Progress in Neurobiology 99 (2012) 179–185

SCA1—Phosphorylation, a regulator of Ataxin-1 function and pathogenesis

Harry T. Orr *

Institute for Translational Neuroscience, Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, MN 55455, USA

Contents

1. SCA1 – overview of the disease and genetics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

2. Of mice and flies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180

3. First steps of relevance to a treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

4. Pathogenic pathways in SCA1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182

5. Ser776 – ATXN1 function and SCA1 pathogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182

6. Pathophysiology of SCA1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

7. Closing comments – future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

A R T I C L E I N F O

Article history:

Received 16 March 2012

Accepted 10 April 2012

Available online 16 April 2012

Keywords:

Polyglutamine disease

SCA1

Phosphorylation

Pathogenesis

A B S T R A C T

Spinocerebellar ataxia type 1 (SCA1) is one an intriguing set of nine neurodegenerative diseases caused by

the expansion of a unstable trinucleotide CAG repeat where the repeat is located within the coding of the

affected gene, i.e. the polyglutamine (polyQ) diseases. A gain-of-function mechanism for toxicity in SCA1,

like the other polyQ diseases, is thought to have a major role in pathogenesis. Yet, the specific nature of this

gain-of-function is a matter of considerable discussion. An issue concerns whether toxicity stems from the

native or normal function of the affected protein versus a novel function induced by polyQ expansion. For

SCA1 considerable evidence is accumulating that pathology is mediated by a polyQ-induced exaggeration

of a native function of the host protein Ataxin-1 (ATXN1) and that phosphorylation of S776 regulates its

interaction with other cellular protein and thereby function. In addition, this posttranslational modification

modulates toxicity of ATXN1 with an expanded polyglutamine.

� 2012 Elsevier Ltd. All rights reserved.

Contents lists available at SciVerse ScienceDirect

Progress in Neurobiology

jo u rn al ho m epag e: ww w.els evier . c om / lo cat e/pn eu ro b io

1. SCA1 – overview of the disease and genetics

Spinocerebellar ataxia type 1 (SCA1) joined the ranks of theunstable nucleotide repeat disorders and specifically the CAG/polyglutamine (polyQ) diseases in 1993 (Orr et al., 1993). SCA1patients have loss of coordination of the limbs and trunk, unstablegait, dysarthric speech, and nystagmus but may have othersymptoms, including extrapyramidal dysfunction, dysautonomia,

Abbreviations: Akt, AKR mouse thymoma kinase; ATXN1, vascular endothelial

growth factor human Ataxin-1; dox, doxycycine; NLS, nuclear localization

sequence; PKA, cyclic AMP-dependent protein kinase; polyQ, polyglutamine;

RBM17, RNA binding motif protein 17; SCA1, Spinocerebellar ataxia type 1; Tta,

tetracycline-transactivator; TRE, tetracycline response element; VEGF, vascular

endothelial growth factor.

* Correspondence address: Institute for Translational Neuroscience, University of

Minnesota, Wallin Medical Bioscience Building, 2101 Sixth Street S.E., Minneapolis,

MN 5545, USA. Tel.: +1 612 625 3647.

E-mail address: orrxx002@umn.edu.

0301-0082/$ – see front matter � 2012 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.pneurobio.2012.04.003

cognitive impairment, and motor and sensory impairments. SCA1is characterized pathologically by loss of Purkinje cells in thecerebellar cortex and neuronal loss in brain stem nuclei andcerebellar dentate nuclei (Koeppen, 2005). Individuals carrying amutant SCA1 allele can have symptoms starting as early as the firstdecade. By the sixth decade disease penetrance is essentiallycomplete.

SCA1 genetics dates back to the mid-late 1970s when use of HLAserological typing as genetic markers revealed an autosomaldominant form of ataxia linked to the HLA complex on chromo-some 6 (Yakura et al., 1974; Jackson et al., 1977). With theapplication of molecular genetic approaches, location of the SCA1

gene was refined on the short arm of chromosome 6 to a regionabout 15 CM distal to HLA-A (Rich et al., 1987; Keats et al., 1991).With its cloning, the SCA1 gene was found to span 450 kb of DNA at6p22.3 and consists of nine exons (Banfi et al., 1994). The SCA1

transcript is 10,660 bases in length with the first seven exonsencoding the 50 untranslated region and exons eight and ninecontaining the Ataxin-1 (ATXN1) coding region (816 amino acids in

Fig. 1. A schematic depiction of the progressive motoric deficits and Purkinje cell

pathology in ATXN1[82Q] mice. Importantly all neurological alterations including

overt ataxia appear prior to the onset of Purkinje cell death.

H.T. Orr / Progress in Neurobiology 99 (2012) 179–185180

a protein having a polyQ tract of 30 residues), and the 7277 base 30

untranslated region. The polyQ stretch begins at amino acid 197and is encoded within exon eight. Interestingly, there are bindingsites for several miRNAs within the 30 untranslated region of theSCA1 transcript that function to down-regulate ATXN1 levels (Leeet al., 2008).

Normal SCA1 alleles contain from 6 to 42 CAG repeats withthose greater than 21 being interrupted with 1–3 CAT trinucleo-tides. Disease alleles, on the other hand, are pure CAG tractsranging from 39 to 82 units. Such interruptions are found in all ofthe longer unaffected alleles. In contrast, all affected alleles arepure CAG tracts (Chung et al., 1993). The presence of repeatinterruptions particularly in the longer wild type allleles leads tothe suggestion that the CAT interruptions have a critical role inmaintaining the relative stability of normal alleles compared tomutant SCA1 alleles.

Disease-related expansions have a direct relationship betweenlength and severity/age-of-onset of disease, i.e. longer the longerthe glutamine tract the more severe and earlier is the age of onsetof disease. Expansion of the CAG repeat into the affected rangebesides encoding a pathogenic protein also enhances theinstability of the DNA repeat such that changes in repeat lengthare found when transmitted from parent to offspring. Overall thisinstability of mutant alleles is the molecular basis of the geneticobservation anticipation, an increase in disease severity/earlier ageof onset as ones follows the disease for generation to generation ina family. In the case of SCA1, anticipation was first noted in 1950 ina large family with an inherited ataxia that subsequently proved tohave SCA1 initially by virtue of a genetic linkage to the HLAcomplex on chromosome 6p (Schut, 1950; Haines et al., 1984). Onepoint worth noting is that large repeat expansions are restricted topaternal transmissions so that in juvenile forms of SCA1 identifiedto date all stem from a father to offspring transmission.

2. Of mice and flies

Over the years several mouse models were generated in aneffort to understand mechanisms underlying SCA1 disease andATXN1 function. In modeling SCA1, as well as other neurodegen-erative disorders, an issue is whether to express the mutanttransgene in all neurons or cells of the CNS, or to target criticalpopulations such as the Purkinje cell in the case of SCA1. Theadvantage of the former approach is that it more closely modelsthe human disease, but in turn it makes interpretation ofmechanisms more complex. The advantage of the latter approachis that Purkinje cells are critical cells for cerebellar function, theyare affected in nearly all forms of SCA, and damage to them usuallycreates a behavioral phenotype.

