Skip Navigation


Human Molecular Genetics Advance Access originally published online on August 8, 2005
Human Molecular Genetics 2005 14(18):2649-2660; doi:10.1093/hmg/ddi299
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
14/18/2649    most recent
ddi299v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Toulouse, A.
Right arrow Articles by Rouleau, G. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Toulouse, A.
Right arrow Articles by Rouleau, G. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2005. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oupjournals.org

Ribosomal frameshifting on MJD-1 transcripts with long CAG tracts

André Toulouse{dagger}, Faith Au-Yeung, Claudia Gaspar, Julie Roussel, Patrick Dion and Guy A. Rouleau*

Department of Medicine and Research Center, Centre Hospitalier de l'Université de Montréal, Hôpital Notre-Dame, Suite Y3616–2, 1560 Sherbrooke Street East, Montreal, Quebec H2L 4M1, Canada

* To whom correspondence should be addressed. Tel: +1 5148908000 ext. 24753; Fax: +1 5144127602; Email: guy.rouleau{at}umontreal.ca

Received June 16, 2005; Revised August 1, 2005; Accepted August 1, 2005


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The expanded CAG tract diseases are a heterogeneous group of late-onset neurodegenerative disorders characterized by the accumulation of insoluble protein material and premature neuronal cell death. Recent work has provided support for several mechanisms that may account for neurodegeneration, but no unifying mechanism has emerged. We have previously demonstrated that in SCA3, the expanded CAG tract in the MJD-1 transcript is prone to frameshifting, which may lead to the production of polyalanine-containing proteins. To further examine the occurrence of frameshifting and understand its mechanism and possible role in pathogenesis, a cellular model was established. We show that this phenomenon results from ribosomal slippage to the –1 frame exclusively, that ribosomal frameshifting depends on the presence of long CAG tracts and that polyalanine-frameshifted proteins may enhance polyglutamine-associated toxicity, possibly contributing to pathogenesis. Finally, we present evidence that anisomycin, a ribosome-interacting drug that reduces –1 frameshifting, also reduces toxicity, suggesting a new therapeutic opportunity for these disorders.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The ‘polyglutamine disorders’ are a group of neurodegenerative diseases including spinocerebellar ataxias 1, 2, 3, 6, 7 and 17, spinobulbar muscular atrophy, dentatorubral–pallidoluysian atrophy and Huntington's disease (1Go–12Go). Although the causative proteins are widely expressed, these disorders are characterized by the degeneration of specific neuronal populations. A pathological hallmark of these diseases is the formation of intranuclear aggregates containing the expanded polyglutamine-bearing protein (13Go). These structures are often characterized by poly-ubiquitination and recruitment of cellular factors, including components of the proteasome and chaperone pathways as well as transcription co-factors (e.g. CREB) (14Go–18Go). Several hypotheses have been proposed to explain degeneration, including transcriptional impairment, altered protein interactions, direct polyglutamine toxicity and accumulation of toxic protein by-products following the alteration of the ubiquitin–proteasome pathway.

Oculopharyngeal muscular dystrophy (OPMD) is caused by the short expansion of a polyalanine tract in the PABPN1 gene and displays pathological characteristics similar to the polyglutamine disorders. It is of late onset and specific cell degeneration is observed despite widespread expression of the mutant protein (19Go). It is accompanied by the formation of ubiquitinated protein aggregates and the potential involvement of the chaperone and proteasome pathways (20Go,21Go). Both in vivo and in vitro evidence suggest that short polyalanine peptides are more prone to aggregation and toxic than polyglutamine stretches of similar or even greater length (22Go,23Go).

The observation that in an altered reading frame, a polyglutamine-encoding CAG tract could generate a polyalanine peptide, prompted us to investigate the presence of frameshifted proteins in expanded CAG tract disorders. Using SCA3 as a model, we generated an antibody against an epitope that is produced only following a frameshift to the polyalanine frame (22Go). We demonstrated the presence of such frameshifted proteins in the intranuclear inclusions of SCA3 affected neurons, lymphoblastoid cells and in a cellular model (22Go). Furthermore, we demonstrated that replacing the polyglutamine tract of ataxin-3 by a polyalanine repeat leads to increased toxicity. Although we showed the existence of frameshifted proteins, we could not establish their contribution to toxicity or determine the mechanism leading to the production of such proteins. In this article, we established a cellular model to further study the mechanism of frameshifting in the SCA3 transcript and to determine its contribution to cell toxicity. Using epitope tagging, we show that only –1 frameshifting occurs, that it is CAG length dependent and that it occurs only in polyglutamine-encoding CAG repeats, and not in polyglutamine-encoding CAA triplets, and results from a translational error. We provide data showing that accumulation of polyalanine-frameshifted peptides contributes to toxicity and that it can be modulated by ribosome-interacting drugs.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Generation of a frameshifting reporter system
We generated a reporter system for frameshifting using the MJD-1 cDNA truncated to the last methionine residue preceding the CAG repeat. A Kozak consensus sequence was generated by mutagenesis. The absence of translation initiation sites in the alternate reading frames ensures that translation initiation cannot occur in frames other than the glutamine frame (frame 0). The C-terminus was modified so that an AU1 epitope is produced in the main reading frame (polyglutamine) and if frameshifted –1 (polyalanine) or +1 (polyserine), proteins will bear either an hemagglutinin (HA) or an AU5 tag, respectively (Fig. 1A). These epitope tags were selected for their minimal cross-reactivity in mammalian cells (24Go). Constructs with CAG repeats ranging from 25 to 78 CAG units were generated and placed under the control of the cytomegalovirus (CMV) immediate early promoter.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 1. Establishment of a frameshifting reporter system. (A) Schematic representation of the constructs used. The MJD-1 cDNA was truncated to the last methionine residue preceding the CAG-encoded polyglutamine repeat. The C-terminal encoding sequences were also removed to allow the insertion of the epitope tags in each of the three possible reading frames. Epitope tags corresponding to each of the reading frames are represented in bold. (B) –1 frameshifted proteins localize to polyglutamine expressing cells. COS-1 cells were transfected with the (CAG)78 construct to test for the presence of frameshifting. Immunocytochemistry was performed 24 h post-transfection using monoclonal anti-AU1 (Gln frame), rabbit polyclonal anti-AU5 (Ser frame) and rabbit polyclonal anti-HA (Ala frame) antibodies. FITC-conjugated anti-mouse and rhodamine-conjugated anti-rabbit secondary antibodies were used. Immunoreactivity against the AU1 and HA epitopes was readily detected, whereas AU5 was absent.

