Human Molecular Genetics, 2000, Vol. 9, No. 1 69-78
© 2000 Oxford University Press
Cell cycle arrest enhances the in vitro cellular toxicity of the truncated Machado–Joseph disease gene product with an expanded polyglutamine stretch
Department of Neurology, Institute of Clinical Medicine, University of Tsukuba, Tsukuba 305-8575, Japan, 1Department of Neurology, University of Tokyo Hospital, Tokyo 113-8655, Japan and 2Department of Molecular Cell Physiology, Tokyo Metropolitan Institute of Medical Science, Tokyo 113-8613, Japan
Received 28 July 1999; Revised and Accepted 15 October 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Machado–Joseph disease (MJD) is an inherited neurodegenerative disorder caused by the expansion of the polyglutamine stretch in the MJD gene-encoded protein, ataxin-3. Using a series of deletion constructs expressing ataxin-3 fragments with expanded polyglutamine stretches, we observed aggregate formation and cell death in cultured BHK-21 cells. The cytotoxic effect of N-terminal-truncated ataxin-3 with the expanded polyglutamine tract was enhanced under serum starvation culture, in which cells were arrested in the G0/G1 phase. Coexpression of p21waf1/cip1/sdi1, a cyclin–Cdk inhibitor that induced cell cycle arrest in the G1 phase, also increased the cell death susceptibility produced by the mutant ataxin-3 fragment in BHK-21 cells. The elevated susceptibility to cell death in the G0/G1 phase was confirmed in nerve growth factor-treated, postmitotic neuronal PC12 cells compared with undifferentiated proliferating PC12 cells. These results strongly suggest that the cellular toxicity of truncated ataxin-3 with an expanded polyglutamine stretch is enhanced by cell cycle arrest in the G0/G1 phase. Mutant ataxin-3 may confer a higher susceptibility to cell death on cells in the G0/G1 phase.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Since identification of the expanded CAG repeat in the androgen receptor cDNA as a pathological gene for spinobulbar muscular atrophy (SBMA) (1), an increasing number of hereditary neurodegenerative disorders have been found to be caused by expansion of the CAG repeat within the protein-coding region of genes (2). Because the CAG repeat encodes a polyglutamine stretch, these disorders are now characterized as polyglutamine diseases (3). In addition to the resemblances in the structural abnormality of genes, a number of similarities have been described among polyglutamine diseases, suggesting the presence of common pathological processes (4). At least four major features appear essential to the pathogenesis of these diseases.
First is the high susceptibility to cell death observed in neurons. Although it has been shown that the mutant protein containing an expanded polyglutamine stretch is expressed throughout the body (5,6), apparent cell death has been described only in neurons.
Second is the selectivity of cell death within neuronal populations. Despite the wide expression of the mutant protein in the nervous system, distinct subsets of neurons have been reported to be affected (7), suggesting that different neuronal selectivity in individual polyglutamine diseases underlies variations in clinical signs and symptoms.
Third is the presence of ubiquitin-positive neuronal intranuclear inclusions (NIIs) described in affected neurons of human patients and various transgenic animal models (8–12). Because neurons possessing NIIs are distributed in the most susceptible areas observed in each of these diseases, NII formation is assumed to be linked to the pathophysiology of cell death (13–16). Several studies have demonstrated intranuclear and cytoplasmic aggregates and cell death caused by the expression of truncated forms of mutant proteins in cultured cells (17–22). Since the mutant protein is reportedly cleaved by caspases in vitro (23,24), these data appear to support the hypothesis that the release of the polyglutamine-containing fragment by proteolytic processing is a prerequisite for cells undergoing death (17). It remains unclear, however, whether the generation of aggregates and/or NIIs is the cause or result of cytotoxicity. Several recent papers suggest that nuclear translocation of mutant proteins rather than aggregate formation is essential to cell death (11,25).
Fourth is the delay in neuronal cell death compared with mutant protein expression. In transgenic animal models, a time lag has been reported between protein expression onset and the first neurological deficit (12,17,26–28), apparently corresponding to the delayed onset of polyglutamine diseases. All of these features strongly suggest the presence of a common pathway that causes cells to undergo death.
Machado–Joseph disease (MJD; spinocerebellar ataxia-3) is an autosomal-dominant spinocerebellar ataxia classified as a polyglutamine disease. Ataxin-3, the protein coded by the MJD gene, has a polyglutamine stretch in the C-terminus (29), with 10–40 repeats in normal subjects (30, and unpublished data). In MJD patients, this repeat range is expanded to 55–84 (30). Similar to other polyglutamine diseases, cell death is confined to unique subsets of neurons (6) and NIIs are present in affected neurons (9). A number of papers have reported the aggregate formation and cell death by truncated ataxin-3 with an expanded polyglutamine stretch in vitro (9,17,31) and in vivo (12,17).
In the present study, using in vitro cell culture, we studied the influence of the size of normal and mutant ataxin-3 proteins on aggregate formation and cytotoxicity, and found that the truncated form with an expanded polyglutamine stretch induced cell death more effectively. We focused on the effects of cell cycle arrest on the toxicity of the truncated mutant ataxin-3 protein. Cell cycle arrest in non-neuronal BHK-21 cells by serum deprivation and by the coexpression of p21waf1/cip1/sdi1 (32), a cyclin–Cdk inhibitor that arrested the cell cycle in the G1 phase (33), both increased cellular toxicity of truncated mutant ataxin-3. Using the PC12 rat pheochromocytoma cell line as a neuronal model, we confirmed that the toxicity of truncated mutant ataxin-3 was enhanced in NGF-treated growth-arrested neuronal PC12 cells compared with undifferentiated proliferating PC12 cells. These data suggest that cellular toxicity of truncated ataxin-3 with an expanded polyglutamine stretch is influenced by the cell cycle and supports the hypothesis that arrested cells in the G0/G1 phase are more susceptible than continuously growing cells to cell death produced by the mutant ataxin-3 fragment.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Aggregate formation induced by full-length or N-terminal-truncated ataxin-3 (Q77)
To evaluate how the protein size of ataxin-3 proteins with normal and expanded polyglutamine stretches influences the aggregate formation and induction of the apoptotic phenotype, we created a series of vectors expressing full-length or N-terminal-truncated ataxin-3 proteins and the C-terminal myc epitope (Fig. 1). These vectors encode proteins possessing either 10 or 77 polyglutamine repeats (Q10 or Q77, respectively), each of which was derived from the normal or mutated allele of one of our MJD patients. BHK-21 cells were transiently transfected with these vectors and the expression of recombinant proteins studied using immunohistochemistry with an anti-myc antibody. Cells transfected with pMJD-FL-Q10-myc encoding full-length ataxin-3 with a wild-type polyglutamine tract showed a strong signal for the myc epitope (Fig. 2a). Approximately 80% of these cells showed both cytoplasmic and nuclear immunoreactivity (data not shown), suggesting transport of the recombinant protein into the nucleus. Similar staining patterns were obtained in cells expressing the series of N-terminal-truncated ataxin-3 proteins with Q10 by anti-myc immunohistochemistry (data not shown). Despite intense immunofluorescence signals, no aggregates formed in cells expressing full-length (Fig. 2a) or N-terminal-truncated ataxin-3 proteins with Q10 (data not shown).
