Identification of a self-association region within the SCA1 gene product, ataxin-1
Identification of a self-association region within the SCA1 gene product, ataxin-1Eric N. Burright1,2, Jennifer D. Davidson2, Lisa A. Duvick2, Beena Koshy4, Huda Y. Zoghbi4,5 and Harry T. Orr1,2,3,*
1Institute of Human Genetics, 2Department of Laboratory Medicine and Pathology and 3Department of Biochemistry, University of Minnesota, Minneapolis, MN 55455, USA, 4Departments of Pediatrics and Molecular and Human Genetics and 5Howard Hughes Medical Institute, Baylor College of Medicine, Houston, TX 77030, USA
Received October 1, 1996;Revised and Accepted January 24, 1997
Spinocerebellar ataxia type 1 (SCA1) is an autosomal dominant neurodegenerative disorder caused by the expansion of a polyglutamine tract within the SCA1 gene product, ataxin-1. Expansion of this tract is believed to result in a gain of function by the mutant protein, perhaps through altered self-associations or interactions with other cellular proteins. We have used the yeast two hybrid system to determine if ataxin-1 is capable of multimerization. This analysis revealed that ataxin-1 does have the ability to self-associate, however, this association does not appear to be influenced by expansion of the polyglutamine tract. Consistent with this finding, deletion analysis excluded the involvement of the polyglutamine tract in ataxin-1 self-association, and instead localized the multimerization region to amino acids 495-605 of the wild type protein. These results, while identifying an ataxin-1 self-interaction region, fail to support a proposed model of polar-zipper mediated multimerization involving the ataxin-1 polyglutamine tract.
Spinocerebellar ataxia type 1 (SCA1) is a progressive neurological disorder characterized by the degeneration of cerebellar Purkinje cells and the selective loss of neurons within the brainstem and spinocerebellar tracts (1 ). The mutational basis of SCA1 was determined to be the expansion of an unstable CAG trinucleotide repeat, encoding glutamine, within the SCA1 gene product (2 ,3 ). This repeat is highly polymorphic within the normal population, ranging in size between 6 and 40 repeat units, and is expanded in length, ranging from 40 to 83 units, in SCA1-affected individuals (2 ,4 -7 ). The onset and severity of symptoms in SCA1 patients is variable and correlates with the length of the CAG repeat on the mutant allele (1 ,2 ,5 -7 ). Juvenile onset cases follow a more rapid progression of symptoms and are associated with the largest repeat expansions. The SCA1 gene product, ataxin-1, is a novel protein which contains no known functional motifs and has no homologies with previously identified molecules (2 ,3 ).
SCA1, like other CAG repeat expansion disorders, is believed to be caused by a dominant gain of function by the mutant protein containing the expanded polyglutamine tract (8 ). We have recently reported the establishment of transgenic mice that express human SCA1 cDNA transgenes with either a normal or an expanded CAG tract within the cerebellar Purkinje cells (9 ). While transgenic animals expressing the unexpanded SCA1 allele had morphologically and functionally normal Purkinje cells, transgenic animals expressing the expanded SCA1 allele developed ataxia and Purkinje cell degeneration, demonstrating a gain of function mutational mechanism. The molecular basis of how expanded polyglutamine tracts within proteins lead to specific neuronal loss is yet to be determined. Perutz and colleagues (10 ) have shown by molecular modeling that poly-(L-glutamine) stretches form [beta]-pleated sheets that might act as polar zippers between proteins, linking [beta]-strands by hydrogen bonds between their main-chain and side-chain amides. These investigators speculated that pathologic effects might arise if extended repeats cause proteins to acquire high affinity for each other and perhaps, for other non-specific proteins.
