H/ACA guide RNAs, proteins and complexes
Keqiong Ye
H/ACA guide RNAs direct site-specific pseudouridylation of
substrate RNAs by forming ribonucleoprotein (RNP) complexes
with pseudouridine synthase Cbf5 and three accessory
proteins. Recently determined crystal structures of H/ACA
protein complexes and a fully assembled H/ACA RNP complex
have provided significant insights into the architecture,
assembly and mechanism of action of RNA-guided
pseudouridine synthase. The binding of guide RNA is directed
by its conserved secondary structure and sequence motifs,
which enables guide RNA with different sequences to be
incorporated into the same protein complex. Accessory
proteins and peripheral domains crucially coordinate the
position of guide RNA, and possibly regulate the reaction
process.
Addresses
National Institute of Biological Sciences, 7 Science Park Road,
Zhongguancun Life Science Park, Beijing, China 102206
Corresponding author: Ye, Keqiong (yekeqiong@nibs.ac.cn)
Current Opinion in Structural Biology 2007, 17:287–292
This review comes from a themed issue on
Nucleic acids
Edited by Dinshaw J Patel and Eric Westhof
Available online 15th June 2007
0959-440X/$ – see front matter
# 2007 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.sbi.2007.05.012
Introduction
Numerous chemical changes are introduced at particular
sites of cellular RNAs after transcription. RNA modification enzymes often decide where to act by recognizing
certain structural and/or sequence features of substrate
RNAs through protein interaction. However, in eukarya
and archaea, two types of modifications of rRNAs and
eukaryotic small nuclear RNAs are specified by two
classes of guide RNAs [1,2]. H/ACA guide RNAs direct
the conversion of uridine into pseudouridine (C) and C/D
guide RNAs direct 20 -O-methylation of ribose. They
bind with a complementary region of substrate RNA
and instruct a protein catalyst to modify a specific target
nucleotide.
It has been known for some time that each H/ACA RNA
associates with a set of core proteins: the C synthase Cbf5
(dyskerin in humans) and three accessory proteins, Nop10,
Gar1 and Nhp2 (L7Ae in archaea) [3]. Only recently have
enzymatically active ribonucleoprotein (RNP) complexes
www.sciencedirect.com
been reconstituted from recombinant archaeal proteins
and in vitro transcribed RNAs [4,5]. The availability of
reconstituted complexes greatly accelerated high-resolution structure determination. In a short period, crystal
structures have been determined for Cbf5–Nop10 [6,7]
and Cbf5–Nop10–Gar1 [8] protein complexes, and a fully
assembled RNP complex [9] (Figure 1). This review will
discuss recent structural insights into the architecture,
assembly and mechanism of action of H/ACA RNA-guided
C synthase.
H/ACA RNAs and proteins
H/ACA RNAs consist of one to four hairpin domains, each
about 65–75 nucleotides (nt) in size [1–3]. The hairpin
domain contains a large internal loop and a 30 -ACA
sequence motif comprising two absolutely conserved adenines and a less conserved cytosine. The internal loop (also
known as the pseudouridylation pocket), if functional,
contains guide sequences that recognize the substrate by
forming two duplexes with regions of the substrate flanking
the target uridine. The target site is positioned 14–16
nucleotides upstream of the ACA motif. These features
are conserved in all H/ACA RNAs and are now understood
structurally [9]. Archaeal H/ACA RNAs additionally contain a kink (K)-turn motif in the upper stem, a widespread
RNA motif also present in rRNAs, C/D RNAs and
many other RNAs [2,10,11]. In eukaryotic cells, H/ACA
and C/D RNAs are located in the nucleolus or Cajal
bodies, and have been called small nucleolar (sno) or small
Cajal-body-specific (sca) RNAs. The vast majority of
eukaryotic H/ACA guide RNAs contain two hairpin
domains. The single-hairpin RNAs have been used in
recent reconstitutions and structural studies because of
their simpler structure [4–6,9].