The first SCA1 mouse model was generated in 1995 to test theCAG expansion, gain-of-function hypothesis. Burright et al. (1995)directed expression of ATXN1 in mouse Purkinje cells using aPurkinje cell-specific Pcp2/L7 promoter. Two types of lines werecreated; control lines with over-expression of the normalinterrupted allele (ATXN1[30Q]) containing 30 repeats (CAG)12

CATCAGCAT(CAG)15, and experimental lines expressing theexpansion construct (ATXN1[82Q]) containing 82 uninterruptedCAG repeats. Progeny of the control ATXN1[30Q] lines, and theexpansion ATXN1[82Q] lines inherited the transgene eitherpaternally or maternally and in all animals the repeat was stableand showed no change in repeat size.

At one year of age, ATXN1[30Q] mice had a normal neurologicalphenotype, suggesting intact Purkinje cell function. In contrast, themutant ATXN1[82Q] mice beginning around five weeks of agedeveloped neurological abnormalities that progressed with age(Fig. 1). These included reduced cage activity and generaluncoordinated movements. Footprint patterns indicated gait

abnormalities and motor performance measured using an acceler-ating rotating-rod apparatus was abnormal as early as five weeksof age (Clark et al., 1997). The motoric deficits progressed with ageuntil mice were clearly ataxic by cage behavior at 12 weeks of age.

Pathologic changes were observed in the ATXN1[82Q] cerebel-lum as early as P25 when clear cytoplasmic vacuoles were found insome of the cell bodies of Purkinje cells (Clark et al., 1997). Thedegeneration progressed with age and by 27 weeks the majority ofPurkinje cells had stunted, atrophic dendritic morphology.Notably, Purkinje cell loss was preceded by the ataxic phenotype(Fig. 1). At 1 year, the cerebellar cortex was dramatically decreasedin size secondary to both atrophy and loss of Purkinje cells,frequent heterotopic localization of Purkinje cells in the molecularlayer, and diminished calbindin immunoreactivity in survivingPurkinje cells. In addition, ATXN1[82Q] expressing Purkinje cellsaccumulate ATXN1 and ubiquitin-positive aggregates, or neuronalintranuclear inclusions, that co-localize with the proteasome andthe molecular chaperone, HDJ-2/HSDJ, demonstrating that there isATXN1 misfolding in SCA1 disease (Cummings et al., 1998).

To determine if the neurological symptoms in SCA1 might becaused by an ATXN1 loss of function, Sca1 null mice weregenerated (Matilla et al., 1998). Sca1 exon 8 was targeted anddeleted as the majority of the coding sequence resides in this exon.Sca1(�/�) mice were viable, and displayed both normal develop-ment and a normal life span. In addition, they failed to displayataxic symptoms or neurodegeneration even out to 30 months ofage. Further behavioral studies, including the open field test,elevated plus maze, and Morris water maze suggest Sca1(�/�) micehave spatial learning deficits.

In order to generate a more accurate model of the effects ofmutant Ataxin-1 expression under control of its endogenouspromoter, a knock-in model was generated by homologousrecombination to inset an expanded CAG trinucleotide repeat intothe mouse Sca1 locus. The initial line generated, Sca178Q/2Q, whileexpressing endogenous levels of expanded Ataxin-1 in theexpected temporal and spatial patterns, lacked any ataxicphenotype or neuropathological abnormalities (Lorenzetti et al.,2000). A second model was generated with 154 CAGs (Wataseet al., 2002). Sca1154Q/2Q mice had an altered phenotype thatincluded muscle wasting, ataxia, abnormal gait, severe kyphosisand premature death between 35 and 45 weeks. Sca1154Q/2Q

animals displayed significant motor deficits by rotating-rodanalysis at 5 and 7 weeks as well as impaired spatial learningdeficits. In addition, the Sca1154Q/2Q mice exhibited a decrease inlong-term potentiation at 24 weeks. The Sca1154Q/2Q animals weredistinct from the ATXN1[82Q] model in that they exhibited repeat

H.T. Orr / Progress in Neurobiology 99 (2012) 179–185 181

instability similar to human patients (Watase et al., 2002). Brainsfrom Sca1154Q/2Q animals were smaller than controls, though brainsections showed uniform atrophy rather than specific atrophy ofthe cerebellum more characteristic of SCA1 disease in humans.While slight Purkinje cell loss was observed, the molecular layerthickness was relatively intact even in 24-week-old animals.

Overall, pathology and disease in Pcp2-ATXN1[82Q] mice issimilar to the typical mid to late-onset clinical SCA1 phenotypethat arises from moderate CAG expansions. Neuronal damage isconfined to specific neuronal groups (e.g. Purkinje cells in thedisease. In contrast mice for the Sca1154Q/2Q knockin line have amore extreme CAG repeat expansion and exhibit a disease moresimilar to SCA1 juvenile or even infantile-onset disease whereneuronal damage is much more widespread than in adult-onsetdisease, reaching into brain regions completely unaffected in olderpatients. This loss of specificity might result from the mutationbeing so severe that protein function and/or clearance is impairedeven in neurons that normally express low levels of ATXN1.

3. First steps of relevance to a treatment

As one envisions moving towards a treatment for a neurode-generative disease, the extent to which the disease is reversibleafter onset is of importance. To determine whether SCA1 disease isreversible, a conditional mouse model of SCA1 was developedusing the tetracycline-regulated system (Zu et al., 2004). Theapproach used the Pcp2/L7 promoter to drive a tetracycline-transactivator (Tta) transgene specifically in Purkinje cells. Ttafunctions to bind to a tetracycline response element (TRE) toinduce expression of a second TRE-ATXN11[82Q] transgene solelyin Purkinje cells. Administration of doxycycine (dox), a tetracyclinederivative that crosses the blood brain barrier, abolishes the Tta/TRE interaction and turns off expression of ATXN1. Effects ofreversal of transgene expression were assessed at early, mid, andlate stages of disease (Zu et al., 2004). At the early stage, 6-week-old mice demonstrated the rotating-rod deficit could be reversedafter 6 weeks of dox treatment. In addition, Purkinje cell pathologyimproved by assessment of molecular layer thickness anddendritic arborization, although heterotopic Purkinje cells werestill present. At mid stage disease, 12 week on–4 week off animalscontinued to have rotating-rod deficits similar to 12 and 16 week-on animals. After 8 weeks and 12 weeks of dox treatment, however,rotating-rod performance improved, but only partially. Interest-ingly, this 8-week timepoint coincided with the return of mGluR1aglutamate receptors at the Purkinje cell-parallel fiber synapses,albeit at a lower expression level. There was recovery of molecularlayer thickness and improved arborization of Purkinje celldendrites demonstrating that halting ATXN1[82Q] expression atthe time of typical onset of ataxia prevents the progression of andpartially reverses degenerative changes in Purkinje cells. Whendox was given to animals at 32 weeks of age, there was nosignificant improvement of ataxia, but there was some improve-ment of Purkinje cell pathology. At all disease stages, Purkinje cellsrapidly eliminated ATXN1[82Q]-containing nuclear inclusionsupon administration of dox and cessation of ATXN1 expression.Thus, even at a late stage of disease Purkinje cells are able to clearATXN1 with an expanded polyQ. Overall, these studies revealedthat in fact Purkinje cell pathology induced by ATXN1 with anexpanded polyQ could be reversed with cessation of ATXN1protein synthesis. Interestingly, the capability of Purkinje cells torecover decreases with increasing age. It remains unclear whetherthis age dependence of recovery is due to affects of prolongedexposure to mutant ATXN1 and/or that older neurons in generalare less able to repair damage.