 
Only –1 frameshifting occurs
The (CAG)78 construct was transfected into COS-1 cells and frameshifting was assessed by immunocytochemistry using antibodies against each of the epitopes. At 24 h post-transfection, only frameshifting to the –1 frame occurs (Fig. 1B); immunoreactivity against the AU5 epitope was very rarely detected (Fig. 1B). We determined that this absence of immunoreactivity was not due to a problem with the anti-AU5 antibody or the respective epitope by moving this tag to the main translational frame (frame 0) by mutagenesis. Results showed the reactivity in immunocytochemistry, demonstrating the functionality of this epitope and its antibody (data not shown). In addition, results obtained from another set of constructs, where a His-tag epitope was placed in the serine (+1) frame, confirmed the absence of +1 frameshifting in long CAG tracts (Gaspar and Rouleau, in preparation).

A time-course assay revealed that –1 frameshifting is detectable 12 h post-transfection, peaks ~24–48 h and decreases until the last time point examined (120 h). None of the time points examined revealed +1 frameshifting (data not shown). These experiments also revealed that –1 frameshifted proteins (HA+) were only found in cells expressing the polyglutamine-bearing proteins (AU1+), and especially in cells harboring polyglutamine aggregates (Fig. 1B). Cells expressing only the frameshifted molecule were never observed. Furthermore, not all cells expressing the polyglutamine protein displayed frameshifted molecules (Fig. 1B). These results suggest that frameshifting occurs only in a fraction of the molecules and that expression in the polyglutamine frame is necessary for frameshifted molecules to appear.

This phenomenon is not restricted to COS-1 cells. The (CAG)78 construct was transfected into other cell lines, namely, T47-D, HEK293, SKMES-1, SkBr3 and COS-7, and frameshifting was assessed by immunocytochemistry (data not shown). A similar pattern of frameshifting was observed, although the expression levels and the frameshifting frequency were different. Transfection efficiency was low in SKMES-1 and SkBr3: these cells displayed low levels of expression, with very little frameshifting. In contrast, T47-D and HEK293 cells displayed frameshifting but over a longer time period (starting ~24 h and peaking ~72 h). Finally, COS-7 cells behaved similar to COS-1 cells. Considering these differences, we decided to continue with COS-1 cells.

Frameshifting is CAG length dependent
In the ‘polyglutamine disorders’, the length of the CAG repeat plays a critical role in determining toxicity. The length of the CAG repeat in part predicts the disease severity and early age of onset (7Go,9Go–11Go,25Go–30Go). Constructs with various CAG lengths were transfected into COS-1 cells to assess the CAG length dependency in our model. Immunocytochemistry was performed using antibodies against the three epitopes. The results confirmed that +1 frameshifting is very rare (data not shown). The formation of aggregates of proteins in the polyglutamine frame (AU1+) is dependent on the repeat length; as the size of the CAG repeat increases, more inclusions are detected (Fig. 2A). Immunodetection using the anti-HA antibody (–1 frame) revealed that frameshifting is also CAG length dependent (Fig. 2A); as the size of the repeat reaches the pathological threshold (approximately 60 CAG), frameshifted molecules are detected and their abundance increases with the length of the CAG repeat in the transfected construct. Northern blot hybridization (data not shown) demonstrated similar transcription levels for all constructs, suggesting that the increased AU1 immunofluorescence seen when using long CAG constructs in Figure 2A is due either to protein accumulation or to increased translation of the messenger RNA, but not to different transfection efficiencies.



View larger version (64K):
[in this window]
[in a new window]
  		Figure 2. Polyglutamine accumulation and frameshifting are dependent on the length of the CAG repeat in truncated MJD-1 cDNA constructs. (A) COS-1 cells transfected with constructs bearing various lengths of CAG repeats. Immunocytochemistry was performed 24 h post-transfection using monoclonal anti-AU1 (Gln frame) and rabbit polyclonal anti-HA (Ala frame) antibodies. FITC-conjugated anti-mouse and rhodamine-conjugated anti-rabbit secondary antibodies were used. (B) Western blot analysis of protein expression. About 100 µg of the soluble fraction of a protein extract was electrophoresed on a 10% SDS–polyacrylamide gel and was transblotted onto nitrocellulose membranes. The membranes were probed with anti-AU1, anti-HA or anti-actin antibodies. (C) Protein dot-blot assay. Twenty-five micrograms of a total protein extract from transfected COS-1 cells was dot blotted onto nitrocellulose membranes. The blots were probed with anti-AU1, anti-HA or anti-actin antibodies. HRP-conjugated secondary antibodies were used for chemiluminescent detection. (D) Frameshifted molecules are detected only in the polyglutamine expressing cells. The (CAG)78 construct was transfected into COS-1 cells, and flow cytometry was performed 24 h post-transfection. The first four panels display various controls (unstained cells, isotypic antibodies and single labeling), whereas the lower panel shows typical results obtained in the presence of both antibodies. (E) Frameshifting frequencies increase with the length of the CAG repeat. Flow cytometry was performed on transfected COS-1 cells using FITC-conjugated anti-AU1 and R-phycoerythrin-conjugated anti-HA antibodies. The percentage of frameshifting is expressed as the proportion of HA-positive events (upper-right quadrant in Fig. 2C) reported over the number of AU1 events (upper and lower right quadrants in Fig. 2C). The results presented here are the mean of at least five independent readings from three different experiments±SEM.

 
Western blots were performed to confirm expression of the various constructs (Fig. 2B). The soluble fraction (supernatant) of the protein extracts was used to assess expression levels of proteins in the 0 and the –1 open-reading frames. Using the anti-AU1 antibody (polyglutamine frame), we detected similar levels of expression for all constructs with levels slightly lower for the (CAG)38 construct (Fig. 2B). The presence of polyglutamine stretches altered the electrophoretic properties of the proteins, as previously described by others (31Go,32Go). Using the anti-HA antibody (–1 frame), we confirmed that frameshifting occurs primarily once the pathological CAG length is reached and that as the repeat size further increases, more frameshifted proteins are detected (Fig. 2B). Western blot analysis also revealed that while the frameshifted proteins are produced primarily as a discrete band, they also form a smear [most prominent with the (CAG)78 construct] in the separating gel, suggesting that the presence of proteins harboring a variable mixture of glutamines and alanines, where the glutamines would lead to species that migrate at a higher molecular weight (Fig. 2B). Note that the molecular weight of alanine is 89 Da, whereas glutamine is 146 Da, explaining the observed band sizes.

