Singly and Bifurcated Hydrogen-bonded Base-pairs …...COMMUNICATION Singly and Bifurcated...

Article No. jmbi.1999.3080 available online at http://www.idealibrary.com on J. Mol. Biol. (1999) 292, 467±483

COMMUNICATION

Singly and Bifurcated Hydrogen-bonded Base-pairs intRNA Anticodon Hairpins and Ribozymes

Pascal Auffinger and Eric Westhof*

ModeÂlisations et Simulationsdes Acides NucleÂiques, UPR9002, Institut de BiologieMoleÂculaire et Cellulaire duCNRS, 15 rue ReneÂ Descartes67084 Strasbourg CedexFrance

E-mail address of the [email protected]

0022-2836/99/380467±17 $30.00/0

The tRNA anticodon loops always comprise seven nucleotides and isinvolved in many recognition processes with proteins and RNA frag-ments. We have investigated the nature and the possible interactionsbetween the ®rst (32) and last (38) residues of the loop on the basis ofthe available sequences and crystal structures. The data demonstrate theconservation of a bifurcated hydrogen bond interaction between residues32 and 38, located at the stem/loop junction. This interaction leads to theformation of a non-canonical base-pair which is preserved in the knowncrystal structures of tRNA/synthetase complexes. Among the tRNA andtDNA sequences, 93 % of the 32 �38 oppositions can be assigned to twofamilies of isosteric base-pairs, one with a large (86 %) and the other witha much smaller (7 %) population. The remainder (7 %) of the oppositionshave been assigned to a third family due to the lack of evidence forassigning them into the ®rst two sets. In all families, the Y32 �R38 base-pairs are not isosteric upon reversal (like the sheared G �A or wobbleG �U pairs), explaining the strong conservation of a pyrimidine at pos-ition 32. Thus, the 32 �38 interaction extends the sequence signature of theanticodon loop beyond the conserved U-turn at position 33 and theusually modi®ed purine at position 37. A comparison with other loopscontaining both a singly hydrogen-bonded base-pair and a U-turnsuggests that the 32 �38 pair could be involved in the formation of a basetriple with a residue in a ribosomal RNA component. It is also observedthat two crystal structures of ribozymes (hammerhead and leadzyme)present similar base-pairs at the cleavage site.

# 1999 Academic Press

Keywords: Nucleic acid conformation; tRNA; anticodon; leadzyme;hammerhead
*Corresponding author
Introduction

Hydrogen bonds between bases constitute thecement of nucleic acid structures. Among nucleicacid base-pairs, the well-known Watson-Crick G �Cand A �U pairs involve three and two hydrogenbonds, respectively. Besides these regular Watson-Crick interactions, a large array of possible base-pair interactions involving two hydrogen bondshave been enumerated and many of them havebeen observed in crystal structures (Tinoco, 1993;Dirheimer et al., 1995; Leontis & Westhof, 1999a).Such non-Watson-Crick interactions subtend thegreat architectural diversity of nucleic acids, andespecially of RNA molecules where they are recur-rently observed. Further, non-Watson-Crick pairs

ing author:

contribute to the structural ¯exibility of thesemolecules, since they are often strongly contextdependent. A less recognized class of base-pairscomprises those in which the two bases are linkedby a single or a bifurcated hydrogen bond.

In transfer RNAs, the anticodon loop plays acentral role in processes associated with proteinsynthesis. On the one hand, it must interact speci®-cally with its cognate synthetase, and on the otherhand, it must ®t in the common tRNA binding siteof the ribosome for codon recognition on the mes-senger RNA. Over the years, rather extensive data-bases at the tRNA level (546 sequences) or at thetDNA level (2726 sequences) have been compiled(Sprinzl et al., 1998). Several years ago, it wasnoticed that the ®rst (32) and last (38) residues ofthe seven-membered anticodon loop are involvedin a bifurcated hydrogen bond contact (Quigley &

# 1999 Academic Press

Figure 1. Secondary structures of tRNA anticodon hairpins for which crystal structures are accessible in the NDB.The 32 �38 base-pairs are shown in bold.

468 Singly and Bifurcated Hydrogen-bonded Base-pairs

Rich, 1976; Westhof et al., 1985). Later, multiplemolecular dynamics simulations performed withappropriate electrostatic treatment showed a sur-prising stability for this 32 �38 contact (Auf®nger &Westhof, 1996, 1998a; Auf®nger et al., 1999). Stillmore surprising, in the crystal structure of thespeci®c complexes between tRNAs and their cog-nate synthetase (Rould et al., 1991; Ruff et al., 1991;Cusack et al., 1996a, 1998), similar contacts betweenresidues 32 and 38 occur despite a severe unfold-ing of the rest of the anticodon loop. Therefore, wedecided to investigate the available crystal struc-tures of tRNAs and their complexes with tRNA-synthetases, in the light of their base distribution inthe tRNA sequence database (Sprinzl et al., 1998),in order to rationalize the possible interactionsbetween residues 32 and 38 (Figure 1).

Results

Crystallographic structures of free tRNAs

In all tRNA crystal structures present in theNDB database (see Table 1), the ®rst (residue 32)and last (residue 38) bases of the anticodon loop(Figure 1), are linked through a bifurcated hydro-gen bond and form a ``pseudo'' base-pair(Figure 2). The only exception is the trna03 struc-ture (Brown et al., 1985) obtained at low pH in thepresence of Pb2� (in trna03, a Pb2� is intercalatedbetween C32 and the hypermodi®ed Y37 base,possibly preventing the formation of a 32 �38 inter-action). This particular anticodon loop confor-mation is probably not adopted by the tRNAmolecule under physiological conditions and willnot be further discussed.

In tRNAAsp, a 32 �C38 and, in tRNAPhe, aC32 �A38 base-pair are present ( stands for apseudouridine where the C5-H group is replacedby a N1-H group and the C10-N1 link between thebase and the sugar is replaced by a C10-C5 bond;see Figure 2). The 32 and 38 bases are linked by abifurcated hydrogen bond established between acarbonyl oxygen atom and an amino group, i.e.O4(32) . . . N4(C38) in tRNAAsp andO2(C32) . . . N6(A38) in tRNAPhe. Interestingly, the

carbonyl and amino groups as well as the glycosi-dic bonds of both base-pairs superimpose verywell. Thus, although a signi®cant out-of-planedeviation is observed (Figure 2), these 32 �38 pairscan be considered as isosteric (Figure 2). It is note-worthy that the observed out-of-plane deviationdoes not affect the interaction scheme of the twobases or the inter-base C10(32) . . . C10(38) distance,close to 11 AÊ as in Watson-Crick pairs. Yet, thedeviation of these isosteric pairs from a Watson-Crick arrangement is clearly illustrated by thelarge distance (�7 AÊ ) between the C10 atoms ofresidue 38 in the bifurcated hydrogen bonded32 �38 pairs and in the corresponding Watson-Crickarrangement (Figure 2). Interestingly, a U �C pairidentical with that observed in yeast tRNAAsp hasbeen observed in the crystal structure of the alpha-sarcin loop of 23 S rRNA (Correll et al., 1998). Adifferent situation is observed in the crystal struc-ture of yeast tRNAi

Met (Basavappa & Sigler, 1991),where the C32 �A38 pair (Figure 3) is not isostericto the 32 �38 pairs described above. The authorsreported that the resolution of the anticodon hair-pin is very low. Hence, this interaction scheme isprobably of poor structural relevance and will notbe further discussed. This view is also supportedby recent NMR studies of two transcripts of yeasttRNAi

Met and Escherichia coli tRNAmMet anticodon

hairpins (Schweisguth & Moore, 1997) from whichit was concluded that the observed non-canonicalC32 �A38 NMR base-pairing are much closer tothose observed in yeast tRNAPhe crystal structuresthan the one observed in the yeast tRNAi

Met struc-ture (Schweisguth & Moore, 1997).

A recent crystal structure of the complexbetween E. coli tRNACys and the translationelongation factor EF-Tu reveals the arrangement ofa 32 �A38 base-pair (pr0004; Nissen et al., 1999).Similarly to what is observed in the crystal struc-ture of the complex of tRNAPhe with EF-Tu(ptr012; Nissen et al., 1995), the anticodon loop oftRNACys adopts a conformation close to that seenin the crystal structures of uncomplexed tRNAswith a non-Watson-Crick 32 �A38 pair. Again the �A pair is isosteric to the �C pair observed in

Figure 2. The 32 �38 base-pairs observed in crystal structures of uncomplexed tRNAs (see Table 1). (a) Averagestructures of the 32 �C38 and C32 �A38 pairs extracted from the crystal structures of uncomplexed tRNA molecules,including also the structure of the yeast tRNAPhe/RS complex (ptr012) where the anticodon loop does not interactwith the protein. The individual structures used for the averages are drawn with small lines. (b) Superposition (left)of the 32 �C38 (yellow) and C32 �A38 (red) average structures showing the isostericity of these base-pairs in thetRNA anticodon loop structural context. For comparitive purposes, a regular A �U Watson-Crick pair is also drawn.Orthogonal view (right) showing the observed out-of-plane deviation of these pairs. In all these Figures, the glycosi-dic bonds of nucleotides at position 32 have been superimposed.