The Tta/TRE conditional SCA1 mice were also used to demon-strate a link between proper Purkinje cell development and

susceptibility of adult neurons to toxic insults. The basis for thisstudy was a finding that one of the initial transgenic mouse lines ofthe ATXN1[82Q] transgenic series (the so-called B04 line) whileexpressing less ATXN1[82Q] than other lines, had a more severeataxia (Burright et al., 1995). Intriguingly, in the B04 miceATXN1[82Q] expression begins 10 days earlier during postnataldevelopment than in the other lines of the ATXN1[82Q] series, onpostnatal day 2, raising the possibility that onset of transgeneexpression impacts disease severity in an adult. This idea wastested directly by delaying ATXN1[82Q] transgene expressionduring early postnatal development in the conditional mice.

In mice, the first three postnatal weeks are a critical period forcerebellar development. It is during this period that one of the twomajor inputs to Purkinje cells matures, climbing fibers from theinferior olives, and the cerebellum undergoes a dramatic increasein size being transformed from an immature smooth surfacedanloga at birth to the adult highly foliated structure (Goldowitzand Hamre, 1998). ATXN1[82Q] mice that do not express ATXN1during this developmental period failed to develop the neurologi-cal and pathological features seen in animals that expressATXN1[82Q] during postnatal development (Serra et al., 2006).Additional studies implicated an alteration in function of theorphan nuclear receptor Rora in ATXN1[82Q], perhaps via aninteraction of ATXN1 with the Rora cofactor Tip60 (Gehrking et al.,2011), as being key for this developmental affect on SCA1pathogenesis in Purkinje cells. Mouse models expressing mutantforms of ATXN1 expressed decreased levels of Rora and thereforedecreased levels of Rora-mediated genes. To evaluate the role ofRora in SCA1 pathogenesis, ATXN1[82Q] animals were crossed tostaggerer heterozygous mice, which lack one functional copy ofRora (sg/+). ATXN1[82Q]:sg/+ mice had more severe Purkinje cellpathology suggesting that depletion of Rora enhances SCA1disease. Crossing ATXN1[82Q] animals to Tip60+/� animalstransiently delayed the ATXN1[82Q] cerebellar degeneration byincreasing Rora and Rora-mediated gene expression (Gehrkinget al., 2011).

One of the first potential pathogenic mechanism examinedusing SCA1 transgenic mice was the role of the ATXN1 nuclearinclusions/aggregates. Neuronal aggregates containing the mutantprotein occur in many of the inherited polyglutamine disordersand were raised as a possible common mechanism for neurode-generation (Ross and Poirier, 2004). To examine the role themacroscopic aggregates play in SCA1, ATXN11[77Q]D mice weredeveloped. ATXN11[77Q]D was generated by deleting the self-association region of Ataxin-1 (Burright et al., 1997). In tissueculture cells ATXN1[77Q]D failed to form macroscopic inclusionsas seen with ATXN1[82Q]. ATXN1[77Q]D transgenic miceexpressed ATXN1 in the Purkinje cell cytoplasm and nucleus,similar to ATXN1[82Q] animals. Like as seen in tissue culture cells,ATXN1[77Q]D mice did not develop inclusions in their Purkinjecells. The absence of inclusions, however, did not protectATXN1[77Q]D mice from developing Purkinje cell pathology orataxia. Indeed, ATXN1[77]D animals had the same Purkinje cellatrophy seen in ATXN1[82Q] mice and performed as poorly asATXN1[82Q] animals in motor performance tasks (Klement et al.,1998). Moreover, Purkinje cells that express ATXN1[82Q] but notthe E6-AP ubiquitin-protein ligase Ube3A also have far fewernuclear inclusions but markedly worse pathology (Cummingset al., 1999). Thus, several lines of investigation show that Purkinjecell pathology and motor deficits associated with SCA1 are notdependent on the formation of mutant ATXN1 inclusions.

Expressing full-length human SCA1 genes in Drosophila isanother strategy used to generate SCA1 animal models. Thisapproach is particularly useful in the elucidation and dissection ofgenetic pathways that impact SCA1 pathogenesis. For example,genetic screens in the fly were the first to reveal that genes

H.T. Orr / Progress in Neurobiology 99 (2012) 179–185182

encoding proteins involved in RNA metabolism and transcriptionregulation modify development of SCA1 (Fernandez-Funez et al.,2000). As will become evident below, transcription and RNAprocessing are now thought to be two of the major pathways inwhich ATXN1 functions and in which alterations lead to diseaseinduced by ATXN1 with an expanded polyQ tract.

4. Pathogenic pathways in SCA1

An important step forward in dissecting SCA1 pathogenesis wastaken when the role of the cellular localization of mutant ATXN1 ondisease was assessed. Using the ATXN1[82Q] transgene, a lysine-to-threonine substitution was introduced at amino acid residue 772 todisrupt the nuclear localization signal (NLS) function (Klement et al.,1998). In mice expressing ATXN1[82Q]-K772T, ATXN1 localizedprimarily to the cytoplasm of Purkinje cells in contrast toATXN1[82Q] mice with a functional NLS where significant levelsof ATXN1 localized to the nucleus as well as to the cytoplasm.Importantly, ATXN1[82Q]-K772T animals failed to develop ataxia,as measured by rotating-rod deficits and ataxic cage behavior. Inaddition ATXN1[82Q]-K772T mice had no cerebellar pathology outto one year of age. This result lead to what has subsequently provento be important hypothesis regarding the molecular basis SCA1pathogenesis – localization of mutant ATXN1 to the nucleus iscritical for disease indicating that it is a function(s) of ATXN1 in thenucleus that is altered in disease. With regards to ATXN1 in thenucleus, further studies showed that its dynamics is altered byexpansion of the polyQ tract (Krol et al., 2008). Moreover, whilemutant ATXN1 with an expanded polyQ tract is able to enter thenucleus, it is ability to be transported back into the cytoplasm isdramatically reduced (Irwin et al., 2005). Also, the length of thepolyQ tract negatively affects the SUMOylation, a commonposttranslational modification of nuclear proteins having a role intranscription (Muller et al., 2004), of ATXN1, which occurs whenATXN1 can be transported to the nucleus (Riley et al., 2005).

With regards to possible functions in the nucleus, ATXN1interacts with RNA (Yue et al., 2001), and several regulators oftranscription, SMRT (Tsai et al., 2004), Capicua (Lam et al., 2006),Gfi-1 (Tsuda et al., 2005), and the Ror&/Tip60 complex describedabove (Serra et al., 2006). In the case of Capicua and the Ror&/Tip60 complex, mutant ATXN1 seems to enhance their degrada-tion. For Capicua, loss of ATXN1 decreases its steady-state level(Lam et al., 2006). ATXN1 also interacts with two RNA-splicingfactors, RBM17 (Lim et al., 2008) and U2AF65 (de Chiara et al.,2009). Another ATXN1 interacting protein is the ATXN1 paralogBrother of ATXN1, BOAT or ATXN1L (Mitzutani et al., 2005).