Protein dot blots were then performed in an attempt to quantify the amount of soluble and aggregated protein in both the 0 and –1 frames (Fig. 2C). Twenty-five micrograms of total protein extracts were blotted onto nitrocellulose membranes. Results showed similar levels of AU1 immunoreactivity with constructs (CAG)25 and (CAG)38. A 2-fold increase was noted with the (CAG)63 construct when compared with the level of the constructs in the normal repeat range and an 8-fold increase was observed with construct (CAG)78.

Considering that mRNA levels detected on northern blots were similar for all constructs and that western blots of soluble fractions revealed similar levels of expression for these constructs, this is suggestive of protein aggregation in the cells. When the expression of the HA epitope was assessed (–1 frame), all constructs displayed similar levels of protein (background, similar to constructs with normal CAG length) with the exception of (CAG)78, which was nine times higher than any other construct (Fig. 2C). This is in accordance with western blot results, showing low levels of frameshifted proteins with constructs (CAG)59 and (CAG)63 and a dramatic increase in cells transfected with (CAG)78.

The results mentioned earlier showed that as the size of the CAG repeat increases, more –1 frameshifted molecules (HA+) are being produced, but that this is unlikely to be due to increased transcription or translation (Fig. 2B). We further characterized frameshifting by determining the proportion of transfected cells harboring HA reactive molecules. Flow cytometry was performed using FITC-conjugated anti-AU1 and R-phycoerythrin-conjugated anti-HA antibodies. The frameshifting frequency was calculated as the number of flow cytometry events positive for HA immunofluorescence reported over the population of transfected cells (AU1+). The proportion of transfected cells in the samples was 20–25% for all constructs in all experiments. Results first confirmed that –1 frameshifted proteins are only present in polyglutamine containing cells (Figs 1B and 2A and D). Figure 2D shows the results of a flow cytometry experiment where the lack of HA-positive events in the upper-left quadrant and their presence in the upper-right quadrant indicate that the presence of frameshifted molecules is linked to the expression of the polyglutamine frame. In addition, the presence of events labeled only for AU1 (frame 0) (lower-right quadrant) confirms that not all transfected cells are prone to frameshifting. When results were compiled for cells transfected with different CAG lengths, we found that as the size of the repeat increases, frameshifting frequencies increase (Fig. 2E). CAG repeats in the normal range for MJD-1 (less than 40 CAG) displayed few HA-positive events (at levels similar to that of isotypic controls, ~0.7% of the total population or ~5% of the transfected cells in Figs 2D and E), whereas with repeats close to or above the pathological threshold (approximately 60 CAG), frequencies of transfected cells expressing frameshifted molecules increased dramatically, peaking at 47% of transfected cells with 78 CAG repeats (Fig. 2E). The frequencies shown in Figure 2E are consistent with western blot results where we showed the presence of –1 frameshifted proteins in cells transfected with constructs primarily above the pathological threshold. These observations are consistent with the notion that –1 frameshifting might play a role in pathogenesis, as it appears mainly when the pathological CAG size threshold is reached and its frequency, as measured at the flow cytometry and western blot levels, increases with the size of the CAG repeat.

Frameshifting is specific to CAG-encoded polyglutamine repeats
We hypothesized that –1 frameshifting in long CAG tracts leads to the formation of polyalanine-containing proteins and that these contribute to the cellular toxicity. To discriminate between the toxicity caused by proteins containing exclusively polyglutamines and those frameshifted to polyalanine, we modified our constructs by replacing the CAG repeat by a polyglutamine-encoding (CAA)73 repeat. The rationale behind this comparison is that in their main frame (0 frame), both encode polyglutamine repeats but if frameshifted, only the CAG repeat codes for polyalanine, whereas the CAA repeat does not (Fig. 3A). The same epitopes were placed in each of the reading frames. The constructs were transfected into COS-1 cells, and frameshifting was assessed as described earlier.



View larger version (45K):
[in this window]
[in a new window]
  		Figure 3. Frameshifting does not occur in long polyglutamine-encoding CAA tracts. (A) Polyglutamine-encoding repeats and their respective reading frames. (B) (CAG)78 and (CAA)73 constructs were transfected into COS-1 cells. Immunocytochemistry was performed 24 h post-transfection using monoclonal anti-AU1 (Gln frame) and rabbit polyclonal anti-AU5 (+1 frame) and anti-HA (–1 frame) antibodies. FITC-conjugated anti-mouse and rhodamine-conjugated anti-rabbit secondary antibodies were used. (C) Flow cytometry was performed on transfected COS-1 cells using FITC-conjugated anti-AU1 and R-phycoerythrin-conjugated anti-HA antibodies. The percentage of frameshifting is expressed as the proportion of HA-positive events reported over the number of AU1 events. The results presented here are the mean of at least four independent readings from three different experiments±SEM.

 
Immunocytochemistry performed using antibodies against each of the epitopes showed that slippage to the +1 frame does not occur when cells are transfected with this construct and confirmed that –1 frameshifting is observed almost exclusively with the construct containing long CAG repeats (Fig. 3B). A western blot was performed using the soluble fraction of a protein extract. Results showed that the polyglutamine-containing protein is produced from the CAA construct at levels similar to those obtained for the CAG-containing constructs (Fig 2B). However, when total protein extracts were dot blotted onto nitrocellulose, we observed a 6-fold increase in AU1 immunoreactivity when compared with extracts of cells transfected with short CAG repeats (Fig. 2C). As the (CAA)73 construct is transcribed and translated at rates similar to short CAG constructs, this suggests that long polyglutamine stretches produced by CAA repeats accumulate in a fashion similar to their long CAG counterpart. It is critical that the levels of expression of (CAG)78 and (CAA)73 constructs be similar. Data shown in Figure 2B and C are representative of all the experiments and show similar levels of expression. Using densitometry, we find that when expression is calculated relative to actin levels, both constructs are expressed at the same level. In Figure 2B, the (CAG)78 is expressed 1.3 times higher than the (CAA)73 construct, whereas on the dot blot (Fig. 2C), this ratio is 0.97. Overall, these results show that both constructs are expressed at similar levels. When the membranes were probed with an antibody against the HA epitope (–1 frame), the signal was similar to that detected for short CAG repeats, confirming that –1 frameshifting does not occur within long CAA repeats (Fig. 2B and C).

Flow cytometry experiments revealed that the transfection efficiency of the constructs with long CAG and long CAA repeats were similar (data not shown). Despite its long polyglutamine-encoding repeat, the (CAA)73 construct showed no difference in –1 frameshifting frequency when compared with those observed with the shortest CAG repeats (7.3 versus 5.2% of the transfected population, P>0.3) (Figs 2E and 3C). Taken together, these results strongly argue that –1 frameshifting is sequence specific in that it only occurs in cells transfected with a long polyglutamine-encoding CAG repeat, but not in cells transfected with CAA repeats of a similar length.