Singly and Bifurcated Hydrogen-bonded Base-pairs 469

yeast tRNAAsp and to the C �A pair observed inyeast tRNAPhe (Figure 2). Interestingly, in bothtRNACys and tRNAPhe (trna06) structures, a watermolecule is seen in the vicinity of the N7 and N6-H groups of A38. In trna09, there is a comparablewater molecule which belongs, however, to thecoordination sphere of a Mg2�.

Hence, from the available tRNA crystal struc-tures where the anticodon hairpin keeps its foldedstructure, it can be deduced that a set of non-cano-nical but isosteric base-pairs involving a bifurcatedhydrogen bond between a carbonyl oxygen atomand an amino group is recurrently present at thejunction between the stem and the loop of theanticodon hairpin. On a thermodynamic basis,base-pairs involving a single or bifurcated hydro-gen bond should be more labile compared to thoseinvolving two or three hydrogen bonds. However,as described below, formation of speci®c tRNA/synthetase complexes, which involve the splayingout toward the exterior of ®ve (33 to 37) of thebases of the loop, do not disrupt the direct hydro-gen bonding contacts established between bases 32and 38.

Crystallographic structures of tRNA/synthetase complexes

The coordinates of several crystal structures oftRNAs complexed with their cognate synthetase

are available (Table 1). Upon formation of thetRNA/synthetase complexes, slight conformationalrearrangements of the 32 �38 base-pair areobserved.

32 �C38 pairs. Two structures of the complex ofthe wild-type yeast tRNAAsp, including all themodi®ed bases with its cognate synthetase(tRNAAsp/RS) are available (Table 1). In thesestructures, the 32 �C38 pair adopts a confor-mation different from that observed in the uncom-plexed tRNA structures (Figure 4). The C38 aminogroup forms a single hydrogen bond rather than abifurcated hydrogen bond with the O4(32) atom.Further, a signi®cant out-of-plane orientation ofbase C38 is observed. These arrangements appearless stable than those formed in uncomplexedtRNAs and are, possibly, stabilized by additionalprotein/tRNA interactions.

U32 �U38 and Um32 �38 pairs. No structure hasbeen solved of an uncomplexed tRNA having aU32 �U38 pair. However, in the structures of thetranscripts of E. coli tRNAGln/RS complex (Table 1),a U32 �U38 pair is observed (Figure 4). Again, asingle hydrogen bond links the two bases. Notethat this base-pair is approximately isosteric to aWatson-Crick pair (see the bottom drawing inright-hand column of Figure 6). Thus, its geometry

Table 1. Description of the crystal structures of complexed and uncomplexed tRNA molecules referenced in the NDB(Berman et al., 1992) before the 1st August 1999

NDB codea PDB code Res. (AÊ ) 32 �38 Transcript Coord.

Free tRNAstrna01 / yeast tRNAPhe 2.5 Cm �A N Ntrna02 / yeast tRNAPhe 3.0 Cm �A N Ntrna03 1TN2 yeast tRNAPhe 3.0 Cm �A N Ytrna04 6TNA yeast tRNAPhe 2.7 Cm �A N Ytrna05 / yeast tRNAAsp 3.0 �C N Ytrna06 1TRA yeast tRNAPhe 3.0 Cm �A N Ytrna07 2TRA yeast tRNAAsp 3.0 �C N Ytrna08 3TRA yeast tRNAAsp 3.0 �C N Ytrna09 4TRA yeast tRNAPhe 3.0 Cm �A N Ytrna10 4TNA yeast tRNAPhe 2.5 Cm �A N Ytrna12 1YFG yeast tRNAMet

i 3.0 C �A N YtRNA/EF-TU complexespr0004 1B23 E. coli tRNACys 2.6 �A N Yptr012 1TTT yeast tRNAPhe 2.7 Cm �A N YtRNA/RS complexespte001 / T. thermo. tRNAPro(GGG) 3.6 U �U N Npte002 / T. thermo. tRNAPro(CGG) 3.5 U �A N Npte003 1QTQ E. coli tRNAGln 2.2 U �U Y Yptr001 1GSG E. coli tRNAGln 2.8 Um � N Nb

ptr002 1GTS E. coli tRNAGln 2.8 U �U Y Yptr003 1GTR E. coli tRNAGln 2.5 U �U Y Yptr004 1SER T. thermo. tRNASer 2.9 C �A N Yc

ptr005 1ASY yeast tRNAPhe 2.9 �C N Yptr007 T. thermo. tRNASer 2.7 C �A N Nptr008 1ASZ yeast tRNAAsp 3.0 �C N Yptr009 1QRS E. coli tRNAGln 2.6 U �U Y Yptr010 1QRT E. coli tRNAGln 2.7 U �U Y Yptr011 1QRU E. coli tRNAGln 3.0 U �U Y Yptr014 / T. thermo. tRNALys 2.7 C �A N Nptr015 / T. thermo. tRNALys 2.9 C �A Y N

aCrystal structure references are given next: (trna01, Quigley et al. (1978); trna02, trna03, Brown et al. (1985); trna04, Sussman et al.(1978); trna05, Comarmond et al. (1986); trna06, Westhof & Sundaralingam (1986); trana07, trna08, trna09, Westhof et al. (1988);trna10, Hingerty et al. (1978); trna12, Basavappa & Sigler (1991); pr0004, Nissen et al. (1999); ptr012, Nissen et al. (1995); pte001,pte002, Cusack et al. (1998); pte003, Rath et al. (1998); ptr001, Rould et al. (1989), ptr002, Perona et al. (1993); ptr003, Rould et al.(1991); ptr004, Biou et al. (1994); ptr005, Ruff et al. (1991); ptr007, Cusack et al. (1996a,b); ptr008, Cavarelli et al. (1994); ptr009, ptr010,ptr011, Arnez & Steitz (1996); ptr014, ptr015, Cusack et al. (1996a,b).

b Only backbone coordinates have been deposited in the NDB.c The coordinates for the atoms of the anticodon hairpin could not be derived from the crystal data.


is clearly distinct from that observed in the crystalstructures of uncomplexed tRNAs for C �A, U �A,or U �C pairs. The stabilization of such base-pairsinvolves a network of water molecules, some link-ing one base to the other and some linking a baseto its backbone (see Figure 3 by Rould et al., 1991).An additional contact between Asn370 and U38 ispresent in these structures. The Um32 �38 base-pair of the E. coli tRNAGln/RS complex, whichincludes all tRNA modi®ed residues, displays ageometry identical with that observed for thetRNA transcripts (Rould et al., 1991).

Other 32 �38 pairs. In the structures of the twocomplexes formed between Thermus thermophilustRNALys synthetase and transcripts of E. coli andT. thermophilus tRNALys, the occurrence of a non-standard C32 �A38 base-pair has been reported(Cusack et al., 1996a; S. Cusack, personal communi-cation). Beside contacts made by A38 with the pro-tein, a N3(C32) . . . N6(A38) is formed. This contactis different from those observed in the structures ofuncomplexed tRNAPhe. Furthermore, in the struc-

tures of the two complexes of T. thermophilustRNAPro synthetase with two tRNAPro isoacceptors,determined at 3.5 AÊ resolution, the occurrence of anon-standard U32 �U38 and of a Watson-CrickU32 �A38 base-pair is mentioned (Cusack et al.,1998; S. Cusack, personal communication).

Phylogenetic analysis

tDNA gene sequences. In order to detect anysequence preferences for base-pairs at position32 �38, the tRNA gene database which contains upto 2726 tDNA sequences (Sprinzl et al., 1998) wasanalyzed. First, there is a very large proportion ofpyrimidines at position 32 (�98 %, see Table 2) anda signi®cant number of purines at position 38(�71 %). Secondly, sequences which could formC �G (six occurrences), G �C (no occurrence) or A �U(ten occurrences) Watson-Crick pairs are very rare.Nevertheless, there is a large proportion ofU32 �A38 pairs (�18 %). Note that the number ofU �A pairs might be slightly lower since it has been

Figure 3. The C32 �A38 pair observed in the 3.0 AÊ

crystal structure of yeast tRNAMeti (Basavappa & Sigler,

1991). This pair displays a unique conformation whichcannot be easily compared to the base pair arrange-ments shown in Figure 2. For comparison, a C �G Wat-son-Crick pair and a C32 �A38 pair extracted from yeasttRNAPhe (see Figure 2) are also represented. The glycosi-dic bonds of 32 have been superimposed.


shown that U! C editing can occur at position 32(Beier et al., 1992).