We used the yeast two hybrid system (11 ,12 ) to test whether ataxin-1 has the ability to self-associate and to determine if such an interaction is influenced by glutamine repeat expansion. Using this approach, we now show that ataxin-1 does have the ability to self-associate. This interaction does not, however, appear to be influenced by glutamine repeat expansion. Consistent with this finding, we have mapped the multimerization domain to amino acids 495-605 of the wild type protein, a region that does not contain the polyglutamine tract. These results suggest that the polyglutamine tract within ataxin-1 does not promote polar zipper-mediated multimerization of the ataxin-1 protein.
To test for ataxin-1 self-association, the yeast two hybrid system was employed. First, two sets of fusion proteins were constructed; one generates a hybrid between sequences encoding the GAL4 DNA binding domain (13 ) and the ataxin-1 protein. The second plasmid contains sequences for a hybrid between the GAL4 activation domain (14 ) and the ataxin-1 protein. If ataxin-1 is able to self-associate when expressed in yeast, the resulting complex of fusion proteins will reconstitute the ability of GAL4 to activate transcription from reporter genes containing its DNA binding sites within their promoter elements.
To test for intrinsic activation activity of the fusion constructs, DNA binding domain and activation domain plasmids containing ataxin-1 sequences were singly transformed into yeast strain HF7c and tested for both lacZ and HIS3 reporter gene activities. In addition, these constructs were transformed in combination with plasmids encoding non-specific proteins including SV40 T antigen, yeast proteins SNF1 and SNF4, human lamin C, and murine CDK2 and p53. None of the ataxin-1 fusion contructs had intrinsic activation activity or interacted non-specifically with these unrelated proteins as demonstrated by the lack of reporter gene activity (Table 1 ).
. Detection of ataxin-1 self-association using the yeast two hybrid system
DNA binding domain fusion construct
Activation domain fusion construct
[beta]-gal activity (filter assay)
Histidine selection
ataxin-1 [30]
-
white
no growth
ataxin-1 [82]
-
white
no growth
m. ataxin-1
-
white
no growth
-
ataxin-1 [30]
white
no growth
-
ataxin-1 [82]
white
no growth
ataxin-1 [30]
SV40 T
white
no growth
ataxin-1 [30]
SNF4
white
no growth
ataxin-1 [82]
SV40 T
white
no growth
ataxin-1 [82]
SNF4
white
no growth
m. ataxin-1
SV40 T
white
no growth
lamin C
ataxin-1 [30]
white
no growth
CDK2
ataxin-1 [30]
white
no growth
SNF1
ataxin-1 [30]
white
no growth
lamin C
ataxin-1 [82]
white
no growth
CDK2
ataxin-1 [82]
white
no growth
SNF1
ataxin-1 [82]
white
no growth
ataxin-1 [30]
ataxin-1 [30]
blue
growth
ataxin-1 [30]
ataxin-1 [82]
blue
growth
ataxin-1 [82]
ataxin-1 [30]
blue
growth
ataxin-1 [82]
ataxin-1 [82]
blue
growth
m. ataxin-1
ataxin-1 [30]
blue
growth
m. ataxin-1
ataxin-1 [82]
blue
growth
SNF1
SNF4
blue
growth
p53
SV40 T
blue
growth
The HF7 reporter strain was transformed with the indicated GAL4 DNA binding and/or activation domain fusion constructs. To assay for [beta]-galactosidase activity, individual transformants were grown on synthetic medium plates lacking leucine and/or tryptophan for 2 or 3 days at 30oC; [beta]-galactosidase activity was determined qualitatively by a filter assay; blue or white indicates the presence or absence of activity, respectively. To assay for histidine prototrophy, individual transformants were streaked onto synthetic medium plates lacking histidine and incubated for 3 or 4 days. GAL4 DNA binding domain and activation domain fusion constructs contained full length human ataxin-1 with either 30 or 82 glutamine residues (ataxin-1 [30] and [82], respectively) or truncated murine ataxin-1 (m. ataxin-1) containing only two glutamine residues in the corresponding region. Control GAL4 DNA binding domain and activation domain fusion constructs encoding non-specific proteins included SV40 T antigen, yeast proteins SNF1 and SNF4, human lamin C, and murine CDK2 and p53.