Frameshifted proteins contribute to cell toxicity
Examination of cell morphology showed that cells transfected with the (CAG)78 construct displayed an altered morphology and showed vacuolization more frequently than cells transfected with the (CAA)73 construct (data not shown). Nuclear morphology is frequently used as an indicator for cell toxicity. DAPI staining and immunocytochemistry using anti-AU1 and anti-HA antibodies were performed to assess nuclear morphology. Figure 4A shows that expression of a long CAA repeat, although resulting in the formation of AU1+ intranuclear aggregates, does not affect the nuclear morphology (Fig. 4Aa and b). Transfection of the (CAG)38 construct resulted in diffuse AU1 staining and did not alter nuclear morphology (Fig. 4Ac and d). In contrast, cells transfected with the long (CAG)78 repeat displayed altered nuclei (Fig. 4Ae–h) that are preferentially associated with protein accumulation in the cytoplasm. Affected nuclei show an invaginated shape and are surrounded by aggregates containing the –1 frameshifted proteins (Fig. 4Af and h), whereas cells harboring very large polyglutamine inclusions still display normal nuclear morphology (Fig. 4Ae and g). This suggests that polyalanine accumulation resulting from –1 frameshifting in CAG repeats might contribute to toxicity.



View larger version (35K):
[in this window]
[in a new window]
  		Figure 4. Polyalanine-containing proteins contribute to cell toxicity. (A) Polyalanine-frameshifted proteins are associated with altered nuclear morphology. COS-1 cells were transfected with constructs (CAA)73 (a and b), (CAG)38 (c and d) or (CAG)78 (e–h). Immunocytochemistry was performed with antibodies against AU1 (a, c, e and g) or HA (b, d, f and h), and nuclei were counterstained with DAPI. (a–f) at 200x magnification, whereas (g and h) at 400x magnification. (B) Increased cell toxicity in cells expressing (CAG)78 repeat. Stable ecdysone-inducible COS-1 cells were serum deprived and induced for 24 and 48 h with ponasterone A (5 µM) to express the (CAG)78, (CAA)73 or (CAG)27 transgenes. Cells were stained with propidium iodide, and random fields were counted. Data are represented as fold increase in the cell death fraction of induced cells over the cell death fraction of non-induced cells. Results from three individual experiments with cell counts of 600–3000 cells per treatment for each cell line were pooled together and calculated.

 
Light scattering properties of a cell are directly related to its surface characteristics (side scatter channel, SSC) and its size (forward scatter channel, FSC). Increased values in both channels are a measure of morphological changes in dying cells such as cell swelling, chromatin condensation and nuclear fragmentation (33Go–35Go). We compared the light scattering properties of cells transfected with various polyglutamine-encoding constructs from at least five readings from three independent experiments. Data are expressed as the mean value of the population. As illustrated in Figure 2D, cells can be divided into distinct populations according to their antigen reactivity: untransfected cells (population 1, AU1–, HA–), cells expressing polyglutamine only (population 2, AU1+, HA–) and cells expressing frameshifted molecules (population 3, AU1+, HA+). Using the gating function of the flow cytometry software, we computed the mean light scattering value of each of the three populations in both the FSC and the SSC (Table 1). We first analyzed each population of cells individually. For each construct, values were compared with those obtained using the construct with a CAG repeat of a normal length, (CAG)25, which were arbitrarily set at 100. Results for population 1 (AU1–, HA–) showed no statistical difference between the various constructs. When the same comparisons were performed for population 2 (AU1+, HA–), results once again showed no significant difference between the various constructs. This suggests that the length of the polyglutamine repeat alone has no effect on cell toxicity as measured in this assay.


View this table:
[in this window]
[in a new window]
  		Table 1. Mean light scattering values of transfected COS-1 cells
 
We next compared the values obtained for population 2 (AU1+, HA–) with untransfected cells (population 1). The results for all the constructs showed that the values in both FSC and SSC tend to be higher in population 2 than in population 1 (AU1–, HA–) (P-values: 0.008–0.182). As these were similar for all constructs, it is most likely that they are a result of cell transfection rather than a reflection of a toxic effect of polyglutamine expression.

The only constructs that lead to a significant number of cells in population 3 (AU1+, HA+) were the constructs with long CAG repeats; therefore, we were not able to compare their light scattering values with that of a normal CAG repeat. However, when we compared the light scattering properties of population 3 (AU1+, HA+) with that of the two other populations, the values for cells expressing frameshifted molecules tend to be significantly higher than those for untransfected cells or cells expressing only polyglutamine-containing proteins. The most significant differences were obtained when comparing the different populations of the construct with the longest CAG. The P-values obtained were P<0.001 (population 1 versus population 3) and P<0.0001 (population 2 versus population 3) when comparing FSC values; and P<0.003 (population 1 versus population 3) and P<0.0001 (population 2 versus population 3) when comparing SSC values.

The experiments described earlier were performed using transient transfections and are only indicative of potential cell toxicity. The presence of a significant proportion of untransfected cells in the samples hinders accurate toxicity measurements. To overcome this obstacle, we established stable transfectants to assay cell toxicity using a more stringent approach. COS-1 cells were selected to stably express our truncated MJD-1 cDNA transgenes (CAG)25, (CAG)78 or (CAA)73 placed under the control of the Ecdysone-Inducible Mammalian Expression System (Invitrogen). Inducibility and levels of transgene expression were confirmed by western blots (data not shown). Recent studies have demonstrated that serum starvation and cell cycle arrest enhance cellular toxicity in models of polyglutamine disorders (36Go,37Go). In particular, Yoshizawa et al. (36Go) showed that aggregation of ataxin-3 bearing 77 glutamines was similar in the presence or in the absence of serum, but that toxicity was very low under normal growth conditions. When cells were deprived of serum, the observed toxicity increased by up to 6-fold. In our assay, stable ecdysone-inducible COS-1 cells were serum deprived, and transgene expression was induced for 24 and 48 h with ponasterone A (5 µM). Cells were stained with propidium iodide, and the ratio of dead cells in post-/pre-induction samples was calculated. Results show that cells expressing (CAG)25 display little change in cell death when compared with induced and non-induced populations, confirming that the expression of a short CAG repeat has little toxic effects (Fig. 4B). An increase in cell death over time was noted for constructs (CAG)78 and (CAA)73, but it was much stronger in (CAG)78-expressing cells than in the (CAA)73 samples (Fig. 4B, P<0.0001), suggesting that accumulation of proteins resulting from –1 frameshifting in the CAG repeats plays a role in cellular toxicity. Altogether, these results support our hypothesis that polyalanine-containing proteins contribute to toxicity in expanded CAG tract disorders.