The preferred base-pair sequence is C �A(�48 %), which is followed by the T �A (�18 %),T �T (�11 %), C �C (�8 %) and T �C (�8 %) pairs.Most of these pairs, with the exception of the T �Apair, cannot form Watson-Crick arrangements.Note that these proportions do not vary signi®-cantly if one considers the two following subsets:subset 1 (932 sequences of tDNA elongators) andsubset 2 (352 sequences of eukaryotic tDNA elon-gators) of the whole 2726 tDNA sequences.

Figure 4. The 32 �38 base-pairs observed in crystal structuThe �C base-pairs are shown on the left and U �U base-paregular U �A Watson-Crick pair as well as the average �C psidic bonds of 32 have been superimposed.

tRNA sequences including modified bases. ThetRNA sequence database is approximately ®vetimes smaller (546 sequences) than the tDNA data-base (2726 sequences). Despite this reduced num-ber of sequences, useful information can begathered due to the presence of modi®ed nucleo-tides (Table 3). It appears that, at least for C �A orU �C pairs, only modi®ed bases compatible withthe formation of the non-canonical interactionshown in Figure 2 are reported. For example, theoccurrence of C32 �A38 pairs is close to 42 %; thatof C �A pairs close to 11 % (a methylation of the20-hydroxyl group does not prevent the formationof a bifurcated hydrogen-bonded pair as shown inyeast tRNAPhe); that of m3C� close to 4 % (themethylation of the N3 atom of a cytosine base iscompatible with the C �A arrangement representedFigure 2); and that of a s2C �A pair close to 1 % (asulfur atom replacing the O2 atom of a cytosinebase could also form a S2(s2C32) . . . N6(A38) bifur-cated hydrogen bond). Thus, 59 % of C �A pairs areobserved at position 32 �38. The majority of thesemodi®ed C �A pairs can adopt the conformationshown in Figure 2. Similarly, all the modi®ed U �Upairs reported in Table 3 are compatible with theWatson-Crick like arrangement shown in Figure 4.

Is the sequence at position 32 �38 specific for agiven tRNA type? In order to detect a possible cor-relation between the sequence of the 32 �38 pairand either the amino acid type of the tRNA or thephylogenetic group in which they occur, the familyof tDNAs containing a T32 �T38 pair was analyzed(Sprinzl et al., 1998). The T �T pairs are found intDNAs coding for 14 different amino acid residues,in eukaryotic, eubacterial, mitochondrial, andchloroplastic tRNAs (in archaea, only one occur-rence is reported). Further, no correlation betweenthe ``base type'' at positions 37 and 38 could benoted (146 A37 �T38 and 140 G37 �T38 sequences).As in the case of tDNAs containing a T32 �T38 pair,

res of tRNAs complexed with synthetases (see Table 1).irs are shown on the right. For comparative purposes, aair of free tRNAs shown Figure 1 are drawn. The glyco-

Table 2. Percentages of occurrence of the four A, G, C and T bases at the non-canonical 32 �38 base-paired positionfor all 2726 tDNAs contained in the tRNA database (Sprinzl et al., 1998), a subset of 932 elongator tDNAs, and a sub-set of 352 eukaryotic elongator tDNAs (percentages larger than 5 % are idicated in bold)

Base 38All tDNAs (2726)

Base 32 A G C T Tot. R/Ya

A 1 �0 �0 �0 2G �0 �0 �0 1 1 3C 48 - 8 1 56T 18 3 8 11 39 97Total: 66 3 16 13 100R/Ya 69 30 100

All tDNA elongators (932)Base 32 A G C T Tot. R/Ya

A 1 �0 �0 �0 1G - - - - - 1C 54 1 12 1 68T 13 1 7 10 30 99Total: 68 2 18 11 100R/Ya 69 30 100

All tDNA eukaryotic elongators (352)Base 32 A G C T Tot. R/Ya

A 1 �0 - - 1G - - - - - 1C 58 �0 14 2 74T 2 �0 8 15 25 99Total: 89 1 22 17 100R/Ya 61 39 100

a Indicates the percentages of purine (R) and pyrimidine (Y) bases at position 32 and 38.


no apparent correlation exists for tDNAs display-ing other types of 32 �38 pairs.

Model structures for the 32 �38 interaction in tRNAs

C �A, U �A, U �C, C �C and U �U base-pairs. Fromthe preceding structural and phylogenetic data, itis possible to propose models for base-pairs locatedat position 32 �38 of tRNA anticodon hairpins(Figure 5). As the vast majority of 32 �38 pairs areof the C �A type, the arrangement observed in thecrystal structures of yeast tRNAPhe (see Figure 2)seems the most plausible (family I). The same con-®guration can be adopted by U �A pairs whichmay be linked by a bifurcated rather than by tworegular hydrogen bonds as in Watson-Crick pairs(Figure 5). As shown before, the much rarer U �Cand C �C pairs could adopt conformations isostericto C �A and U �A pairs. Interestingly, the shallowgroove patterns of C �A and U �A pairs are similar,as are the shallow groove patterns of U �C andC �C pairs. Note that the U �C pair shown inFigure 5 is different from the U �C pair observed inthe crystal structure of several RNA oligomers(Holbrook et al., 1991; Cruse et al., 1994; Tanakaet al., 1999) where the two bases are linked by aO4(U) . . . N4(C) instead of an O2(U) . . . N4(C)hydrogen bond with a water molecule bridging thetwo N3 sites of both pyrimidines.

However U �U pairs, which account for approxi-mately 7 % of the total number of pairs, cannotadopt a conformation isosteric to that observed forC �A pairs of family I. The arrangement shown inFigure 5 is the one observed in the crystal struc-

tures of the wild-type and transcripts of E. colitRNAGln complexed with its cognate synthetase(see Figure 4). It is proposed to be a representativeof family II. tRNAs with a U �U pair at position32 �38 may also belong to a separate family oftRNAs with speci®c and unknown features.

Rare 32 �38 base-pairs. Besides the pairs infamilies I and II shown in Figure 5, several rarebase oppositions are found at positions 32 �38.They account for 7 % of the total number of tDNA(Table 2) and tRNA (Table 3) sequences and aregathered in family III. Table 2 indicates that onlyG �C pairs are completely avoided in the 2726sequences reported in the tRNA database (Sprinzlet al., 1998). For several of these pairs, pairingarrangements approximately isosteric to family I orto family II can be suggested (Figure 6). In the fol-lowing, only cis pairing types will be considered.

A �A (1.1 %, see Table 2), G �U (0.6 %), and G �A(0.1 %) pairs are roughly isosteric to C �A pairs(family I of Figure 5), while U �G (2.7 %), C �U(1.3 %), and A �U (0.4 %) pairs are approximatelyisosteric to U �U pairs (family II shown in Figure 5).For the ®rst family, the A13 �A22 pair, found in thecrystal structure of E. coli tRNAGln (pte003), is closeto the bifurcated C �A pair. For G �U pairs, nostructures close to that of the C �A pair have beenobserved and, thus, the G �U pair shown in Figure 6is suggested. The crystal structure of the 5 S rRNAloop E (ur1064; Correll et al., 1997) contains asheared G98 �A78 pair superposable to the singlyhydrogen bonded C �A pair. For the second family,the G �U and A �U pairs shown in Figure 6 have

Table 3. Percentages of occurrence of the four A, G, C and U bases at the non-canonical 32 �38 base-paired positionfor all 546 tRNAs contained in the tRNA database (Sprinzl et al., 1998), a subset of 367 elongator tRNAs, and a subsetof 191 eukaryotic elongator tRNAs

Base 38All tRNAs (546)

Base 32 A G C m5C U Tot. R/Ya

A 1 - - - �0 - 1G �0 - - - - �0 �0 1C 42 �0 7 1 1 2 53Cm 11 - �0 1 �0 - 12m3C� 4 - - - - - 4s2C 1 - - - - - 1U 5 2 4 �0 2 2 16Um 1 - - - �0 1 2 3 - 5 - �0 1 2m - - �0 - - - �0 99Total: 68 2 16 2 4 7 100R/Ya 70 30 100

All tRNA elongators (367)Base 32 A G C m5C U Tot. R/Ya

A - - - - 1 - 1G - - - - - �0 �0 1C 35 1 7 2 1 2 47Cm 14 - �0 1 �0 - 16m3C� 5 - - - - - 5s2C 1 - - - - - 1U 7 1 4 1 3 3 17Um 1 - - - �0 2 3 2 5 - - - 1 8m - - 1 - - - 1Total: 66 2 16 3 4 8 100R/Ya 68 32 100