Three types of ataxin-1 sequences were used to examine ataxin-1 self-association; full length human ataxin-1 containing a wild type, unexpanded repeat with an overall length of 30 (ataxin-1 [30]), full length human ataxin-1 with a mutant, expanded repeat with an overall length of 82 (ataxin-1 [82]), and a nearly full length murine ataxin-1 which contains only two glutamine residues in the region corresponding to the polyglutamine tract in human ataxin-1 (15 ). Both [beta]-galactosidase activity and histidine prototrophy were observed in yeast transformed with DNA binding and activation domain plasmids each harboring wild type ataxin-1 [30] sequences, indicating ataxin-1 self-association (Table 1 ). This ability to multimerize was maintained when co-transformations were performed with ataxin-1 [30] and [82], and also ataxin-1 [82] and [82] constructs. Furthermore, murine ataxin-1 also had the ability to interact with human ataxin-1 containing either the unexpanded or expanded polyglutamine tract. These results suggest that the polyglutamine tract within ataxin-1 is not likely to be involved in the formation of ataxin-1 multimers.
To determine if expansion of the polyglutamine tract affected the affinity of ataxin-1 self-interactions, we performed quantitative o-nitrophenyl-[beta]-galactopyranoside (ONPG) assays. GAL4 DNA binding domain or activation domain fusion constructs containing ataxin-1 sequences singly transformed into SFY526 cells had negligible levels of [beta]-galactosidase activity (Table 2 ). In contrast, yeast co-transformed with various combinations of the ataxin-1 [30] and [82] fusion constructs produced 8-16 units of [beta]-galactosidase activity. As a reference for comparison, ONPG assays performed on SFY526 cells co-transformed with constructs encoding known interactors, murine p53 and SV40 T antigen, produced ~350 units of [beta]-galactosidase activity. These results indicate that the affinity for ataxin-1 self-association is ~20- to 30-fold less than that of the murine p53 and SV40 T antigen interaction and is not significantly influenced by polyglutamine repeat expansion.
. Quantitative ONPG assays of ataxin-1 self-association
DNA binding domain fusion construct
Activation domain fusion construct
[beta]-gal units (ONPG assay)
h. ataxin-1 [30]
-
0.007 ± 0.0005
h. ataxin-1 [82]
-
0.03 ± 0.01
-
h. ataxin-1 [30]
0.05 ± 0.02
-
h. ataxin-1 [82]
0.06 ± 0.03
h. ataxin-1 [30]
h. ataxin-1 [30]
13.9 ± 6.8
h. ataxin-1 [30]
h. ataxin-1 [82]
8.8 ± 9.2
h. ataxin-1 [82]
h. ataxin-1 [30]
16.0 ± 7.2
h. ataxin-1 [82]
h. ataxin-1 [82]
9.5 ± 4.3
p53
SV40 T
356.0 ± 42.1
The SFY526 reporter strain was transformed with the indicated fusion constructs. [beta]-galactosidase activity was measured by ONPG assays and is reported in units as described previously (22).
To identify ataxin-1 sequences required for the observed self-association, a set of 13 ataxin-1 deletion constructs were prepared in the GAL4 DNA binding domain plasmid (Fig. 1 ). HF7c cells were co-transformed with each of the deletion constructs along with the GAL4 transcriptional activation domain plasmid containing the full length ataxin-1 [30] coding sequence and assayed for [beta]-galactosidase activity and histidine prototrophy. The smallest truncated form of ataxin-1 able to associate with the full length ataxin-1 protein contained amino acids 495-605. All six clones containing this region produced both lacZ and HIS3 reporter gene activity. To further demonstrate that amino acids 495-605 are sufficient for ataxin-1 self-association, sequences encoding this region were placed in the activation domain fusion construct and co-transformed with the 495-605 DNA binding construct into HF7c cells. The resulting co-transformants also produced detectable levels of reporter gene activity indicating residues 495-605 are sufficient to mediate ataxin-1 self-association. Efforts to further define this domain using deletion analyses resulted in the loss of reporter gene activity, suggesting this entire region is either directly involved in the protein-protein interaction, or is necessary to allow the truncated ataxin-1 assume the conformation necessary for such an interaction to occur.