Mechanism of frameshifting
Experiments with the (CAG)78 construct revealed that –1 frameshifting events occur in 47% of the transfected cells, whereas +1 frameshifting was not detectable. Several mechanisms could account for frameshifted molecules in our model. The existence of –1 frameshifted proteins can result from contamination of the plasmid with frameshifted clones. Prior to transfection, all the constructs were sequenced and no mutations were detected. In addition, the following observations strongly argue for the absence of such frameshifted DNA mutants. First, significant levels of +1 frameshifting were never observed. If mutations were to occur randomly in plasmid DNA and account for –1 frameshifting, one would expect to detect as much +1 frameshifting. Secondly, frameshifting is CAG repeat length dependent (discussed earlier). If mutations were to occur in the DNA, the odds of detecting frameshifting would be equal for all constructs. Finally, cells expressing the –1 frameshifted protein alone were not observed. Proteins in the polyglutamine frame were always detectable in those cells. If frameshifted molecules were produced by mutant plasmids, one would expect to find cells expressing only the frameshifted proteins.

If the DNA is void of mutations, frameshifting must be due to transcriptional or translational errors. To determine whether transcriptional errors occur, the construct (CAG)78 was transfected into COS-1 cells and RNA was isolated after 30 h. RT–PCR was performed on the truncated MJD-1 cDNA using a proof-read Taq DNA polymerase. The PCR product was introduced into a vector and individual clones were isolated. A total of 119 clones were sequenced. Eighteen clones (15%) displayed mutations: six missense, seven silent, three nonsense, one rearranged repeat and one causing +1 frameshifting. None of the 119 clones analyzed showed a –1 frameshifting mutation, despite the fact that –1 frameshifting was observed in 47% of the cells transfected with the very same construct. The absence of –1 frameshifting mutations in 119 clones argues that transcriptional errors do not account for the observed frequency of frameshifting.

Programed ribosomal frameshifting (PRF) is a mechanism used to encode two peptides from the same RNA sequence. It is usually mediated by the presence of a strong mRNA secondary structure and a ‘slippery’ sequence. The secondary structures involved are usually pseudo-knots, although some reports suggest that other structures, such as hairpins, might also play a role (38Go,39Go). During translation, these structures induce the ribosome to stall over the slippery sequence, allowing the ribosome-bound tRNA to slip and pair its non-wobble bases with the out-of-frame codon (40Go).

We hypothesize that frameshifting occurs following translational slippage due to secondary structures in the CAG repeat. To address this question, we first examined the predicted occurrence of RNA secondary structures, using the MFOLD program (41Go): results showed that all the CAG repeat constructs adopted a similar structure and that the free energy values (dG) linearly increased as the length of the CAG repeat increased (Fig. 5A). When we replaced the CAG repeat by a long CAA repeat, the energy requirements were reduced below the level of constructs with short CAG repeat (Fig. 5A). As illustrated in Figure 5B, the MFOLD program predicts that only the construct with the CAG repeat forms a hairpin, whereas the CAA repeat forms a single large loop. A recent study by Michlewski and Krzyzosiak confirmed the existence of such secondary structures in the SCA3 transcript. Their results showed that MJD-1 transcripts with short and expanded CAG repeats adopt several conformations suggesting that they are dynamic structures, but as the free energy increases with the length of the repeat, these structures might become more stable (42Go). This supports our hypothesis and suggests that the CAG repeat might form relatively stable structures and therefore might be able to stall the ribosome and induce –1 frameshifting.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 5. Secondary structure and energy requirement of polyglutamine encoding repeats. (A) Energy requirements were calculated for all constructs using the MFOLD program and were plotted as a function of the length of the CAG repeat. The (CAA)73 construct was also computed. Frameshifting frequencies are plotted against the secondary axis for comparison. (B) Secondary structure predictions. The mRNA produced by the (CAG)78 and (CAA)73 constructs was analyzed using the MFOLD program. The structures represented here are the optimal predictions.

 
Key steps have been identified in PRF; drugs inhibiting or stimulating –1 frameshifting through different ribosomal mechanisms have also been identified. Anisomycin is a ribosome-specific inhibitor, which decreases –1 PRF by inhibiting the accommodation of the frameshifted tRNA to its new codon (40Go,43Go,44Go). Sparsomycin stimulates –1 frameshifting by slowing peptidyl transfer, thus causing the ribosome to slow down and change frame (40Go,43Go).

To determine whether the observed frameshifting is due to ribosomal frameshifting, COS-1 cells were transfected with the (CAG)78 construct, grown for 24 h to allow expression of the transgene and then treated with concentrations of anisomycin or sparsomycin ranging from 0.5 to 4 µM for 24 h. Initial experiments (data not shown) demonstrated that a 2 µM concentration had optimal effects, although having minimal toxic effects. The rate of frameshifting was measured by flow cytometry using untreated cells as control. Anisomycin-treated cells displayed a 31% reduction in the frequency of –1 frameshifting when compared with untreated cells (P<0.001). In comparison, cells treated with sparsomycin displayed a 26% increase in their frequency of frameshifting when compared with controls (P<0.05) (Fig. 6A). Immunocytochemistry experiments confirmed these observations (data not shown). These data support our contention that –1 frameshifting results from translational slippage of the ribosome.



View larger version (21K):
[in this window]
[in a new window]
  		Figure 6. Anisomycin and sparsomycin treatment affects –1 frameshifting frequency and toxicity. (A) COS-1 cells were transfected with the (CAG)78 construct and treated with 2 µM anisomycin or sparsomycin for 24 h. Flow cytometry was performed on a FACSCAN apparatus using FITC-conjugated anti-AU1 and R-phycoerythrin-conjugated anti-HA antibodies. Frameshifting frequencies were normalized relative to the value observed for untreated cells. The values represent the mean of at least four independent readings from at least two experiments±SEM. (B) Stable ecdysone-inducible COS-1 cells were serum deprived and induced for 24 and 48 h with ponasterone A (5 µM) to express the (CAG)78 construct. Where indicated, cells were also treated with 2 µM anisomycin. Cells were stained with propidium iodide, and random fields were counted. Data are represented as fold increase in the cell death fraction of induced cells over the cell death fraction of non-induced cells. Results from three individual experiments with cell counts of 600–3000 cells per treatment for each cell line were pooled together and calculated.