Currently known C synthases are divided into five families,
classified as RluA, RsuA, TruA, TruB and TruD [12]. Cbf5
is the C synthase in H/ACA RNPs and belongs to the TruB
family, whose archetypical member, TruB, is responsible
for the synthesis of C55 in the T-loop of elongator tRNAs.
Structures have been determined for TruB alone and in
complex with a hairpin substrate RNA [13–15], providing a
good opportunity for structure comparison with Cbf5. In
contrast to closely related TruB and all other C synthases,
which recognize the substrate through protein interaction,
Cbf5 relies on guide RNA to recognize the substrate.
Moreover, all other C synthases are composed of single
polypeptide chains, whereas Cbf5 needs three accessory
proteins. How Cbf5 cooperates with guide RNA and
accessory proteins to achieve RNA-guided pseudouridylation is a central question that the H/ACA RNP structure is
able to address.
Current Opinion in Structural Biology 2007, 17:287–292
288 Nucleic acids
Figure 1
Structure of the fully assembled H/ACA RNP in (a) front and (b) side views (PDB code 2HVY). The corresponding structures of the Cbf5–Nop10
and Cbf5–Nop10–Gar1 complexes [6–8] are similar. The Cbf5 catalytic (Cat) domain is colored green, the Cbf5 PUA domain is light green,
Gar1 is cyan, L7Ae is blue, Nop10 and its zinc ion are violet, the RNA K-turn and ACA motif are red, the guide regions are orange and the rest
of the RNA is yellow. The same color coding is used for the other figures, unless indicated otherwise. The N and C termini of the proteins are
labeled when appropriate. The asterisk denotes the active site, dots represent disordered protein residues, and yellow and orange spheres
represent disordered RNA residues.
H/ACA protein complexes
In the absence of H/ACA RNA, the four H/ACA proteins
can form a four-subunit complex in eukaryotes [16,17] or a
maximal three-subunit Cbf5–Nop10–Gar1 complex in
archaea [4]. Crystal structures of heterodimeric Cbf5–
Nop10 complexes from Methanococcus jannaschii [6] and
Pyrococcus abyssi [7], and a heterotrimeric Cbf5–Nop10–
Gar1 complex from Pyrococcus furiosus [8] have provided
the first insights into the organization of the H/ACA complex. Cbf5 contains a catalytic domain, which is divided by
the central active site cleft into roughly two equal parts, and
a pseudouridine synthase and archaeosine transglycosylase
(PUA) domain (Figure 1). All structurally characterized C
synthases, including Cbf5 and members of each of the five
families, share the same catalytic domain fold, a similar
configuration of the active site and an absolutely conserved
catalytic aspartate [12]. It is very likely that pseudouridylation occurs through a common mechanism, although the
exact reaction pathway has not been well defined.
C synthases often contain add-ons, such as insertions
and peripheral domains, that confer various substrate
Current Opinion in Structural Biology 2007, 17:287–292
specificities [12]. Two such add-ons of Cbf5, the peripheral PUA domain and a thumb loop, are also shared by
TruB. In contrast to TruB, Cbf5 has a large PUA domain,
which is wrapped around by a stretch of 30 N-terminal
residues and extended by five C-terminal residues. The
latter critically bind RNA, as shown in the RNP structure
[9]. In addition, the thumb loop and insertion I of TruB,
which together bind substrate RNA, are shorter or not
present in Cbf5, respectively (Figure 2a,b), consistent
with the different substrate RNAs that TruB and Cbf5
recognize.
When in complex with Cbf5, the 60-residue protein
Nop10 folds into an elongated two-domain structure
and binds along its whole length to the Cbf5 catalytic
domain. The N-terminal half of Nop10 forms a zinccoordinating ribbon domain and the C-terminal half forms
a linker region and an a helix [6–8]. The formation of
the elongated structure critically depends on its extensive
interaction with Cbf5, as isolated Nop10 has a disordered
linker region and a much shorter C-terminal a helix [6].