In an effort to identify proteins that interact with ataxin-1, we performed yeast two hybrid screens of an adult human brain cDNA library (Clontech Inc.) expressed as GAL4 activation domain fusion proteins. HF7c cells were sequentially transformed with a fusion construct encoding the GAL4 DNA binding domain and either ataxin-1 [30] or ataxin-1 [82] sequences, followed by the GAL4 activation domain fusion library. Of ~2.5 × 106 co-transformants plated under tryptophan, leucine and histidine selective conditions, 171 colonies displayed histidine prototrophy (Table 3 ). One hundred and forty-two of these Trp+, Leu+, His+ colonies were also positive for [beta]-galactosidase activity using a filter assay (see Materials and Methods). The activation domain plasmids containing the rescued library cDNAs from these 142 His+, lacZ+ clones were recovered by transformation into Escherichia coli strain HB101. Plasmids containing identical inserts were determined by comparative restriction analysis and cross-hybridization studies. A representative plasmid from each group was tested for specific interaction with ataxin-1 by transformation singly and in combination with DNA binding domain fusion constructs encoding ataxin-1 [30], ataxin-1 [82], lamin C, CDK2, SNF1 and p53 sequences. Plasmids representing 140 of the original 142 His+, lacZ+ clones produced reporter gene activity only when co-transformed with ataxin-1 fusion constructs indicating specific interaction with ataxin-1. Eighteen of these 140 clones (12.8%) contained ataxin-1 sequences, all of which included sequences encoding amino acids 495-605 of the wild type protein (3 ). These findings provide additional evidence of ataxin-1 self-association, and are consistent with the involvement of amino acids 495-605 in mediating this interaction.
The remaining 122 clones represented 21 different rescued cDNAs encoding proteins that specifically interact with ataxin-1 using the yeast two hybrid system. ONPG assays indicated that none of these cDNAs encoded a protein which differentially interacted with wild type and mutant ataxin-1. Analysis of partial DNA sequence obtained for each clone failed to reveal significant homologies to previously characterized molecules. Further characterization of these clones is ongoing.
Summary of yeast two hybrid screens of a human cDNA library
DNA binding domain fusion construct
Number of transformants screened
Number of histidine prototrophs
Number of [beta]-gal positive clones
Number of specific clones
Number of ataxin-1 clones
ataxin-1 [30]
1.0 × 106
75
61
59
9
ataxin-1 [82]
1.5 × 106
96
81
81
9
Combined totals
2.5 × 106
171
142
140
18
An adult human brain cDNA library was screened with the indicated GAL4 DNA binding domain/ataxin-1 fusion constructs. Transformants surviving histidine selection were assayed for [beta]-galactosidase activity. The activation domain plasmids from His+, LacZ+ clones were isolated and tested for specific interaction with ataxin-1. Those clones encoding ataxin-1 are also indicated.
Ataxin-1 is a novel protein with no known regions of homology to other previously characterized molecules (2 ,3 ). Comparison of the human (2 ,3 ), murine (15 ) and rat (16 ) homologs of SCA1 indicate that the sequence is both highly and nearly uniformly conserved across the entire coding region (~90% amino acid identity overall), thereby revealing little information with regard to its putative functional domains. Using the yeast two hybrid system, we have detected the ability of ataxin-1 to self-associate and have mapped the domain mediating this interaction to amino acids 495-605 of the human wild type protein (3 ). We have also identified a number of ataxin-1 clones, each containing sequences encoding amino acids 495-605, in two hybrid screens of a human brain cDNA library using full length ataxin-1 clones as bait. This finding provides additional evidence of ataxin-1 self-association and supports amino acids 495-605 as the minimal region involved. Interestingly, the identified interaction domain is extremely highly conserved. A significant portion of this region, amino acids 513-605, differs from the corresponding regions of rat and mouse ataxin-1 by only one and four amino acids, respectively. These findings define a functional segment within ataxin-1 and suggest that self-association may be important for wild type activity.