All tRNA eukaryotic elongators (191)Base 32 A G C m5C U Tot. R/Ya

A - - - - �0 - �0G - - - - - 1 1C 27 - 5 3 1 4 40Cm 23 - 1 2 - - 26m3C� 10 - - - - - 10s2C - - - - - - -U - - 1 1 - 2 4Um - - - - �0 2 3 2 - 9 - - 2 13m - - 1 - - - 1 99Total: 64 - 18 6 1 10 100R/Ya 64 36 100

a Indicates the percentages of purine (R) and pyrimidine (Y) bases at position 32 and 38.


been modeled in order to be roughly isosteric tothe U �U pair while interacting only through asingle hydrogen bond. The C �U pair is extractedfrom the same crystal structure of an RNA duplex(ar0005; Tanaka et al., 1999). Note that NMR dataobtained for E. coli tRNAPhe revealed signalsassociated to a rare A32 �38 interaction (Hyde &Reid, 1985a,b). For all these base-pairs, slightchanges in the positions of the bases may``improve'' their isosteric character. For A �G(0.3 %), A �C (0.2 %), C �G (0.2 %), G �C (0.04 %),and G �C (0 %) oppositions, no arrangement closeto either the C �A or the U �U pair shown inFigure 5 can be proposed. Yet, for these rare base-pairs, the possibility of base editing (Price & Gray,1998) and the probable existence of a noise level inthe tRNA sequence database leading to a smallnumber of incorrect sequences should be takeninto account. Therefore, the sequences and model

structures of family III should be considered withcaution.

Discussion

A conserved non-canonical 32 �38 base-pair in theanticodon loop of tRNAs

The preceding structural and phylogenetic datalead to the conclusion that particular types of pair-ing occur systematically between the ®rst (32) andlast (38) residues in the anticodon loop of tRNAs.The largest family of base-pairs (family I) com-prises the C �A pair, where the two bases are linkedby a bifurcated hydrogen bond, and the isostericU �A, U �C and C �C pairs (Figure 5). This categoryrepresents 86 % of the total number of sequences.A second family of base-pairs (family II) comprisesthe U �U pairs (7 %), which are not isosteric to the

Figure 5. The possible singlehydrogen-bonded arrangements of32 �38 base-pairs in anticodon loopof tRNAs organized in threefamilies. In family I, the structureof the U �A pair is inferred fromthat of the C �A pair extracted fromyeast tRNAPhe, and the structure ofthe C �C pair is inferred from thatof the U �C pair extracted fromyeast tRNAAsp (see Figure 2). Infamily II, the structure of the U �Upair is extracted from the E. colitRNAGln/RS complex (see Figure 4).Family III gathers the rare opposi-tions. The glycosidic bonds of base32 are similarly oriented in all thebase-pair oppositions.


pairs of family I. Finally, a third family (family III)comprises a set of rare base-pairs (7 %) for whichfew structural data are available (Figure 6), butwhich could be roughly distributed among the ®rsttwo families.

Importance of bifurcated hydrogen bonds

Up to now, pairs where the bases are linked bya single or a bifurcated hydrogen-bonding inter-action have encountered limited interest, probablybecause of their apparent lability. Yet, such pairswere observed in the crystal structures of yeasttRNAPhe (Quigley & Rich, 1976) and yeast tRNAAsp

(Westhof et al., 1985) at position 32 �38 (where twochemical groups are involved) and at positionG18 �55 (where three chemical groups areinvolved). Furthermore, subsequent moleculardynamics simulations have shown that they arestable over nanosecond time scales in yeasttRNAAsp (Auf®nger & Westhof, 1996; Auf®ngeret al., 1999). Since the resolution of the ®rst tRNAcrystal structures, the bestiary of non-canonicalbase-pairs including bifurcated hydrogen bondshas been extended and comprises, among others,G �G and G �U pairs (Correll et al., 1997). Thus, theconsideration of singly or bifurcated hydrogen-

bonded pairs should improve the structures result-ing from crystallographic, NMR, and modelingre®nement processes. The fact that such pairs areadditionally stabilized by water-mediated inter-actions should also be taken into account (Rouldet al., 1991; Auf®nger & Westhof, 1998c).

Why a conserved pyrimidine at position 32in tRNAs?

The non-canonical C32 �A38 base-pair is charac-terized by a Watson-Crick-like distance betweenthe C10 atoms and an asymmetric disposition ofthe angles at the C10 atoms (Figure 2). The lattergeometrical characteristic leads to a pronouncednon-isostericity upon reversal of the base-pair.Thus, C �A pairs are not isosteric to A �C pairs(Figure 7), U �A pairs are not isosteric to A �Upairs, and U �C pairs are not isosteric to C �U pairs.The C �C and U �U pairs shown in Figure 5 areobviously self-isosteric. Thus, the anticodon loopmust begin with a pyrimidine at position 32 inorder to adopt its characteristic functional shape.These structural considerations are in agreementwith phylogenetic data indicating that, in tRNAs, apyrimidine is present in 98 % of the sequences atposition 32 (Tables 2 and 3).

Figure 6. Possible arrangementsof the rare 32 �38 oppositions gath-ered in family III. Only thosearrangements which can approxi-mately be consisdered as isostericto the base-pairs of families I and IIhave been represented. The struc-ture of the C �A pair (family I),extracted from yeast tRNAPhe, isshown in blue and the structure ofthe U �U pair (family II), extractedfrom the E. coli tRNAGln/RS com-plex, is shown in red (see Figure 5).The glycosidic bonds of residues 32are similarly oriented in all thebase-pair oppositions.


Is isostericity always required?

As demonstrated by the Watson-Crick pairs, iso-stericity is a central concept in nucleic acid struc-ture. From crystal and phylogenetic data gatheredfrom a set of sequences of the 5 S rRNA loop E,isostericity has also been proposed for several non-canonical base-pairs (Leontis & Westhof, 1998a,b,1999b). In the present study, a certain number ofisosteric arrangements have been observed at pos-ition 32 �38. Such isosteric pairs may result in asimilar structure of the anticodon loop for freetRNAs. However, the occurrence of a non-negli-gible number of U �U and related pairs (7 %) which

are not isosteric to C �A pairs indicates that,depending on the context, isosteric base-pairs maynot be systematically required. Thus, such sets ofnon-isosteric base-pairs may be characteristic ofdifferent tRNA families, for which adjustments forinteractions with proteins, other RNA motifs ordivalent ions are required. Signi®cantly, whenbase-pairs are not strictly isosteric to the C �A orU �U pairs, their percentage of occurrence dropsconsiderably (see Figure 6).

Several biochemical studies have evaluated theeffects of changing the 32 �38 base-pair type. Yarusand co-workers (1986b) have studied the suppres-sion ef®ciency of a large number of mutants of the

Figure 7. Bifurcated C �A pairs are not isosteric toA �C pairs as shown by the large distance separating theglycosidic bonds. Thus, a A �C pair cannot replace aC �A pair at position 32 �38, indicating why a pyrimidineat position 32 is required in tRNAs. The glycosidicbonds of 32 have been superimposed.


amber suppressor Su7 and have shown that nomutation at position 32 or 38 was neutral. Forexample, Um � and Um �C pairs at position 32 �38were found to be less ef®cient than Um �A orCm �A pairs (Yarus et al., 1986a). Interestingly, themeasured suppression ef®ciency is Um � 'Um �C < Um �A < Cm �A and, thus, follows the pro-portion of 32 �38 pairs derived from phylogeneticanalysis (Figure 5). The substitution of the Cm32nucleotide by any purine residue reduces the sup-pression ef®ciency more than tenfold (Smith &Yarus, 1989) in agreement with the fact that pur-ines are strongly disfavored at position 32.Mutations of base 38 show less dramatic effectsindicating that a larger mutational variability isallowed at position 38 than at position 32 (seeTables 2 and 3). In transcripts of E. coli tRNAGly,with a U32 �A38 pair the anticodon discriminatesthe glycine codons according to the wobble rules,but with a C �A pair it loses its discriminatingpower (Lustig et al., 1993). In transcripts of Myco-plasma myciodes tRNAGly, the substitution of thewild-type C32 �A38) by a U �A pair (Claesson et al.,1995) resulted in similar effects (but the transform-ation was the reverse of that performed in E. colitRNAGly). Since in both cases an isosteric substi-tution was performed (C �A$U �A, see Figure 5),these data indicate that the nature of the contactsbetween the tRNA and the ribosome may involvethe 32 �38 base-pair and that such a substitutionaffects the recognition pattern of the anticodonhairpin more than its three-dimensional fold. It hasalso been demonstrated that substitution ofU32 �A38 by a C �A pair in M. myciodes tRNAGly

results in a substantial increase in its frameshiftingef®ciency (O'Connor, 1998). In E. coli tRNAGln tran-scripts, substitution of the U32 �C38 opposition bya C �C or C �A pair resulted in an increase of bothaminoacylation and ribosome performance, butenhanced the latter function to a greater extent(McClain et al., 1998). It is possible that the replace-

ment of a rare U32 �U38 by a non-isosteric C �Apair has as a consequence the formation of a more``anticodon'' like hairpin. Besides the isostericity ofseveral of these non-canonical base-pair types, eachof these base-pairs may interact speci®cally withproteins and ribosomal elements (see below). Iso-stericity, thus, is a generally valid concept indicat-ing that it is possible to substitute, with minorstructural changes, base-pairs displaying differentshallow and deep groove recognition patterns.However, each of these isosteric base-pairs hasspeci®c functional roles, experimentally character-ized by different translational ef®ciency. Further-more, in some occurrences non-isostericsubstitutions have been observed like the U �U pairof the rare base-pairs found in Tables 2 and 3. Forexample, in E. coli, a minor tRNAAla isoacceptordiffers from the other isoacceptors in possessing arare purine instead of a pyrimidine at position 32.However, the structure and functional role of suchbase-pairs is not yet known.

