Báo cáo khoa học: Structural and functional analysis of ataxin-2 and ataxin-3 potx

Structural and functional analysis of ataxin-2 and ataxin-3
Mario Albrecht
1,
*, Michael Golatta
2,
*, Ullrich Wu¨ llner
3
and Thomas Lengauer
1
1
Max-Planck-Institute for Informatics, Saarbru
¨
cken, Germany;
2
Institute for Medical Biometry, Informatics, and Epidemiology,
University of Bonn, Germany;
3
Department of Neurology, University of Bonn, Germany
Spinocerebellar ataxia types 2 ( SCA2) and 3 ( SCA3) are
autosomal-dominantly inherited, neurodegenerative dis-
eases caused by C AG repeat expansions in the coding
regions of t he genes encoding ataxin-2 a nd ataxin-3,
respectively. To provide a rationale for further functional
experiments, we explored the p rotein architectures of ataxin-
2 and ataxin-3. Using structure-based multiple se quence
alignments of homologous proteins, we investigated
domains, s equence motifs, and interaction partners. Our
analyses focused o n p resumably f unctional a mino acids and
the construction of tertiary structure models of the RNA-
binding Lsm domain of a taxin-2 and the deubiquitinating
Josephin domain o f ataxin-3. We also speculate about dis-

tant evolutionary relationships of ubiquitin-binding UIM,
GAT, UBA and CUE domains and helical ANTH and
UBX domain extensions.
Keywords: spinocerebellar a taxia; Machado–Joseph dis-
ease; polyglutamine disorder; ubiquitin; valosin-containing
protein.
Spinocerebellar ataxia types 2 ( SCA2) and 3 ( SCA3) are
autosomal-dominantly inherited, neurodegenerative dis-
orders [1,2]. SCA3 has also been known as Machado–
Joseph disease ( MJD), and SCA2 and SCA3 belong to a
heterogeneous group of trinucleotide repeat disorders. This
group includes Huntington disease ( HD), d entatorubral-
pallidoluysian atrophy (DRPLA), and other spinocerebellar
ataxia t ypes such as SCA1, SCA7 and SCA 17 [3–7]. T he a ge
of onset of SCA2 and SCA3 is in the third to fourth decade
[8]. The disorders share common phenotypic features such
as the degeneration of s pecific vulnerable neuron popula-
tions and the presence of intracellular aggregations of the
mutant protein s in affected neurons. In contrast, the
expression of the disease-associated genes occurs in a g reat
variety of tissues and is not restricted to neuronal cells.
The SCA2 and SCA3/MJD genes h ave been mapped to
chromosomes 12q24.1 and 14q32.1 [1,2]. T he common
underlying genetic basis of SCA2 and SCA3 is the
expansion of a CAG repeat region beyond a certain thresh-
old. These CAG repeats encode a polyglutamine (polyQ)
tract in the respective proteins a taxin-2 and ataxin-3. The
polyQ stretch in ataxin-2 lies near the N-terminus at the 5¢-
end of the coding region of exon 1 [9], but the polyQ region
of ataxin-3 is contained in e xon 10 close to the C -terminus

[10]. While ataxin-2 is located p redominantly in t he Golgi
apparatus [11], ataxin-3 is found in both t he nucleus and the
cytoplasm of cells [12].
To provide a rationale for further experiments, we
characterized the protein architecture s of ataxin-2 and
ataxin-3 a nd investigated domains, sequence m otifs, and
interaction partners. To explore t he functional i mplications,
we asse mb led a multiple sequence alignmen t for the Lsm
domain of ataxin-2 homologues including the yeast
homologue Pbp1. We also constructed a 3D structural
model f or the RNA-binding Lsm domain of ataxin-2.
Similarly, we used a structure-based multiple sequence
alignment of t he Josephin domain o f ataxin-3 homologues
to derive a 3D model of this domain and to analyse specific
residues involved in deubiquitination.
Materials and methods
Protein sequences were retrieved from the NCBI [13],
Ensembl [14], and SWISS-PROT/TrEMBL (SPTrEMBL)
[15] databases and protein domain architectures from the
Pfam [16] an d SCOP [17] d ataba ses. Sequence a ccession
numbers are given in the respective figure legends and
Tables S1 and S2. Species names a re abbreviated by first
letters ( Table S3). Protein str uctures wer e obtained from t he
PDB database [18]. The secondary structure assignments of
PDB structures were taken from the DSSP database [19]. A
single capital letter appended to the a ctual PDB identifier
denotes the chosen structure chain. We used the
PSI
-
BLAST

suite of p rograms [20] t o search f or homologues ( E-value
cut-off 0 .005) and the web servers PSIPRED [21], SAM-T99
[22], a nd SSpro2 [23] to predict the secondary structure o f
proteins and t o f orm a consensus prediction by majority
voting [24]. To predict intrinsically unstructured and
disordered regions in proteins, we e xplored the consens us
of the results returned by the DisEMBL [25], D ISOPRED
[26], GlobPlot [27], NORSp [28] and P ONDR [10] online
Correspondence to M. Albrecht, Max-Planck-Institute for Informatics,
Stuhlsatzenhausweg 85, 66123 Saarbru
¨
cken, Germany.
E-mail:
Abbreviations: A2BP1, ataxin-2 binding protein 1; D RPLA, denta-
torubral pallidoluysian atrophy; DUB, de ubiquitinating enzymes;
HD, Huntington disease; MJD, Machado–Joseph disease; NLS,
nuclear localization signal; OTU, otubains; PABP, poly(A)-binding
protein; RMSD, root mean square deviati on; SCA, spinocerebellar
ataxia; SnRNPs, small nuclear ribonucleoproteins; UBP, ub iquitin-
specific protease; UCH, ubiquitin C-terminal hydrolase; UIM,
ubiquitin-interacting motif; V CP, valosin-containing protein.
*Note: M . A lb recht and M. G olatta co ntribut ed equally to t his w or k.
(Received 6 April 2004, accepted 7 June 2004)
Eur. J. Biochem. 271, 3155–3170 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04245.x
servers. The nuclear localization s ignals in ataxin-3 homo-
logues were discovered with help of the prediction server
PSORT II [29].
Multiple sequence alignments were assembled by m eans
of T-COFFEE [30] and improved manually by minor
adjustments based on structure prediction results and pair-

wise structure superpositions computed by the program
CE
[31]. The root m ean square deviations (RMSDs) were taken
from t he
CE
superpositions. W e i nvestigated the results of all
state-of-the-art fold recognitio n methods available via the
online meta-server BioInfo.PL [32], which contacts a dozen
other sta te-of-the-art p redict ion servers (the names of which
are listed on t he web s ite The
associated 3D-Jury system allows for the comparison and
evaluation of the predicted 3D models in a consensus view
[33]. To model the protein structure of ataxin-2 and ataxin-
3, we submitted the constructed sequence–structure align-
ments to t he 3D mo delling server W HAT IF [ 34]. The
sequence a lignments depicted in the figures were prepared in
the
SEAVIEW
editor [35] and illustrated by the web s ervice
ESPript [36]. The protein structure images were drawn
in the Accelrys Discovery Stud io ViewerLight. The online
version of t his m anuscript c ontains supplementar y material,
and our web site will provide additional pictures.
Results and discussion
Protein architecture of ataxin-2
Ataxin-2 has 1 312 residues (including 22 glutamines of the
polyQ stretch) and a molecular m ass o f  140 kDa. A taxin-
2 is a highly bas ic protein except f or one acidic region
(amino acid 254–475) containing 46 acidic amino acids
(Fig. 1 ). This region covers roughly e xons 2–7 and is

predicted to consist of two g lobular domains named Lsm
(Like S m, amino acid 254–345) [37] and LsmAD (Lsm-
associated doma in, amino acid 353–475). The LsmAD
domain of ataxin-2 contains both a clathrin-mediated
trans-Golgi signal (YDS, amino acid 414–416) and an
endoplasmic reticulum (ER) exit signal (ERD, amino acid
426–428) [11,38]. It is composed mainly of a-helices
according to t he results f rom s econdary structure p rediction
servers.
The r est of ataxin-2 outside of the L sm and LsmAD
domains is only weakly conserved i n eukaryotic ataxin-2
homologues and is predicted to b e i ntrinsically unstru ctured
according to the consensus result from the DisEMBL,
DISOPRED, GlobPlot, NORSp and PONDR online
servers. These nonglobular, flexible N- and C-terminal tails
(amino acid 1–253 and 476–1312) contain the polyQ region
(amino acid 166–187), several highly conserved short
sequence motifs as possible p rotein interaction sites, and
conspicuous (R)RG peptides a t t he C-terminus of the
LsmAD domain. One of the sequence motifs constitutes a
putative PABP [poly(A)-binding protein] interacting motif
PAM2 (amino acid 908–925) [39], and (R)RG peptides are
well-known to bind RNA in other proteins [40]. The N- and
C-terminal tails of ataxin-2 also have a high c ontent of
proline (179 prolines out of 1090 amino acids, 16.4%).
This property and the low complexity of unstructured
sequence regions may lead to several significant, but
probably f alse-positive, hits during a
PSI
-

BLAST
search for
homologues of ataxin-2. For instance, despite the use of the
standard low complexity fi lter, our
PSI
-
BLAST
search with
human ataxin-2 homologues found se veral questionable hits
outside globular domains to homologu es of t he poly-
glutamine DRPLA gene product atrophin. For instance,
starting the
PSI
-
BLAST
search wit h an Arabidopsis thaliana
ataxin-2 homologue (SPTrEMBL: Q94AM9), human
atrophin is r etrieved in the third iteration w ith an E-value
of 5 · 10
)11
. Conversely, using the rat atrophin homologue
(SPTrEMBL: Q62901) as t he start s equence, hum an ataxin-
2 was detecte d in the s econd iteration with an E-value of
8 · 10
)04
.
Fig. 1. Protein architectures of human ataxin-2, its yeast homologue Pbp1, and the P. falciparum homologue PF13_0048 of the decapping enzyme
DCP2 (DCP2_Pf).
3156 M. Albrecht et al. (Eur. J. Biochem. 271) Ó FEBS 2004
RNA binding of ataxin-2