SCA1 is one of several adult-onset neurodegenerative disorders caused by the expansion of a polyglutamine stretch within the affected protein (for review, see 8 ). Although a variety of evidence suggests that these disorders share a common pathway of neuronal degeneration, the molecular mechanism leading to cell death is yet to be determined. Perutz and colleagues (10 ) have proposed that the polyglutamine tracts within proteins may form [beta]-strands and function as polar zippers that promote protein oligomerization. These investigators speculated that the cell death and pathogenic effects seen in neurodegenerative diseases might arise if extended repeats cause proteins to self-aggregate or associate with other non-specific proteins. This model provides a plausible explanation of how longer repeat tracts might cause early onset of disease through increased affinity of the proteins for each other or for other proteins with glutamine repeats. Support for this polar zipper hypothesis includes molecular modeling using polyglutamine peptides demonstrating the formation of [beta]-sheets held together by hydrogen bonds (10 ) and the detection of glutamine repeat-mediated oligomerization of chymotrypsin inhibitor 2 (17 ). At present, however, there is a lack of in vivo evidence supporting this model.
We have used the yeast two hybrid system as an in vivo test for ataxin-1 self-association and to examine the possibility that such an interaction might be influenced by polyglutamine repeat length. Although the data presented here indicate that ataxin-1 has the ability to self-associate, the sequences involved in this interaction do not include the polyglutamine tract. Furthermore, the affinity of this association does not appear to be influenced by polyglutamine repeat length. These data suggest that ataxin-1 does not associate via its polyglutamine tracts when expressed in yeast, but does not formally rule out such an interaction in human neurons. The polar zipper model proposes that expanded polyglutamine repeats cause protein multimerization leading to the accumulation of that protein, and ultimately, to neuronal death. This possibility seems unlikely considering the inability to detect alterations in polyglutamine protein levels and/or subcellular distribution using immunohistochemical staining of SCA1 patient (18 ), SCA1 transgenic animal (Servadio and Zoghbi, unpublished data), and Huntington disease patient necropsy material (19 ). Therefore, the present data suggest that the aggregation of proteins containing expanded polyglutamine tracts is not an element of the pathogenic mechanism involved in these neurodegenerative disorders.
Growth and manipulation of yeast strains were done according to standard procedures (20 ). All experiments involving the two-hybrid system were performed in strains HF7c [MATa, ura3-52, his3-200, ade 2-101, lys 2-801, trp 1-901, leu 2-3, 112, gal4-542, gal80-538, LYS2::GAL1-HIS3, URA3::(GAL4 17mers)3-CYC-lacZ] and SFY526 (MATa, ura3-52, his3-200, ade 2-101, lys 2-801, trp 1-901, leu 2-3, 112, canr, gal4-542, gal80-538, URA3::GAL1-lacZ). Yeast plasmids pGBT9, pGAD424, pLAM5' (human lamin C), pVA3 (murine p53), and pTD1 (SV40 T antigen) were from Clontech Inc. The coding sequence of human ataxin-1 was amplified from clone 31-5 (3 ) using PCR with primers containing EcoRI and HindIII recognition sequences (GCGAATTCATGAAATCCAAC and CGAAGCTTCTACTTGCCTACATT) and directionally subcloned into pBluescript KS+ (pBS) (Stratagene). A human ataxin-1 cDNA containing an expanded CAG tract was constructed by exchanging the SfiI fragment harboring the 30 repeats with a fragment containing an 82 repeat tract as described (9 ). The full length ataxin-1 [30] and [82] cDNAs were recovered from pBS