|
|
Most cells expressing full-length or N-terminal-truncated ataxin-3 proteins with an expanded polyglutamine stretch (Q77) showed both cytoplasmic and nuclear immunostaining for myc (Fig. 2c–p). In addition, myc-positive cells had distinct aggregate bodies in the cytoplasmic, perinuclear and nuclear areas (Fig. 2e, g, i, k, m and o).
We determined the frequency of intracellular aggregate formation in these cells expressing full-length or truncated ataxin-3 with Q77 (Fig. 3a). The frequency correlated inversely with the length of mutant ataxin-3 proteins. Cells expressing a shorter product showed a higher frequency of aggregate formation. Cells transfected with pMJD-Q77(
N286)-myc encoding the mutant ataxin-3 protein with an N-terminal deletion of 286 amino acids showed a particularly high frequency of aggregate formation (80.0%).
|
In addition to aggregate frequency, a considerable difference was observed in the nature of aggregates. After pMJD-FL-Q77-myc transfection, aggregates showed small, dot-like features (data not shown). After transfection of pMJD-Q77(
N94)-myc (Fig. 2e), pMJD-Q77(N163)-myc (Fig. 2g) or pMJD-Q77(N222)-myc (Fig. 2i), larger aggregates ~1–2 µm in diameter were seen together with small aggregates. These inclusions were ovoid or spherical with a smooth surface. The center often showed weak labeling by anti-myc antibody compared with the periphery. After transfection of pMJD-Q77(N268)-myc (Fig. 2k) or pMJD-Q77(N286)-myc (Fig. 2m), myc-positive inclusions formed multiply in cells, sometimes fusing together to form a large irregularly shaped mass (Fig. 2o).
Subcellular localization of aggregates was evaluated in transfection of the Q77 series of plasmids (Table 1). When full-length mutant ataxin-3 was expressed in BHK-21 cells, the frequency of intracellular aggregates was 5.3% (Fig. 3a). Of these, 16.9% showed nuclear localization. When truncated forms were expressed, intranuclear aggregates were detected in ~40–50% of cells having aggregates, except in the expression of the Q77(N222) protein. In cells expressing the Q77(N222) protein, the frequency of intranuclear aggregates decreased to 24.3%. These data suggest that N-terminal residues may influence the subcellular localization of aggregates.
|
Apoptotic phenotype induced by full-length or N-terminal-truncated ataxin-3 (Q77)
Apoptotic features were often observed by DNA staining (Fig. 2o and p) and TUNEL assay (data not shown) in cells transfected with pMJD-Q77(
N286)-myc. We quantitated apoptotic phenotypes by studying the morphology of nuclei stained with Hoechst dye (Fig. 3b and c). Expression of full-length and N-terminal-truncated ataxin-3 proteins with Q10 had no significant effect on apoptosis frequency (Fig. 3c). N-terminal truncation of ataxin-3 with Q77 up to 268 amino acids did not significantly affect apoptosis frequency (Fig. 3b). In contrast, pMJD-Q77(N286)-myc induced apoptosis at a significantly higher frequency (16%) than that observed in cells expressing LacZ (6%) under normal culture (Fig. 3b)
Serum starvation enhances cellular toxicity induced by truncated ataxin-3 (Q77)
Though mutated ataxin-3 is known to be widely expressed in the brain and elsewhere in the body in MJD, cell death has only been reported in neurons (6). One important feature of neurons is the state of the cell cycle. Neurons exit their cell cycle during development and remain in the G0 phase. We hypothesized that the cell cycle phase influenced the cellular toxicity induced by truncated ataxin-3 (Q77). In BHK-21 cells, serum starvation in culture media is known to inhibit growth and arrest the cell cycle in the G0/G1 phase (34). In asynchronously growing BHK-21 cells, cyclin D-like immunoreactivity was localized predominantly in the nucleus at different intensities (Fig. 4a). In contrast, serum starvation for 36 h dramatically reduced nuclear signals of cyclin D-like immunoreactivity (Fig. 4b), suggesting that ~70% of cells were in the quiescent G0 phase under serum starvation (35) (Fig. 4c). We therefore studied aggregate formation and cellular toxicity under this condition.
|
The frequency of aggregate formation in cells expressing full-length or N-terminal-truncated ataxin-3 proteins with Q77 showed a slightly higher value under serum starvation than under normal culture (Fig. 3d), but this difference only became statistically significant in transfection of pMJD-Q77(
N286)-myc (P = 0.0443). Marked effects were observed in the determination of cellular toxicity (Fig. 3e). Under serum starvation, apoptosis frequency induced by pMJD-Q77(N286)-myc greatly increased (16.7 ± 2.2% under normal culture versus 54.0 ± 3.5% under serum starvation, P = 0.0009). Because the frequency of apoptosis in LacZ transfection increased slightly under serum starvation condition (5.9 ± 0.9% in normal culture versus 12.7 ± 1.4% in serum starvation, P = 0.0154), we calculated the ratio of apoptosis frequency (N286/LacZ). The ratio under serum starvation (4.3) was higher than that under normal culture (2.8), suggesting that the cellular toxicity of pMJD-Q77(N286)-myc was more enhanced in serum starvation. pMJD-Q77(N222)-myc and pMJDQ77(N222)-myc, which did not show any toxicity under normal culture, also produced significant cell death compared with LacZ (24.6 ± 1.0% in pMJD-Q77(N222)-myc, P = 0.023; 25.8 ± 0.7% in pMJD-Q77(N268)-myc, P = 0.011), but to a lesser extent. Figure 5 shows pMJD-Q77(N286)-myc transfection under serum starvation. Nuclear fragmentation and condensation were often seen in cells expressing Q77(N286) protein (Fig. 5).