 
Treatment with an inhibitor of ribosomal frameshifting reduces cell toxicity
Using stable transfectants expressing an inducible (CAG)78 transgene, we assessed cell toxicity (as described earlier) in samples treated with 2 µM anisomycin or untreated. Results showed that treatment with anisomycin reduced toxicity to levels comparable with those observed in non-induced cells (Fig. 6B). This suggests that at least part of the toxicity observed in cells expressing very long CAG tracts is indeed due to ribosomal slippage to the –1 frame.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Gaspar et al. (22Go) reported the identification of –1 frameshifted proteins in samples from SCA3 patients and MJD-1 transfected cells. The experiments presented here were aimed at characterizing this observation. Our results show that only –1 frameshifting occurs, as we have not been able to detect +1 frameshifting. Furthermore, –1 frameshifting proved to be specific to CAG repeats and to be dependent on the length of the repeat. This is in agreement with clinical observations, where the length of the CAG expansion correlated with the disease severity and age of onset (5Go).

To determine the contribution of polyalanine accumulation to toxicity, we modified the MJD-1 construct by replacing the longest CAG repeat by a polyglutamine-encoding CAA repeat of similar size. While encoding for polyglutamine in the 0 frame, the CAA repeat does not encode for polyalanine stretches upon frameshifting. Therefore, comparison of toxicity in cells harboring long CAG or CAA repeats should give indications of the contribution of frameshifting to toxicity. We show that (CAA)73 did not frameshift more than constructs with very short CAG repeats. Furthermore, morphological observations suggest increased toxicity of the CAG repeat over the CAA repeat of similar size. In our cell system, nuclear staining experiments demonstrated that toxicity is effectively associated with the expression of long CAG tracts and not with the long CAA repeat or the short CAG repeat. Only cells expressing –1 frameshifted proteins often showed nuclei with a lobular fragmented shape, suggesting an increased toxicity due to the expression of the potential polyalanine tracts. Using the light scattering properties of the cells as an indicator of toxicity, we showed that there is no significant difference between cells expressing various lengths of polyglutamine repeats, but the cells expressing –1 frameshifted molecules display increased values. Furthermore, results from a propidium iodide staining experiment also showed higher cell toxicity in cells expressing a long CAG tract over cells expressing a short CAG tract or a long polyglutamine-encoding CAA repeat. The observation that all three assays revealed increased physiological alterations in the group of cells expressing frameshifted molecules suggests that the expression of polyalanine-containing proteins resulting from that frameshifting does contribute to cell toxicity. This is in agreement with previous results from our group and other laboratories showing higher toxicity of polyalanine over polyglutamine (22Go,23Go).

The predicted formation of potential ribosome-stalling mRNA secondary structures and our experiments with anisomycin and sparsomycin provide a strong indication that this frameshift happens during translation and may involve ribosome pausing with slippage into the –1 frame. We propose a model similar to that seen in –1 programed ribosome frameshifting. During translation, expanded CAG repeats form transient hairpins of varying lengths, which would sometimes be of sufficient stability to stall the ribosome. Depending on the site of pausing and the surrounding sequences, some ribosomes will have frameshifted –1 prior to resuming translation. Although we have not unequivocally demonstrated the presence of polyalanine-frameshifted residues, it is interesting to speculate that pausing and frameshifting should occur most frequently at the beginning of the CAG tract, where the secondary mRNA structure formed downstream of the ribosome would be most stable. Figure 2B suggests that this may be so, as there seems to be a more prominent band [especially in the (CAG)59 and (CAG)63 constructs] at the predicted molecular weight for a protein containing mostly alanines, which is compatible with preferential frameshifting at the beginning of the CAG tract. Drugs, such as anisomycin, would reduce frameshifting by preventing the frameshifted ribosome-bound tRNA to accommodate a new codon. In contrast, sparsomycin would favor frameshifting by delaying the transfer of the elongating chain. Finally, our experiments showed that cells expressing long CAG repeats and treated with anisomycin display toxicity levels very similar to those expressing short CAG tracts or CAA repeats. As –1 frameshifting is only observed in cells expressing long CAG tract, and that anisomycin reduces both –1 frameshifting and cell toxicity in those cells, we conclude that toxicity must at least partly arise from –1 frameshifting and the resulting expression of polyalanine-containing proteins. Modulation of ribosomal frameshifting and therefore of polyalanine toxicity with drugs, such as anisomycin, could lead to new therapeutic opportunities.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
cDNA constructs
The constructs used are based on plasmids previously described (22Go). All constructs were placed under the control of the CMV promoter of plasmid pEGFP-N1. The enhanced green fluorescent protein (EGFP) encoding sequences were removed by BamHI–NotI digestion. The MJD-1 cDNA was truncated by PCR to the last methionine residue preceding the CAG repeat. A Kozak consensus sequence was introduced by site-directed mutagenesis using primer MJDsh–BamHI (5'-GGGCAGAAGAAACAGATCCAAACCATGGCACAGACATCAGGTACAAATC-3') and its complement. PCR was also used to introduce the epitope tags at the 3' end. Sequential nested amplifications were performed to introduce the AU1 (DTYRYI) and AU5 (TDFYLK) tags 3' of the sequences encoding the HA tag in pMJD-11.

Modification of the length of the CAG repeat was performed by transforming XL-1 blue bacteria with a construct with 78 CAG repeats. Bacteria were grown overnight at 37°C and the plates were left at 4°C for one night. Colonies were picked and plasmids were isolated and sequenced. Once clones were identified, large-scale purification of plasmid DNA was performed and the products sequenced.

Pure CAA repeats were generated by PCR in the absence of exogenous template. Briefly, 4 µM of complementary primers (CAA)10 and (TTG)10 were mixed in the presence of 1.5 mM MgCl2, 50 mM KCl, 10 mM Tris–HCl, pH 8.3, 200 µM dNTP and 1 U of Taq DNA polymerase. The products were cloned into EcoRV digested pBluescript SKII+ (Stratagene). A clone containing (CAA)73 was isolated and the repeats were introduced back into the truncated MJD constructs to replace the CAG repeat.

For the ecdysone-inducible system, cDNAs were excised with BamHI and NotI from the constructs described earlier and cloned in the pIND vector (Invitrogen).

Cell culture and transfection
All cell lines were cultured at 37°C in a humid atmosphere provided with 5% CO2 in recommended medium. Transfections were performed using Lipofectamine Plus Reagent (Invitrogen), according to the manufacturer's instructions. For immunocytochemistry, 3x105 cells were plated in six-well plates 24 h prior to the transfection. For flow cytometry, 1x106 cells were plated in 100 mm dishes and transfected the next day. Stable transfectants were created by transfecting pIND-MJD constructs (described earlier) in pVgRXR-transfected COS-1 cells. Individual clones were selected and maintained in geneticin/zeocin containing medium. Where indicated, cells were treated with 5 µM ponasterone A, 2 µM anisomycin or 2 µM sparsomycin.