as EcoRI-SalI fragments and directionally cloned into pGBT9 (containing amino acids 1-147 of the DNA binding domain of GAL4) and pGAD424 (containing amino acids 768-881 of the activation domain of GAL4). Ataxin-1 cDNA deletion fragments were prepared from pBluescript clones and subcloned into pGBT9 using standard methods (21 ). EcoRI-BamHI, EcoRI/PstI and EcoRI/StuI fragments containing cDNAs encoding ataxin-1 amino acids 1-702, 1-316 and 1-176 were subcloned into EcoRI and BamHI, PstI or SmaI prepared pGBT9, respectively. cDNAs containing ataxin-1 amino acids 176-816 and 176-702 were recovered as StuI/SalI and StuI/BamHI fragments, respectively, and subcloned into pGBT9 sequentially digested with EcoRI, blunted with mung bean nuclease, and then digested with either SalI or BamHI. Ataxin-1 deletion clone containing amino acids 176-549 was prepared by subcloning the StuI/NcoI (blunt-ended with Klenow) into pGBT9 digested with EcoRI and blunted with mung bean nuclease. The pGBT9 clones containing ataxin-1 amino acids 428-549 and 428-702 were constructed by subcloning the BstUI-SmaI or BstUI-BamHI ataxin-1 fragments into vector prepared by EcoRI digestion followed by blunting with mung bean nuclease, sequentially followed by SmaI or BamHI digestion, respectively. The pGBT9 construct containing ataxin-1 amino acids 549-702 was made by subcloning the StuI-BamHI ataxin-1 fragment into pGBT9 sequentially digested with EcoRI, blunted with mung bean nuclease, and digested with BamHI. Inserts of the remaining deletion clones (containing amino acids 495-605, 495-576, 529-605 and 529-576) were amplified by PCR using a 5' primer containing an EcoRI recognition sequence (RI-495, ATAGAATTCGTCGGCAGCACTGAC; or RI-529, ATAGAATTCCCCAAGAGCGAGAAC) in combination with a 3' primer containing a BamHI recognition sequence (B-605, AATGGATCCTATCTCTGCACTCTG; or B-576, AATGGATCCCATGAAGTAGGGAGG). The amplified products were then digested with EcoRI and BamHI and subcloned into EcoRI/BamHI prepared pGBT9. A murine ataxin-1 cDNA containing amino acids 13-738 (15 ) was recovered as a BsmI-BamHI fragment from clone 19a (Banfi and Zoghbi, unpublished data) and subcloned into EcoRI (mung bean nuclease blunted)/BamHI digested pGBT9. Sequence analysis was performed on all two hybrid fusion constructs to confirm maintenance of the proper reading frame. DNA binding domain plasmids pAS1-CDK2 and pAS1-SNF1 and activation domain plasmid pACT-SNF4 were generously provided by S. J. Elledge.
HF7c cells were sequentially transformed with the pGBT9 plasmid containing ataxin-1 [30] or [82] sequences followed by an adult human brain cDNA library in pGAD10 (Clontech Inc.) with conditions recommended by the manufacturer. Following the second transformation step, cells were allowed to recover by growth in YPD non-selective media and then plated on synthetic medium plates lacking tryptophan, leucine and histidine. As low basal levels of HIS3 expression in the HF7c host strain were observed, 25 mM 3-aminotriazole (3-AT) was added to the selection medium plates to suppress growth of transformants containing non-interacting proteins. Transformants (~1-1.5 × 106) were plated in each library screen. [beta]-Galactosidase filter assays were performed on histidine prototrophs as described below. The library plasmid was recovered from lacZ+ His+ colonies by transformation into E.coli strain HB101. Specific interaction of the recovered clone with ataxin-1 was tested by a series of yeast co-transformations with GAL4 DNA binding domain plasmids containing ataxin-1 or unrelated, non-specific proteins (lamin C, CDK2, SNF1 and p53). True positives were confirmed by their ability to activate reporter gene activity only when co-transformed with plasmids encoding ataxin-1.