The Lsm domain of ataxin-2 is typical of RNA-binding Sm
and Sm-like proteins, which o ften form cyclic 6-, 7- or even
14-oligomers [41–43]. Generally, L sm domain proteins are
involved in a varie ty of essential RNA processing events
including RNA modification, pre-mRNA splicing, and
mRNA decapping and degradation. Some of them are also
important componen ts of spliceosomal small nuclear ribo-
nucleoproteins (snRNPs).
The L smAD do main is contained i n t he Pfam database
with the name Ataxin-2_N and also occurs in another, as
yet uncharacterized Plasmodium falciparum/yoelii yoelii
gene products PF13_0048/PY07327 without an Lsm
domain (Fig. 1). Both Plasm odium gene products have an
additional N-terminal DCP2 domain (also termed box A),
which is always followed by a NUDIX domain [ 44] in a ll
known D CP2 homologues. This NUDIX domain consti-
tutes the catalytic subunit of the mRNA decapping
holoenzyme DCP1–DCP2 [45,46].
The physiological function of ataxin-2 and closely related
eukaryotic homologues in RNA processing is as y et quite
unexplored [47–50]. Interestingly, ataxin-2 has been
observed to interact with A2BP1 (ataxin -2 binding pro tein
1) [38], whose RNA-binding Caenorhabditis elegans homo-
logue, fox-1, regulates t issue-specific alternative sp licing [ 51].
Disruption o f the human A2BP1 gene may c ause epilepsy or
mental retardation [52]. In addition, ataxin-2 shows signi-
ficant homology to the yeast p rotein Pbp1 (Pab1/PABP-
binding protein 1), which also contains the Lsm and
LsmAD domains; regions ou tside o f t hese two globular
domains are predicted to be mainly unstructured in Pbp1 as

in ataxin-2.
Although the C-terminal tail of Pbp1 does not contain a
PAM2 motif [39], this yeast protein regulates polyadenyla-
tion afte r pre-mRNA splicing and interacts with the
C-terminal part of the yeast homologue PAB1 of the
human PABP [53]. A2BP1 and PABP are also evolutionar-
ily related and possess RNA recognition motifs [38]. These
observations strongly suggest that ataxin-2 is involved in
similar mRNA processing tasks.
Structural modelling of ataxin-2
First, we compiled a list of a taxin-2 homologues including
the yeast homologue Pbp1 and several Lsm domains of
snRNPs and other Sm and Sm-like proteins f rom v arious
species. Then, we assembled a structure-based multiple
sequence alignmen t o f the Lsm domains, crystallographi-
cally de termined structu res of which reveal a close structural
homology b etween archaeal an d eukaryotic proteins
(Fig. 2 ) [42,43,54–65]. T his suggests that the function and
the RNA-binding mode of the Lsm domain have been
preserved during evolution.
The RNA-binding Lsm domain is characterized by a
conserved sequence motif consisting of two short segments
known as Sm1 and Sm2, which are separated by a variable
linker [66,67]. The very stron g conservation of certain
glycine residues is especially striking and also demonstrates
the evolutionary relationship of ataxin-2 to Lsm domain
proteins. T he amide group s o f the glycines are known to
stabilize the protein structure when forming hydrogen
bonds to adjacent b-strands [55]. T he secondary structure
predictions of ataxin-2 and its yeast homologue Pbp1 are

also in good agr eement with the known s tructure of the Lsm
domain as open b-barrel, consisting of an N-terminal
a-he lix followed by a strongly bent five-stranded a ntiparallel
b-shee t with a 3
10
helical turn in some cases before the fifth
b-strand.
The top two alignment rows in Fig. 2 show human
ataxin-2 aligned w ith the Pyrococcus abyssi Sm1 protein
(PDB ide ntifier 1 m8v, cha in A), the c rystal structure of
which consists of a heptameric r ing w ith a central cavity like
other Lsm domain oligomers [65]. This Sm1 protein
provides the only Lsm domain structure, which is bound
to RNA inside and outside of the doughnut-shaped ring at
an internal and an external binding site. Therefore, we used
this alignment o f ataxin-2 t o Sm1 to model the 3D structure
of the L sm domain of a taxin-2 in complex with RNA and
Lsm domains of ataxin-2 protomers (Fig. 3).
Functional analysis of the Lsm domain
We applied the same colour scheme to fu nctionally relevant
residues shown in the multiple sequence a lignment and the
3D model of ataxin-2 ( Figs 2 and 3). Based on the crystal
structure o f Sm1 from P. abyssi bound to uridine heptamers
(U
7
), we marked several amino acids in Sm1, which are
involved in RNA binding [65] and are mostly physico-
chemically conserved in ataxin-2 (Sm1/ataxin-2 residue
numbers). The residues forming the internal U
7

binding site
are H37/K299, N39/L302 and R63/K330, while ionic
interactions between K22/K284, R63/K330 and D 65/S332
stabilize t he RNA-binding area. The residues involved in t he
external U
7
binding site are R4/R266, H10/T272 and Y34/
Y296, stabilized by a hydrogen bond between H10/T272
and Y34/Y296. It is interesting to note that Sm1 from
P. abyssi and from Archaeoglobus fulgidus (PDB identifier
1i4k, chain A) share identical RNA-binding residues except
for H10, which is replaced by an asparagine [59,65].
Furthermore, we investigated whether ataxin-2 may also
form oligomers through t he Lsm domain. To this end, we
used the detailed crystal structure a nalyses of the very
similar s nRNP heterodimers D
1
–D
2
and D
3
–B [55]. Because
of analogous intermolecular interactions in both dimers, we
focusedonthecomplexofD
3
with B. This complex is
stabilized mainly by the pairing of the fifth b-strand ( b5)
from D
3
with the f ourth b-strand (b4) from B (D

3
/ataxin-2–
B/ataxin-2): R69/V335–R73/K330, L71/V337–L71/L328,
and L73/F339–L69/S326. In addition, two hydrophobic
clusters formed by residues of D
3
and B contribute to the
stability of t he dimer. The first c luster includes F70/V336
and I72/Q338 ( both in b5 s trand) of D
3
and F 27/Y289
(b2 strand), L67/M324, V70/I327 and L72/L328 (all in b4
strand) of B. The second cluster c onsists of P6/M267, L10/
L271 (both i n a-helix), V18/C279 (b1 s trand), L32/F293 (b2
strand), I33/K294 (loop after b2 strand), I68/F334, L71/
V337 and L73/F339 (all in b5 strand) of D
3
and I41/L304,
C43/A306 ( both in b3), L69/S326 and L71/L328 (both in
b4) of B. Stacking interactions between guanidinium groups
of arginines R69/V335 of D
3
and R25/G287 and R49/T312
of B a s well a s an i onic interaction between E21/Q282 of D
3
and R 65/S322 o f B stabilize the dimer further. However, the
latter salt bridge is not observed in the D
1
–D
2

complex
Ó FEBS 2004 Analysis of ataxins 2 and 3 (Eur. J. Biochem. 271) 3157
Fig. 2. Structure-based multiple s equence alignment of the Lsm domains of ataxin-2 homologues including th e yeast h omologue Pbp1 (upper part) with Sm and Sm-like proteins (lower p art). The known D S SP
secondary structure assignment of the Sm1 protein from P. abyssi isshownatthetopofthealignment(cylinderfora-helix, arrow for b-strand), and the amino acid sequences of crystallographically
determined PDB structures of Lsm domains are underlined a ccordingly (curled line for a-helix, straight line for b-strand). The corresponding secondary structure predictions for the Lsm domains of ataxin-
2 and Pbp1 are also given. Physico-chemically similar amino acids are coloured identically. T he high ly co nse rved g lycines c haracteristic o f L sm d omains are indicated. I n t he up per p art , blue text boxes
point to functionally relevant amino acids forming an i n ternal binding site for u ridin e heptamers bound to Sm1 from P. abyssi, and green text boxes mark amino acids of t he external RNA binding site. In
the lower part, orange text labels annotate how the dimerization of the snRNPs D
3
and B is stabilized by i n termolecular interactions. PDB identifier s and corresponding SPTrEMBL accession numbers for
Lsm proteins are given in Table S2.
3158 M. Albrecht et al. (Eur. J. Biochem. 271) Ó FEBS 2004
despite identical amino acids. Altogether, the degree of
conservation of amino acids relevant for heterodimerization
is only mod erate, but may still suggest t hat ataxin-2 m ay
form Lsm domain oligomers.
Protein architecture of ataxin-3
The longest splice variant of ataxin-3 possesses 376 amino
acids (including 22 g lutamines of the polyQ stretch , amino
acid 296–317) and an approximate molecular weight of
42 kDa. Ataxin-3 consists of a globular deubiquitinating
N-terminal Josephin domain (amino acid 1–170) [68,69] and
a flexible C-terminal tail containing two ubiquitin-interact-
ing motifs (UIMs) [70] (also termed LALAL motifs and
PUBs [ 71], a mino acid 223–240 a nd 243– 260) and the polyQ
region (amino acid 296–317) (Fig. 4) [72]. A slightly shorter
alternative splice variant of ataxin-3 with 373 amino acids
has a third UIM (amino acid 334–351) at the C-terminus.
An as yet uncharacterized ataxin-3 paralogue on the X
chromosome (sequence identity 70%) is expressed in testis

(ataxin-3t) [10]. The Josephin domain is also found without
a C-terminal t ail i n other, a s y et uncharacterized, proteins
named josephins (Fig. 5) [73].
A highly c onserved, putative nuclear localiz ation signal
(NLS) i s found upstream of the polyQ stretch (RKRR,
amino acid 282–285), which may be bipartite in the
Caenorhabditis e legans homologue of ataxin-3, consisting
of 17 residues (RRDRQKFLERFEKKKEE, amino acid
296–312). This NLS follows a potential casein kinase II
(CK-II) phosphorylation site (TSEE, amino acid 277–280),
which may determine the rate of the observed a taxin-3
transport into the nucleus [74]. Ataxin-3 m ay also contain
a nuclear export signal (NES) following the Josephin
domain (ADQLLQMIRV, a mino ac id 174–183) based on
our comparis on with a published sequence profile of nuclear
export signals [75]. F urthermore, ataxin-3 contains several
conserved sequence motifs similar to N R- and CoRNR-
boxes L-x-x-L-L/[IL]-x-x-[IV]- I of transcriptional coactiva-
tors and corepressors, respectively [73]. Indeed, ataxin-3
interacts w ith histones a nd the histone acetyltransferases
CBP, p300, and PCAF, which w ork as transcriptional
coactivators. In particular, dependent on these cofactors,
ataxin-3 represses histone acetylation a nd transcription [76],
and altered protein acetylation has already been implicated
in polyglutamine disease processes [77]. Generally, the
(de-)ubiquitination of histones has been linked to transcrip-
tional regulation [78], which may also explain the observed
interactions of ataxin-3.
Ataxin-3 is evolutionarily conserved in eukaryotes inclu-
ding P. falciparum and plants, but not yeast. The P. falci-

parum homologue PFL1295w of ataxin-3 (ataxin-3_Pf),
whose gene expression is upregulated similarly to the
P. falc iparum josephin homologue PF11 _0125 i n gameto-
cytes [79–81], c onstitutes an exception because it has only
the second UIM conserved (amino acid 250–267) a nd has
an additional ubiquitin-like U BX domain [82–85] a t the
Fig. 3. 3D model of t he Lsm domain of a taxin-2 using three adja cent protomers of the Sm1 protein f rom P. abyssi as template (PDB identifier 1m8v,
chain A, B and G). The model illustrates predicted internal (blue) and external (green) binding sites of ataxin-2 to RNA (grey). a-Helices are in
shown in r ed, b-strands are shown in cyan . Only functionally relevant r esidues of the central a taxin-2 protomer are anno tated as follows: dark blue
boxes point to residues forming the internal s ite, and light blue boxes mark amino acids stabilizing the RNA b inding area; dark g reen boxes
highlight residues involved in the external site, and light green ones indicate stabilizing hydrogen bonds.
Ó FEBS 2004 Analysis of ataxins 2 and 3 (Eur. J. Biochem. 271) 3159
C-terminus (amino acid 271–381) instead of the polyQ-
containing region [69]. Like human ataxin-3, this ataxin-3
homologue PFL1295w also has a potential casein kinase II
phosphorylation site (TSDE, amino acid 278–281) close to
basic amino acids, which can be indicative of an NLS
(KKIH, amino acid 293–296) near the N-terminus of the
UBX domain. In contrast, t he prediction server
PSORT II
returns another region inside the U BX domain a s a possible
NLS ( PRRK, a mino acid 339–342). I t is unclear which NLS
motif may be functionally more relevant because both
NLS m otifs correspond to amino acids at solvent exposed
N-termini of the second and fourth b-strand in the crystal
structure of the UBX domain o f the cofactor p47 ( PDB
identifier 1 s3s) [86]. Similar to the P. falciparum homologue,
the Cryptosporidium parvu m homologue of ataxin-3 also
possesses only one UIM motif (amino acid 266–283) and a
C-terminal UBX domain (amino acid 288–397) instead of a