Immunocytochemistry
Cells were fixed in 4% paraformaldehyde followed by permeabilization in 0.2% Triton-X-100. Immunodetection was performed using mouse monoclonal anti-AU1 (1:2000, Covance Research Products), rabbit polyclonal anti-AU5 (1:2000, Abcam) and rabbit polyclonal anti-HA (1:1000, Santa Cruz Biotechnology) antibodies. FITC-, Cy3- or rhodamine red-X-conjugated secondary antibodies (Jackson Immunoresearch) were used at 1:300. Where indicated, cells were mounted in DAPI-containing mounting medium (Vector Laboratories).

Immunoblotting
Total protein extracts were prepared by homogenizing cells in protein extraction buffer [70 mM Tris–HCl, pH 6.8, 10% glycerol, 3% sodium dodecly sulfate (SDS) and 700 mM 2-mercaptoethanol] followed by centrifugation at 14 000g. About 100 µg was electrophoresed on 10% SDS–polyacrylamide gels and was transblotted onto nitrocellulose membranes. Immunodetection was performed using mouse monoclonal anti-AU1 (1:5000), rabbit polyclonal anti-HA (1:1000) or rabbit polyclonal anti-actin (1:1000, Santa Cruz Biotechnology) antibodies. Results were visualized by chemiluminescence. Dot blots were prepared by blotting 25 µg of total proteins onto nitrocellulose membranes using a vacuum manifold and were processed as described for western blots.

Flow cytometry
Flow cytometry was performed on 5x105 cells transfected 24 h prior to the experiment. Cells were fixed and permeabilized using the Cytofix/Cytoperm kit (BD Biosciences), and subsequent steps were carried according to the manufacturer's instructions. The antibodies used were a goat polyclonal FITC-conjugated anti-AU1 (Bethyl Laboratories) and a rabbit polyclonal R-phycoerythrin-conjugated anti-HA (Santa Cruz Biotechnology). Matching isotypic antibodies were used as controls (Southern Biotechnology Associates). The samples were read on a FACSCAN apparatus (BD Biosciences). All cytometric measurements were performed at least three times.

Cell death assay
The day before induction, cells were plated in six-well plates. After 24 h, medium was aspirated and cells were washed with Hank's buffer to remove fetal bovine serum. The medium was replaced with serum-free medium and cells were induced with 5 µM ponasterone A for 24 or 48 h. Where indicated, cells were treated with 2 µM anisomycin. Non-induced controls were included for all cell lines at every time point. After induction, cells were harvested, washed and stained with propidium iodide (50 ng/ml, Sigma). Cells were washed and observed under a fluorescence microscope. Sample cell populations were chosen randomly. The number of cells with propidium iodide uptake over the total number of cells in the sample population was calculated and the proportions were compared between the non-induced and induced cells for each cell line as described by Giuliano et al.


   ACKNOWLEDGEMENTS
 
The authors would like to thank Dr Jiangping Wu for helpful discussion, Dr Marie-Pierre Dubé for assistance with statistical analysis and Daniel Rochefort, Eric Massicotte and Martine Dupuis for technical assistance. This work was supported by grants from the Canadian Institutes of Health Research and from the N.C.E. Human Genetics to G.A.R. A.T. was the recipient of a post-doctoral fellowship from the Fonds de Recherche en Santé du Québec. C.G. was the recipient of a post-doctoral fellowship from Fundação para a Ciência e Tecnologia, Portugal. F.A.-Y. was the recipient of a McGill University Scholarship.

Conflict of Interest statement. The authors have no conflict of interest.


   FOOTNOTES
 
{dagger} Present address: Department of Anatomy/Neuroscience, University College Cork, Cork, Ireland. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. 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. and Zoghbi, H.Y. (1993) Expansion of an unstable trinucleotide CAG repeat in spinocerebellar ataxia type 1. Nat. Genet., 4, 221–226.[CrossRef][ISI][Medline]

  2. Imbert, G., Saudou, F., Yvert, G., Devys, D., Trottier, Y., Garnier, J.M., Weber, C., Mandel, J.L., Cancel, G., Abbas, N. et al. (1996) Cloning of the gene for spinocerebellar ataxia 2 reveals a locus with high sensitivity to expanded CAG/glutamine repeats. Nat. Genet., 14, 285–291.[CrossRef][ISI][Medline]

  3. Pulst, S.M., Nechiporuk, A., Nechiporuk, T., Gispert, S., Chen, X.N., Lopes-Cendes, I., Pearlman, S., Starkman, S., Orozco-Diaz, G., Lunkes, A. et al. (1996) Moderate expansion of a normally biallelic trinucleotide repeat in spinocerebellar ataxia type 2. Nat. Genet., 14, 269–276.[CrossRef][ISI][Medline]

  4. Sanpei, K., Takano, H., Igarashi, S., Sato, T., Oyake, M., Sasaki, H., Wakisaka, A., Tashiro, K., Ishida, Y., Ikeuchi, T. et al. (1996) Identification of the spinocerebellar ataxia type 2 gene using a direct identification of repeat expansion and cloning technique, DIRECT. Nat. Genet., 14, 277–284.[CrossRef][ISI][Medline]

  5. 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. Nat. Genet., 8, 221–228.[CrossRef][ISI][Medline]

  6. Zhuchenko, O., Bailey, J., Bonnen, P., Ashizawa, T., Stockton, D.W., Amos, C., Dobyns, W.B., Subramony, S.H., Zoghbi, H.Y. and Lee, C.C. (1997) Autosomal dominant cerebellar ataxia (SCA6) associated with small polyglutamine expansions in the alpha 1A-voltage-dependent calcium channel. Nat. Genet., 15, 62–69.[CrossRef][ISI][Medline]

  7. David, G., Abbas, N., Stevanin, G., Durr, A., Yvert, G., Cancel, G., Weber, C., Imbert, G., Saudou, F., Antoniou, E. et al. (1997) Cloning of the SCA7 gene reveals a highly unstable CAG repeat expansion. Nat. Genet., 17, 65–70.[CrossRef][ISI][Medline]

  8. Nakamura, K., Jeong, S.Y., Uchihara, T., Anno, M., Nagashima, K., Nagashima, T., Ikeda, S., Tsuji, S. and Kanazawa, I. (2001) SCA17, a novel autosomal dominant cerebellar ataxia caused by an expanded polyglutamine in TATA-binding protein. Hum. Mol. Genet., 10, 1441–1448.[Abstract/Free Full Text]