|
In contrast to the Q77 series of transfection, no difference was seen in the frequency of apoptosis among the Q10 series of transfection and LacZ under serum starvation (Fig. 3f). These data show that ataxin-3 cytotoxicity was specific to the truncated form with Q77, not to truncated forms with Q10, and that serum starvation greatly enhanced cytotoxicity produced by Q77(
N286) protein expression.
Overexpression of cyclin–Cdk inhibitor p21waf1/cip1/sdi1 enhances cellular toxicity induced by truncated ataxin-3 (Q77)
Given the above results, we could not determine whether cell cycle arrest directly affected the viability of transfected cells and/or loss of factors in serum-enhanced cellular toxicity. To determine this, we focused on the cellular toxicity of pMJD-Q77(N286)-myc and investigated the effect of overexpression of cyclin–Cdk inhibitor p21waf1/cip1/sdi1, which has been shown to regulate the kinase activity of G1 cyclin–Cdk complexes (32), and its transient overexpression in BHK-21 cells has been reported to induce cell cycle arrest in the G1 phase (33). Bromodeoxyuridine (BrdU) uptake was greatly suppressed in p21waf1/cip1/sdi1-transfected BHK-21 cells (Fig. 6), indicating that the cell cycle was arrested in p21waf1/cip1/sdi1-expressing cells.
|
The human p21waf1/cip1/sdi1 expression vector was cotransfected in BHK-21 cells either with pMJD-Q77(
N286)-myc or pMJD-Q10(N286)-myc. Separate transfection of the individual plasmid was used as a control. Apoptotic features were determined 24 h after transfection (Fig. 7d). The overexpression of p21 did not appear to induce a high frequency of cell death under our culture conditions (7.6 ± 1.2%). No significant cell death was seen in the expression of Q10(N286) (4.8 ± 0.5%) or the coexpression of p21waf1/cip1/sdi1 and Q10(N286) (5.5 ± 0.8%). In contrast, Q77(N286) expression caused prominent cell death (24.5 ± 1.7%). Coexpression of p21waf1/cip1/sdi1 and Q77(N286) significantly increased cellular toxicity (43.5 ± 2.7%, P = 0.0001). Figure 7a–c shows a representative apoptotic cell demonstrated by double indirect immunofluorescence for human p21waf1/cip1/sdi1 (a) and myc-tag (b), and by nuclear staining (c). Note the presence of human p21waf1/cip1/sdi1 immunofluorescence, intracellular aggregate formation containing the truncated ataxin-3, and nuclear fragmentation. Results of serum starvation and p21waf1/cip1/sdi1 overexpression strongly suggest the direct influence of cell cycle arrest on cellular toxicity by truncated ataxin-3 with Q77 in non-neuronal BHK-21 cells.
|
Nerve growth factor (NGF)-treated neuronal PC12 cells show higher susceptibility to cell death by truncated ataxin-3 (Q77) than undifferentiated proliferating PC12 cells
Proliferating PC12 rat pheochromocytoma cells continue to drive the cell cycle. Once treated with NGF, PC12 cells start a process similar to the differentiation of sympathetic neurons and exit the cell cycle (36). We chose this system as a model to study the influence of cell cycle arrest on the ataxin-3-related toxicity in neurons.
Undifferentiated and NGF-treated PC12 cells were transiently transfected with pMJD-Q77(N286)-myc, pMJD-Q10(N286)-myc, or the LacZ-expressing vector. Aggregate formation and apoptotic features in transfected cells were evaluated by anti-myc immunohistochemistry and nuclear staining 24 and 48 h after transfection (Fig. 8). Figure 9 shows representative undifferentiated and NGF-treated PC12 cells expressing Q10(N286) or Q77(N286).
|
|
Aggregate formation was seen only in the transfection of pMJD-Q77(
N286)-myc (Fig. 9). Both in undifferentiated and NGF-treated PC12 cells, the frequency of aggregates increased markedly 48 h after transfection compared with 24 h post-transfection (Fig. 8a). These aggregates were located in the nucleus and cytoplasmic regions (Fig. 9c and g).
Among undifferentiated proliferating PC12 cells expressing Q77(N286), ~15% showed cell death 24 h after transfection. A similar frequency of cell death was seen 48 h after transfection. These values were significantly higher than those in undifferentiated PC12 cells expressing LacZ or Q10(N286), indicating that truncated ataxin-3 with an expanded polyglutamine stretch exerts toxicity in proliferating PC12 cells, but with relatively low frequency (Fig. 8b).
In contrast, among NGF-treated neuronal PC12 cells expressing Q77(N286), the frequency of cell death was apparently augmented time-dependently compared with undifferentiated PC12 cells (Fig. 8b). Approximately 35% of neuronal PC12 cells expressing Q77(N286) showed cell death 24 h and ~48% 48 h post-transfection. Only 6–7% of LacZ-expressing neuronal PC12 cells showed cell death. The Q10(N286) expression did not produce a high frequency of cell death in NGF-treated PC12 cells. These data clearly show that cell death produced by Q77(N286) expression was enhanced in PC12 cells after NGF-induced neuronal differentiation. Because NGF treatment is known to induce PC12 cells to arrest the cell cycle and differentiate to neurons, these data appear to confirm results in non-neuronal BHK-21 cells. Taken together, our results strongly suggest that cellular toxicity produced by truncated ataxin-3 with an expanded polyglutamine stretch is enhanced by cell cycle arrest in the G0/G1 phase. These findings may explain why neurons are more susceptible to cell death in polyglutamine diseases.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
We studied truncated ataxin-3 proteins with an expanded polyglutamine stretch to determine their aggregate formation and cytotoxic effect on BHK-21 cells. Protein truncation of ataxin-3 from the N-terminus greatly affected the formation of intracellular aggregate bodies. Shorter fragments containing an expanded polyglutamine tract generated more aggregate bodies than did the full-length protein. In contrast to the apparent correlation between N-terminal truncation and aggregate formation frequency, cytotoxicity did not appear to correlate with N-terminal truncation under normal culture, where the low level of cell death showed very little change. The only truncated ataxin-3 significantly increasing cell death was the shortest truncated ataxin-3 with a pathological polyglutamine expansion, Q77(
N286), in which 286 amino acids were deleted from the N-terminus. Even under serum starvation, in which N222(Q77) and N268(Q77) appeared to increase cell death to a lesser extent, essentially only N286(Q77) was able to induce cell death at a high frequency.