polyQ region.
Ubiquitin binding of ataxin-3
Ubiquitination fulfills many cellular functions in cytoplas-
mic trafficking, guiding specific proteins through the
endocytic pathways, and targeting proteins t o the protea-
some [84,87–93]. Above all, the ubiquitin–proteasomal
pathway is involved i n processing mutant or damaged
proteins that cause neurodegenerative diseases. The small
ubiquitin protein can be covalently linked to other proteins
as single molecule or polyubiquitin chain.
Recently, the two UIMs between the Josephin domain
and the polyQ stretch o f ataxin-3 have been shown to b e
capable of binding tetraubiquitin and polyubiquitinated
proteins [68,94–97]. In our previous study, we used the
C-terminal ANTH domain extension, which consists of an
antiparallel three-helix bundle, to model t he structure of the
UIMs in the C-terminal t ail of ataxin-3 [73]. In fact, novel
structure determinations have shown t hat U IM peptides are
a-h elices and can form helix bundles in the crystal structure
[98]. In contrast, the NMR s olution structures of UIM
peptides reveal that they are single amphiphatic a-helices
connected by unstructured linkers [99,100]. The latter
observation is in agreement with the observed flexibility of
the C-terminal tail of ataxin-3 [72].
Furthermore, the ANTH domain itself is evolutionarily,
structurally, and functionally related to a VHS domain
[101]. Lately, the structure of the GAT (GGAs and Tom1)
domain directly following the V HS domain of Tom1 and
GGAs (Golgi-associate, c-adaptin ear-containing, Arf-
binding proteins) w as determined crystallographically

[102–105]. The GAT domain contains a three-helix bundle,
which we found to superimpose very well with the helical
bundle o f t he C-terminal ANTH domain e xtension (RMSD
3.1 A
˚
, PDB identifiers 1o3x and 1hx8, A chains).
Interestingly, the GAT domains of GGAs a nd Tom1
have been report ed t o i nteract with ubiquitin [106–108].
The corresponding ubiquitin binding site was located to the
third a-helix of the GAT three-helix bundle, and h ydro-
phobic amino acids like leucines are important for the
interaction (Fig. 5). The same residue type also plays an
essential r ole in binding ubiquitin to the UIM a-helix [98–
100] and the third a-helix of the helical bundle in the
homologous CUE and UBA domains [109]. However, the
sequence s imilarity is quite low, and thus it is difficult to
deduce an evolutionary relationship, although the ubiquitin
binding sequence of GGAs and Tom1 resembles a
noncanonical UIM whose, otherwise strictly conserved,
serine residue is replaced by an asparagine except in case of
human GGA3 (Fig. 5).
Further interaction partners of ataxin-3
It has b een shown t hat a taxin-3 interacts with the ubiquitin-
like ( UBL) domain of the homologou s ubiquitin- and
proteasome-binding factors hHR23A and hHR23B, whose
yeast orthologue is Rad23 [96,110–112]. The latter factors
are also involved in the nucleotide excision repair pathway
by targeting the ubiquitinated nucleotide excision repair
factor XPC/Rad4 to the proteasome [113]. Their UBL
domain b inds to a UIM helix of the 26S proteasome subunit

S5a, and this interaction disr upts the interdomain con tacts
between the N-terminal ubiquitin-mimicking UBL domain
and t he two C-terminal ubiquitin-binding UBA domains,
thereby inducing t he change from a closed to an open
protein c onformation [109,111,114,115]. R ad23 and t he
yeast o rthologue Rpn10 o f S 5a serve a s a lternative ubiquitin
receptors for the proteasome [116], and the UBA domains
Fig. 4. Protein a rchitectures of human ataxin-3, its P. falciparum
homologue PFL1295w (ataxin-3_Pf), and human josephin 1.
3160 M. Albrecht et al. (Eur. J. Biochem. 271) Ó FEBS 2004
of Rad23 inhibit proteasome-catalysed proteolysis by
sequestering Lys48-linked polyubiquitin chains [117,118].
In particular, the NMR solution structures o f the U BL
domain of hHR23A/B bound to a UIM peptide of S 5a
[99,119] could be used to model the complex of hHR23A/B
and ataxin-3. Similarly, the complex of a UIM of ataxin-3
with ubiquitin could b e modelled based on the NMR
solution structure of the UIM of t he Vps27 protein bound
to ubiquitin [100].
The C-terminal region of ataxin-3 i ncluding the polyQ
region i nteracts with the N-terminal cofactor/substrate-
binding adaptor domain of the valosin-containing protein
VCP/p97/Cdc48/VAT/ter94 [96,120–123]. VCP is an
important multifunctional AAA+ ATPase with two C-
terminal ATPase domains after the adaptor domain, which
provide the energy for major conformational changes [124].
VCP forms hexamers and works as molecular chaperone
involved in a variety of intracellular functions including cell
cycle progression, membrane fusion, vesicle-mediated trans-
port, transcription activation, apoptosis prevention, and

ubiquitin-proteasome degradation, modulating polygluta-
mine-induced neurodegeneration [96,120–123,125–127].
VCP binds the ubiquitin E3 ligase and the chain assembly
factor UFD2a/E4B, which is a U box homologue of yeast
Ufd2 [128], a nd intera cts w ith a nd regulates the degradation
of the proteasome-associated ataxin-3, forming a trimeric
complex o f ataxin-3, VCP, and UFD2a [96,127,129–131].
Interestingly, Ufd2 binds the UBL domain of R ad23 and
competes with Rad23 for binding to the Rpn1 proteasome
subunit, while the N-terminal U BL domain o f the ubiquitin
C-terminal hydrolase Ubp6 interacts with Rpn1 without
competition with Rad23 [116,132].
Furthermore, VCP also binds the C -terminal UBX
domain of the membrane fusion adaptor p47/SHP1/EYC/
Ubx3 [85,86,133], w hich consist s of three domains UBA-
SEP-UBX [134]. The c rystallographically determined com-
plex of the N-terminal adaptor domain of VCP with this
UBX domain (PDB identifier 1s3s) indicates the interacting
residues [86] and could be used to model the putative
complex o f V CP with the C-terminal UBX domain o f t he
ataxin-3 homologue from P. falciparum (ataxin-3_Pf). Like
the U BX domain o f p47, a taxin-3_Pf contains the c onserved
loop that is essential for an interaction with VCP because it
inserts into a hydrophobic pocket o f VCP [86]. T he UBX
domain structure of p47 is extended at its N-terminus by a
disordered peptide structure and an additional a-helix of as
yet unknown f unctional relevance [86] . The length of this
a-he lix is similar to a UIM a-helix (Fig. 5), and such a UIM
also precedes the U BX domain o f ataxin-3_Pf. Th erefore,
this a-helix of p47 might be related to t he second UIM in

ataxin-3 homologues (recall that the first UIM is missing in
ataxin-3_Pf). In addition, the a rrangement o f one UIM
helix followed by a C-terminal UBX domain is a lso found in
the cofactor Ubx2 with domain architecture UBA-UAS-
Fig. 5. Multiple sequence alignmentofUIMpeptides,dividedinto
groups b y horizontal lines from top to b ottom: UIM sequences of the
Pfam seed alig nment including first, second, and th ird U IMs of ataxin-3
homologues, UIM-like peptides f rom GGAs and Tom1, and re lated
AP180 sequences. The latter a re derive d from t he 3D struc ture sup er-
position of the G AT do main of hu man G GA1 with t he AP180 exten-
sions from Rattus norvegicus and D. melanogaster (PDB identifiers 1hf8
and 1hx8, respectively). The sec ond group o f UIMs i n ataxin-3 h omo -
logues also includes the similar N -terminal a-helix of the UBX domain
extension of p 47 (PDB identifier 1s3s). F or each group, amino acids in
alignment co lumns with a majority of identical residue s are printed on a
black background, and similar amino acids are highlighted in grey.
Ó FEBS 2004 Analysis of ataxins 2 and 3 (Eur. J. Biochem. 271) 3161
UIM-UBX [133]. The UIM o f U bx2 binds ubiquitin c hains,
and the UBX domain i nteracts with VCP. T hus the same
interactions can be expected for ataxin-3_Pf.
The C-terminal, presumably VCP-binding, UBX domain
of ataxin-3_Pf appears to correspond to the V CP-binding
C-terminal part of human ataxin-3, which follows the
second UIM and includes the polyQ region [120,123,131].
In addition, the polyQ tract of ataxin-3 has been shown to
be indispensable for the interaction with VCP, and i ts length
correlates with the strength of the interaction. These obser-
vations raise the question how human ataxin-3 binds VCP
in contrast to its P. falciparum homologue. This is partic-
ularly interesting because VCP may suppress polyQ induced

neurodegeneration, and mutations in VCP have been
observed t o c ause cyto plasmic vacuoles followed by cell
death b ecause of a dysfunctional s econd ATPase domain
and inclusion body formation [ 120–123,127,135,136]. We
also observed that all VCP sequence variations associated
with Paget d isease of bone and frontotemporal dementia
(IBMPFD) [135] are not located in the binding interface of a
UBX domain w ith the N- terminal adaptor domain o f V CP,
but are involved in i nteractions between protein region s
(for details see the online supplement). Therefore, motions
of the adaptor domain, which are essential for proper VCP
function [124,127], may be impaired by IBMPFD-associ-
ated mutations.
According to a recent yeast-2-hybrid s creen [137], a
josephin homologue from Drosophila melanogaster
(CG3781) on t he X chromosome i nteracts with the heat
shock protein HSP60b (CG2830), which is involved in
spermatogenesis [138,139], suppresses ubiquitination [140]
and associates with 38 further proteins including a ubiquitin
E3 ligase, but no other deubiquitinating enzyme except
josephin (CG8184). Interestingly, HSP40 and HSP70 chap-
erones h ave already been observed to a ssociate with VCP,
and t hey also colocalize with intranucle ar ataxin-3 aggre-
gates and may play an important role in the disease process
and the impairment of the ubiquitin-proteasome system
[121,141–149].
Structural modelling of the Josephin domain
Recently, it has been observed t hat t he Josephin domain
contains highly conserved amino acids reminiscent of the
catalytic residues o f a deubiquitinating cysteine protease