Importance of modified nucleotides at positions 32and 38

Modi®ed nucleotides are found in many RNAmolecules and especially in tRNAs where theyaccount for 12 % of the total number of residues(Grosjean & Benne, 1998). A large number of modi-®ed nucleotides are located at the stem/loop junc-tion of the anticodon hairpin in tRNAs (Auf®nger& Westhof, 1998b). It has been proposed, in agree-ment with a large body of experimental evidenceand molecular dynamics simulations, that some ofthese modi®cations, and especially pseudouridyla-tion (Auf®nger & Westhof, 1998a), are required tostabilize the three-dimensional structure of thefunctionally important anticodon loop. At position32, the stabilization mechanism involves the for-mation of a pseudouridine/water complex wherethe water molecule mediates base to backboneinteractions, and thus helps to increase the stabilityof the 32 �38 interaction pairs (Davis & Poulter,1991; Arnez & Steitz, 1994; Auf®nger & Westhof,1997, 1998a). The same nucleotide/water com-plexes can occur after pseudouridylation at pos-itions 31, 38, and 39. Such modi®cations may thushelp to strengthen the structure of the tRNA anti-codon hairpin in the ribosome and prevent exces-sive unfolding of the anticodon loop wheninteracting with synthetase proteins.

Indeed, as noted above, in a few of the crystalstructures of tRNAs complexed with their cognatesynthetase, the 32 �38 interaction is systematicallymaintained even after partial unfolding of theloop, and is stabilized by water mediated inter-actions involving modi®ed nucleotides (Roud et al.,1989). Thus, one possible role of the 32 �38 pair andof associated modi®ed nucleotides is to limit the``unfolding'' of the loop to the ®ve bases, 33 to 37.The structure of base-pairs 31 �39, 32 �38, and thestacking of U33 below base 32 being at least par-tially preserved, as shown in Figure 8(a) for the

Figure 8. Turns in RNA motifs involving non-canonical base-pairs. (a) Views extracted from crystal structures ofyeast tRNAPhe (trna09) and tRNAAsp (trna07) as well as from the crystal structures of the complex of E. coli tRNAGln

with its cognate synthetase (pte003). For comparison, a fragment of a conventional Watson-Crick helix is drawn. Theposition of the glycosidic bond, for a hypothetical 32-38 Watson-Crick base-pair in the tRNAPhe anticodon loop ismarked by a green arrow. (b) Fragments of the crystal structures of the TC loop of yeast tRNAPhe (trna09) and ofthe GNRA tetraloop of the hammerhead ribozyme (uhx026). (c) Fragments of the crystal structure of the leadzyme(ur0001) and the hammerhead ribozyme (uhx026) showing the C23 �A45 and the C3 �C17 base-pairs. In all theseFigures, the glycosidic bonds of 32 have been superimposed.


E. coli tRNAGln/RS complex, a subsequent ``refold-ing'' to the ribosomal active form, after decom-plexation with the synthetase, may be facilitated.Besides a structural role, the question of a possiblefunctional role of the non-canonical 32 �38 base-

pair can be raised. It has been noted that in someinstances bases 32 and 38 can interact with the cog-nate synthetase and, thus, act as structural determi-nants in the recognition process. Biochemical datafully support this view (for a review, see GiegeÂ


et al., 1998). Nevertheless, the largest number oftRNA/synthetases do not use these bases as struc-tural determinants and, thus, their structural roleseems to prevail over their direct implications in arecognition process with a synthetase.

Is there a relationship between the 32 �38interaction and the U-turn?

The active three-dimensional structure of tRNAanticodon loops is characterized by a U-turnassociated with the conserved base U33 and the A-form helical stack of loop residues 34 to 38 in conti-nuity with the helical stem residues 39 to 44 (asshown in Figure 8(a) for yeast tRNAAsp and yeasttRNAPhe). While both residues 32 and 38 arelocated in a helical track, the rotation of the glyco-sidic bonds on the side of residue 32 is much morepronounced than in regular helical A-form stacks(Figure 8(a)). Furthermore, between residues 34and 38, the rotation angles are rather regular andnot far from the helical case. But, clearly, residue38 has not rotated suf®ciently to form a Watson-Crick pair with residue 32. Hence, a substitution ofthe non-canonical 32 �38 pair by a regular Watson-Crick pair would hinder the formation of the U-turn and the regularity of the loop. This explainswhy no G �C and very few C �G pairs are reportedat these positions (Table 2). Therefore, the isostericC �A, U �A, U �C, and C �C pairs at position 32 �38,and possibly also U �U pairs, appear as a crucialtransition element linking the stem to the loop ofthe anticodon hairpin. Residues 32 and 38 should

Figure 9. tRNA anticodon loop signature showing the struThe distribution of the four A, G, C, and U nucleotides is esent in the tRNA database (Auf®nger & Westhof, 1998b; Spri

therefore be considered, along with the conservedbase U33 and likely a conserved purine at position37, as a signature for tRNA anticodon loops(Figure 9). The view that only speci®c base combi-nations occur at position 32 �38 is in agreementwith the extended anticodon concept, indicatingthat the coding performance of the triplet antico-don is enhanced by the appropriate loop and stemsequence (Yarus, 1982; Yarus et al., 1986b; Yarus &Smith, 1995).

In the seven base T-loops of tRNAs, a non-cano-nical two hydrogen-bonded trans HoogsteenT54 �A58 pair precedes the U-turn. This base-pairappears to play a similar role in T-loops as thatplayed by the 32 �38 pair in anticodon loops(Figure 8(b)). Note that a T54 �A58 pair is reportedin 78 % of tRNA gene sequences (Sprinzl et al.,1998). Similarly, in the recent crystal structure ofthe L11 binding region in LSU RNA, a trans Wat-son-Crick A �U pair ¯anks the U-turn of a ®venucleotide loop (Conn et al., 1999). GNRA tetra-loops are known to form U-turns as well (Westhofet al., 1989; Jucker & Pardi, 1995). In such motifs,G �A sheared pairs, which involve two hydrogenbonds, are used to close the loop (Figure 8(b)).Thus, sheared G �A pairs, almost isosteric to C �Apairs (Figure 6), are used in short loops, presum-ably because of their intrinsic higher stability andbecause they induce a sharper turn, while theapparently more labile singly hydrogen bondedbase-pairs found in anticodon motifs are used inlonger loops.

cture/sequence relationship at the 32 �38 base-pair level.xtracted from the set of 2726 tDNA gene sequences pre-nzl et al., 1998).


Other examples of interfacial non-canonical base-pairs

In hairpins. A recent NMR structure of thetRNALys,3 anticodon hairpin obtained at low pHrevealed the formation of a C32 �A�38 base-pair(Durant & Davis, 1999). The pKa of the adeninebase was estimated to be close to 6 for A38. Like-wise, evidence for the formation of a non-canonicalbifurcated U �C pair have been obtained by NMRfor the central hairpin of the HDV ribozyme con-taining a seven nucleotide loop (Kolk et al., 1997).In this structure, the sharp turn which is associatedwith the change of direction of the backbone isshifted in the 30 direction by one nucleotide, andcharacterized as a reversed U-turn. Interestingly,an independent NMR determination of the struc-ture of the same hepatitis delta virus hairpin pro-posed a different hydrogen-bonding scheme forthe U �C pair of the nucleotides on the 50 and 30-ends of the loop (Lynch & Tinoco, 1998). In this lat-ter structure, the cytosine amino group interactswith O4 instead of O2 of U resulting in a waterinserted U �C pair similar to that observed in thecrystal structure of RNA oligonucleotides(Holbrook et al., 1991; Cruse et al., 1994; Tanakaet al., 1999). In any case, in each example of struc-turally determined hairpins containing loops withseven bases, a non-canonical base-pair is observedat the junction between the stem and the loop.