We do not yet know why a <20 amino acid difference between N268(Q77) and N286(Q77) significantly influences cell death. Substantially no difference in subcellular distribution of recombinant proteins (data not shown) or aggregates (Table 1) was seen between N268(Q77) and N286(Q77). Although several recent studies have shown the nuclear transport of ataxin-3 (37) and the importance of nuclear translocation of pathological proteins on cellular toxicity (11,25), our data suggested that the difference in toxicity between N268(Q77) and N286(Q77) was not explained simply by nuclear translocation of recombinant proteins.
Our data, in which essentially only N286(Q77) was able to produce cell death, strongly suggested that proteolytic cleavage around amino acid position 286 and liberation of the polyglutamine-containing fragment could be required for cells to die. Within the amino acid sequence of ataxin-3, four potential caspase cleavage sites have been predicted (24). Because each of these is located more N-terminally from amino acid position 286, it seems unlikely that a fragment similar to N286 is generated by caspases. One paper suggests the presence of proteases other than caspase-1 and -3 in the osteosarcoma apoptotic extract that generated a 18 kDa fragment from ataxin-3(Q79) (24). Additional proteases that cleave ataxin-3 near amino acid position 286 may thus exist and play a role in initiating cell death.
In cultured cells, aggregates were found in cytoplasm and intranuclear regions. In affected neurons in MJD, however, aggregate bodies were reported localized almost exclusively in intranuclear regions (9). The difference in subcellular localization may derive from the expression of the mutant protein and/or cell types expressing the mutant protein.
We showed that serum starvation in culture media significantly increased cell death caused by the N286(Q77) expression. BHK-21 cell cycles are synchronized at the G0/G1 phase by serum starvation (34). It thus seems likely that cell susceptibility to the mutant ataxin-3 fragment is influenced by the cell cycle phase. Susceptibility was particularly enhanced in the G0/G1 phase. Our experiment on the coexpression of cyclin–Cdk inhibitor p21waf1/cip1/sdi1 and N286(Q77) directly demonstrated the effect of cell cycle arrest in the G1 phase on cytotoxicity. Previous papers reported that cells treated with a low dose of antiestrogen, tamoxifen (19) or a Ser/Thr protein kinase inhibitor, staurosporine (18), increased cellular toxicity induced by truncated huntingtin containing an expanded polyglutamine stretch. Although both tamoxifen and staurosporine are apoptosis-inducing agents, they also affect the cell cycle and produce cell cycle arrest in the G1 phase (38,39). Susceptibility to cell death increased by these agents may be exerted in part through the effect on the cell cycle.
Using the rat PC12 pheochromocytoma cell line, we confirmed that NGF-treated neuronal PC12 cells were more susceptible to cell death induced by N286(Q77) than untreated cells. In PC12 cells, NGF treatment is known to produce cell cycle arrest in G0/G1 phases and to cause a process similar to neuronal differentiation (36). It is possible that differentiation-related changes in cellular physiology other than cell cycle arrest significantly influence the toxicity of N286(Q77) in neuronal PC12 cells. Our data from the non-neuronal model, however, showed that cell cycle arrest was sufficient to increase the toxicity of N286(Q77). In the transgenic fly expressing the mutant ataxin-3 fragment in its eyes, cell death was observed in posteriorly situated postmitotic cells, not in the dividing cells of imaginal discs (12). Given these results, we propose that the cytotoxicity of truncated ataxin-3 with an expanded polyglutamine stretch increases in G0/G1 phases.
Currently we cannot explain why the susceptibility to cell death by the mutant ataxin-3 fragment is influenced by the cell cycle. One possibility is that the mutant protein may interact with G0/G1-specific proteins and disturb their function. A second possibility is that the alteration of protein degradation during the cell cycle may influence cytotoxicity. Since the cell cycle of neurons was arrested in the G0 phase, our results from non-neuronal and neuronal models both appear to show that neurons, i.e. postmitotic cells, are more vulnerable than other types of cell in MJD. Our data do not explain, however, why only specific subsets of neurons undergo cell death in MJD, and further study is needed to explain this neuronal selectivity.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Cell culture
BHK-21 cells were grown in a 5% CO2 atmosphere at 37°C in Dulbecco’s modified Eagle’s medium (Gibco BRL, Gaithersburg, MD) supplemented with 10% fetal bovine serum (FBS; Gibco BRL), 2 mM glutamine (Gibco BRL) and 100 µg/ml of penicillin–streptomycin (Gibco BRL) (40). PC12 cells were obtained from Dr Y. Kasuya (Department of Pharmacology, University of Tsukuba, Japan), and cultured in the same atmosphere in RPMI medium (Gibco BRL) supplemented with 10% equine serum (Gibco BRL), 5% FBS and 100 µg/ml of penicillin–streptomycin.