Yeast harboring GAL4 DNA binding and activation fusion proteins were monitored for [beta]-galactosidase activity using plate and liquid assay methods. Yeast transformants were transferred to Whatmann No. 3 filter paper, permeabilized in liquid nitrogen, and placed on Whatmann No. 3 paper saturated with Z-buffer (60 mM Na2HPO4, 40 mM NaH2PO4, 10 mM KCl, 1 mM MgSO4, 50 mM [beta]-mercaptoethanol) containing 1.0 mg/ml 5-bromo-4-chloro-3-indolyl [beta]-D-galactopyranoside at 30oC. Positive colonies appeared in 10 min to 12 h. Liquid o-nitrophenyl-[beta]-galactopyranoside (ONPG) assays were performed to quantitate two hybrid fusion protein interactions. SFY526 cells containing the plasmids of interest were grown overnight in SD media lacking the appropriate amino acids to maintain selective pressure. Two milliliters of this culture was used to inoculate 10 ml of YPD non-selective medium. Cells were grown at 30oC to OD600 of 0.5-1.0, pelleted, washed in Z-buffer, and resuspended in 300 µl Z-buffer. Cells were then permeabilized in liquid nitrogen, thawed, mixed with 700 µl Z-buffer containing 0.5 mg/ml ONPG, and incubated at 30oC for between 15 min and 12 h. Reactions were stopped by the addition of 400 µl Na2CO3, and the OD420 of the supernatant measured. These assays were performed on a single day, in triplicate, on four independent colonies of each of the co-transformants to control for experimental variation. [beta]-Galactosidase units were calculated according to Miller (22 ).
This work was supported by grants from the NINDS, NIH to H.T.O. (NS22920), to H.Y.Z. (NS27699) and to E.N.B. (NS09724-01). H.Y.Z. is an investigator with the Howard Hughes Medical Institute.
1 Zoghbi, H.Y. and Orr, H.T. (1995) Spinocerebellar ataxia type 1. Sem. Cell Biol., 6, 29-35.
2 Orr, H.T., Chung, M.-y., Banfi, S., Kwiatkowski, T.J., Jr., Servadio, A., Beaudet, A.L., McCall, A.E., Duvick, L.A., Ranum, L.P.W. and Zoghbi, H.Y. (1993) Expansion of an unstable trinucleotide CAG repeat in spinocerebellar ataxia type 1. Nature Genet., 4, 221-226.MEDLINE Abstract
3 Banfi, S., Servadio, A., Chung, M.-y., Kwiatkowski, T.J., Jr., McCall, A.E., Duvick, L.A., Shen, Y., Roth, E.J., Orr, H.T. and Zoghbi, H.Y. (1994) Identification and characterization of the gene causing type 1 spinocerebellar ataxia. Nature Genet.,7, 513-519.MEDLINE Abstract
4 Chung, M.-y, Ranum, L.P.W., Duvick, L.A., Servadio, A., Zoghbi, H.Y. and Orr, H.T. (1993) Evidence for a mechanism predisposing to intergenerational CAG repeat instability in spinocerebellar ataxia type 1. Nature Genet., 5, 254-258.
5 Matilla, T., Volpini, V., Genis, D. Rosell, J., Corral, J., Davalos, A., Molins, A. and Estivill, X. (1993) Presymptomatic analysis of spinocerebellar ataxia type 1 (SCA1) via the expansion of the SCA1 CAG-repeat in a large pedigree displaying anticipation and parental male bias. Hum. Mol. Genet., 2, 2123-2128.MEDLINE Abstract
6 Ranum, L.P.W., Chung, M.-y., Banfi, S., Bryer, A., Schut, L.J., Ramesar, R., Duvick, L.A., McCall, A.E., Subramony, S.H., Goldfarb, L., Gomez, C., Sandkuijl, L.A., Orr, H.T. and Zoghbi, H.Y. (1994) Molecular and clinical correlations in spinocerebellar ataxia type 1 (SCA1): evidence for familial effects on the age of onset. Am. J. Hum. Genet., 55, 244-252.