[69], and first experimental results s upport this function
hypothesis [68]: decrease of polyubiquitination of
125
I-
labelled lysozyme by removal of ubiquitin, cleavage of the
ubiquitin p rotease substrate ubiquitin-AMC, and b inding of
the s pecific ubiquitin protease i nhibitor ubiquitin-aldehyde
(Ubal). Mutating t he catalytic c ysteine in a taxin-3 i nhibits
these functions [68].
Previously, we modeled the 3D structure of ataxin-3
based o n the A NTH domain [ 150] of the adaptin AP180
as structural template [73]. However, this prediction has to
be revised with r egard to t he N-terminal Josephin domain
because of t he identified cysteine protease signature [69].
In contrast to our previous pr ediction [73], which relied
on the secondary structure prediction from a single server,
we now formed the consensus result of the three s tate-of-
the-art s econdary structure prediction servers PSIPRED,
SAM-T99, and SSpro2. A ll three online servers basically
returned the same secondary structure for human ataxin-3
and josephin 1, resulting in a much more reliable
secondary structure prediction o f b-strands besides a-heli-
ces. We propose that the increased accuracy of this
prediction is due, a t least in part, to a substantial growth
of protein sequence and structure databases. The predic-
ted b-strands in the Josephin domain corroborate a
cysteine protease fold of deubiquitinating enzymes
(DUBs) and do not support the ANTH domain structure
consisting solely of a-helices. In hindsight, the fold
recognition methods applied in the past to predict the

structure of ataxin-3 may have been misguided by the
pronounced prediction of a-helices only.
DUBs process ubiquitin proteolytically at the C -terminus
and can be divided into a t l east two evolutionarily related
families of cysteine proteases, U BPs (ubiquitin-specific
proteases) and UCHs (ubiquitin C-terminal hydrolases)
[151,152]. However, new ubiquitin-specific families such as
otubains (OTU) and JAMMs with low sequence similarity
to known DUBs are still being discovered [151]. A
consensus of fold recognition servers now selects both
available UCH domain structures of human UCH-L3 [153]
and yeast YUH1 [154], which superimpose with a low
RMSD of 2.0 A
˚
(PDB identifiers 1uch and 1cmxA,
respectively), as best modelling templates with a moderate
confidence score for human josephin 1, but still with only a
weak score for ataxin-3. T he pairwise sequence–structure
alignments r eturned by the structure prediction servers for
3D modelling differ mainly in the central part of the
Josephin domain ( amino a cid 47–117 in at axin-3) aligned to
DUBs. This finding underpins the distant relationship o f the
Josephin domain t o known DUBs. The centr al part does
not contain catalytic residues a nd is thus less co nserved,
containing insertions of variable length and st ructure in
other cysteine proteases [155].
Based on a multiple sequence alignment of Josephin
domain homologues (Fig. 6), we used the crystallographi-
cally determ ined structure of Y UH1 bound to the ubiquitin-
like i nhibitor Ubal (PDB i dentifier 1cmx, chains A and B,

respectively) to model the tertiary structure of the Josephin
domain o f a taxin-3 i n c omplex with Ubal (Fig. 7). Thus, the
structure of ataxin-3 is predicted to be distinct from the
finger–palm–thumb architecture of U BPs such as USP7/
HAUSP [156]. Because of the low degree of conservation in
the central part, w e believe that ataxin-3 and josephin 1
adopt slightly different structures in this part, which are not
very sim ilar t o YUH1. In addition, we observed t hat the
Josephin domain also resembles the O TU domain because
both have a h ighly conserved histidine three residues
downstream of t he catalytic c ysteine. Interestingly, like
ataxin-3, the deubiquitinating OTU domain protein
VCIP135 interacts with the N-terminal adaptor domain of
VCP through the C-terminal tail including a UBL domain
and d issociates p47 f rom the complex with V CP during
ATP hydrolysis o f VCP [157,158]. This observation also
indicates a close functional relationship of the homologous
ubiquitin-like UBL and UBX domains.
Functional analysis of the Josephin domain
The active site of UCHs is divided into two parts as
follows (YUH1/ataxin-3 r esidue numbers) [153,154]: The
3162 M. Albrecht et al. (Eur. J. Biochem. 271) Ó FEBS 2004
N-terminal part consists of a glutamine (Q84/Q9) upstream
of a cystein e (C90/C14), both of which form an oxyanion
hole to accommodate the n egative charge on the substrate
carbonyl oxygen during catalysis. The C-terminal part
contains a histidine (H166/H119), which is thought to be
deprotonated, and an asparagine or aspartate ( D181/N134),
both o f which activate the side chain of the cysteine to
unleash a nucleophilic attack on the carbonyl carbon atom

of the scissile peptide bond. The cysteine, histidine, and
asparagine/aspartate const itute the catalytic triad c harac-
teristic of cysteine proteases such as papain.
While all four discussed catalytic residues are strictly
conserved in the Josephin domain (Fig. 6), a function-
ally relevant d isordered loop (E144–N164/V79–Q100)
Fig. 6. Structure-based multiple sequence alignment of the Josephin domains of ataxin-3 homologues with the crystallographically determined UCH
domains of hum an UCH -L3 and yeast Y UH1. The known DSSP secondary structure assignments of UCH -L3 a nd Y UH1 a re sh own a t t he t o p o f
the alignment (curled lines for a-helix, arrows for b-strands). The corresponding consensus secondary structure predictions for h uman ataxin-3 and
josephin 1 are also depicted. Alignment columns with id en tical residues are highlighted in purple-coloure d boxes, those with m ore than 50%
physico-chemically similar amino acids i n yellow box es (bold-printed letters). Text labels (including U CH-L3/YUH1 and ataxin -3/josephin 1
residue numbers) point to catalytic residues (four grey-shaded b o xes) and to other highly c onserved amino acids in the Josephin d omain. The PDB/
SPTrEMBL identifiers of UCH-L3 and Y UH1 a re 1uch/P15374 and 1cmxA/P35127, respectively. NCBI or Ensembl accession numbers for
Josephin domain homolo gues are given in Table S3.
Ó FEBS 2004 Analysis of ataxins 2 and 3 (Eur. J. Biochem. 271) 3163
positioned over the catalytic cleft is aligned in the less
conserved central part. This loop maintains an inaccessible
active s ite, but becomes o rdered upon binding of Ubal [154].
Therefore, it may control substrate specificity together with
further s trongly conserved amino acids s uch as N88/L13,
which forms hydrogen bonds with main chain groups of the
loop, and Y167/W120 next to the catalytic histidine [154].
Unfortunately, the structure of the central part and the
loop function remains unclear for the Josephin domain
because of insufficien t sequence similarity to UCHs. The
Josephin domain is also missing the N-terminal extension s
of UCHs, which are involved in substrate recognition [154].
In addition, a functional relevance of a second strictly
conserved histidine H17, two highly c onserved asparagines
N20 and N21, and another identical glutamine Q24

downstream of the catalytic cysteine C14 cannot be derived
either from the s tructural model o f the Jos ephin domain
(Figs 6 and 7 ). However, considering t heir distance from the
active s ite and location i nside the protein, they may be solely
important for the stability of t he domain fold. This may a lso
hold true for the strictly conserved S135 and P140 after the
catalytically active N134. In contrast, it i s easy to interpret
an alternative splice variant of ataxin-3 [10], which consists
of a deletion of the residues from E10 to Q64 including
the catalytic cysteine and thus cannot possess proteolytic
activity.
Fig. 6. (Continued).
3164 M. Albrecht et al. (Eur. J. Biochem. 271) Ó FEBS 2004
Comparison to other polyQ proteins
The polyQ stretch of both a taxin-2 a nd ataxin-3 lies in
sequence regions whose degree of conservation is very low
in contrast to the globular domains and which are
predicted to be in trinsically unstructured. This pred iction
has been confirmed experimentally for a taxin-3 [72], and
polyQ tracts themselves also adopt a random coil
conformation [159]. So we decided to investigate other
polyglutamine disease proteins such as ataxin-1 (SCA1),
ataxin-7 (SCA7), atrophin (DRPLA) and huntingtin (HD)
as to whether their polyQ regions are also p redicted to be
surrounded by d isordered structure. F or this purpose, we
used several online p rediction servers (Dis EMBL, DISO-
PRED, GlobPlot, NORSp, PONDR), which consensus
basically indicates that the polyQ tracts are generally
located in unstructured regions (Table S4). This is also in
agreement with secondary structure prediction r esults,

which do not indicate globular domains consisting of
a-he lices or b-strands (data not shown), and other
computational predictions of locally unfolded regions
[160]. T herefore, i t i s not surprising that mutant polyglu-
tamine proteins can readily form aggregates via the
solvent-exposed polyQ region.
Conclusions
We presented a detailed analysis of ataxin-2 h omologues
including the yeast homologue Pbp1, using a structure-
based multiple s equence alignment of Sm a nd Sm-like
proteins and a 3D model of t he Lsm domain of ataxin-2.
Our comparison revealed a high degree of conservation of
chemical properties for RNA-binding residues in t he aligned
Lsm domains in general and between Sm1 from P. abyssi
and human ataxin-2 in particular. Based on this observa-
tion, we propose t hat a taxin-2 i s capable of binding RNA
by the identified residues. Therefore, an essential f unction of
ataxin-2 homologues in RNA processing should b e
explored experimentally and could implicate the regulation
of polyadenylation o f mRNA a s it i s known for Pbp1. I n
addition, the similarity of amino acids involved in the
formation o f Lsm domain oligomers a s derived from the
D
1
–D
2
and D
3
–B heterodimers may suggest that ataxin-2
may also form such complexes.