An evolutionary conserved region within ATXN1 that drivessome of these interactions is a 120-amino acid AXH domain,Ataxin-1/HBP1 (HMG box-containing protein-1 transcriptionfactor), crystalizes as a dimer that contains an OB-fold (de Chiaraet al., 2003). Presence in ATXN1 of an OB-fold, a structural motiffound in many oligonucleotide-binding proteins, is consistent withit ability to bind RNA. In addition to the OB-fold, the AXH domain ofATXN1 was postulated to contain a protein interacting surface (deChiara et al., 2003). Subsequently, the AXH domain was shown tobe important for the interaction of ATXN1 with the Drosophia/mammalian transcription factor Senseless/Gfi-1 (Tsuda et al.,2005), the transcription repressor Capicua (Lam et al., 2006), andthe transcription factors Ror&/Tip60 (Gehrking et al., 2011), aswell as the ATXN1 parolog BOAT/ATXN1L (Mitzutani et al., 2005).

An indication that the AXH domain of ATXN1[82Q] has a role inSCA1 pathogenesis came for studies in Drosophila where expres-sion of ATXN1[82Q] lacking the AXH domain no longer induced thepathogenic effects on the Senseless (Tsuda et al., 2005). As part ofan effort to assess whether ATXN1 with an expanded polyQinduces neurotoxicity via formation of novel, aberrant versus

native protein complexes, it was found that the majority of wildtype Atxn1 as well as expanded ATXN1 assembled into large stablecomplexes (Lam et al., 2006). ATXN1 was shown to also interactwith the transcription repressor Capicua through the AXH domainand modulate its activity in Drosophila and mammalian cells.Furthermore, Capicua co-eluted with the ATXN1 containingcomplexes by sizing chromatography. Lastly, this study foundthat ATXN1[82Q] with an alanine residue instead of the normalserine at position 776 no longer formed high molecular weightcomplexes, suggesting that it is mutant ATXN1 in a nativemultiprotein complex that is pathogenic.

Recently, it was demonstrated that reducing the amount ofCapicua decreases severity of disease in Sca1154Q/2Q mice (Fryeret al., 2011). Fryer et al. showed that a 50% reduction in Capicuagenerated by crossing Sca1154Q/2Q animals with Capicua+/� micewas sufficient to improve motor performance as well as learningand memory and extend life span of mice carrying on mutant alleleof Sca1. Remarkably, mild exercise also reduced Capicua levels andextended life span of Sca1154Q/2Q mice. The effect of exercise on lifespan last for an extended period of time, lasting well after exercisewas ceased. This latter affect is due to the ability of exercise toincrease concentrations of epidermal growth factor in brains thatfunctions to reduce Capicua. Molecularly, Fryer et al. found that theeffect of mutant ATXN1 on Capicua regulation transcription varieddepending on the target gene. They suggested that polyglutamine-expanded Atxn1 causes Capicua to bind more tightly to certaintranscriptional targets (hyper-repressing them) and at othertranscriptional targets causing Cic to bind less to—and thusupregulate (derepress) them. It is the relief of the hyper-repressionthat seems to underlie the protective effect of a reduction inCapicua either genetically or by exercise.

Two other functions shown to impact SCA1 pathogenesis inmice are the neurotrophic and angiogenic vascular endothelialgrowth factor (VEGF) and calbindin-D28k pathways (Vig et al.,2011). In the case of VEGF, ATXN1 with an expanded polyQrepressed its transcription and genetic overexpression or infusionof recombinant of VEGF dampened pathology in Sca1154Q/2Q mice(Cvetanovic et al., 2011). Maintaining calcium homeostasis iscritical for proper neuronal function and downregulation of genesencoding proteins involved in calcium homeostasis and signaling,including calbindin-D28k, is a feature of ATXN1[82Q] expressingPurkinje cells (Lin et al., 2000; Serra et al., 2004). The importance ofCalbindin-D28k function in SCA1 was directly demonstrated by thefinding that a partial genetic reduction of calbindin-D28k in SCA1mice exacerbated disease severity (Vig et al., 2011).

5. Ser776 – ATXN1 function and SCA1 pathogenesis

Ser776 (S776) is one of seven endogenous sites of phosphor-ylation in ATXN1 (Emamian et al., 2003; Huttlin et al., 2010).Importantly, S776 is located within another highly conservedportion of ATXN1, a 12 amino acid segment (residues 768–780)towards the C-terminus of ATXN1 (Carlson et al., 2009; de Chiaraet al., 2009). This segment of ATXN1 contains three overlappingfunctional motifs among which are the previously describednuclear localization sequence (NLS), amino acids 771–774 (Kle-ment et al., 1998), a 14-3-3 binding motif, amino acids 774–778(Chen et al., 2003), and a UHM ligand motif (ULM), amino acids771–776 (de Chiara et al., 2009). ULMs are a protein–proteininteraction motif found exclusively in proteins involved with RNAsplicing (Corsini et al., 2007).

Interest in S776 and the role its phosphorylation may have inATXN1 function was first stirred with the finding that in mice withPurkinje cell expression of ATXN1[82Q]-A776, a phosphorylationblocking amino acid substitution, there is substantially fewernuclear inclusions compared to ATXN1[82Q] transgenic mice at 32

H.T. Orr / Progress in Neurobiology 99 (2012) 179–185 183

weeks. In addition, ATXN1[82Q]-A776 mice do not develop ataxiccage behavior, do not show deficits by rotating-rod testing, andhave only mild Purkinje cell pathology (Emamian et al., 2003).Recently, it was shown that animals with a phospho-mimickingAsp776 (ATXN1[82Q]-D776) had a more severe disease thananimals expressing ATXN1[82Q]-S776 in their Purkinje cells. Moreintriguingly, ATXN1[30Q]-D776 also had impaired motor coordi-nation on the Rotarod compared to wild type and ATXN1[30Q]control animals (Duvick et al., 2010). The level of impairment wassimilar to that of ATXN1[82Q] animals and progressed as animalsaged until they could no longer perform the task. Similarly,ATXN1[30Q]-D776 animals had a significantly widened gaittypical of ATXN1[82Q] animals. Both ATXN1[30Q]-D776 andATXN1[82Q]-S776 mice at 12 weeks had Purkinje cell atrophy.Thus by a single, possibly mimicking phosphorylation amino acidsubstitution at residue 776, an ATXN1 with a wild type polyQ tractwas converted into a pathogenic protein.

So how does phosphorylation of S776 affect ATXN1 and to whatextent does having an Asp at position 776 have a similar effect?Phosphorylation of S776 was first suggested to stabilize ATXN1 intissue culture cells (Chen et al., 2003). In support of this idea aredata from mice expressing various forms of ATXN1 (Jorgensenet al., 2009; Lai et al., 2011). In these studies the relative ratios ofSCA1 mRNA to ATXN1 in mice expressing Pcp2-ATXN1 transgeneon a Sca1�/� background were used as a surrogate for proteinstability. Based on this approach it was concluded that thephospho-resistant ATXN1-A776 (Ala) is far less stable than eitherphosphorylatible ATXN1-S776 or phospho-mimicking ATXN1-D776. These results are consistent with phosphorylation of S776stabilizing ATXN1 in vivo in Purkinje cells. More recent in vitro datashowing that in the absence of protease inhibitors, inhibition ofS776 phosphorylation results in a decrease in ATXN1 level overtime. In contrast when protease activity is blocked, inhibition ofS776 phosphorylation had no affect on ATXN1 levels (Lagalwarunpublished data). In sum, there is considerable and consistentevidence that phosphorylation of S776 stabilizes ATXN1 byblocking its proteolysis.