Several examples of loops closed by non-canoni-cal base-pairs can be found in crystal structures(Pley et al., 1994; Scott et al., 1995; Cate et al., 1996;Correll et al., 1998; Perbandt et al., 1998). A hairpinclosing wobble U �U pair is found in the crystalstructure of the complex formed between a frag-ment of the U2 snRNA and a spliceosomal protein(Price et al., 1998). Interestingly, in the crystal struc-ture of the RNA-binding domain of the U1Aspliceosomal protein complexed with an RNA hair-pin, no loop closing C �A base-pair is observed(Oubridge et al., 1994) indicating that, in this otherstructural context, a possible non-canonical base-pair is disrupted after complex formation.

In ribozymes. Interestingly, the crystal structure ofthe leadzyme (Wedekind & McKay, 1999), aC23 �A45 pair, identical with the C �A pairobserved in tRNAs, is present (Figure 8(c)). It hasbeen proposed, on the basis of crystallographicdata, that this non-canonical pair is involved in thebinding of a Pb2� in the RNA deep groove whichmay be important for catalysis. In one motif of thecrystal structure of a Ba2� occupies this site (intRNA crystal structures, a related Mg2� bindingsite is located in the deep groove of the 32 �38pair). Interestingly, the formation of a C �A pair isnot mandatory for catalysis, since substitutionexperiments have shown that the ribozyme is stillactive after replacement of the adenine residue byan abasic ribose (Chartrand et al., 1997).

In the hammerhead ribozyme, a U �A non-cano-nical base-pair involving a single hydrogen bond

closes stem III (Pley et al., 1994; Scott et al., 1995).This hydrogen bond, which is not a bifurcated oneas in tRNAs, is well maintained during an MDsimulation of the ribozyme (Hermann et al., 1998).Also, and most interestingly, the base precedingthe U-turn in the hammerhead ribozyme is nor-mally a C and it forms a wobble base-pair with thecleavable C residue (C3 �C17: see Figure 8(c)). Thisbase-pairs differ from the singly hydrogen-bondedbase-pair shown in Figure 5. Nevertheless, it seemsessential for the ribozyme as a substitution of thisC �C pair by a G �C pair results in a considerableloss of activity (McKay, 1996).

Could the 32 �38 base-pair participate in RNA/RNArecognition in the ribosome?

As emphasized above, several biochemical stu-dies have demonstrated that modi®cations of the32 �38 base-pair can alter both aminoacylation andribosome performance suggesting that this base-pair is involved in direct or indirect recognitionfeatures with their cognate synthetase and with theribosome (Lustig et al., 1993; Claesson et al., 1995;Yarus & Smith, 1995; GiegeÂ et al., 1998; McClainet al., 1998; O'Conner, 1998) or, as proposed byYarus and co-workers (Smith & Yarus, 1989; Yarus& Smith, 1995), with a second tRNA inside theribosome. For RNA/RNA interactions within agiven organism, the Watson-Crick binding sites ofresidue 38, located in the shallow groove of theanticodon stem, are not blocked and could beinvolved in the formation of a triple interactionwith a ribosomal residue (Figure 10). Thus, theadenine base of C �A and U �A pairs and the cyto-sine base of U �C and C �C pairs could be involvedin Watson-Crick interactions with U or G residues,respectively. Yet, other interaction types cannot beexcluded. For example, the formation of a transA38 �U or C38 �U pair could be proposed. Thisinteraction scheme presents the advantage of invol-ving a uridine base in all four cases (Figure 10).Further, a uridine base could also be placed in theshallow groove of the U32 �U38 pair. Interestingly,GNRA tetraloops in which the Watson-Crick sitesof the guanine or the adenine residue of the closingG �A pair are available, are involved in RNA/RNArecognition via the Watson-Crick sites of the A inthe shallow groove side of G �C pairs in helices(Michel & Westhof, 1990; Jaeger et al., 1994). Crys-tal structures have demonstrated that the G �A pairof a GNRA tetraloop can indeed form of a base tri-ple with a helix (Pley et al., 1994).

However, RNA/RNA interactions could alsotake place on the less accessible deep groove side,especially at the border of a helix. Biochemical dataindicate that the substitution of a C �A by an isos-teric U �A pair affects the discriminating ability oftRNA transcripts (Lustig et al., 1993; Claesson et al.,1995) and their frameshifting ef®ciency (O'Connor,198). This seems only possible if recognitionphenomena involve the deep groove side whichpresents different shapes. However, interaction

Figure 10. Possible interaction of a ribosomal RNA residue with the shallow groove of the set of four isosteric32 �38 base-pairs in the largest Family. Left: Given the accessibility of the Watson-Crick sites of residue 38, formationof Watson-Crick pairs involving either a U or a G residue can be proposed. Right: Base triples involving an uridineresidue in trans with respect to residue 38 are possible in all four cases.


patterns involving a third residue are less obvious.Further, the deep groove side of the non-canonicalC �A pair in yeast tRNAPhe ( �C in yeast tRNAAsp)has been described as forming an ion binding sitein tRNAs and in the leadzyme (Wedekind &McKay, 1999). Several rare modi®cations were alsofound to block the access to the deep groove side.These modi®cations are m3C� at position 32 andm5C at position 38. Note that m3C� has only beendetected at position 32 in tRNAs (Auf®nger &Westhof, 1998b). In order to test the precedinghypothesis, several biochemical assays can be pro-posed. For example, a methylation of the N1 site ofA38 or of the N3 site of C38 should block access tothe shallow groove, while a methylation of the N7site of A38, the C5 site of C38, the N3 site of C32,or the N3 site of U32 would at least partially hin-der access to the deep groove. Interestingly, in therecent crystal structure of a conserved ribosomalRNA/protein domain, a U �A pair adopts the con-formation of the C �A pair observed in yeasttRNAPhe (Conn et al., 1999). Here, the hydroxylgroup of a distant adenine residue seems hydrogenbonded to the uridine residue on the deep grooveside N3(U) . . . O20(A) � 2.6 AÊ ). Although such ahydrogen bond is not compatible with a methyl-ation at the N3 site of the uridine base, it doespoint to another potential binding site in the deepgroove. Furthermore, binding of ribosomal proteinresidues cannot be excluded.

Conclusions

Biomolecular motifs with speci®c structure andfunctions can be recognized by a set of sequence

characteristics constituting their molecular signa-tures. The knowledge of these signatures has manyimplications such as (i) the ability to recognizemotifs with speci®c structure in sequence data-bases; and (ii) the possibility to derive a 3D modelfrom the sequence. Up to now, the signature of thetRNA anticodon hairpins comprised the phylogen-etically conserved U33 residue associated with aconserved pyrimidine at position 32 and a con-served purine at position 37. A strong preferencefor a purine at position 38 had also been notedwhile an almost regular distribution of the fourbases is observed at the anticodon positions 35 to36. At position 34 adenine bases are poorly rep-resented (Grosjean et al., 1982; Auf®nger &Westhof, 1998b). An analysis of available data ledto the re®nement of the signature of the tRNAanticodon loop. tRNA anticodon loops are gener-ally closed by a set of non-canonical isosteric base-pairs involving a single inter-residue hydrogenbond at position 32 �38, which can be distributed inthree families. Family I (86 % of the sequences)comprises the four C �A, U �A, U �C, and C �C pairs,i.e. Y �C/A, base-pairs (interestingly, the U �A pairis not of the Watson-Crick type). Family II (7 %)gathers the set of U �U base-pairs, not isosteric tothe preceding ones, also observed at position32 �38. Thus, sequences with U32 �U38 pair maycomprise a tRNA family with distinct properties.Family III (7 %) comprises a set of rare base opposi-tions for which only a limited number of approxi-mately isosteric base-pairs could be tentativelyproposed. Besides the structural aspects, it is pro-posed that conserved patterns associated with theset of the four isosteric 32 �38 base-pairs of family I,


could be involved in the formation of base triplesmediating tertiary interactions of the tRNA with aribosomal RNA residue.

Acknowledgements

The authors thank Stephen Cusack for providing coor-dinate sets for the tRNA anticodon hairpin of structurespte002 and ptr014.

References

Arnez, J. G. & Steitz, T. A. (1994). Crystal structure ofunmodi®ed tRNAGln complexed with glutaminyl-tRNA synthetase and ATP suggests a possible rolefor pseudo-uridines in stabilization of RNA struc-ture. Biochemistry, 33, 7560-7567.

Arnez, J. G. & Steitz, T. A. (1996). Crystal structure ofthree mysacylating mutants of Escherichia coli gluta-minyl-tRNA synthetase complexed with tRNAGln

and ATP. Biochemistry, 35, 14725-14733.Auf®nger, P. & Westhof, E. (1996). H-bond stability in

the tRNAAsp anticodon hairpin: 3 ns of multiplemolecular dynamics simulations. Biophys. J. 71, 940-954.