Our structural model of the Josephin domain of ataxin-3
confirms the e volutionary relationship with deubiquitinat-
ing cysteine proteases of the UCH family. Interestingly, this
relates ataxin-3 to another ubiquitin hydrolase termed
USP14, which is involved in synaptic dysfunction in ataxic
Fig. 7. 3D model o f t he deubiquitinating Josephin d omain o f ataxin-3 u sing th e structure of yea st YUH1 b ound to the ubiquitin-like inhibitor Ubal (in
CPK view mode) as t emplate (PDB i dentifier 1cmx, chains A an d B, r espectively). Grey-shaded t ext l abels i ndicate t he f our catalytic residues (ball-
and-stick view) forming the active site of the ubiquitin hydrolase. The remaining text boxes point tootherresidues,whicharehighlyconservedinthe
Josephin domain. Residues a re coloured in agreement with t he alignment c olum ns in Fig. 6. Th e N-terminal e xtension of Y UH1, which i s missing
in ataxin-3 homologues, is depicted in the background as thin dark brown protein backbone only. The less conserved central part of ataxin-3 is
shown in green; it could not be modelled reliably using YUH1 as template beca use of low sequence similarity.
Ó FEBS 2004 Analysis of ataxins 2 and 3 (Eur. J. Biochem. 271) 3165
mice [161,162]. Moreover, t he polyglutamine disease protein
ataxin-1 interacts with the ubiquitin-specific protease USP7/
HAUSP, and the length of the polyQ region influences the
strength of the interaction [163]. Unfortunately, the central
part of the Josephin domain is difficult to model because of
low sequence similarity. Therefore, i t cannot be deduced
whether the ataxin-3 mechanism of ubiqu itin recognition
works similarly to UCHs.
It is striking th at both human ataxin-3 and its P. falci-
parum homologue ataxin-3_Pf can bind the N-terminal
adaptor domain of t he molecular chaperone VCP at their
C-termini, although they differ considerably and a n evolu-
tionary relationship is not apparent: ataxin-3 contains the
polyQ region, but the P. falciparum homologue has a
ubiquitin-like UBX domain. Another open question is
whether a deubiquitinating analogue of ataxin-3 exists in
yeast, since the Josephin domain is not found in any y east
protein, but complexes of the ataxin-3 and proteasome

binding proteins hHR23A/B, VCP and U FD2a are con-
served in yeast. Generally, it r emains to be s een how the
normal f unctions of ataxin-2 and ataxin-3 a re affected in
mutant proteins with an expanded polyglutamine tract.
Acknowledgements
Part of this research was fu nded b y t he G erman Research Foundation
(DFG) under contract no. LE 491/14–1, by the F ederal Ministry of
Education and Research (BMBF) under contract no. 01gs0115-NV-
S02T12, and by the European Commission through the EUROSCA
project under contract no. LSHM-CT-2004-503304.
References
1. Ka wagu chi, Y., O kamoto, T., Taniwaki, M., Aizawa, M.,
Inoue, M., K atayama, S., Kawakami, H., N akamura, S.,
Nishimura, M., Akiguchi, I. et al. (1994) CAG expansions in a
novel gene for Mach ad o-Joseph disease at chromosome 14q32.1.
Nat. Genet. 8, 221–228.
2. Pulst, S.M., N echiporuk, A., Nechiporuk, T., Gispert, S., Chen,
X.N., Lopes-Cendes, I., Pearlman, S., S tark man, S., O rozco -
Diaz, G ., Lunkes, A ., DeJong, P ., Rouleau, G.A ., Auburger, G.,
Korenberg, J.R., F igueroa, C. & Sahba, S . (1996) Moderate
expansion o f a normally biallelic trinucleotide repeat i n spino-
cerebellar ataxia type 2. Nat. Genet. 14, 269–276.
3. Evert, B.O., W u
¨
llner, U. & Klockgether, T. (2000) Cell de ath in
polyglutamine diseases. Cell Tissue Res. 301, 189–204.
4. Zoghbi, H.Y. & Orr, H.T. (2000) Glutamine repeats and neu-
rodegeneration. Annu.Rev.Neurosci.23, 217–247.
5. Margolis, R.L. ( 2002) The spinocerebellar ataxias: order e merges
from chaos. Curr.Neurol.Neurosci.Report2, 447–456.

6. Stevanin, G., Du
¨
rr, A. & Brice, A. ( 2002) Spinocerebellar a taxias
caused by polyglutamine expansions. Adv. Exp. Med. Biol. 516,
47–77.
7. Wu
¨
llner, U. (2003) Genes implicated in the pathogenesis of
spinocerebellar ataxias. Drugs Today (Barc) 39, 927–937.
8. Gusella, J.F. & MacDonald, M.E. (2000) Molecular genetics:
unmasking polyglutamine triggers in neurodegenerative disease.
Nat. Rev. Neurosci. 1, 109–115.
9. Sahb a, S., Nec hiporuk, A., F igueroa, K.P., N echipo ruk, T. &
Pulst, S.M. (1998) Ge nom ic structure of the human gene for
spinocerebellar ataxia type 2 (SCA2) on chromosome 12q24.1.
Genomics 47, 359–364.
10. Ichikawa,Y.,Goto,J.,Hattori,M.,Toyoda,A.,Ishii,K.,Jeong,
S.Y.,Hashida,H.,Masuda,N.,Ogata,K.,Kasai,F.,Hirai,M.,
Maciel, P., Rouleau, G.A., Sakaki, Y. & Kanazawa, I. (2001)
The g enomic structure a nd expression of MJD, the Machado-
Joseph disease gene. J. Hum. Genet. 46, 413–422.
11. Huynh, D.P., Yang, H.T., Vakharia, H., Nguyen, D. & P ulst,
S.M. (2003) Expansion of the polyQ repeat in ataxin-2 alters its
Golgi localization, disrupts the Golgi complex and c auses cell
death. Hum. Mol. Genet. 12, 1485–1496.
12. Trott ier, Y ., Canc el, G., An-Gourfinkel, I ., Lutz, Y., Weber, C.,
Brice, A., Hirsch, E. & Mandel, J.L. (1998) H eterogeneous
intracellular localization and expression of ataxin-3. N eurobiol.
Dis. 5, 335–347.
13. Wheeler, D.L., Church, D.M.,Edgar,R.,Federhen,S.,

Helmberg, W., Madden, T.L., Pontius, J.U., Schuler, G.D.,
Schriml, L.M., Sequeira, E. et al. (2004) Database resources of
the National Center for Biotechnology – information update.
Nucleic Acids Res. 32 (Database issue), D35–40.
14. Birney, E., A ndrews, D., B evan, P., Cac camo, M., C ameron, G.,
Chen, Y., Clarke, L., Coates, G., Cox, T., Cuff, J. et al. (2004)
Ensembl. Nuceic Acids Res. 32 (Database issue), D468–470.
15. Apwe iler, R., B airoch, A., Wu, C.H ., Barker, W.C.,
Boeckmann, B., Ferro, S., Gasteiger, E., Huang, H., Lopez, R.,
Magrane, M. et al. (2004) UniProt: the Universal Protein
knowledgebase. Nucle ic Acids Res. 32 (Database issue), D115–
119.
16. Batem an, A., C oin, L., D urbin , R., F inn, R.D., H o llich, V.,
Griffiths-Jones, S., Khanna, A., Marshall, M., Moxon, S.,
Sonnhammer, E.L, et al. (2004) The Pfam protein families
database. Nucleic Acids Res. 32 (Database issue), D138–141.
17. Andre eva, A., Howorth, D., Bren ner, S.E., Hu bbard, T.J.,
Chothia, C. & Murzin, A.G. (2004) (2004) SCOP database in:
refinements integrate structure and sequence family data. Nucleic
Acids Res. 32 (Database issue), D226–229.
18. Bo urne, P.E., Ad dess, K.J.,Bluhm,W.F.,Chen,L.,
Deshpande, N., Feng, Z., F leri, W., Green, R., Merino-Ott, J.C.,
Townsend-Merino, W., et al. (2004) The distribution and query
systems of the RCSB Protein Data Bank. Nucleic Acids Res. 32
(Database issue), D223–225.
19. Kabsc h, W. & Sander, C. (1983) Dic tionary of protein secondary
structure: patte rn recognition of hydrogen-bonded a nd geo-
metrical features. Biopolymers 22, 2577–2637.
20. Alt schul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J.,
Zhang, Z., Miller, W. & Lipman, D.J. (1997) Gapped BLAST

and PSI-BLAST: a new generation of p rotein database search
programs. Nuceic. Acids Res. 25, 3389–402.
21. McGuffin,L.J.,Bryson,K.&Jones,D.T.(2000)ThePSIPRED
protein structure pred iction server. Bioinformatics 16, 404–
405.
22. Karplus, K., Barrett, C. & Hughey, R. (1998) Hidden Markov
models for detecting remote protein homologies. Bioinformatics
14, 846–856.
23. Pollastri, G., Przybylski, D., Rost, B . & Baldi, P. (2002)
Improving th e p rediction of protein s ec ondary struc ture i n three
and eight classes using recurrent ne ural networks and profiles.
Proteins 47, 228–235.
24. Albrecht, M., Tosatto, S.C., Lengauer, T. & Valle, G. (2003)
Simple consensus procedures are effective and sufficient in sec-
ondary structure prediction. Protein Eng. 16, 459–462.
25.Linding,R.,Jensen,L.J.,Diella,F.,Bork,P.,Gibson,T.J.&
Russell, R.B. (2003) Protein d isorder prediction: implications for
structural proteomics. Structure (Camb) 11, 1453–1459.
26. Ward, J.J.,.McGuffin, L.J., Bryson, K., Buxton, B.F. & J ones,
D.T. (200 4) The D ISOPRED server for the prediction o f protein
disorder. Bioinformatics (in press) .
27. Linding, R., Russell, R.B., Neduva, V. & Gibson, T.J. (2003)
GlobPlot: Exploring protein sequences for globularity and d is-
order. Nucleic Acids Res. 31, 3701–3708.
3166 M. Albrecht et al. (Eur. J. Biochem. 271) Ó FEBS 2004
28. Liu, J. & Rost, B . (2003) NORSp: Pred ictions of long r egions
without regular secondary structure. Nucl. Acids Res. 31, 3833–
3835.
29. Nakai, K. & H orton, P. (1999) PSOR T: a program for detecting
sorting signals in proteins and predicting their subcellular loca-

lization. Trends Biochem. Sci. 24, 34–36.
30. Poirot, O., O’Toole, E. & Notredame, C. (2003) Tcoffee@igs: a
web server f or c omput ing, evaluating and combining multip le
sequence alignment s. Nucleic Acids Res. 31, 3503–3506.
31. Shindyalov, I.N. & Bourne, P.E. (1998) Protein structure align-
ment by incremen tal combinatorial e xtension (CE) o f the op ti-
mal path. Protein Eng. 11, 739–747.
32. Bujnicki, J.M., Elofsson, A., Fischer, D. & R ychlewski, L. ( 2001)
Structure prediction meta server. Bioinformatics 17, 750–1.
33. Ginalski,K.&Rychlewski,L.(20 03) Detection of reliable and
unexpected protein fold predictions usi ng 3D-Jury. N ucleic Acids
Res. 31, 3291–3292.
34. Ro driguez, R., Chine a, G., Lop ez, N., Pon s, T. & Vriend, G.
(1998) Ho mology modeling, model and software e valuation:
three related resources. Bioinformatics 14, 523–528.
35. Galtier, N., G ouy, M. & Gautier, C. (1996) SEAVIEW a nd
PHYLO_WIN. two graphic t ools for se quence alignment and
molecular p hy log eny. Comput. Appl. Biosci. 12, 543–548.
36. Gouet, P., Robert, X. & Courcelle, E. (2003) ESPript/ENDscript.
extracting an d rendering se qu ence and 3D infor mation from
atomic structures of proteins. Nucleic Acids Res. 31, 3320–3323.
37. Ne uwald, A.F . & Koonin, E.V. (1998) Ataxin-2, g lobal reg-
ulators of bacterial gene expression, and spliceosomal snRNP
proteins share a conserved domain. J. Mol Med. 76,3–5.
38. Sh ibata, H., Huynh, D.P. & Pulst, S .M. (2000) A novel protein
with RNA-binding motifs i nteracts with ataxin-2. Hum. Mol.
Genet. 9, 1303–1313.
39. Albrecht, M . & Lengauer, T. ( 2004) Surve y o n the PABC
recognition motif PAM2. Bioc hem. Biophys. R es. Commun. 316,
129–138.