Yeast Nop10 lacks zinc coordination and is even more
www.sciencedirect.com
H/ACA guide RNAs Ye 289
Figure 2
Proposed mechanism of RNA-guided pseudouridylation. (a) Structure of the TruB–RNA complex [13] (PDB code 1K8W) used to model the
substrate-bound state of H/ACA RNP shown in (b). TruB was aligned with Cbf5. The RNA regions used in modeling are colored the same as
the corresponding regions in the H/ACA substrate model and other regions are in gray. TruB is colored like Cbf5, with the thumb loop and
insertion I in gray. (b) Model of the substrate-bound state of H/ACA RNP. (c) Two states of the Cbf5 thumb loop. The thumb loop docks against
Gar1 in the open state (gray) and appears to be unable to bind substrate RNA. The loop from the Cbf5–Nop10–Gar1 structure (orange) is shown
as an approximation of the RNA-bound state [8]. (d) Schematic of the archaeal H/ACA RNP structure. The thumb loop is shown in open (grey)
and closed (orange) states. Paired RNA regions are named P1 and P2 for the lower and upper stems, and PS1 and PS2 for the substrate
duplexes formed with the 50 and 30 guides, respectively (purple). The modification target is shown as product pseudouridine (C, red). N, any
nucleotide; R, purine.
unfolded in the unbound state [6,18], but surprisingly
can assemble with archaeal Cbf5, indicating that the
Cbf5–Nop10 interaction interface is conserved among
archaeal and eukaryotic H/ACA RNPs [6].
The secondary accessory protein, Gar1, folds into a compact b barrel and binds at one end of the Cbf5 catalytic
www.sciencedirect.com
domain, distant from the Nop10-binding site [8]. The
independent interaction of Nop10 and Gar1 with Cbf5 is
consistent with biochemical results [4,5]. Overall,
Nop10, Gar1 and the PUA domain extend the catalytic
core in three directions. The functional consequences of
such an arrangement became apparent following the
determination of the entire RNP structure [9].
Current Opinion in Structural Biology 2007, 17:287–292
290 Nucleic acids
H/ACA RNP structure
Figure 3
The structure of the entire P. furiosus H/ACA RNP
includes proteins Cbf5, Nop10, Gar1 and L7Ae and a
single-hairpin H/ACA RNA [9] (Figure 1), which are
all required to achieve optimal pseudouridylation activity
[4,5]. The RNP structure resembles a triangle with the
catalytic domain located in the center. L7Ae docks against
Nop10 and occupies one corner. The PUA domain and
Gar1 take up the other two corners, as already revealed by
the protein complex structures. The H/ACA hairpin adopts
an extended structure, lying along one side of the active
site cleft. The upper stem (P2) of RNA is bound jointly by
L7Ae, which recognizes the K-turn as expected, Nop10
and Cbf5. The lower stem (P1) and ACA motif are recognized by the Cbf5 PUA domain. The guide sequences,
flanked by the lower and upper stems, are thus placed in
the vicinity of the active site cleft, an appropriate site for
substrate recruitment.
The interactions of Nop10 with RNA and L7Ae escaped
previous biochemical detection [4,5], and apparently
only occur in the context of the RNP complex. The
C-terminal half of Nop10 crucially organizes the RNP
structure by interacting with Cbf5, L7Ae and guide RNA.
Consistent with this structural observation, a fragment
consisting of only the C-terminal half of Nop10 was
capable of reconstituting an active RNP [7]. By contrast,
the N-terminal half of Nop10, the zinc-ribbon domain,
contacts solely Cbf5, which probably serves to increase
the binding affinity of Nop10. Compared with the
C-terminal half, the N-terminal half of Nop10 would
experience fewer evolutionary constraints and thus
becomes more divergent in structure; eukaryotic Nop10
no longer coordinates a zinc ion.