S776 and perhaps its phosphorylation also has a crucial role inregulating interactions of wild type and mutant ATXN1 with at leastthree cellular proteins. These include the phospho-serine/phospho-threonine binding protein 14-3-3 (Chen et al., 2003) – a regulator ofmany signal transduction pathways (Morrison, 2008), and splicingfactors RBM17 (Lim et al., 2008) and U2AF65 (de Chiara et al., 2009).As of now characterization of the interaction of ATXN1 with U2AF65,constitutive component of the spliceosome (Hartmuth et al., 2002),is limited to one study (de Chiara et al., 2009). In a peptide–peptideinteraction assay, phosphorylation of S776 in a nine-residue ATXN1peptide reduced but did not eliminate binding to a U2AF65 UHMpeptide. This study also showed that overexpression of wild typeenhances U2AF65 splicing function in a transfected cells co-expressing splicing reporter minigene. Interestingly, ATXN1 withan expanded polyQ was not able to promote U2AF65 splicingfunction, suggesting that polyQ expansion in some way disrupts theATXN1/U2AF65 interaction.

The interaction of ATXN1-pS776 with 14-3-3 is important fromseveral perspectives. First, the ATXN1-pS776/14-3-3 complex isstable such that the two proteins co-immunoprecipitate (Chenet al., 2003). Moreover, the interaction of 14-3-3 with ATXN1seems to be restricted to the cytoplasm of cells in the cerebellum(Lai et al., 2011). Binding of 14-3-3 to ATXN1 in the cytoplasm hastwo identified outcomes; it blocks the dephosphorylation ofATXN1-S776 in the cytoplasm and it covers the NLS therebyblocking transport of ATXN1 to the nucleus (Lai et al., 2011). In thelatter case, it means that for ATXN1 to be transported to thenucleus 14-3-3 must first disassociate presumably by some yet tobe identified regulated process. Notably, 14-3-3 does not bind to

ATXN1-D776 (de Chiara et al., 2009). Yet, ATXN1-D776 is resistantto proteolysis similar to ATXN1-pS776 (Lai et al., 2011). Thus, 14-3-3 binding is not required for stabilization of ATXN1 but does actindirectly to stabilize ATXN1-pS776 by preventing its dephos-phorylation. The finding that the relative amount of ATXN1-D776found in the nucleus is increased is consistent with the inability of14-3-3 to form a complex with ATXN1-D776, i.e. the NLS in ATXN1-D776 is not masked by 14-3-3 (Lai et al., 2011).

Another protein whose interaction with ATXN1 involves S776 isRBM17, RNA-binding motif protein 17/spf45 in Drosophila (Limet al., 2006; de Chiara et al., 2009). To date, RBM17 is the onlyprotein whose interaction with ATXN1 is modified by the length ofATXN1s polyQ tract and the amino acid at position 776 (Lim et al.,2008). The ATXN1/RBM17 interaction increases with increasinglength of the polyQ tract. In addition, the ATXN1/RBM17 isdecreased dramatically with ATXN1-A776 regardless of polyQtract length. Interestingly, replacing S776 with an aspartic acidreside results in a form of ATXN1-30Q, ATXN1[30Q]-D776, that hasan enhanced interaction with RBM17 similar to that seen withATXN1[82Q]-S776. Thus, the two forms of ATXN1 that are able tocause disease in Purkinje cells of transgenic mice, ATXN1[82Q]-S776 and ATXN1[30Q]-D776 (Burright et al., 1995; Duvick et al.,2010), are the both forms of ATXN1 that interact strongly withRBM17 (Lim et al., 2008). Moreover, a form of ATXN1–ATXN1-A776– that interacts weakly with RBM17 is unable to cause disease(Emamian et al., 2003; Lim et al., 2008). These results indicate thata gain of function pathogenic mechanism of mutant ATXN1involves its interaction with RBM17, and suggests that thephosphorylation state of serine 776 is critical for the strength ofthis interaction.

The phosphorylation state of a protein is a dynamic processdictated by both protein kinases and phosphatases. S776 of ATXN1lies within strong consensus phosphorylation sites for the kinasesAKR mouse thymoma (Akt) and cyclic AMP-dependent proteinkinase (PKA). While initial data using tissue culture cells and aDrosophila model of SCA1 indicated that Akt could phosphorylateS776 (Chen et al., 2003), more recent data favors PKA as the ATXN1-S776 kinase in the mammalian cerebellum (Jorgensen et al., 2009).Inhibition of Akt either in vivo or in a cerebellar extract-basedphosphorylation assay did not reduce the amount of phospho-S776ATXN1. Moreover, reduction of Akt in SCA1 transgenic mice was notprotective against disease. These results argue against Akt as akinase that phosphorylates S776 of ATXN1. In contrast, drugs thatselectively inhibit PKA or immunodepletion of PKA blocked theability of cerebellar extracts to phosphorylate ATXN1-S776. Thus,PKA is likely to be one kinase that phosphorylates ATXN1 at S776 inthe cerebellum. Together the data suggest that the phosphorylationof S776 in ATXN1 by PKA and the subsequent interaction ofATXN1[82Q]-pS776 with RBM17 in the nucleus is a majorcontributor to at least Purkinje cell pathogenesis in SCA1.

6. Pathophysiology of SCA1

One fundamental question facing SCA1 and many of theneurodegenerative disorders concerns the pathophysiology ofneuronal dysfunction. Is there dysfunction in a specific neuronalpopulation or circuit initially that contributes the onset ofbehavioral abnormalities? One study examined Ca2+ dynamicsand electrophysiological properties of Purkinje cells from adult (3–7 months) ATXN1[82Q] mice in slice preparations (Inoue et al.,2001). After adjustment for neuronal volume, the basicelectrophysiological properties of mutant Purkinje cells werefound to be essentially intact, including the responses to bothmajor afferent projections, the climbing fibers (CFs) and parallelfibers (PFs). This study, in the end, suggested that the functioningof the intact cerebellar neuronal circuitry is affected in SCA1 mice.

Fig. 2. The cellular pathway by which S776 phosphorylation of ATXN1 impacts its

function and pathogenesis in the nucleus. In the nucleus two pathways involving

the interaction of ATXN1 with either the transcriptional repressor Capicua or the

RNA splicing factor RBM17 contribute to pathogenesis. It is the RBM17-mediated

pathway that appears to be the pathway that is impacted by S776 phosphorylation.

