and Kaziro. Y. (1978) Ear. J. Biochem. 92. 521-531. [6i Wells. J.. Henkler. F.. Leversha. M. and Koshy. R. (Iqq5) FEBS I.ctt. 358. 119-125. [7] Woriax. V.. Burkhart. W. and Spremulli. 1.. (It)t)5) Biochim. Biophy, Acta 1264. 347-356. [8] Bourne. H.R.. Sanders. D.A. and McCormick. F. ( 1991 ) Nature 348. 117-127. [91 RajBIm.dary. U.I.. a.d Chow. C.M. (Iq95) in tRNA: Structure. Biosynthesis. and Function (Soil. D.. and RajBhandary. LI.L.. eds.), pp, 511-528, ASM Press. Washington, [1()] Andersoll. S,, de Briujn. M,H.L.. Coulson, A,R,. Epertm. i.C,. Sanger. t:. and Young. I.G. (1982) J, Mol, Biol, 15f). 683-717. [I I] Roe. B,A.. Wong. F.F.H.. Chen. E.Y.a.d Arm,,trong. P,A, (1981) m Recombinant DNA. Proceedings of tile Tllird Clevehmd Symposium on Macromolecules (Walton. A,G,, ed.). pp, 167=17I). Elsevier. Amsterdam. 1121 I.iao. IIoX, mid Sprcmulli. !., {1~)91)J, Biol, ('llem. 2~)¢), 21)714. 207 It), 113] Ra~,eh J,M, aml Shorc~. R,I. (IqTI) M¢thotl~ F.luymt)l, 2I)(', ,~()t')= 3 It,~, 1141 Mucilch, K, alld I]cql, I), lib)Of))ill I)l'O¢¢thll'c~ lit Nt)¢1¢)~' Acid I,h'~carch I('alltOlii, (h mid I).'tvic~, I),, ed~,), lip, }71 ililrlwr alld Row. Ne~v York, II5] (h'ave,,. M, a,d Sprcmtdli, I., (It~83) Arch, IIi.~'llcm, Bioph?~, 2-32, I112~ Ith~, 111~1 I-bcrly, S,I... I..cklear, V, atld Spre.itdli, I.,I. (IqS~J J. Biol, Chclll, 201), 8721-8725, [17] Walker. R, alld Ra,jBImndary, t!, (It)72)J, Biol, ('l~cm, -347, 48711~ 4892. [18] (h'ave~. M,. lh'eitenbelgcr. (', a11d Spl'Clmdli. I.,I., l lqgI)) Arcll, Biochem. Biophy,,,, 204..ld.4-454, [It}] la ('otn'. I',F,M,. Nyborg. J,. Tllirup. S, alltl Clark. B,F, (It)85) I-MI]() J, 4. 2385=2388. [20] ,hlrtlak. I', 11985) Science 2::41.32-~3f~, 1211 de Vos. A,, Tong, I.... Millmrl. M,. Matias. P,, Ja.carik, J,, No[luchi. S,. Nishimura. S.. Miura. K. ()iltsuka. E,. Kim. S-K 11q88) Scie.ce 239. 888-81)3, [22] Kjeldgaard, M, and Nyb.rg. J, (It}t)2) J, Mol, Biol. 223. 721-742. [23] Kjcldgaard. M., Ni~,,en. P,. Thirup. S, and Nyborg. J, (Itlt)3) Structure I. 35-50. [24] Lapadat. M,. Deerlield. D.. Pederscn. 1., alld Sprcmulli. h, (It]90) Prof. Strt|ct. I-unct. Genet. 8. 237-250. I25] I:.cc1¢:~1o.. J.. Kanagasabai. T.. Molloy. D,. Ncal. S, a.d Webb. M, (I t)St)) in The Guanine Nucleotide Bi.ding Protei.s: COlllnlon Structural and Functiomd Properties (Bosch. L.. Kraal. H, and Parmcggiani. A,. eds.), NAT() ASI series 165. pp 87-97. Plenum. New York.

-/2

V.L Win'lax et al./ Bi~,himi¢'~a el iti~q)hysi¢', A(,to 1307 ¢1996~ 66-72

[26] Carvalho. M.. Carvalho. J. and Merrick. W. (1984) Arch. Bit,:hem. Biophys. 234, ~3-61 I. [27] Berchtold, H., Reshetnikova, L.. Reiser. C.O.A., Schirmer, N,K,, Sprinzl, M. and Hilgenfekl. K. (1993) Nature 365. 126-132, [28] Wakita, K,, Watanabe. Y.. Yokogawa, T., Kumazawa. Y., Ueda, T., Watanabe, K. and Nishikawa, K. (1994) Nucleic Acids Res. 22, 347-353. [29] Benkowski, L., Takemoto. C., Ott. G., Beikman, M., Ueda, T., Watanabe, K., Sprinzl, M. and Spremulli, L. (1995) Nucleic Acids Symposium Series, 33, 163-166. [30] Kum~awa, Y.. Schwanzbach, C., Liao. H-X,, Mizumoto, K., Kaziro, Y., Miura, K., and Watanabe, K. and Spremulli (1991) Biochim. Biophys. Acta 1090. 167-172. [31] Janiak, F.. Dell. V.. Abrahamson. J., Watson, B.. Miller. D. and Johnson. A. (1990) Biochemistry 29. 4268-4277. [32] Jerel. C., Sandoval. A., Allende. J., Henes, C. and Ofengand. J. (1969) Biochemistry 8. 3~)6-3014.

[33] Louie, A,, Ribcim, N,S,, Reid, B.R. and Jurnak, F. (1984) J, Biol. Chem, 259, 5010-5016. [34] Spremulli, L. and Ravel, J. (1976) Arch. Biochem. Biophys. 172. 261-269. [35] Dyson, M,, Mandal, N, and RajBhandary, U. (1993) Biochimie 75, 105 ! - 1060. [36] Tokogawa, T., Watanabe, Y., Kumazawa, U., Ueda, T., Hirao, I., Miura, K. and Watanabe, K, (1991) Nucleic Acids Res. 19, 61016105. [37] Basavappa, R. and Sigler, P,B. (1991) EMBO J. i0. 3105-3111, [38] Desgres, J., Keith, G., Kuo, K.C. and Gehrke, C.W. (1989) Nucleic Acids Res. 17, 865-882. [39] Rosenthal, L.P. and Bodley, J.W. (1987) J. Biol. Chem. 262, !0955-10959. [40] Kiesewetter, S., Ott, G. and Sprinzl, M. (1990) Nucleic Acids Res. 18, 4677-4682.