The K-turn motif in the RNP structure is placed at
the terminus of the hairpin, unlike previously characterized canonical K-turn structures, which are embedded within a duplex. However, the structure of the
terminal K-turn and its interaction with L7Ae are essentially the same as that of the canonical type [9,
19,20].
The duplex structure of the lower stem and the ACA
sequence motif are critical for H/ACA RNA localization,
stability and function [21]. Their interactions with the
Cbf5 PUA domain provide a structural explanation for
their importance [9] (Figure 3). Numerous interactions involving the minor groove of the lower stem
underscore the importance of maintaining the duplex
structure. In the ACA motif, the two adenines are
largely buried and specifically recognized by directional
hydrogen bonds along base edges, whereas the middle
cytosine is more exposed. These interactions correlate
with the universal conservation of the two adenines and
the increased sequence variability associated with the
second base.
Current Opinion in Structural Biology 2007, 17:287–292
Binding of the RNA lower stem (P1) and ACA motif by the Cbf5 PUA
domain. The RNA is colored gray, except the first nucleotide in the ACA
motif is colored purple, the second is orange and the third is red.
The PUA domain of archaeosine tRNA-guanine transglycosylase (ArcTGT), a tRNA modification enzyme, has also
been shown to bind an RNA duplex (the tRNA acceptor
stem) followed by a 30 triplet (CCA) [22], but the details of
the interaction are markedly different in the two cases
[9]. Additionally, how the PUA domain of TruB interacts
with tRNA remains unclear, as current structures of TruB–
RNA complexes include only a short hairpin mimicking
the tRNA T-stem loop [13–15]. Nevertheless, in the model
of the TruB–tRNA complex [13], the acceptor stem of
tRNA lies over the TruB PUA domain and would align well
with the lower stem of H/ACA RNA. The tRNA CCA
motif is also in place to contact the PUA domain, similar to
the ACA motif in the H/ACA RNP structure, although
whether the contact occurs and how remains unknown.
The RNP structure showed that Gar1 does not contact the
H/ACA RNA. This is a surprising result given that yeast
Gar1 was the first protein associated with H/ACA RNAs to
be identified [23] and has been characterized as an H/ACA
RNA-binding protein [24]. Eukaryotic Gar1 contains a
central domain homologous to archaeal Gar1 and two
additional glycine/arginine-rich (GAR) domains. It is
possible that the GAR domains interact with guide or
substrate RNAs, giving rise to cross-linking between
Gar1 and substrate RNA [16], but these interactions do
not appear to be essential, as only the central domain is
required for yeast cell viability [25]. The non-involvement
www.sciencedirect.com
H/ACA guide RNAs Ye 291
of Gar1 in RNA assembly correlates with the observation
that all H/ACA proteins, except for Gar1, are critical to H/
ACA RNA stability [3].
Biogenesis of eukaryotic H/ACA RNP requires a Naf1
protein that probably assembles into H/ACA RNPs during
the early stages of biogenesis and is then replaced by Gar1
in mature RNPs (see review in [26]). Naf1 shares sequence
homology with the central domain of Gar1, including those
residues that interact with Cbf5, and probably binds Cbf5
in a similar way to Gar1 [8]. As Gar1 only contacts Cbf5,
exchange between Gar1 and Naf1 would be unlikely to
interfere with H/ACA RNA association and the rest of the
structure of the RNP.
Mechanism of RNA-guided
pseudouridylation
The RNP structure represents the enzyme in the rest
state. Before substrate binding, the two guide sequences
are in an irregular and flexible conformation (Figure 1).
This flexibility might facilitate their hybridization with
substrate RNA without the energy cost of breaking a
preformed structure.