H.T. Orr / Progress in Neurobiology 99 (2012) 179–185184

Barnes et al. (2011) utilized flavoprotein autofluorescence opticalimaging and field potential recordings in vivo to examine whetherthere are alterations in the climbing fiber-Purkinje cell and parallelfiber-Purkinje cell circuitry in early-stage (6-week old) and mid-stage (12-week old) disease mice prior to gross Purkinje cellpathology and death. The results show that ATXN1[82Q] preferen-tially affects the climbing fiber-Purkinje cell synapse early in thedisease process while parallel fiber-Purkinje cell synaptic transmis-sion is unaltered. As disease progresses out to 40 weeks of age (late-stage disease) there are signs of altered parallel fiber-Purkinje cellsynaptic activity. The functional and morphological assessmentsrevealed that the abnormalities in the climbing fiber-Purkinje cellcircuit that are dependent on ATXN1[82Q] expression and entranceinto the nucleus of PCs. Mice that express ATXN1[82Q] that fails toenter the nucleus of Purkinje cells show no signs of an alteredclimbing fiber-Purkinje cell circuit Moreover, the deficits in climbingfiber-Purkinje cell synaptic transmission required mutant ATXN1expression during early postnatal development. These findingssupport a model of early SCA1 disease where abnormalities at thesynapse contribute to disease initiation and motoric deficits. Theyimplicate a compromised function at a specific type of synapse in thedevelopment of SCA1 and highlight the importance of identifyingfactors unique to the climbing fiber-Purkinje cell synapse thatcontribute to its dysfunction.

7. Closing comments – future directions

The data generated to date strongly support a model of SCA1pathogenesis where four molecular features of mutant ATXN1 allmust be present for disease; an expanded polyQ tract, a functionalNLS, the AXH domain, and phosphorylatible Ser at residue 776.Together these contribute to critical events that occur in thenucleus of an affected neuron. Expansion of the polyQ tractdifferentially affects the function of ATXN1 depending on thecontext of the different native complexes it forms with othernuclear proteins. PolyQ tract expansion shifts the status quo byenhancing certain functional complexes/pathways perhaps at theexpense of others. At the center of the postulated imbalance innative complex formation induced by mutant ATXN1 are thecomplexes it forms with the transcriptional repressor Capicua andthe splicing protein RBM17 (Fig. 2). While changes in both Capicuaand RBM17 complexes contribute to disease, it is striking that asingle amino change, S776 to D776, converts ATXN1 with a wildtype polyQ tract to a protein that shares two features of ATXN1with an expanded polyQ – a protein that interacts strongly withRBM17 and causes disease in murine Purkinje cells. Thus, it seemsthat an enhanced interaction of mutant ATXN1 with RBM17 is amajor driver of disease at least in Purkinje cells. Since RBM17 isknow to regulate alternative RNA splicing it seems very likely thatalterations in splicing underlie the Purkinje aspect of SCA1.

Participation of S776 of ATXN1 and its phosphorylation in RBM17interaction and SCA1 pathogenesis provides one potential thera-peutic target for SCA1. Much effort has gone into and is going intounderstanding signaling pathways regulated by protein kinases.Protein kinases are one of the most active groups under investigationas drug target owing to their involvement in many pathologicalconditions. Yet there are substantial challenges that face targetingprotein kinases in disorders of the brain like SCA1 (Chico et al., 2009).Not the least of which are penetration of the blood–brain barrier andselectively targeting a multi-functional kinase like PKA. Thus, it isimportant not to depend on any one single therapeutic target.Rather, multiple therapeutic avenues need to be explored. In thisregard identification of the kinase(s) responsible for ATXN1-S776phosphorylation in affected neurons is of importance as it affords anopportunity to elucidate processes/signals upstream to the kinasethat regulate its action on ATXN1 that could result in viable drug

targets. Likewise, elucidation of downstream events triggered byATXN1-S776 phosphorylation is of value.

It is worth noting that all of the evidence on SCA1 usingexperimental models show a direct relationship between diseaseseverity and level of mutant ATXN1 expression. Thus, identifyingother pathways or strategies by which the levels of the disease-causing protein could be decreased would serve as additional noveldrug targets and or therapeutic approaches. For example, theability of RNA interference to inhibit polyQ-induced neurodegen-eration caused by mutant ATXN1 in a mouse model of SCA1 (Xiaet al., 2004) or the use of allele specific oligonucleotides as has beensued in a mouse model of ALS (Smith et al., 2006) offer twopotential genetic therapeutic approaches for decreasing thefunction of mutant ATXN1.

References

Banfi, S., Servadio, A., Chung, M.-y., Kwiatkowski Jr., T.J., McCall, A.E., Duvick, L.A.,Shen, Y., Roth, E.J., Orr, H.T., Zoghbi, H.Y., 1994. Identification and characteriza-tion of the gene causing type 1 spinocerebellar ataxia. Nature Genetics 7: 513–520.

Barnes, J., Ebner, B.A., Duvick, L.A., Gao, W., Chen, G., Orr, H.T., Ebner, T.J., 2011.Abnormalities in the climbing fiber-Purkinje cell circuitry contribute to neuro-nal dysfunction in ATXN1[82Q] mice. Journal of Neuroscience 31, 12778–12789.

Burright, E.N., Clark, H.B., Servadio, A., Matilla, T., Feddersen, R.M., Yunis, W.S.,Duvick, L.A., Zoghbi, H.Y., Orr, H.T., 1995. SCA1 transgenic mice: a model forneurodegeneration caused by an expanded CAG trinucleotide repeat. Cell 82,937–948.

Burright, E.N., Davidson, J.D., Duvick, L.A., Koshy, B., Zoghbi, H.Y., Orr, H.T., 1997.Identification of a self-association region within the SCA1 gene product, ataxin-1. Human Molecular Genetics 6, 513–518.

Carlson, K.M., Melcher, L., Lai, S., Zoghbi, H.Y., Clark, H.B., Orr, H.T., 2009. Characteri-zation of the Zebrafish atxn1/axh Gene Family. Journal of Neurogenetics 23,313–323.

H.T. Orr / Progress in Neurobiology 99 (2012) 179–185 185

Chen, H.-K, Fernandez-Funez, P., Acevedo, S.F., Lam, Y.C., Kaytor, M.D., Fernandez,M.H., Aitken, A., Skoulakis, E.M.C., Orr, H.T., Botas, J., Zoghbi, H.Y., 2003.Interaction of Akt-phosphorylated ataxin-1 with 14-3-3 mediates neurodegen-eration in spinocerebellar ataxia type 1. Cell 113, 457–468.

Chico, L.K., Van Eldik, L.J., Watterson, D.M., 2009. Targeting protein kinases incentral nervous system disorders. Nature Reviews Drug Discovery 8, 892–909.

Chung, M.Y., Ranum, L.P., Duvick, L.A., Servadio, A., Zoghbi, H.Y., Orr, H.T., 1993.Evidence for a mechanism predisposing to intergenerational CAG repeat insta-bility in spinocerebellar ataxia type I. Nature Genetics 5, 254–258.

Clark, H.B., Burright, E.N., Yunis, W.S., Larson, S., Wilcox, C., Hartman, B., Matilla, A.,Zoghbi, H.Y., Orr, H.T., 1997. Purkinje cell expression of a mutant allele of SCA1in transgenic mice leads to disparate effects on motor behaviors followed by aprogressive cerebellar dysfunction and histological alterations. Journal of Neu-roscience 17, 7385–7395.

Corsini, L., Bonnal, S., Basquin, J., Hothorn, M., Scheffzek, K., et al., 2007. U2AFho-mology motif interactions are required for alternative splicing regulation bySPF45. Nature Structural & Molecular Biology 14, 620–629.