40. Dreyfuss, G., Matunis, M.J., Pinol-Roma, S. & Burd, C.G.
(1993) hnRNP proteins and the biogenesis o f mRNA. Annu. Rev.
Biochem. 62, 289–321.
41. He, W. & Parker, R. (2000) Functions o f Lsm protei ns in
mRNA de gradatio n a nd splicing. C urr. Opin. Cell Biol. 12,346–
350.
42. Mura, C., Phillips, M., Kozhukhovsky, A. & Eisenberg, D.
(2003) Structure and assembly of an augmented Sm-like arc h-
aeal protein 14-mer . Proc. Natl Acad. Sci. USA 100 , 4539–
4544.
43. Sauter, C ., Basquin, J. & Suck, D . (2003) S m-like proteins in
Eubacteria: the crystal structure of the Hfq protein from
Escherichia coli. Nucleic Acids Res. 31, 4091–4098.
44. Bessman, M.J., Frick, D .N. & O’Handley, S.F. (1996) The MutT
proteins or ÔNudixÕ hydrolases, a family of versatile, widely dis-
tributed, ÔhousecleaningÕ enzyme s. J. Biol. Chem. 271, 25059–
25062.
45. Piccirillo, C., Khanna, R. & Kiledjian, M. (2003 ) Functional
characterization of the mammalian m RNA decapping enzyme
hDcp2. RNA 9, 1138–1147.
46. Sh e, M., D ecker, C.J., Sundramurthy, K., Liu, Y., C hen, N.,
Parker, R. & Song, H. (2004) Crystal s tructure of Dc p1p a nd its
functional implications in mRNA decapping. Nat. Struct. Mol.
Biol. 11, 249–256.
47. Kie hl, T.R., Shibata, H. & Pulst, S.M. (2000) The ortholog of
human ataxin-2 is e ssential for e arly embryonic p atterning in
C. elegans. J. Mol. Neurosci. 15, 231–241.
48. Meunier, C., Bordereaux, D., Porteu, F., G isselbrecht, S.,
Chretien,S.&Courtois,G.(2002) Cloning and c haracterization
of a family of proteins associated with Mpl. J. Biol. Chem. 277,

9139–9147.
49. Satterfield, T.F., Jackson, S.M. & Pallanck, L.J. (2002) A
Drosophila homolog of the polyglutamine disease gene SCA2 is a
dosage-sensitive regulator of actin filament formation. Genetics
162, 1687–1702.
50. Figueroa, K.P. & Pulst, S.M. (2003) Identification and e xpres-
sion of the gene for human ataxin-2-related protein on chro-
mosome 16. Ex p Neurol. 184, 669–678.
51. Jin , Y., Suzuki, H., Maegawa, S., Endo, H., Sugano, S., Hashi-
moto,K.,Yasuda,K.&Inoue,K. (2003) A vertebrate RNA-
binding protein Fox-1 regulates tissue-specific splicing via the
pentanucleotide GCAUG. EMBO J. 22, 905–912.
52. Bhalla, K., Phillips, H.A., Crawford, J., McKenzie, O.L., Mul-
ley, J.C., Eyre, H., Gardner, A.E., Kremmidiotis, G. & Callen,
D.F. (2004) The de novo chromoso me 16 transloc atio ns of two
patients with abnormal phenotypes ( mental retardation and
epilepsy) disrupt the A2BP1 gene, J. Hum. Genet. 49, 308–311.
53. Mangus, D.A., Amrani, N. & Jacobson, A. (1998) Pbp1p, a
factor interacting with Saccharomyces ce revisiae poly(A)-binding
protein, regulates polyade nylation. Mol. Cell Biol. 18 , 7383–
7396.
54. Achsel, T., Brahms, H ., Kastner, B., Bachi, A ., Wilm, M. &
Lu
¨
hrmann, R. (1999) A doughnut-shaped heteromer of human
Sm-like proteins binds to th e 3 ¢-end of U6 snRNA, thereby
facilitating U4/U6 duplex formation in vitro. EMBO J. 18, 5789–
5802.
55. Kambach, C., Walke, S., Young, R., Avis, J.M., de la Fortelle,
E.,Raker,V.A.,Lu

¨
hrmann,R.,Li,J.&Nagai,K.(1999)Crystal
structures of two S m protein complexes and their implications
for the assembly of the spliceosomal snRNPs. Cell 96, 375–387.
56. Achse l, T ., Stark, H. & Lu
¨
hrmann, R. (2001) The Sm domain is
an ancient RNA-binding motif with o ligo(U) specificity. Proc.
NatlAcad.Sci.USA98, 3685–3689.
57. Collins, B.M., Harrop, S.J., Kornfeld, G.D., Dawes, I.W.,
Curmi, P.M. & M abbutt, B.C. (2001) Crystal structure of a
heptameric Sm-like protein complex from archaea: implications
for the structure and evolution of snRNPs. J. Mol. Biol. 309,
915–923.
58. Mura, C., Cascio, D., Sawaya, M.R. & Eisenberg, D.S. (2001)
The crystal structure of a heptameric archaeal Sm protein: Im-
plications for the eukaryotic snRNP core. Proc. Natl Acad. Sci.
USA 98, 5532–5537.
59. To
¨
ro
¨
, I., Thore, S., Mayer, C ., Basquin, J., S e
´
raphin,B.&
Suck, D. (2001) RNA binding in an Sm core domain: X-ray
structure and functional analysis of an archaeal Sm protein
complex . EMBO J. 20, 2293–2303.
60. Urlaub, H., R aker, V.A., Kostka, S. & Lu
¨

hrmann, R . (2001) Sm
protein–Sm site RN A interactions within th e inner ring of the
spliceosomal snRNP core structure. EMBO J. 20, 187–196.
61. Sc humacher, M.A., Pearson, R.F.,Møller,T.,Valentin-
Hansen, P. & Brennan, R.G. (2002) Structures of the pleiotropic
translational regulator Hfq and an Hfq-RNA complex: a bac-
terial Sm-like protein. EMBO J. 21 , 3546–3556.
62. To
¨
ro
¨
, I., Basquin, J., Teo-Dreher, H. & Suck, D. (2002) Archaeal
Sm proteins form heptameric and h exameric complexes: crystal
structures of the Sm1 and Sm2 proteins from the hyperthermo-
phile Archaeoglobus fulgidus. J. Mol. Biol. 320, 129–142.
63. Collins, B.M., Cubeddu, L., Naidoo, N., Harrop, S.J., Kornfeld,
G.D., Dawes, I.W., Curmi, P.M. & Mabbutt, B.C. (2003)
Homomeric ring assemblies o f eukaryotic Sm proteins have
affinity for both RNA and DNA. Crystal structure of an
oligomeric complex of yeast SmF. J. Biol. Chem. 278, 17291–
17298.
64. Mura, C., Kozhukhovsky, A., Gingery, M., Phillips, M. &
Eisenberg, D. (2 003) T he ol igomerization a nd ligand-binding
properties of Sm -like archaeal p roteins (SmAPs). P rotein Sci. 12,
832–847.
Ó FEBS 2004 Analysis of ataxins 2 and 3 (Eur. J. Biochem. 271) 3167
65. Th ore, S., Mayer , C., Sauter, C., Week s, S. & S uck, D. (2003)
Crystal structures of t he Pyrococcus abyssi Sm core and its
complex with RNA. C ommon f eatures of RNA b inding in
archaea and eukarya. J. Biol. Chem. 278, 1239–1247.

66. Hermann, H., Fabrizio, P., Raker, V.A., Foulaki, K., Hornig, H.,
Brahms,H.&Lu
¨
hrmann, R. (1995) snRNP Sm proteins s hare
two evolutionarily co nserved sequence motifs which are i nvolved
in Sm protein–protein interactions. EM BO J. 14, 2076–2088.
67. Seraphin, B. (1995) Sm and Sm-like proteins b elong to a large
family: identification o f proteins of the U6 as well as t he U1, U2,
U4 and U5 snRNPs. EMBO J. 14, 2089–2098.
68. Burnett, B., Li, F. & Pittman, R .N. (2003) The polyglutamine
neurodegenerative protein ataxin-3 binds polyubiquitylated
proteins and has ubiquitin protease activity. Hum. Mol. Genet.
12, 3195–3205.
69. Scheel, H., Tomiuk, S . & Hofmann, K. (2003) Elucid ation of
ataxin-3 and ataxin-7 function by integrative bioinformatics.
Hum. Mol. Genet. 12, 2845–2852.
70. Hofmann, K. & F alquet, L . (2001) A u biquitin-interacting motif
conserved in components of the proteasomal and lysosomal
protein degradation systems. Trend s Biochem . Sci. 26, 347–350.
71. Suzuki, T., Park, H ., Till, E. A. & L ennarz, W.J. (2001) The PUB
domain: a putative protein–protein interaction domain implica-
ted in the ubiquitin-proteasome pathway. Biochem. Bioph ys. Res.
Commun. 287, 1083–1087.
72. Masino, L., Musi , V., Menon, R.P., Fusi, P., K elly, G., Frenkiel,
T.A., Trottier, Y. & Pastore, A. (2003) Domain architecture of
the polyglutamine protein ataxin-3: a globular domain followed
by a flexible tail. FEBS Lett. 549, 21–25.
73. Albrecht, M., Hoffmann, D., Evert, B.O., Schmitt, I., Wu
¨
llner,

U. & L engauer, T. (2003) Structural m odeling of ata xin-3 reveals
distant homology to adaptins. Proteins 50, 355–370.
74. Tait, D ., Ricci o, M ., Sittler, A., Scherzinger, E., S anti, S .,
Ognibene,A.,Maraldi,N.M.,Lehrach,H.&Wanker,E.E.
(1998) Ataxin-3 is transported into the nucleus and associates
with the nuclear matrix. Hum. Mol. Genet. 7, 991–997.
75. la Co ur, T ., G upta, R., Rapacki, K., Sk river, K ., P oulsen, F.M.
& Brunak, S. (2003) N ESbase, Version 1.0. a database o f nuclear
export signals. Nucleic Acids R es. 31, 393–396.
76. Li, F., Macfarlan, T., Pittman, R.N. & Chakravarti, D. (2002)
Ataxin-3 is a histone-binding protein with two independent
transcriptional corepressor activities. J. Biol. Chem. 277, 45004–
45012.
77. Bodai, L., P allos, J., T hompson, L.M. & Marsh, J.L. (2003)
Alteredproteinacetylationinpolyglutaminediseases.Curr. Med.
Chem. 10, 2577–2587.
78. Zhang, Y. (2003) Transcriptional regulation by histone ubiqui-
tination and deubiquitination. Genes Dev. 17, 2733–2740.
79. Flore ns, L., Washburn, M.P., R aine, J.D., An thony, R.M.,
Grainger, M ., Haynes, J.D., Moc h , J.K., Muster, N., Sacci, J.B.,
Tabb, D.L., et al. (2002) A proteomic view of the Plasmodium
falciparum life cycle. Nature 419, 520–526.
80. Bozdech, Z., Llinas, M., Pulliam, B.L., Wong, E.D., Zhu, J. &
DeRisi, J.L. (2003) The transc riptome of the intraerythroc ytic
developmental cycle of Plasmodium falciparum. PLoS Biol.
1,E5.
81. Le Roch, K.G., Zhou, Y., Blair, P.L., Grai nger, M., Moch, J.K.,
Haynes, J.D., De La Vega, P., Holder, A.A., Batalov, S.,
Carucci, D.J. & Winzeler, E.A . (2003) Discov ery of gene fun ction
by expression profiling of the malaria parasite life cycle. Science