Auf®nger, P. & Westhof, E. (1997). RNA hydration:three nanoseconds of multiple molecular dynamicssimulations of the solvated tRNAAsp anticodon hair-pin. J. Mol. Biol. 269, 326-341.

Auf®nger, P. & Westhof, E. (1998a). Effects of pseudour-idylation on tRNA hydration and dynamics: atheoretical approach. In Modi®cation and Editing ofRNA (Grosjean, H. & Benne, R., eds), pp. 103-112,American Society for Microbiology, Washington,DC.

Auf®nger, P. & Westhof, E. (1998b). Location and distri-bution of modi®ed nucleotides in tRNA. In Modi®-cation and Editing of RNA (Grosjean, H. & Benne, R.,eds), pp. 569-576, American Society forMicrobiology, Washington, DC.

Auf®nger, P. & Westhof, E. (1998c). RNA base pairhydration. J. Biomol. Struct. Dynam. 16, 693-707.

Auf®nger, P., Louise-May, S. & Westhof, E. (1999). Mol-ecular dynamics simulations of the solvated yeasttRNAAsp. Biophys. J. 76, 50-64.

Basavappa, R. & Sigler, P. B. (1991). The 3 AÊ crystalstructure of yeast initiator tRNA: functional impli-cations in initiator/elongator discrimination. EMBOJ. 10, 3105-3111.

Beier, H., Lee, M. C., Sekiya, T., Kuchino, Y. &Nishimura, S. (1992). Two nucleotides next to theanticodon of cytoplasmic rat tRNA(Asp) are likelygenerated by RNA editing. Nucl. Acids Res. 20,2679-2683.

Berman, H. M., Olson, W. K., Beveridge, D. L.,Westbrook, J., Gelbin, A., Demeny, T., Hsieh, S. H.& Srinivasan, A. R. (1992). The nucleic acid data-base: a comprehensive relational database of three-dimensional structures of nucleic acids. Biophys. J.63, 751-759.

Biou, V., Yaremchuk, A., Tukalo, M. & Cusack, S.(1994). The 2.9 AÊ crystal structure of T. thermophilusseryl-tRNA synthetase complexed with tRNASer.Science, 263, 1404-1436.

Brown, R. S., Dewan, J. C. & Klug, A. (1985). Crystallo-graphic and biochemical investigation of the lea-

d(II)-catalyzed hydrolisis of yeast phenylalaninetRNA. Biochemistry, 24, 4785-4801.

Cate, J. H., Gooding, A. R., Podell, E., Zhou, K. H.,Golden, B. L., Kundrot, C. E., Cech, T. R. &Doudna, J. A. (1996). Crystal structure of a group Iribozyme domain-Principles of RNA packing.Science, 273, 1678-1685.

Cavarelli, J., Eriani, G., Rees, B., Ruff, M., Boeglin, M.,Mitschler, A., Martin, F., Gangloff, J., Thierry, J.-C.& Moras, D. (1994). The active site of yeast aspar-tyl-tRNA synthetase: structural and functionalaspects of the aminoacylation reaction. EMBO J. 13,327-337.

Chartrand, P., Usman, N. & Cedergreen, R. (1997).Effects of structural modi®cations on the activity ofthe leadzyme. Biochemistry, 36, 3145-3150.

Claesson, C., Lustig, F., BoreÂn, T., Simonsson, C.,Barciszewska, M. & Lagerkvist, U. (1995). Glycinecodon discrimination and the nucleotide in position32 of the anticodon loop. J. Mol. Biol. 247, 191-196.

Comarmond, M. B., GiegeÂ, R., Thierry, J. C. & Moras, D.(1986). Three-dimensional structure of yeasttRNAAsp. I. Structure determination. Acta Crystallog.sect. B, 42, 272-280.

Conn, G. L., Draper, D. E., Lattman, E. E. & Gittis, A. G.(1999). Crystal structure of a conserved ribosomalprotein-RNA complex. Science, 284, 1171-1174.

Correll, C. C., Freeborn, B., Moore, P. B. & Steitz, T. A.(1997). Metals, motifs and recognition in the crystalstructure of a 5S rRNA domain. Cell, 91, 705-712.

Correll, C. C., Munishkin, A., Chan, Y. L., Ren, Z.,Wool, I. G. & Steitz, T. A. (1998). Crystal structureof the ribosomal RNA domain essential for bindingelongation factors. Proc. Natl Acad. Sci. USA, 95,13436-13441.

Cruse, W. B. T., Saludjian, P., Biala, E., Strazewski, P. &PrangeÂ, T. (1994). Structure of a mispaired RNAdouble helix at 1.6-AÊ resolution and implicationsfor the prediction of RNA secondary structure. Proc.Natl Acad. Sci. USA, 91, 4160-4164.

Cusack, S., Yaremchuk, A. & Tukalo, M. (1996a). Thecrystal structure of T. thermophilus lysyl-tRNAsynthetase complexed with E. coli tRNALys and aT. thermophilus tRNALys transcript: anticodon recog-nition and conformational changes upon binding ofa lysyl-adenylate analogue. EMBO J. 15, 6321-6334.

Cusack, S., Yaremchuk, A. & Tukalo, M. (1996b). Thecrystal structure of the ternary complex of T. ther-mophilus seryl-tRNA synthetase with tRNASer and aseryl-adenylate analogue reveals a conformationalswitch in the active site. EMBO J. 15, 2834-2842.

Cusack, S., Yaremchuk, A., Krikliviy, I. & Tukalo, M.(1998). tRNAPro anticodon recognition by Thermustermophylus prolyl-tRNA synthetase. Structure, 6,101-108.

Davis, D. R. & Poulter, C. D. (1991). 1H-15N NMR stu-dies of escherichia coli tRNAPhe from hisT mutants:a structural role for pseudouridine. Biochemistry, 30,4223-4231.

Dirheimer, G., Keith, G., Dumas, P. & Westhof, E.(1995). Primary, secondary, and tertiary structuresof tRNAs. In tRNA: Structure, Biosynthesis, and Func-tion (SoÈ ll, D. & RajBhandary, U., eds), pp. 93-126,American Society for Microbiology, Washington.

Durant, P. C. & Davis, D. R. (1999). Stabilization of theanticodon stem-loop of tRNALys,3 by A��C base-pair and by pseudouridine. J. Mol. Biol. 285, 115-131.


GiegeÂ, R., Sissler, M. & Florentz, C. (1998). Universalrules and idiosyncratic features in tRNA identity.Nucl. Acids Res. 26, 5017-5035.

Grosjean, H. & Benne, R. (1998). Editors of Modi®cationand Editing of RNA, American Society for Micro-biology, Washington, DC.

Grosjean, H., Cedergreen, R. J. & McKay, W. (1982).Structure in tRNA data. Biochimie, 64, 387-397.

Hermann, T., Auf®nger, P. & Westhof, E. (1998). Mol-ecular dynamics investigations of the hammerheadribozyme RNA. Eur. J. Biophys. 27, 153-165.

Hingerty, B., Brown, R. S. & Jack, A. (1978). Furtherre®nement of the structure of yeast tRNAPhe. J. Mol.Biol. 124, 523-534.

Holbrook, S. R., Cheong, C., Tinoco, I. & Kim, S. H.(1991). Crystal structure of an RNA double helixincorporating a track of non-Watson-Crick basepairs. Nature, 353, 579-581.

Hyde, E. I. & Reid, B. R. (1985a). Assignment of thelow-®eld 1H NMR spectrum of Escherichia colitRNAPhe using nuclear overhauser effects. Biochemis-try, 24, 4307-4314.

Hyde, E. I. & Reid, B. R. (1985b). NMR studies of ionbinding to Escherichia coli tRNAPhe. Biochemistry, 24,4315-4325.

Jaeger, L., Michel, F. & Westhof, E. (1994). Involvementof a GNRA tetraloop in long-range RNA tertiaryinteractions. J. Mol. Biol. 236, 1271-1276.

Jucker, F. M. & Pardi, A. (1995). GNRA tetraloops makea U-turn. RNA, 1, 219-222.

Kolk, M. H., Heus, H. A. & Hilbers, C. W. (1997). Thestructure of the isolated, central hairpin of the HDVantigenomic ribozyme: novel structurl features andsimilarity of the loop in the ribozyme and free insolution. EMBO J. 16, 3685-3692.

Leontis, N. B. & Westhof, E. (1998a). The 5S rRNA loopE: chemical probing and phylogenetic data versuscrystal structure. RNA, 4, 1134-1153.

Leontis, N. B. & Westhof, E. (1998b). A common motiforganizes the structure of multi-helix loops in 16Sand 23S ribosomal RNAs. J. Mol. Biol. 283, 571-583.

Leontis, N. B. & Westhof, E. (1999a). Conserved geo-metrical base pairing patterns in RNA. Quart. Rev.Biophysics, in the press.