Guide sequences are expected to form two duplexes with
substrate RNA, resulting in a three-helix junction around
the target uridine [27]. The conformation of the substrate-bound state of H/ACA RNP must await further
structural study. Nevertheless, the previously determined
structures of TruB in complex with a hairpin substrate
RNA provide important hints [13–15] (Figure 2a). In
these structures, the target uridine was flipped out from
its normal structural context and inserted into the active
site cleft. Such a conformation of the target uridine was
also observed in a more recent structure of RluA in
complex with RNA [28] and is likely to be conserved
for other C synthases, including Cbf5. Moreover, Hoang
and Ferré-D’Amaré have pointed out the analogy between the hairpin structure recognized by TruB and one
of the substrate–guide RNA duplexes in substrate-bound
H/ACA RNP (PS2 in Figure 2d). The model based on this
analogy illustrates the possible conformation of the substrate (Figure 2b). In the substrate-loaded RNP, the
target uridine and the ACA motif would be bound to
their respective sites in Cbf5, which would impose a
length constraint (14–16 nt, as observed) on the structural
elements (PS2, P1 and any remaining unpaired region)
connecting them.
The function of Gar1 remains mysterious. Although Gar1
does not assemble guide RNA, it increases the reaction
efficiency [4,5]. Clues about its function came from the
observation that Gar1 interacts with the thumb loop of Cbf5
in the RNP structure (Figure 2c). Such interactions were
not observed in the structure of the Cbf5–Nop10–Gar1
complex, probably due to a different crystal packing
environment [8]. The corresponding thumb loop interacts
www.sciencedirect.com
extensively with loaded substrate RNA in the TruB and
RluA structures [13–15,28]; therefore, the thumb loop of
Cbf5 probably assumes a similar role in locking substrate in
place. However, in the substrate-free RNP structure, the
partially disordered thumb loop docks against Gar1 and
appears to be too far away from the modeled substrate for
interaction. Thus, Gar1 may hold the thumb loop in an
open conformation to aid substrate loading and release.
Conclusions and future directions
Recent structures have revealed the intricate organization
of H/ACA-RNA-guided C synthase. The binding of H/
ACA RNA is directed by its conserved secondary structure and sequence motifs, and thus guide RNA with
different sequences can be incorporated into the same
protein complex. Accessory proteins and peripheral
domains critically coordinate the position of the guide
RNA, and probably regulate the reaction process. An
understanding of H/ACA RNP in action will require more
structural studies of various functional states.
It will be more challenging to study the structure and
function of eukaryotic H/ACA RNPs, which have no
enzymatically active reconstituted system reported to date.
The structure of archaeal H/ACA RNP has provided an
excellent model for understanding eukaryotic RNP
because of the general conservation of protein sequences
and RNA features. The unique features of eukaryotic RNP
remain to be explored. Most eukaryotic H/ACA RNAs
strictly contain two hairpins. This hairpin duality is functionally important, but has not been understood structurally. Eukaryotic H/ACA RNAs lack the K-turn motif and
the L7Ae counterpart, Nhp2, likewise lacks binding specificity for the K-turn. Moreover, Nhp2, unlike L7Ae, interacts stably with the Cbf5–Nop10 complex [16,17]. How
these differences are accommodated in the eukaryotic
RNP structure remains unclear.
In addition to pseudouridylation, eukaryotic H/ACA RNPs
participate in the processing of precursor rRNAs (U17/
SnR30 RNP) and telomere replication (vertebrate telomerase) [3]. An understanding of the structure and function
of these special H/ACA RNPs is still at a very early stage.
Mutations in dyskerin (human Cbf5) have been linked to a
rare genetic disease, dyskeratosis congenita, which probably results from telomerase and ribosome disorder [29].
The archaeal structural model seems inadequate to explain
the effect of these mutations. About half of these mutations
can be mapped onto a clustered area of the archaeal PUA
domain and have been suggested to affect RNA association
because of the RNA-binding function of the PUA domain.
However, the H/ACA RNP structure indicates that the
corresponding archaeal residues have no apparent role in
binding RNA or maintaining structure [9]. Structural
studies of the human H/ACA RNP complex will surely
shed new light on these mutations.
Current Opinion in Structural Biology 2007, 17:287–292
292 Nucleic acids
Acknowledgements
I would like to thank Ling Li for many discussions. This work was
supported by the Chinese Ministry of Science and Technology.