7 Jodice, C., Malaspina, P., Persichetti, F., Noveletto, A., Spadaro, M., Giunti, P., Morocutti, C., Terrenato, L., Harding, A.E. and Frontali, M. (1994) Effect of trinucleotide repeat length and parental sex on phenotypic variation in spinocerebellar ataxia 1. Am. J. Hum. Genet., 54, 959-965.MEDLINE Abstract
8 Paulson, H.L. and Fischbeck, K.H. (1996) Trinucleotide repeats in neurogenetic disorders. Annu. Rev. Neurosci., 19, 79-107.MEDLINE Abstract
9 Burright, E.N., Clark, H.B., Servadio, A., Matilla, T., Feddersen, R.M., Yunis, W.S., Duvick, L.A., Zoghbi, H.Y. and Orr, H.T. (1995) SCA1 transgenic mice: A model for neurodegeneration caused by an expanded CAG trinucleotide repeat. Cell,82, 937-948.MEDLINE Abstract
10 Perutz, M.F., Johnson, T., Suzuki, M. and Finch, J.T. (1994) Glutamine repeats as polar zippers: Their possible role in inherited neurodegenerative diseases. Proc. Natl. Acad. Sci. USA,91, 5535-5538.
11 Fields, S. and Song, O. (1989) A novel genetic system to detect protein-protein interactions. Nature,340, 245-246.MEDLINE Abstract
12 Chien, C.-T., Bartel, P.L., Sternglanz, R. and Fields, S. (1991) A method to identify and clone genes for proteins that interact with a protein of interest. Proc. Natl. Acad. Sci. USA,88, 9578-9582.
13 Keegan, L., Gill, G. and Ptashne, M. (1986) Separation of DNA binding from the transcription-activating function of a eukaryotic regulatory protein. Science,231, 699-704.MEDLINE Abstract
14 Ma, J. and Ptashne, M. (1987) Deletion analysis of GAL4 defines two transcriptional activating sequences. Cell,48, 847-853.MEDLINE Abstract
15 Banfi, S., Servadio, A., Chung, M.-Y., Capozzoli, F., Duvick, L.A., Elde, R., Zoghbi, H.Y. and Orr, H.T. (1996) Cloning and developmental expression analysis of the murine homolog of the spinocerebellar ataxia type 1 gene (Sca1). Hum. Mol. Genet.,5, 33-40.
16 Gossen, M., Schmitt, I., Obst, K., Wahle, P., Epplen, J. T. and Riess, O. (1996) cDNA cloning and expression of rsca1, the rat counterpart of the human spinocerebellar ataxia type 1 gene. Hum. Mol. Genet.,5, 381-389.MEDLINE Abstract
17 Stott, K., Blackburn, J.M., Butler, P.J.G. and Perutz, M. (1995) Incorporation of glutamine repeats makes protein oligomerize: Implications for neurodegenerative diseases. Proc. Natl. Acad. Sci. USA, 92, 6509-6513.MEDLINE Abstract
18 Servadio, A., Koshy, B., Armstrong, D, Anatakfy, B., Orr, H.T. and Zoghbi, H.Y. (1995) Expression analysis of the ataxin-1 protein in tissues from normal and spinocerebellar ataxia type 1 individuals. Nature Genet.,10, 94-98.MEDLINE Abstract
19 Schilling, G., Sharp, A.H., Loev S.J., Wagster, M.V., Li, S.-H., Stine, O.C. and Ross, C.A. (1995) Expression of the Huntington's disease (IT15) protein product in HD patients. Hum. Mol. Genet.,4, 1365-1371.MEDLINE Abstract
20 Sherman, F., Fink, G.R., and Hicks, J.B. (1983) Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
21 Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
22 Miller, J.H. (1972) Experiments in Molecular Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
*To whom correspondence should be addressed
This page is maintained by OUP admin. Last updated Thu Mar 13 17:29:21 GMT 1997. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