301, 1503–1508.
82. Buchberger, A., Howard, M.J., Proctor, M. & Bycroft, M. (2001)
The UBX domain: a widespread ubiquitin-like module. J. Mol.
Biol. 307, 17–24.
83. Yuan, X., Shaw, A., Zhang, X., Kondo, H ., Lally, J., Freemont,
P.S. & Matthews, S. (2 001) Solution structure and interaction
surface of the C-terminal domain from p47: a major p97-cofactor
involved in SNARE disassembly. J. Mol. Biol. 311, 255–263.
84. Buchberger, A. ( 2002) From UBA to UBX: new words in the
ubiquitin vocabulary. Trends Cell Biol. 12, 216–221.
85. Yuan,X.,Simpson,P.,Kondo,H.,McKeown,C.,Dreveny,I.,
Zhang, X., Freemont, P.S. & M atthews, S. (2004) Com plet e
Backbone Resonance Assignments of p47: The 41kDa Adaptor
Protein of the AAA ATPase p97. J. Biomol N MR 28, 309–310.
86. Dreveny,I.,Kondo,H.,Uchiyama,K.,Shaw,A.,Zhang,X.&
Freemont, P.S. ( 2004) Structural basis of the interaction between
the A AA ATPase p97/VCP and i ts adaptor p rotein p47. EMBO
J. 23, 1030–1039.
87. Aguilar, R.C. & W endland, B. (2003) Ubiquitin: not just for
proteasomes anymore. Curr.Opin.CellBiol.15, 184–190.
88. Bach, I. & Ostendorff, H.P. (2003) Orchestrating nuclear
functions: ubiquitin s ets t he rh ythm. Trends Biochem. Sci. 28,
189–195.
89. Baek, K.H. (2003) Conjugation and deconjugation of ubiquitin
regulating the destiny of proteins. Exp. Mol Med. 35,1–7.
90. Berke, S.J. & Paulson, H.L. (2003) Protein aggregation and the
ubiquitin proteasome pathway: gaining the U PPer hand o n
neurodegeneration. Curr. Opin. Genet Dev. 13, 253–261.
91. Di Fiore, P.P., Polo, S. & Hofmann, K. (2003) When ubiquitin
meets ubiquitin receptors: a signalling co nnection. Nat. Rev. Mol.

Cell Biol. 4, 491–497.
92. Hicke, L. & Dunn, R. (2003) Regulation of membrane protein
transport by ubiquitin and ubiquitin-bind ing proteins. Annu.
Rev. Cell Dev. Biol. 19, 141–172.
93. Passm ore, L .A. & Barford, D. (2004) G etting in to p osition: T he
catalytic m echanisms o f protein ubiquitylation. Bioche m. J. 379,
513–525.
94. Chai, Y., Berke, S.S., Cohen, R.E. & Paulson, H.L. (2003) Poly-
ubiquitin binding by the polyglut amine disease protein ataxin-3
links its norm al function to p rotein surveillance pathways.
J. Biol. Chem. 279, 3605–3611.
95. Donaldson, K.M., Li, W., Ching, K.A., Batalov, S., Tsai, C.C. &
Joazeiro, C.A. (2003) Ubiquitin-mediated sequestration of nor-
mal cellular proteins into polyglutamine aggregates. Proc. Natl
Acad. Sci. USA 100, 8892–8897.
96. Do ss-Pepe, E.W., Stenroos, E.S., J ohn son, W.G . & Madura, K.
(2003) Ataxin)3 interaction s with rad23 and valosin-containing
protein and its associations with ubiquitin chains and the
proteasome are c onsiste nt with a role in ubiquitin-mediated
proteolysis. Mol. Cell Biol. 23, 6469–6483.
97. Miller, S.L.,.Malotky, E . & O’Brya n, J.P. (2004) Analysis of t he
role of ubiquitin-interacting motifs (UIMs) in ubiquitin binding
and ubiquitylation. J. Biol. Chem. (in press).
98. Fisher, R.D., Wang, B ., Alam, S.L., Higginson, D.S.,
Robinson, H ., Sundquist, W .I. & Hill, C.P. (2003) Structure a nd
ubiquitin binding of the ubiquitin-interacting motif. J. Biol.
Chem. 278, 28976–28984.
99. Mueller, T.D. & Feigon, J. (200 3) Structural determinants for the
binding o f ubiquitin-like domains to the proteasome. EMBO
J. 22, 4634–4645.

100. Swanson, K.A., Kang, R.S., S tamenova, S.D., H icke, L. &
Radhakrishnan, I. (2003) Solution structure of Vps27 UIM-
ubiquitin complex important f or endosomal sorting and receptor
downregulation. EMBO J. 22, 4597–4606.
101. Evans, P.R. & Owen, D.J. (2002) Endocytosis and vesicle traf-
ficking. Curr. Opin. Struct. Biol. 12, 814–821.
102. Collins, B.M., Watson, P.J. & Owen, D.J. (2003 ) The structure of
the G GA1-GAT d omain reveals th e molecular basis for ARF
binding and me mbran e asso ciation o f GGAs. Dev. Cell. 4, 321–
332.
103. Shiba, T., Kawasaki, M., Takatsu, H., Nogi, T ., Mats ugaki, N .,
Igarashi,N.,Suzuki,M.,Kato,R.,Nakayama,K.&
3168 M. Albrecht et al. (Eur. J. Biochem. 271) Ó FEBS 2004
Wakatsuki, S . (2003) Molecular mechanism of membrane
recruitment of GG A by ARF in lysosomal protein tran sport.
Nat. Struct. Biol. 10, 386–393.
104. Suer, S., Misr a, S., S aidi, L.F. & Hurley, J .H. (2003) S tructure of
the GAT domain of human GGA1: a syntaxin amino-terminal
domain fold in an endosomal t raffi cking adaptor, Proc. Natl
Acad. Sci. USA 100, 4451–4456.
105. Zh u, G., Zhai, P., He, X., Terzyan, S., Zhang, R., Joachimiak,
A., Tang, J. & Zhang, X.C. (2003) Crystal structure of the hum an
GGA1 GAT domain. Biochemistry 42, 6392–6399.
106. Ka toh , Y., Shiba, Y., Mitsuhashi, H., Yanagida, Y., Takatsu, H.
& Nakayama, K. (2004) Tollip and Tom1 form a complex and
recruit ubiquitin-conjugated proteins onto early e ndosomes.
J. Biol. Chem. 279, 24435–24443.
107. Mattera, R., Puertollano, R ., Smith, W.J. & Bonifacino, J.S.
(2004) T he tri-helical bund le subdomain of the GGA proteins
interacts w ith multiple p artners through ove rlapping but distinc t

sites. J. Biol. Chem. 14, R357–359.
108. Pelham, H.R. (2004) Membrane traffic: G GAs sort ubiquitin.
Curr. Biol. 14, R357–R359.
109. Mueller, T.D., Kamionka, M. & Feigon, J. (2004) Specificity of
the interaction between UBA domains and ubiquitin. J. Biol.
Chem. 279, 11926–11936.
110. Wang,G.,Sawai,N.,Kotliarova,S.,Kanazawa,I.&Nukina,N.
(2000) Ataxin-3, the MJD 1 ge ne prod uct, in teracts with t he two
human homologs of yeast DNA repair protein RAD23,
HHR23A and HHR23B. Hum. Mol. Genet. 9, 1795–1803.
111. Walters,K.J.,Lech,P.J.,Goh,A.M.,Wang,Q.&Howley,P.M.
(2003) DNA-repair protein hH R23a alters its protein structure
upon binding pro teasomal subu nit S5a. Proc. Natl Acad. Sci.
USA 100, 12694–12699.
112. Upadhya, S.C. & Hegde, A.N. (2003) A potential proteasome-
interacting motif within the u biquitin-like domain of p arkin and
other proteins. Trends Biochem. Sci. 28, 280–283.
113. Sweder,K.&Madura,K.(2002)Regulationofrepairbythe26S
proteasome. J. Biomed. Biotechnol. 2, 94–105.
114. Ryu, K.S., Lee, K.J., Bae, S.H., Kim, B.K., Kim, K.A. & C hoi,
B.S. (2003) Bindin g surface mapping of intra– and interdomain
interactions among h HR23B, ubi quitin , and po lyubiqu itin
binding site 2 of S5a. J. Biol. Chem. 278, 36621–36627.
115. Wang, Q., Goh, A.M., Howley, P.M. & Walters, K.J. (2003)
Ubiquitin recognition by t he DNA repair p rotein h HR23a.
Biochemistry 42, 13529–13535.
116. Elsasser, S., Chandler-Militello, D., Mueller, B., Hanna, J. &
Finley, D. (2004) Rad23 and Rpn10 serve as alternative ubiquitin
receptors for th e pro teasom e. J. Biol. Chem. 279, 26817–26822.
117. Lambertson, D., Chen, L. & Madur a, K. (2003) Investigating the

importance of proteasome interaction for Rad23 function. Curr.
Genet. 42, 199–208.
118. Rassi, S. & Pickart, C.M. (2003) Rad23 ubiquitin-associated
domains (UBA) inhibit 26 S proteasome-catalyzed proteolysis by
sequestering lysine 48-linked polyubiquitin chains. J. Biol. Chem.
278, 8951–8959.
119.Fujiwara,K.,Tenno,T.,Sugasawa,K.,Jee,J.G.,Ohki,I.,
Kojima, C., Tochio, H., Hiroaki, H. , Hanaoka, F. & Shirakawa,
M. (2004 ) Structure of the ubiquitin-interacting motif of S5a
bound to the ubiqu itin-like domain of H R23B. J. Biol. Chem.
279, 4760–4767.
120. Hirabayashi, M., Inoue, K., Tanaka, K., Nakadate, K.,
Ohsawa, Y., Kamei, Y., Popiel, A.H., Sinohara, A., Iwamatsu,
A., Kimura, Y., Uchiyama, Y., Hori, S. & Kakizuka, A. (2001)
VCP/p97 in abnormal protein aggregates, cytoplasmic vacuoles,
and cell death, phenotypes relevant to neurodegeneration. Cell
Death Differ. 8, 977–984.
121. Higashiyama, H., H irose, F., Y amaguchi, M., Inoue, Y.H.,
Fujikake, N., Matsukage, A. & Kakizuka, A. (2002) Identifica-
tion of ter94, Drosophila VCP, as a modulator of p olyglutamine -
induced neurodege neration. Cell Death Differ. 9, 264–273.
122. Kobayashi, T., Tanaka, K., Inoue, K. & Kakizuka, A . (2002)
Functional ATPase activity of p97/valo sin-containing protein
(VCP) is r equired for the quality control of endop lasmic reticu-
lum in neuronally differentiated mammalian PC12 cells. J. Biol.
Chem. 277, 47358–47365.
123. Kobayashi, T. & Kakizuka, A. (2003) Molecular a nalyses of
Machado-Joseph disease. Cytogenet. Genome Res. 100, 261 –
275.
124. DeLaBarre, B. & Brunger, A.T. (2003) Complete structure of