References and recommended reading
Papers of particular interest, published within the annual period of
review, have been highlighted as:
of special interest
of outstanding interest
1.
Kiss T: Small nucleolar RNAs: an abundant group of noncoding
RNAs with diverse cellular functions. Cell 2002, 109:145-148.
2.
Dennis PP, Omer A: Small non-coding RNAs in Archaea.
Curr Opin Microbiol 2005, 8:685-694.
3.
Meier UT: The many facets of H/ACA ribonucleoproteins.
Chromosoma 2005, 114:1-14.
4.
Baker DL, Youssef OA, Chastkofsky MI, Dy DA, Terns RM,
Terns MP: RNA-guided RNA modification: functional
organization of the archaeal H/ACA RNP. Genes Dev 2005,
19:1238-1248.
The authors assembled, for the first time, an enzymatically active H/ACA
RNP from four recombinant proteins (Cbf5, Nop10, L7Ae and Gar1) and
guide RNA. This study was a major step towards unveiling the structural
organization of H/ACA RNP.
5.
Charpentier B, Muller S, Branlant C: Reconstitution of archaeal
H/ACA small ribonucleoprotein complexes active in
pseudouridylation. Nucleic Acids Res 2005, 33:3133-3144.
As in [4], the authors reconstituted an active H/ACA RNP. They also
showed that Cbf5 and Nop10 are minimally required for activity.
6.
Hamma T, Reichow SL, Varani G, Ferre-D’Amare AR:
The Cbf5-Nop10 complex is a molecular bracket that
organizes box H/ACA RNPs. Nat Struct Mol Biol 2005,
12:1101-1107.
The structure of the Cbf5–Nop10 complex was determined, revealing
extensive interactions between them. Nop10 is partially disordered in
isolation.
Manival X, Charron C, Fourmann JB, Godard F, Charpentier B,
Branlant C: Crystal structure determination and site-directed
mutagenesis of the Pyrococcus abyssi aCBF5-aNOP10
complex reveal crucial roles of the C-terminal domains of both
proteins in H/ACA sRNP activity. Nucleic Acids Res 2006,
34:826-839.
As in [6], the structure of the Cbf5–Nop10 complex was solved. The
authors also carefully analyzed the contribution of the two Nop10
domains to the assembly and activity of H/ACA RNP.
7.
8.
Rashid R, Liang B, Baker DL, Youssef OA, He Y, Phipps K, Terns RM,
Terns MP, Li H: Crystal structure of a Cbf5-Nop10-Gar1 complex
and implications in RNA-guided pseudouridylation and
dyskeratosis congenita. Mol Cell 2006, 21:249-260.
A structure comprising Cbf5, Nop10 and a third protein, Gar1, represents
the largest protein-only complex that could be assembled by archaeal H/
ACA proteins. Gar1 binds Cbf5 independently of Nop10.
9. Li L, Ye K: Crystal structure of an H/ACA box ribonucleoprotein
particle. Nature 2006, 443:302-307.
This study presents the first structure of an entire H/ACA RNP. The
structure reveals how H/ACA RNA is assembled into the complex and
how RNA-guided pseudouridylation could occur.
10. Klein DJ, Schmeing TM, Moore PB, Steitz TA: The kink-turn: a
new RNA secondary structure motif. EMBO J 2001,
20:4214-4221.
11. Rozhdestvensky TS, Tang TH, Tchirkova IV, Brosius J,
Bachellerie JP, Huttenhofer A: Binding of L7Ae protein to the
K-turn of archaeal snoRNAs: a shared RNA binding motif for
C/D and H/ACA box snoRNAs in Archaea. Nucleic Acids Res
2003, 31:869-877.
Current Opinion in Structural Biology 2007, 17:287–292
12. Hamma T, Ferre-D’Amare AR: Pseudouridine synthases.
Chem Biol 2006, 13:1125-1135.