Cummings, C.J., Mancini, M.A., Antalffy, B., DeFranco, D.B., Orr, H.T., Zoghbi, H.Y.,1998. Chaperone suppression of ataxin-1 aggregation and altered subcellularproteasome localization imply protein misfolding in SCA1. Nature Genetics 19,148–154.

Cummings, C.J., Reinstein, E., Sun, Y., Antalffy, B., Jiang, Y.-H., Ciechanover, A., Orr,H.T., Arthur, L., Beaudet, A.L., Zoghbi, H.Y., 1999. Mutation of the E6-AP ubiquitinligase reduces nuclear inclusion frequency while accelerating polyglutamine-induced pathology in SCA1 transgenic mice. Neuron 24, 879–892.

Cvetanovic, M., Patel, J.M., Marti, H.H., Kini, A.R., Opal, P., 2011. Vascular endothelialgrowth factor ameliorates the ataxic phenotype in a mouse model of spinocer-ebellar ataxia type 1. Nature Medicine 17, 1445–1447.

de Chiara, C., Giannini, C., Adinolfi, S., de Boer, J., Guida, S., Ramos, A., Jodice, C.,Kioussis, D., Pastore, A., 2003. The AXH module: an independently foldeddomain common to ataxin-1 and HBP. FEBS Letters 551, 107–112.

de Chiara, C., Menon, R.P., Strom, M., Gibson, T.J., Pastore, A., 2009. Phosphorylationof S776 and 14-3-3 binding modulate ataxin-1 interaction with splicing factors.PLoS One 4, e8372.

Duvick, L., Barnes, J., Ebner, B., Agrawal, S., Andresen, M., Lim, J., Giesler, G.J., Zoghbi,H.Y., Orr, H.T., 2010. SCA1-like disease in mice expressing wild type ataxin-1with a serine to aspartic acid replacement at residue 776. Neuron 67, 929–935.

Emamian, E.S., Kaytor, M.D., Duvick, L.A., Zu, T., Susan, K., Tousey, S.K., Zoghbi, H.Y.,Clark, H.B., Orr, H.T., 2003. Serine 776 of ataxin-1 is critical for polyglutamine-induced disease in SCA1 transgenic mice. Neuron 38, 375–387.

Fernandez-Funez, P., Rosales, M.L.N., de Gouyon, B., She, W.-C., Luchack, J., Martinez,P., Turiegano, E., Benito, J., Capovilla, M., Skinner, P., McCal, A., Canal, I., Orr, H.T.,Zoghbi, H.Y., Botas, J., 2000. Identification of genes that modify ataxin-1 inducedneurodegeneration. Nature 408, 101–106.

Fryer, J.D., Yu, P., Kang, H., Mandel-Brehm, C., Carter, A., Crespo-Barreto, J., Gao, Y.,Flora, A., Shaw, C., Orr, H.T., Zoghbi, H.Y., 2011. Exercise and genetic rescue ofSCA1 via the transcriptional repressor Capicua. Science 334, 690–693.

Goldowitz, D., Hamre, K., 1998. The cells and molecules that make a cerebellum.Trends in Neuroscience 21, 375–382.

Gehrking, K.M., Andresen, J.M., Duvick, L., Lough, J., Zoghbi, H.Y., Orr, H.T., 2011.Partial loss of Tip60 slows midstage neurodegeneration in a spinocerebellarataxia type 1 (SCA1) mouse model. Human Molecular Genetics 20, 2204–2212.

Haines, J.L., Schut, L.J., Weitkamp, L.R., Thayer, M., Anderson, V.E., 1984. Spinocer-ebelaar ataxia in a large kindred: age at onset, reproduction and genetic linkagestudies. Neurology 34, 1542–1548.

Hartmuth, K., Urlaub, H., Vornlocher, H.P., Will, C.L., Gentzel, M., et al., 2002. Proteincomposition of human prespliceosomes isolated by a tobramycin affinityselection method. Proceedings of the National Academy of Sciences of theUnited States of America 99, 16719–16724.

Huttlin, E.L., Jedrychowski, M.P., Elias, J.E., Goswami, T., Rad, R., Beausoleil, S.A.,Villen, J., Haas, W., Sowa, M.E., Gygi, S.P., 2010. A tissue-specific atlas of mouseprotein phosphorylation and expression. Cell 143, 1174–1189.

Inoue, T., Lin, X., Kohlmeier, K.A., Orr, H.T., Zoghbi, H.Y., Ross, W.N., 2001. Calciumdynamics and electrophysiological properties of cerebellar Purkinje cells inSCA1 transgenic mice. Journal of Neurophysiology 85, 1750–1760.

Irwin, S., Vandelft, M., Howell, J.L., Graczyk, J., Orr, H.T., Truant, R., 2005. RNAassociation and nucleocytoplasmic shuttling by ataxin-1. Journal of Cell Science118, 233–242.

Jackson, J.F., Currier, R.D., Terasaki, P.I., Morton, N.E., 1977. Spinocerebellarataxiaand HLA linkage: risk prediction by HLA typing. The New England Journal ofMedicine 296, 1138–1141.

Jorgensen, N.D., Andresen, J.M., Lagalwar, S., Armstrong, B., Stevens, S., Byam, C.E.,Duvick, L.A., Lai, S., Zoghbi, H.Y., Clark, H.B., Orr, H.T., 2009. Phosphorylation ofATXN1 at Ser776 in the cerebellum. Journal of Neurochemistry 110, 675–686.

Keats, B.J.B., Pollack, M.S., McCall, A., Wilensky, M.A., Ward, L.J., Lu, M., Zoghbi, H.Y.,1991. Tight linkage of the gene for spinocerebellar ataxia to D6S89 on the shortarm of chromosome 6 in a kindred for which close linkage to both HLA andF13A1 is excluded. American Journal of Human Genetics 49, 972–977.

Klement, I.A., Skinner, P.J., Kaytor, M.D., Yi, H., Hersch, S.M., Clark, H.B., Zoghbi, H.Y.,Orr, H.T., 1998. Ataxin-1 nuclear localization and aggregation: role in poly-glutamine-induced disease in SCA1 transgenic mice. Cell 95, 41–53.

Koeppen, A.H., 2005. The pathogenesis of spinocerebellar ataxia. Cerebellum 4, 62–73.Krol, H.A., Krawczk, P.M., Bosch, K.S., Aten, J.A., Hol, E.M., Reits, E.A., 2008. Poly-

glutamine expansion accelerates the dynamics of ataxin-1 and does not resultin aggregate formation. PLoS One 3, e1503.

Lai, S., O’Callaghan, B., Zoghbi, H.Y., Orr, H.T., 2011. 14-3-3 binding to ataxin-1(ATXN1) regulates its dephosphorylation at S776 and transport to the nucleus.Journal of Biological Chemistry 286, 34606–34616.

Lam, Y.C., Bowman, A.B., Jafar-Nejad, P., Lim, J., Richman, R., Fryer, J.D., Hyun, E.,Duvick, L.A., Orr, H.T., Botas, J., Zoghbi, H.Y., 2006. Mutant Ataxin-1 interactswith the repressor Capicua in its native complex to cause SCA1 neuropathology.Cell 127, 1335–1347.