Construction of full-length and truncated ataxin-3 expression vectors containing Q10 or Q77
The human MJD cDNA clone is described elsewhere (pMJD2-1) (41). Total RNA was extracted with an RNA extraction solution (Wako Chemicals, Osaka, Japan) from white blood cells of an MJD patient and reverse-transcribed into cDNA with random hexamers and reverse transcriptase (Perkin-Elmer, Foster City, CA). Using the above cDNA as a template, the DNA fragment containing wild-type or expanded polyglutamine tract was amplified by nested PCR with a high-fidelity enzyme (Takara, Kyoto, Japan). Primers for the first PCR were 5'-GAATTCTCTCTTGACGGGTCCAG-3' (sense) and 5'-ACTGCTCCTTAATCCAG-3' (antisense). Primers for the second PCR were 5'-GTCGTTAAGGGTGATCTGCCAG-3' (sense) and 5'-CGGGGTACCGAGGGAATGAAGAATAATG-3' (TY-148; antisense). The second antisense primer contained a KpnI recognition site. After amplification, DNA fragments were analyzed with 3% NuSieve GTG agarose (FMC, Rockland, ME) and extracted. Each fragment was then cloned into the pCR-Script Amp SK(+) vector (Stratagene, La Jolla, CA) and sequenced. One plasmid had 10 CAG repeats (Q10) from the normal allele and the other possessed 77 CAG repeats (Q77) from the mutant allele. These plasmids were double-digested with BglII and KpnI. To generate full-length MJD sequences, DNA fragments were independently ligated into BglII–KpnI-digested pMJD2-1 and subcloned. The full-length ataxin-3 expression vector with 363 (Q10) or 430 (Q77) amino acids together with myc epitope was generated by subcloning a BamHI–KpnI fragment containing full-length MJD cDNA with 10 or 77 CAG repeats into BamHI–KpnI-digested pcDNA3.1-Myc-HisC (Invitrogen, San Diego, CA) to create a fusion to the myc epitope (pMJD-FL-Q10-myc or pMJD-FL-Q77-myc).
To introduce a hemagglutinin (HA) tag at the N-terminus of pMJD-FL-Q10-myc or pMJD-FL-Q77-myc, we obtained two PCR fragments from each plasmid using two different primer pairs. One pair was composed of 5'-AATTAACCCTTCACTAAAGGG-3' (T3 primer; sense) and 5'-CCCAAGCTTGTTTATTTGTCTGGAGCC-3' (TY-149; antisense). The other pair was composed of 5'-CCCAAGCTTATGTACCCATACGATGTTCCAGATTACGCTGAGTCCATCTTCCACGAG-3' (TY-150; sense) and TY-148 (antisense). TY-149 possessed a HindIII site, and TY-150 possessed a HindIII site and a sequence for the HA tag. PCR was done using pMJD-FL-Q10-myc or pMJD-FL-Q77-myc as a template. After digestion of two PCR products (T3 and TY-149, TY-150 and -148) from each plasmid with HindIII, DNA fragments were mutually ligated and the resultant DNA fragment containing the 5'-non-coding region of MJD cDNA, HA tag sequence and the full-length MJD cDNA (10 or 77 CAG repeats) subcloned into the pCR-Script Amp SK(+) vector (Stratagene). The BamHI–KpnI-digested fragment (10 or 77 CAG repeats) was subcloned again into BamHI–KpnI-digested pMJD-FL-Q10-myc or pMJD-FL-Q77-myc (pMJD-FL-Q10-HA-myc or pMJD-FL-Q77-HA-myc). Sequences were confirmed. These expression vectors expressed full-length ataxin-3 (with Q10 or Q77) together with N-terminal HA tag and C-terminal myc tag.
To create a series of truncated ataxin-3 expression vectors containing Q10 or Q77, we obtained DNA fragments of different sizes by PCR using pMJD-FL-Q10-HA-myc or pMJD-FL-Q77-HA-myc as a template. The sets of sense primers were:
5'-CCCAAGCTTATGAACAGTCCAGAGTATCAG-3',
5'-CCCAAGCTTATGGTCGTTAAGGGTGATCTG-3',
5'-CCCAAGCTTATGGACGAAGATGAGGAGGAT-3',
5'-CCCAAGCTTATGACACAGACATCAGGTACA-3', and
5'-CCCAAGCTTATGGCCTACTTTGAAAAACAGCAG-3'.
A HindIII site was introduced together with the initiation codon in each sense primer. The antisense primer 5'-CGGGGTACCGAGGGAATGAAGAATAATG-3' contained a KpnI recognition site. The series of PCR fragments were isolated and double-digested with HindIII and KpnI, and subcloned into HindIII–KpnI-digested pMJD-FL-Q10-HA-myc or pMJD-FL-Q77-HA-myc. These expression vectors were named pMJD-Q10/77(N94)-myc, pMJD-Q10/77(N163)-myc, pMJD-Q10/77(N222)-myc, pMJD-Q10/77(N268)-myc and pMJD-Q10/77(N286)-myc. All coded truncated ataxin-3 fragments containing Q10 or Q77 with a C-terminal myc epitope. For control experiments, we used pcDNA3.1(–)/Myc-His/LacZ (Invitrogen), an expression vector encoding ß-galactosidase.
DNA transfections
DNA was transfected using calcium phosphate coprecipitation (40). For BHK-21, 25 000 cells were plated onto a well of a poly-D-lysine-coated, two-well slide-glass chamber (Becton-Dickinson, Bedford, MA) in 2 ml of medium. A total of 1.8 µg of plasmid DNA was added to 6.2 µl of 0.2 M CaCl2 and precipitated by adding 50 µl of 2x HEPES-buffered saline over a period of 15 s while stirring. After 12.5 min, 1 ml of complete medium was added to the DNA precipitate. The mixture was applied to cells. Each slide-glass chamber was incubated in an incubator for 4 h, then the culture medium was changed to a complete new medium and cells incubated for another 44 h. To evaluate the influence of serum starvation, the medium was changed again to new medium containing 0.5% FBS 12 h after transfection. Cells were then incubated for an additional 36 h.
In the cotransfection experiment using the human p21waf1/cip1/sdi1 expression vector (32) and pMJD-Q77(N286)-myc, 1.8 µg of each plasmid was added to the reaction, and each slide-glass chamber incubated for 4 h. The culture medium was changed to a complete new medium and cells incubated for another 24 h.