13. Hoang C, Ferre-D’Amare AR: Cocrystal structure of a tRNA
Psi55 pseudouridine synthase: nucleotide flipping by an RNAmodifying enzyme. Cell 2001, 107:929-939.
14. Pan H, Agarwalla S, Moustakas DT, Finer-Moore J, Stroud RM:
Structure of tRNA pseudouridine synthase TruB and its RNA
complex: RNA recognition through a combination of rigid
docking and induced fit. Proc Natl Acad Sci USA 2003,
100:12648-12653.
15. Phannachet K, Huang RH: Conformational change of
pseudouridine 55 synthase upon its association with RNA
substrate. Nucleic Acids Res 2004, 32:1422-1429.
16. Wang C, Meier UT: Architecture and assembly of mammalian
H/ACA small nucleolar and telomerase ribonucleoproteins.
EMBO J 2004, 23:1857-1867.
17. Henras AK, Capeyrou R, Henry Y, Caizergues-Ferrer M: Cbf5p,
the putative pseudouridine synthase of H/ACA-type snoRNPs,
can form a complex with Gar1p and Nop10p in absence of
Nhp2p and box H/ACA snoRNAs. RNA 2004, 10:1704-1712.
18. Khanna M, Wu H, Johansson C, Caizergues-Ferrer M, Feigon J:
Structural study of the H/ACA snoRNP components Nop10p
and the 30 hairpin of U65 snoRNA. RNA 2006, 12:40-52.
19. Hamma T, Ferre-D’Amare AR: Structure of protein L7Ae bound
to a K-turn derived from an archaeal box H/ACA sRNA at 1.8 A
resolution. Structure 2004, 12:893-903.
20. Moore T, Zhang Y, Fenley MO, Li H: Molecular basis of box C/D
RNA-protein interactions; cocrystal structure of archaeal
L7Ae and a box C/D RNA. Structure 2004, 12:807-818.
21. Balakin AG, Smith L, Fournier MJ: The RNA world of the
nucleolus: two major families of small RNAs defined by
different box elements with related functions. Cell 1996,
86:823-834.
22. Ishitani R, Nureki O, Nameki N, Okada N, Nishimura S,
Yokoyama S: Alternative tertiary structure of tRNA for
recognition by a posttranscriptional modification enzyme.
Cell 2003, 113:383-394.
23. Girard JP, Lehtonen H, Caizergues-Ferrer M, Amalric F,
Tollervey D, Lapeyre B: GAR1 is an essential small nucleolar
RNP protein required for pre-rRNA processing in yeast.
EMBO J 1992, 11:673-682.
24. Bagni C, Lapeyre B: Gar1p binds to the small nucleolar RNAs
snR10 and snR30 in vitro through a nontypical RNA binding
element. J Biol Chem 1998, 273:10868-10873.
25. Girard JP, Bagni C, Caizergues-Ferrer M, Amalric F, Lapeyre B:
Identification of a segment of the small nucleolar
ribonucleoprotein-associated protein GAR1 that is
sufficient for nucleolar accumulation. J Biol Chem 1994,
269:18499-18506.
26. Richard P, Kiss T: Integrating snoRNP assembly with mRNA
biogenesis. EMBO Rep 2006, 7:590-592.
27. Ganot P, Bortolin ML, Kiss T: Site-specific pseudouridine
formation in preribosomal RNA is guided by small nucleolar
RNAs. Cell 1997, 89:799-809.
28. Hoang C, Chen J, Vizthum CA, Kandel JM, Hamilton CS,
Mueller EG, Ferre-D’Amare AR: Crystal structure of
pseudouridine synthase RluA: indirect sequence readout
through protein-induced RNA structure. Mol Cell 2006,
24:535-545.
29. Marrone A, Walne A, Dokal I: Dyskeratosis congenita:
telomerase, telomeres and anticipation. Curr Opin Genet Dev
2005, 15:249-257.
www.sciencedirect.com