Lee, Y., Samaco, R.C., Gatchel, J.R., Thaller, C., Orr, H.T., Zoghbi, H.Y., 2008. miR-19,miR-101 and miR-130 co-regulate ATXN1 levels to potentially modulate SCA1pathogenesis. Nature Neuroscience 11, 1137–1139.

Lim, J., hao, T., Shaw, C., Patel, A.J., Szabo, G., Ruai, J.-F., Fisk, C.J., Li, N., Smolyar, A.,Hill, D.E., Barabasi, A.-L., Vidal, M., Zoghbi, H.Y., 2006. A protein-protein inter-action network for human inherited ataxias and disorders of Purkinje celldegeneration. Cell 125, 801–814.

Lim, J., Crespo-Barreto, J., Jafar-Nejad, P., Bowman, A.B., Richman, R., Hill, D.E., Orr,H.T., Zoghbi, H.Y., 2008. Opposing effects of polyglutamine expansion on nativeprotein complexes contribute to SCA1. Nature 452, 713–719.

Lin, X., Antalffy, B., Kang, D., Orr, H.T., Zoghbi, H.Y., 2000. Polyglutamine expansiondownregulates specific neuronal genes before pathologic changes in SCA1.Nature Neuroscience 3, 157–163.

Lorenzetti, D., Watase, K., Xu, B., Antalffy, B., Matzuk, M.M., Guos, Q., Wang, P., Orr,H.T., Zoghbi, H.T., 2000. Repeat instability and motor incoordination in micewith a targeted expanded CAG repeat in the Sca1 locus. Human MolecularGenetics 9, 779–785.

Matilla, A., Roberson, E.D., Banfi, S., Morales, J., Armstrong, D.L., Burright, E.N., Orr,H.T., Sweatt, J.D., Zoghbi, H.Y., Matzuk, M., 1998. Mice lacking ataxin-1 displaylearning deficits and decreased hippocampal paired-pulse facilitation. Journalof Neuroscience 18, 5508–5516.

Mitzutani, A., Wang, L., Rajan, H., Vig, P.J., Alaynick, W.A., et al., 2005. EMBO Journal24, 3339–3351.

Morrison, D.K., 2008. The 14-3-3 proteins: integrators of diverse signaling cues thatimpact cell fate and cancer development. Trend in Cell Biology 19, 16–23.

Muller, S., Ledl, A., Schmidt, D., 2004. Oncogene 23, 1998–2008.Orr, H.T., Chung, M.Y., Banfi, S., Kwiatkowski, T.J., Servadio, A., Beaudet, A.L., McCall,

A.E., Duvick, L.A., Ranum, L.P., Zoghbi, H.Y., 1993. Expansion of an unstabletrinucleotide CAG repeat in spinocerebellar ataxia type 1. Nature Genetics 4,221–226.

Rich, S.S., Wilkie, P.J., Schut, L., Vance, G., Orr, H.T., 1987. Spinocerebellar ataxia:localization of an autosomal dominant locus between two markers on humanchromosome 6. American Journal of Human Genetics 41, 524–531.

Riley, B.E., Zoghbi, H.Y., Orr, H.T., 2005. SUMOylation of the polyglutamine protein,ataxin-1 is dependent on a functional nuclear localization signal. Journal ofBiological Chemistry 280, 21942–21948.

Ross, C.A., Poirier, M.A., 2004. Protein aggregation and neurodegenerative disease.Nature Medicine S10–S17.

Schut, J.W., 1950. Hereditary ataxia, a clinical study through six generations.Archives of Neurology and Psychiatry 63, 535–568.

Serra, H.G., Byam, C.E., Lande, J.D., Tousey, S.K., Zoghbi, H.Y., Orr, H.T., 2004. Geneprofiling links SCA1 pathophysiology to glutamate signaling in Purkinje cells oftransgenic mice. Human Molecular Genetics 13, 2535–2543.

Serra, H.G., Duvick, L., Zu, T., Carlson, K., Stevens, S., Jorgensen, N., Lysholm, A.,Burright, E., Zoghbi, H.Y., Clark, H.B., Andresen, J.M., Orr, H.T., 2006. RORa-mediated Purkinje cell development determines disease severity in adult SCA1mice. Cell 127, 697–708.

Smith, R.A., Miller, T.M., Yamanaka, K., Monia, B.P., Condon, T.P., Hung, G., Lobsiger,C.S., Ward, C.M., McAlonis-Downes, M., Wei, H., Wancewicz, E.V., Bennett, C.F.,Cleveland, D.C., 2006. Antisense oligonucleotide therapy for neurodegenerativedisease. Journal of Clinical Investigation 116, 2290–2296.

Tsai, C.C., Kao, H.Y., Mitzutani, A., Banayo, E., Rajan, H., McKeown, M., Evans, R.M.,2004. Ataxin 1, a SCA1 neurodegenerative disorder protein, is functionallylinked to the silencing mediator of retinoid and thyroid hormone receptors.Proceedings of the National Academy of Sciences of the United States of America101, 4047–4052.

Tsuda, H., Jafar-Nejad, H., Patel, A.J., Sun, Y., Chen, H.-K., Rose, M.F., Venken, K.J.T.,Botas, J., Orr, H.T., Bellen, H.J., Zoghbi, H.Y., 2005. The AXH domain in mamma-lian/Drosophila Ataxin-1 mediates neurodegeneration in spinocerebellar ataxia1 through its interaction with Gfi-1/Senseless proteins. Cell 122, 633–644.

Vig, P.J., Wei, J., Shano, Q., Lopez, M.E., Halperin, R., Gerber, J., 2011. Suppression ofcalbindin-D28k expression exacerbates SCA1 phenotype in a disease mousemodel. Cerebellum Nov 11 [Epub ahead of print].

Watase, K., Weeber, E.J., Xu, B., Antalffy, B., Yuva-Paylor, L., Hashimoto, K., Kano, M.,Atkinson, D.L., Sweatt, J.D., Orr, H.T., Paylor, R., Zoghbi, H.Y., 2002. A long CAGrepeat in the mouse Sca1 locus replicates SCA1 features and reveals the impactof protein solubility on selective neurodegeneration. Neuron 34, 905–919.

Xia, H., Mao, Q., Eliason, S.I., Harper, S.Q., Martins, I.H., Orr, H.T., Paulson, H.L., Yang,L., Kotin, R.M., Davidson, B.I., 2004. RNAi suppresses polyglutamine-inducedneurodegeneration in a model of spinocerebellar ataxia. Nature Medicine 10,816–820.

Yakura, H., Waldsaka, A., Fujimoto, S., Itakura, K., 1974. Hereditary ataxia and HLAgenotypes. The New England Journal of Medicine 291, 154–155.

Yue, S., Serra, H.G., Zogbhi, H.Y., Orr, H.T., 2001. The spinocerebellar ataxia type 1protein, ataxin-1, has RNA-binding activity that is inversely affected by thelength of its polyglutamine tract. Human Molecular Genetics 10, 25–30.

Zu, T., Duvick, L.A., Kaytor, M.D., Berlinger, M., Zoghbi, H.Y., Clark, H.B., Orr, H.T.,2004. Recovery from polyglutamine-Induced neurodegeneration in conditionalSCA1 transgenic mice. Journal of Neuroscience 24, 8853–8861.