For undifferentiated PC12 cells, 1 x 105 cells were plated onto a well of a two-well Permanox slide chamber (Nalge Nunc, Naperville, IL) coated with 50 µg/ml of rat-tail collagen type I (Sigma, St Louis, MO) in 2 ml of RPMI-based medium (RPMI supplemented with 10% equine serum, 5% FBS and 100 µg/ml of penicillin–streptomycin). To induce differentiation, PC12 cells were plated onto a 100 mm dish coated with 50 µg/ml of rat-tail collagen type I at a density of 1 x 106 cells in 10 ml of RPMI-based medium and then treated with 100 ng/ml of NGF-2.5S (Sigma). Half of the culture medium was changed every other day. On day 9, 1 x 105 NGF-treated PC12 cells were plated onto a well of a two-well Permanox slide chamber coated with 50 µg/ml of rat-tail collagen type I in 2 ml of RPMI-based medium with 100 ng/ml of NGF. Both in undifferentiated and differentiated PC12 cells, DNA transfections were conducted the same way as in BHK-21 transfection. Transfection efficiency in undifferentiated PC12 cells was 5.5%, and that in differentiated PC12 cells was 3.0%. Cell death was evaluated 24 and 48 h after transfection.
Antibodies
Anti-myc tag mouse monoclonal antibody (Ab-2; NeoMarkers, Union City, CA), anti-myc tag rabbit polyclonal antibody (MBL, Nagoya, Japan), anti-human p21waf1/cip1/sdi1 mouse monoclonal antibody (187; Santa Cruz Biotechnology, Santa Cruz, CA) and anti-cyclin D1 mouse monoclonal antibody (H-295; Santa Cruz Biotechnology, Santa Cruz, CA) were used in this study. Based on the manufacturer’s documentation, the H-295 antibody directed against human cyclin D1 cross-reacts with mouse and rat cyclin D1, D2 and D3. This antibody also recognizes an ~34 kDa band in BHK-21 whole cell lysate by western blotting (data not shown), suggesting cross-reactivity to hamster cyclin D.
Immunohistochemistry
The expression of myc-tagged full-length ataxin-3 and truncated ataxin-3, and human p21waf1/cip1/sdi1 was evaluated by indirect immunofluorescence with anti-myc and anti-p21waf1/cip1/sdi1 antibodies. The endogenous expression of cyclin D was examined by anti-cyclin D1 antibody. Following the incubation period, cells grown on the slide-glass chambers were washed with phosphate-buffered saline (PBS) and fixed in a 4% paraformaldehyde/PBS solution for 20 min at room temperature. Cells were permeated in PBS containing 0.5% Tween-20 for 10 min. Primary antibodies were diluted with PBS containing 1% normal goat serum (1:250 for monoclonal anti-myc, 1:200 for polyclonal anti-myc, polyclonal anti-p21waf1/cip1/sdi1 and anti-cyclin D1) and applied to cells overnight at 4°C. Next, slides were washed with PBS, and secondary antibodies (anti-mouse or anti-rabbit IgG) conjugated to Cy-3 or FITC were added at a 1:400 dilution for 1 h at room temperature. Cells were washed again and DNA was labeled with Hoechst dye 33258 (Sigma) at 10 µg/ml PBS for 3 min. Slideglasses were mounted with coverslips in 90% glycerol, 20 mM Tris–HCl pH 7.5. Immunofluorescence was observed and photographed at a magnification of 420x using a PROVIS microscope (Olympus, Tokyo, Japan).
Quantitation of aggregate formation and apoptotic phenotype
Aggregate formation was counted by analyzing >150 cells expressing the myc epitope selected by random sweeps at 420x under the fluorescence microscope. The percentage of aggregate-forming cells per total cells with the myc epitope was calculated. Among myc-positive cells, the apoptotic phenotype was counted based on the morphology of the Hoechst 33258-stained nucleus. The presence of a pyknotic or fragmented nucleus was judged as the apoptotic phenotype. In some cases, TUNEL assay (DNA nick-end labeling) was done using an in situ cell death detection kit (Boehringer Mannheim, Mannheim, Germany) based on the manufacturer’s instructions. Results were obtained from three independent experiments and represented as means ± SE. In the cotransfection of pMJD-Q77(N286)-myc and the p21waf1/cip1/sdi1 expression vector, aggregate formation and apoptotic phenotype were evaluated among cells expressing both myc-epitope and the human p21waf1/cip1/sdi1 protein. Each graph represents the mean ± SE from eight independent experiments. Statistical analysis was conducted using Student’s t-test and StatView 4.0 (Abacus Concepts, Berkeley, CA).
Evaluation of BrdU uptake in cells expressing p21waf1/cip1/sdi1
BrdU uptake was assessed in non-transfected BHK-21 cells and cells expressing p21waf1/cip1/sdi1 using an in situ cell proliferation kit, FLUOS (Boehringer Mannheim) based on the manufacturer’s instructions. For cells expressing p21waf1/cip1/sdi1, BrdU labeling was done 24 h after transfection, followed by p21waf1/cip1/sdi1 immunohistochemistry. Immunofluorescence was analyzed by a TCS4D confocal laser microscope (Leica, Heidelberg, Germany). Confocal images were processed by Photoshop 4.0J (Adobe, San Jose, CA).
|
ACKNOWLEDGEMENTS |
|---|
We thank Dr S.J. Elledge for kindly providing the p21waf1/cip1/sdi1 expression vector, Ms Nissato for her technical assistance, Drs K. Goto and Y. Kasuya (Department of Pharmacology, University of Tsukuba) for the use of their facilities, Dr Y. Hoshikawa (Tokyo Metropolitan Research Institute), Dr N. Nakanishi (Harvard University) and Dr H. Okayama (University of Tokyo) for their helpful suggestions. This study was supported in part by a grant from the Ministry of Education, Science and Culture of Japan, a grant from the University of Tsukuba, Japan, and a grant from Dynamics of the Brain and Amenity for the Mind, a project underway at the University of Tsukuba, Japan.
|
FOOTNOTES |
|---|
+ To whom correspondence should be addressed. Tel: +81 298 53 3223; Fax: +81 298 53 3224; Email: toshi-yo@md.tsukuba.ac.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Spada, L., Wilson, E.M., Lubahn, D.B., Harding, A.E. and Fishbeck, K.H. (1991) Androgen receptor gene mutations in X-linked spinal and bulbar muscular atrophy. Nature, 353, 77–79.