Leontis, N. B. & Westhof, E. (1999b). Recurrent RNAmotifs: analysis at the base pair level. In RNA Bio-chemistry and Biotechnology (Barciszewski, J. & Clark,B. F. C., eds), Kluwer Academic Publishers, Boston.

Lustig, F., BoreÂn, T., Claesson, C., Simonsson, C.,Barciszewska, M. & Lagerkvist, U. (1993). Thenucleotide at position 32 of the tRNA anticodonloop determines ability of anticodon UCC to dis-criminate among glycine codons. Proc. Natl Acad.Sci. USA, 90, 3343-3347.

Lynch, S. R. & Tinoco, I. (1998). The structure of the L3loop from the hepatitis delta virus ribozyme: a syncytidine. Nucl. Acids Res. 26, 980-987.

McClain, W. H., Schneider, J., Bhattacharya, S. &Gabriel, K. (1998). The importance of tRNA back-bone-mediated interactions with synthetase for ami-noacylation. Proc. Natl Acad. Sci. USA, 95, 460-465.

McKay, D. B. (1996). Structure and function of the ham-merhead ribozyme: an un®nished story. RNA, 2,395-403.

Michel, F. & Westhof, E. (1990). Modelling of the three-dimensional architecture of group I catalytic intronsbased on comparative sequence analysis. J. Mol.Biol. 216, 585-610.

Nissen, P., Kjeldgaard, M., Thirup, S., Polekhina, G.,Reshetnikova, L., Clark, B. F. C. & Nyborg, J.(1995). Crystal structure of the ternary complex ofPhe-tRNAPhe, EF-Tu and a GTP analog. Science, 270,1464-1472.

Nissen, P., Thirup, S., Kjeldgaard, M. & Nyborg, J.(1999). The crystal structure of Cys-tRNACys-EF-Ti-GDPNP reveals general and speci®c features in theternary complex and in tRNA. Structure, 7, 143-154.

O'Connor, M. (1998). tRNA imbalance promotes-1frameshifting via near-cognate decoding. J. Mol.Biol. 279, 727-736.

Oubridge, C., Nobutoshi, I., Evans, R. P., Teo, C. H. &Nagai, K. (1994). Crystal structure at 1.92 AÊ resol-ution of the RNA-binding domain of the U1A spli-ceosomal protein complexed with an RNA hairpin.Nature, 372, 432-438.

Perbandt, M., Nolte, A., Lorentz, S., Erdmann, V. A. &Betzel, C. (1998). Crystal structure of domain E ofThermus ¯avus rRNA: a helical RNA structureincluding a hairpin loop. FEBS Letters, 429, 211-215.

Perona, J. J., Rould, M. A. & Steitz, T. A. (1993). Struc-tural basis for transfer RNA aminoacylation byEscherichia coli glutaminyl-tRNA synthetase. Bio-chemistry, 32, 8758-8771.

Pley, H. M., Flaherty, K. M. & McKay, D. B. (1994).Three-dimensional structure of a hammerhead ribo-zyme. Nature, 372, 68-74.

Price, D. H. & Gay, M. W. (1998). Editing of tRNA. InModi®cation and Editing of RNA (Grosjean, H. &Benne, R., eds), pp. 103-112, American Society forMicrobiology, Washington, DC.

Price, S. R., Evans, P. R. & Nagai, K. (1998). Crystalstructure of the spliceosomal U2B00-U2A0 proteincomplex bound to a fragment of U2 smal nuclearRNA. Nature, 394, 645-650.

Quigley, G. J. & Rich, A. (1976). Structural domains oftransfer RNA molecules. Science, 194, 796-806.

Quigley, G. J., Teeter, M. M. & Rich, A. (1978). Struc-tural analysis of spermine and magnesium ion bind-ing to yeast phenylalanine transfer RNA. Proc. NatlAcad. Sci. USA, 75, 64-68.

Rath, V. L., Silvian, L. F., Beijer, B., Sproat, B. S. &Steitz, T. A. (1998). How glutaminyl-tRNA synthe-tase selects glutamine. Structure, 6, 439-449.

Rould, M. A., Perona, J. J., SoÈ ll, D. & Steitz, T. A. (1989).Structure of E. coli glutaminyl-tRNA synthetasecomplexed with tRNA(Gln) and ATP at 2.8 AÊ resol-ution. Science, 246, 1135-1142.

Rould, M. A., Perona, J. J. & Steitz, T. A. (1991). Struc-tural basis of anticodon loop recognition by gluta-minyl-tRNA synthetase. Nature, 352, 213-218.

Ruff, M., Krishnaswamy, S., Boeglin, M., Poterszman,A., Mitschler, A., Podjarny, A., Rees, B., Thierry,J. C. & Moras, D. (1991). Class I aminoacyl tRNAsynthetases: crystal structure of yeast aspartyl-tRNA synthetase complexed with tRNAAsp. Science,25, 1682-1689.

Schweisguth, D. C. & Moore, P. B. (1997). On the con-formation of the anticodon loops of initiator andelongator methionine tRNAs. J. Mol. Biol. 267, 505-519.

Scott, W. G., Finch, J. T. & Klug, A. (1995). The crystalstructure of an all-RNA hammerhead ribozyme: aproposed mechanism for RNA catalytic cleavage.Cell, 81, 991-1002.

Smith, D. & Yarus, M. (1989). tRNA-tRNA interactionswithin cellular ribosomes. Proc. Natl Acad. Sci. USA,86, 4397-4401.


Sprinzl, M., Horn, C., Brown, M., Loudovitch, A. &Steinberg, S. (1998). Compilation of tRNA sequencesand sequences of tRNA genes. Nucl. Acids Res. 26,148-153.

Sussman, J. L., Holbrook, S. R., Warrant, R. W., Church,G. M. & Kim, S.-H. (1978). Crystal structure ofyeast phanylalanine transfer RNA. I. Crystallo-graphic re®nement. J. Mol. Biol. 123, 607-630.

Tanaka, Y., Fujii, S., Hiroaki, H., Sakata, T., Tanaka, T.,Uesugi, S., Tomita, K. & Kyogoku, Y. (1999). A0-form RNA double helix in the single crystal struc-ture of r(UGAGCUUCGGCUC). Nucl. Acids Res. 27,949-955.

Tinoco, I. (1993). Structure of base pairs involving atleast two hydrogen bonds. In The RNA World(Gesteland, R. F. & Atkins, J. F., eds), pp. 603-607,Cold Spring Harbor Laboratory Press, Cold SpringHarbor, NY.

Wedekind, J. E. & McKay, D. B. (1999). Crystal structureof a lead-dependent ribozyme revealing metal bind-ing sites relevant to caalysis. Nature Struct. Biol. 6,261-268.

Westhof, E. & Sundaralingam, M. (1986). Restrainedre®nement of the monoclinic form of yeast phenyl-alanine transfer RNA. Temperature factors and dya-mics, coordinated waters, and base-pair propellertwist angles. Biochemistry, 25, 4868-4878.

Westhof, E., Dumas, P. & Moras, D. (1985). Crystallo-graphic re®nement of yeast aspartic acid transferRNA. J. Mol. Biol. 184, 119-145.

Westhof, E., Dumas, P. & Moras, D. (1988). Restrainedre®nement of two crystalline forms of yeast asparticacid and phenylalanine transfer RNA crystals. ActaCrystallog. sect. A, 44, 112-123.

Westhof, E., Romby, P., Romaniuk, P., Ebel, J.-P.,Ehresmann, C. & Ehresmann, B. (1989). Computermodeling from solution data of spinach chloroplastand of Xenopus laevis somatic and oocyte 5S rRNAs.J. Mol. Biol. 207, 417-431.

Yarus, M. (1982). Translational ef®ciency of transferRNA's: uses of an extended anticodon. Science, 218,646-652.

Yarus, M. & Smith, D. (1995). tRNA on the ribosome: awaggle theory. In tRNA: Structure, Biosynthesis, andFunction (SoÈ ll, D. & RajBhandary, U., eds), pp. 443-469, American Society for Microbiology,Washington, DC.

Yarus, M., Cline, S., Raftery, L., Wier, P. & Bradley, D.(1986a). The translational ef®ciency of tRNA is aproperty of the anticodon arm. J. Biol. Chem. 261,10496-10505.

Yarus, M., Cline, S. W., Wier, P., Breeden, L. &Thompson, R. C. (1986b). Actions of the anticodonarm in translation on the phenotypes of RNAmutants. J. Mol. Biol. 192, 235-255.

Edited by D. E. Draper

(Received 7 June 1999; received in revised form 26 July 1999; accepted 30 July 1999)

Singly and Bifurcated Hydrogen-bonded Base-pairs …...COMMUNICATION Singly and Bifurcated...

Documents

Transcript of Singly and Bifurcated Hydrogen-bonded Base-pairs …...COMMUNICATION Singly and Bifurcated...