p97/valosin-con taining protein reveals communic atio n between
nucleotide domains . Nat. Struct. Biol. 10, 856–863.
125. Kimura, Y. & Kakizuka, A. (2003) Polyglutamine diseases and
molecular chapero nes. IUBMB Life. 55 , 337–345.
126. Woodman, P.G. (2003) p97, a p rotein coping with multiple
identities. J. Cell Sci. 116, 4283–4290.
127. Wang,Q.,Song,C.&Li,C.C.(2004)Molecularperspectiveson
p97-VCP: progress i n understanding its st ructu re and d ive rse
biological functions. J. Struct. Biol. 146, 44–57.
128. Hatakeyama, S. & Nakayama, K.I. (2003) Ubiquitylation as a
quality control system for intracellular proteins. J. Bioc hem.
(Tokyo) 134, 1–8.
129. Ko egl, M., Hoppe, T., Sc hlen ker, S., Ulrich, H.D., Maye r, T .U.
& Jentsch, S. (1999) A novel ubiquitination factor, E4, is involved
in multiubiquitin chain assembly. Cell 96 , 635–644.
130. Kane ko, C., Hatakeyama, S., Matsumoto, M., Yada, M.,
Nakayama, K. & Nakayama, K.I. (20 03) Characterization of the
mouse gene for the U-box-type ubiquitin ligase UFD2a. Bio-
chem. Biophys. Res. Commun. 300, 297–304.
131. Matsumoto, M., Yada, M., H atakeyama, S., Ishimoto, H.,
Tanimura, T., Tsuji, S., Kakizuka, A., Kitagawa, M. &
Nakayama, K.I. (2004) Molecular clearance of ataxin-3 is
regulated by a mammalian E4. EMBO J. 23, 659–669.
132. Kim, I., M i, K. & Rao, H. (2004) Multiple interactions of Rad23
suggest a mechanism fo r ubiquitylated substrate delivery
important in proteolysis. Mol. Biol. Cell. 15, 3357–3365.
133. Hartmann -Petersen, R., Wallace, M., Hofmann, K., Koch, G.,
Johnsen, A.H., Hendil, K .B. & Gordon, C. (2004) The U bx2 and
Ubx3 cofactors d irect Cdc48 activity to proteolytic and non-
proteolytic ub iquitin-de pendent processes. Curr. Biol. 14, 824–

828.
134. Yuan, X., Simpson, P., McKeown, C., Kondo, H., Uchiyama,
K.,Wallis,R.,Dreveny,I.,Keetch,C.,Zhang,X.,Robinson,C.,
Freemont, P. & Matthews, S. (2004) Structure, dynamics and
interactions of p47, a major adaptor of the AAA ATPase, p97.
EMBO J. 23, 1463–1473.
135. Watts, G.D., Wymer, J., K ovach, M.J., Mehta, S.G., Mumm, S.,
Darvish, D., Pestronk, A., Whyte, M.P. & Kimonis, V.E. ( 2004)
Inclusion body myopathy associated with Paget disease of bone
and frontotemporal dementia is caused by mutant valosin-
containing protein. Nat. Genet. 36, 377–381.
136. Yam anaka, K., Okubo, Y., Suzaki, T. & Ogura, T. (2004)
Analysis of the two p97/VCP/Cdc48p pro teins of Caeno rhabditis
elegans and their suppression of polyglutamine-induced protein
aggregation. J. Struct. Biol. 146, 242–250.
137. Gio t, L., Bader, J .S., Brouwer, C., Ch audh uri, A., Kuang, B.,
Li, Y., Hao, Y.L., Ooi, C.E., Godwin, B ., Vitols, E., et al. (2003)
A protein interaction map of Dro sophila melanogaster. Science
302, 1727–1736.
138. Castrillon, D.H., Go
¨
nczy, P., Alexander, S., Rawson, R., Eb er-
hart, C.G., Viswanathan, S., D iNardo, S. & Wasserman, S.A.
(1993) Toward a molecular genetic analysis of spermatogenesis
in Drosophila melanogaster: characterization of male-sterile
mutants generated by single P element mu tagene sis. Genetics 135,
489–505.
Ó FEBS 2004 Analysis of ataxins 2 and 3 (Eur. J. Biochem. 271) 3169
139. Spradling, A.C., Stern, D., Beaton, A., Rhem, E.J., Laverty, T.,
Mozden, N ., Mis ra, S. & R ubin, G.M. (1999) The Berkeley

Drosophila Genome Project gene disruption project: Single
P-element i nsertions m utating 25% o f vital D rosophila genes.
Genetics 153, 135–177.
140. Shan, Y.X., Yang, T.L., Mestril, R. & Wang, P.H. (2003) Hsp10
and Hsp60 suppress ubiquitination of in sulin-like growth f actor-
1 r eceptor and aug ment insulin-like gro wth factor-1 receptor
signaling i n cardiac m u scle: implications on decreased myo-
cardial protection i n diabetic cardiomyopathy. J. Bi ol. Chem.
278, 45492–45498.
141. Chai, Y., Koppe nhafer, S.L., B onini, N.M. & Paulson, H.L.
(1999) A nalysis of t he role of h eat shock protein ( Hsp) molecular
chaperones in polyglutamine disease. J. Neurosci. 19, 1 0338–
10347.
142.Chai,Y.,Koppenhafer,S.L.,Shoesmith,S.J.,Perez,M.K.&
Paulson, H .L. (1999) Evidence for proteasome i nvolvement in
polyglutamine disease: l ocalization to nuclear inclusions in
SCA3/MJD and suppression of polyglutamine aggregation
in vitro. Hum. Mol. Genet. 8, 673–682.
143. Warrick, J.M., Chan, H.Y., Gray-Board, G.L., Chai, Y., Paul-
son, H .L. & Bonini, N.M. (1999) Suppression of p olyglutamine-
mediated neurodegeneration in Dr osophila by the mole cular
chaperone HSP70 . Nat. Genet. 23, 425–428.
144. Bonini, N.M. (2002) Chaperoning brain de generation. Proc. Natl
Acad. Sci. USA 99 16407–16411.
145. Sc hmidt, T., Lindenberg, K.S., Krebs, A., Scho
¨
ls, L., Laccone,
F., Herms, J ., Rechsteiner, M., Riess, O. & Landwehrm eyer,
G.B. (2002) Protein surveillanc e machinery in br ains with spi-
nocerebellar ataxia type 3: redistribution and diffe rential

recruitment of 26S proteasome subunits and chaperones to
neuronal intranuclear inclusions. Ann. Neurol. 51, 302–310.
146. Thoms, S. (200 2) Cdc48 c an distinguish b etween native and non-
native proteins in the absence of cofactors. FEBS Lett. 520 , 107–
110.
147. Ciechanover, A . & Brundin, P. (2003) The ubiquitin prote asome
system in neurodegenerative diseases: sometimes th e chic ke n,
sometimes the egg. Neuron 40, 427–446.
148. Jana, N .R. & Nukina, N . ( 2003) Re cent adv ances in under-
standing the pathogenesis of p olyglutamine diseases: involve-
ment of molecular chaperones and ubiquitin-proteasome
pathway. J. Chem. Neuroanat. 26, 95–101.
149. Tarlac, V. & Storey, E . (2003) Role o f proteolysis in poly-
glutamine disorders. J. Neurosci. Res. 74, 406–416.
150. Legendre-Guillemin, V., Wasiak, S., Hussain, N.K., Angers, A.
& McPherson, P .S. ( 2004) ENTH/ANTH p roteins a nd cla thrin-
mediated membrane budding. J. Cell Sci. 117, 9–18.
151. Kim, J.H., Park, K.C., Chung, S.S., Bang, O. & Chung, C.H.
(2003) Deubiquitinating enzymesascellularregulators.J. Bio-
chem. (Tokyo) 134, 9–18.
152. Wing, S.S. ( 2003) De ubiquitinating enzy mes – the importance of
driving in reverse along the ubiquitin-proteasome pathway. Int.
J. Biochem. Cell Biol. 35, 590–605.
153. Johnston, S.C., Larsen, C.N., Cook, W .J., Wilkinson, K.D. &
Hill, C.P. (1997) Crystal structure of a deubiquitinating enzyme
(human UCH-L3) at 1.8 A
˚
resolution. EMBO J. 16, 3787–
3796.
154. Johnston, S.C., Riddle, S.M., Cohen, R.E. & Hill, C.P. (1999)

Structural basis for th e specificity of ubiquitin C-terminal
hydrolases. EMBO J. 18, 3877–3887.
155. Barrett, A.J . & Rawlings, N.D. ( 2001) E volutionary lines of
cysteine peptidases. Biol. Chem. 382, 727–733.
156.Hu,M.,Li,P.,Li,M.,Li,W.,Yao,T.,Wu,J.W.,Gu,W.,
Cohen, R.E. & Shi, Y. (2002) Crystal structure of a UBP-family
deubiquitinating enz yme in isolation and in complex with
ubiquitin aldehyde. Cell 111, 1041–1054.
157. Uchiyama,K.,Jokitalo,E.,Kano,F.,Murata,M.,Zhang,X.,
Canas, B., N ewm an, R., Rabouille, C., Pappin, D ., Freemont, P.
& Kondo, H. (2002) VCIP135, a novel essential factor for p97/
p47-mediated membrane fu sion, is r equired for Golgi and ER
assembly in vivo. J. Cell Biol. 159, 855–866.
158. Wang, Y., Satoh, A., Warren, G. & Meyer, H.H. (2004)
VCIP135 acts as a d eubiquitinating e nzyme during p97-p 47-
mediated reassembly of mitotic Golgi fragments. J. Cell Biol.
164, 973–978.
159. Masino,L.,Kelly,G.,Leonard,K.,Trottier,Y.&Pastore,A.
(2002) Solution structure of polyglutamine tracts in G ST-poly-
glutamine fusion proteins. FEBS Lett. 513, 267–272.
160. Chen, Y.W. ( 2003) Local protein unfoldin g and pa thogenesis of
polyglutamine-expansion dise ases. Proteins 51, 68–73.
161. Wilson, S.M., Bhattacharyya, B., Rachel, R .A., Coppola, V.,
Tessarollo, L., Householder, D.B., Fletcher, C.F., Miller, R .J.,
Copeland, N.G. & Jenkins, N.A. (2002) Synaptic defects i n
ataxia mice result from a mutation in Usp14, encoding a
ubiquitin-specific prote ase. Nat. Genet. 32, 420–425.
162. Ehlers, M.D. (2003) Ubiquitin and synaptic dysfunction: ataxic
mice highlight new common themes in neurological d isease.
Trends Neurosci. 26 ,4–7.

163. Hong, S., Kim, S.J.KaS., Choi, I. & Kang, S. (2002) USP7, a
ubiquitin-specific prote ase, interact s with ataxin-1, the SCA1
gene product. Mol. Cell Neurosci. 20, 298–306.
Supplementary material
The following material is available from
/>ejb/ejb4245/ejb4245sm.htm
Appendix. Supplementary online material.
3170 M. Albrecht et al. (Eur. J. Biochem. 271) Ó FEBS 2004