2 Andrew, S.E., Goldberg, Y.P. and Hayden, M.R. (1997) Rethinking genotype and phenotype correlations in polyglutamine expansion disorders. Hum. Mol. Genet., 6, 2005–2010.
3 Koshy, B.T. and Zoghbi, H.Y. (1997) The CAG/polyglutamine diseases: gene products and molecular pathogenesis. Brain Pathol., 7, 927–942.[ISI][Medline]
4 Perutz, M.F., Johnson, T., Suzuki, M. and Finch, J.T. (1994) Glutamine repeats as polar zippers: their possible role in inherited neurodegenerative disease. Proc. Natl Acad. Sci. USA, 91, 5355–5358.
5 Sharp, A.H., Loev, S.J., Schilling, G., Li, S.H., Li, X.J., Bao, J., Wagster, M.V., Kotzuk, J.A., Steiner, J.P. and Lo, A. (1995) Widespread expression of Huntington’s disease gene (IT15) protein product. Neuron, 14, 1065–1074.[ISI][Medline]
6 Paulson, H.L., Das, S.S., Crino, P.B., Perez, M.K., Patel, S.C., Gotsdiner, D., Fishbeck, K.H. and Pittman, R.N. (1997) Machado–Joseph disease gene product is a cytoplasmic protein widely expressed in brain. Ann. Neurol., 41, 453–462.[ISI][Medline]
7 Reddy, P.S. and Housman, D.E. (1997) The complex pathology of trinucleotide repeats. Curr. Opin. Cell Biol., 9, 364–372.[ISI][Medline]
8 DiFiglia, M., Sapp, E., Chase, K.O., Davis, S.W., Bates, G.P., Vonsattel, J.P. and Aronin, N. (1997) Aggregation of huntingtin in neuronal intranuclear inclusions and dystrophic neurites in brain. Science, 277, 1990–1993.
9 Paulson, H.L., Perez, M.K., Trottier, Y., Trojanowski, J.Q., Subramony, S.H., Das, S.S., Vig, P., Mandel, J.-L., Fishbeck, K.H. and Pittman, R.N. (1997) Intranuclear inclusions of expanded polyglutamine protein in spinocerebellar ataxia type 3. Neuron, 19, 333–344.[ISI][Medline]
10 Ordway, J.M., Tallaksen-Greene, S., Gutekunst, C.-A., Bernstein, E.M., Cearley, J.A., Wiener, H.W., Dure IV, L.S., Lindsey, R., Hersch, S.M., Jope, R.S., Albin, R.L. and Detloff, P.J. (1997) Ectopically expressed CAG repeats cause intranuclear inclusions and a progressive late-onset neurological phenotype in the mouse. Cell, 91, 753–763.[ISI][Medline]
11 Klement, I.A., Skinner, P.J., Kaytor, M.D., Yi, H., Hersch, S.M., Clark, H.B., Zoghbi, H.Y. and Orr, H.T. (1998) Ataxin-1 nuclear localization and aggregation: role in polyglutamine-induced disease in SCA1 transgenic mice. Cell, 95, 41–53.[ISI][Medline]
12 Warrick, J.M., Paulson, H.L., Gray-Board, G.L., Bui, Q.T., Fishbeck, K.H., Pittman, R.N. and Bonini, N.M. (1998) Expanded polyglutamine protein forms nuclear inclusions and causes neural degeneration in Drosophila. Cell, 93, 939–949.[ISI][Medline]
13 Sisodia, S.S. (1998) Nuclear inclusions in glutamine repeat disorders: are they pernicious, coincidental, or beneficial? Cell, 95, 1–4.[ISI][Medline]
14 Ross, C. (1997) Intranuclear neuronal inclusions: a common pathogenic mechanism for glutamine-repeat neurodegenerative diseases? Neuron, 19, 1147–1149.[ISI][Medline]
15 Lunkes, A. and Madel, J.L. (1997) Polyglutamines, nuclear inclusions and neurodegeneration. Nature Med., 3, 1201–1202.[ISI][Medline]
16 Davies, S.W., Beardsall, K., Turmaine, M., DiFiglia, M., Aronin, N. and Bates, G.P. (1998) Are neuronal intranuclear inclusions the common neuropathology of triplet-repeat disorders with polyglutamine-repeat expansion? Lancet, 351, 131–133.[ISI][Medline]
17 Ikeda, H., Yamaguchi, M., Sugai, S., Aze, Y., Narumiya, S. and Kakizuka, A. (1996) Expanded polyglutamine in the Machado–Joseph disease protein induces cell death in vitro and in vivo. Nature Genet., 13, 196–202.[ISI][Medline]
18 Martindale, D., Hackam, A., Wieczorek, A., Ellerby, L., Wellington, C., McCutcheon, K., Singaraja, R., Kazemi-Esfarjani, P., Devon, R., Kim, S.U. et al. (1998) Length of huntingtin and its polyglutamine tract influences localization and frequency of intracellular aggregates. Nature Genet., 18, 150–154.[ISI][Medline]
19 Hackam, A.S., Singaraja, R., Wellington, C.L., Metzler, M., McCutcheon, K., Zhang, T., Kalchman, M. and Hayden, M.R. (1998) The influence of huntingtin protein size on nuclear localization and cellular toxicity. J. Cell Biol., 141, 1097–1105.
20 Cooper, J.K., Schilling, G., Peters, M., Herring, W., Sharp, A.H., Kaminsky, Z., Masene, J., Khan, F.A., Delanoy, M., Borchelt, D.R. et al. (1998) Truncated N-terminal fragments of huntingtin with expanded polyglutamine repeats form nuclear and cytoplasmic aggregates in cell culture. Hum. Mol. Genet., 7, 783–790.
21 Li, S.-H. and Li, X.-J. (1998) Aggregation of N-terminal huntingtin is dependent on the length of its glutamine repeats. Hum. Mol. Genet., 5, 777–782.
22 Igarashi, S., Koide, R., Shiomohata, T., Yamada, M., Hayashi, Y., Takano, H., Date, H., Oakaya, M., Sato, T., Sato, A. et al. (1998) Suppression of aggregate formation and apoptosis by transglutaminase inhibitors in cells expressing truncated DRPLA protein with an expanded polyglutamine stretch. Nature Genet., 18, 111–117.[ISI][Medline]
23 Goldberg, Y.P., Nicholson, D.W., Rasper, D.M., Kalchman, M.A., Koide, H.B., Graham, R.K., Bromm, M., Kazemi-Esfarjani, P., Thornberry, N.A., Vaillancort, J.P. and Hayden, M.R. (1996) Cleavage of huntingtin by apopain, a proapoptotic cysteine protease, is modulated by the polyglutamine tract. Nature Genet., 13, 442–449.[ISI][Medline]
24 Wellington, C.L., Ellerby, L.M., Hackam, A.S., Margolis, R.L., Trifiro, M.A., Singaraja, R., McCutcheon, K., Salvesen, G.S., Prop, S.S., Bromm, M. et al. (1998) Caspase cleavage of gene products associated with triplet expansion disorders generates truncated fragments containing the polyglutamine tract. J. Biol. Chem., 10, 9158–9167.
25 Saudou, F., Finkbeiner, S., Devys, D. and Greenberg, M.E. (1998) Huntingtin acts in the nucleus to induce apoptosis but death does not correlate with the formation of intranuclear inclusions. Cell, 95, 55–66.[ISI][Medline]
26 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.[ISI][Medline]
27 Mangiarini, L., Sathasivam, K., Seller, M., Cozens, B., Harper, A., Hetherington, C., Lawton, M., Trottier, Y., Lehrach, H., Davis, S.W. and Bates, G.P. (1996) Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell, 87, 493–506.[ISI][Medline]
28 Jackson, G.R., Salecker, I., Dong, X.Z., Yao, X., Arnheim, N., Faber, P.W., Macdonald, M.E. and Zipursky, S.L. (1998) Polyglutamine-expanded human huntingtin transgenes induce degeneration of Drosophila photoreceptor neurons. Neuron, 21, 633–642.[ISI][Medline]
29 Kawaguchi, Y., Okamoto, T., Taniwaki, M., Aizawa, M., Inoue, M., Katayama, S., Kawakami, H., Nakamura, S., Nishimura, M., Akiguchi, I. et al. (1994) CAG expansions in a novel gene for Machado–Joseph disease at chromosome 14q32.1. Nature Genet., 8, 221–228.[ISI][Medline]
30 Paulson, H.L. (1998) Analysis of Triplet Repeat Disorders: Spinocerebellar Ataxia Type 3/Machado–Joseph Disease. BIOS Scientific Publishers, Oxford, UK.
31 Perez, M.K., Paulson, H.L., Pendse, S.J., Saionz, S.J., Bonini, N.M. and Pittman, R.N. (1998) Recruitment and the role of nuclear localization in polyglutamine-mediated aggregation. J. Cell Biol., 143, 1457–1470.
32 Harper, J.W., Adami, G.R., Wei, N., Keyomarsi, K. and Elledge, S.J. (1993) The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell, 75, 805–816.[ISI][Medline]
33 Sekiguchi, T. and Hunter, T. (1998) Induction of growth arrest and cell death by overexpression of the cyclin–Cdk inhibitor p21 in hamster BHK21 cells. Oncogene, 16, 369–380.[ISI][Medline]
34 Talavera, A. and Basilico, C. (1978) Requirement of BHK cells for exit from different quiescent states. J. Cell Physiol., 97, 429–439.[ISI][Medline]
35 Baldin, V., Lukas, J., Marcote, M.J., Pagano, M. and Draetta, G. (1993) Cyclin D1 is a nuclear protein required for cell cycle progression in G1. Genes Dev., 7, 812–821.
36 Greene, L.A. and Tischler, A.S. (1976) Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc Natl Acad. Sci. USA, 73, 2424–2428.
37 Tait, D., Riccio, M., Sittler, A., Scherzinger, E., Santi, S., Ognibene, A., Maraldi, N.M., Lehrach, H. and Wanker, E.E. (1998) Ataxin-3 is transported into the nucleus and associates with the nuclear matrix. Hum. Mol. Genet., 6, 991–997.
38 Osborne, C.K., Boldt, D.H., Clark, G.M. and Trent, J.M. (1983) Effects of tamoxifen on breast cancer cell cycle kinetics: accumulation of cells in early G1 phase. Cancer Res., 43, 3583–3585.
39 Orr, M.S., Reinhold, W., Yu, L., Schreiber-Agus, N. and O’Connor, P.M. (1998) An important role for the retinoblastoma protein in staurosporin-induced G1 arrest in murine embryonic fibroblasts. J. Biol. Chem., 273, 3803–3807.
40 Shibasaki, F. and Mckeon, F. (1995) Calcineurin functions in Ca2+-activated cell death in mammalian cells. J. Cell Biol., 131, 735–743.
41 Goto, J., Watanabe, M., Ichikawa, Y., Yee, S.-B., Ihara, N., Endo, K., Igarashi, S., Takiyama, Y., Gaspar, C., Maciel, P. et al. (1997) Machado–Joseph disease gene products carrying different carboxyl termini. Neurosci. Res., 28, 373–377.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Q. Qin, R. Inatome, A. Hotta, M. Kojima, H. Yamamura, H. Hirai, T. Yoshizawa, H. Tanaka, K. Fukami, and S. Yanagi A novel GTPase, CRAG, mediates promyelocytic leukemia protein-associated nuclear body formation and degradation of expanded polyglutamine protein J. Cell Biol., February 13, 2006; 172(4): 497 - 504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Alon and J. M. Friedman Late-Onset Leanness in Mice with Targeted Ablation of Melanin Concentrating Hormone Neurons J. Neurosci., January 11, 2006; 26(2): 389 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Toulouse, F. Au-Yeung, C. Gaspar, J. Roussel, P. Dion, and G. A. Rouleau Ribosomal frameshifting on MJD-1 transcripts with long CAG tracts Hum. Mol. Genet., September 15, 2005; 14(18): 2649 - 2660. [Abstract] [Full Text] [PDF]
|
|
|
|