  9. La Spada, A.R., Wilson, E.M., Lubahn, D.B., Harding, A.E. and Fischbeck, K.H. (1991) Androgen receptor gene mutations in X-linked spinal and bulbar muscular atrophy. Nature, 352, 77–79.[CrossRef][Medline]

  10. Nagafuchi, S., Yanagisawa, H., Sato, K., Shirayama, T., Ohsaki, E., Bundo, M., Takeda, T., Tadokoro, K., Kondo, I., Murayama, N. et al. (1994) Dentatorubral and pallidoluysian atrophy expansion of an unstable CAG trinucleotide on chromosome 12p. Nat. Genet., 6, 14–18.[CrossRef][ISI][Medline]

  11. Koide, R., Ikeuchi, T., Onodera, O., Tanaka, H., Igarashi, S., Endo, K., Takahashi, H., Kondo, R., Ishikawa, A., Hayashi, T. et al. (1994) Unstable expansion of CAG repeat in hereditary dentatorubral–pallidoluysian atrophy (DRPLA). Nat. Genet., 6, 9–13.[CrossRef][ISI][Medline]

  12. The Huntington's Disease Collaborative Research Group. (1993) A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington's disease chromosomes. Cell, 72, 971–983.[CrossRef][ISI][Medline]

  13. Cummings, C.J. and Zoghbi, H.Y. (2000) Trinucleotide repeats: mechanisms and pathophysiology. Annu. Rev. Genomics Hum. Genet., 1, 281–328.[CrossRef][ISI][Medline]

  14. Cummings, C.J., Mancini, M.A., Antalffy, B., DeFranco, D.B., Orr, H.T. and Zoghbi, H.Y. (1998) Chaperone suppression of aggregation and altered subcellular proteasome localization imply protein misfolding in SCA1. Nat. Genet., 19, 148–154.[CrossRef][ISI][Medline]

  15. Chai, Y., Koppenhafer, S.L., Shoesmith, S.J., Perez, M.K. and Paulson, H.L. (1999) Evidence for proteasome involvement in polyglutamine disease: localization to nuclear inclusions in SCA3/MJD and suppression of polyglutamine aggregation in vitro. Hum. Mol. Genet., 8, 673–682.

  16. Kazantsev, A., Preisinger, E., Dranovsky, A., Goldbager, D. and Housman, D. (1999) Insoluble detergent-resistant aggregates form between pathological and non pathological lengths of polyglutamine in mammalian cells. Proc. Natl Acad. Sci. USA, 96, 11404–11409.[Abstract/Free Full Text]

  17. Nucifora, F.C., Sasaki, M., Peters, M.F., Huang, H., Cooper, J.K., Yamada, M., Takahashi, H., Tsuji, S., Transoco, J., Dawson, V.L. et al. (2001) Interference of huntingtin and atrophin-1 with cbp-mediated transcription leading to cellular toxicity. Science, 291, 2423–2428.[Abstract/Free Full Text]

  18. Wyttenbach, A., Sauvageot, O., Carmichael, J., Diaz-Latoud, C., Arrigo, A.P. and Rubinsztein, D.C. (2002) Heat shock protein 27 prevents cellular polyglutamine toxicity and suppresses the increase of reactive oxygen species caused by huntingtin. Hum. Mol. Genet., 11, 1137–1151.[Abstract/Free Full Text]

  19. Brais, B., Bouchard, J.P., Xie, Y.G., Rochefort, D.L., Chretien, N., Tome, F.M., Lafreniere, R.G., Rommens, J.M., Uyama, E., Nohira, O. et al. (1998) Short GCG expansions in the PABP2 gene cause oculopharyngeal muscular dystrophy. Nat. Genet., 18, 164–167.[CrossRef][ISI][Medline]

  20. Shanmugam, V., Dion, P., Rochefort, D., Laganiere, J., Brais, B. and Rouleau, G.A. (2000) PABP2 polyalanine tract expansion causes intranuclear inclusions in oculopharyngeal muscular dystrophy. Ann. Neurol., 48, 798–802.[CrossRef][ISI][Medline]

  21. Abu-Baker, A., Messaed, C., Laganiere, J., Gaspar, C., Brais, B. and Rouleau, G.A. (2003) Involvement of the ubiquitin–proteasome pathway and molecular chaperones in oculopharyngeal muscular dystrophy. Hum. Mol. Genet., 12, 2609–2623.[Abstract/Free Full Text]

  22. Gaspar, C., Jannatipour, M., Dion, P., Laganiere, J., Sequeiros, J., Brais, B. and Rouleau, G.A. (2000) CAG tract of MJD-1 may be prone to frameshifts causing polyalanine accumulation. Hum. Mol. Genet., 9, 1957–1966.[Abstract/Free Full Text]

  23. Ravikumar, B., Duden, R. and Rubinsztein, D.C. (2002) Aggregate-prone proteins with polyglutamine and polyalanine expansions are degraded by autophagy. Hum. Mol. Genet., 11, 1107–1117.[Abstract/Free Full Text]

  24. Jarvik, J.W. and Telmer, C.A. (1998) Epitope tagging. Annu. Rev. Genet., 32, 601–618.[CrossRef][ISI][Medline]

  25. 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.[Abstract/Free Full Text]

  26. Cancel, G., Durr, A., Didierjean, O., Imbert, G., Burk, K., Lezin, A., Belal, S., Benomar, A., Abada-Bendib, M., Vial, C. et al. (1997) Molecular and clinical correlations in spinocerebellar ataxia 2: a study of 32 families. Hum. Mol. Genet., 6, 709–715.[Abstract/Free Full Text]

  27. Maciel, P., Gaspar, C., DeStefano, A.L., Silveira, I., Coutinho, P., Radvany, J., Dawson, D.M., Sudarsky, L., Guimaraes, J., Loureiro, J.E. et al. (1995) Correlation between CAG repeat length and clinical features in Machado–Joseph disease. Am. J. Hum. Genet., 57, 54–61.[ISI][Medline]

  28. Riess, O., Epplen, J.T., Amoiridis, G., Przuntek, H. and Schols, L. (1997) Transmission distortion of the mutant alleles in spinocerebellar ataxia. Hum. Genet., 99, 282–284.[CrossRef][ISI][Medline]

  29. Illiaroshkin, S.N., Igarashi, S., Onedara, O., Markova, E.D., Nikolskaya, N.N., Tanaka, H., Chabrashwili, T.Z., Insarova, N.G., Endo, K., Ivanova-Smolenskaya, I.A. et al. (1994) Trinucleotide repeat length and rate of progression of Huntington's disease. Ann. Neurol., 36, 630–635.[CrossRef][ISI][Medline]

  30. Trottier, Y., Biancalana, V. and Mandel, J.-L. (1994) Instability of CAG repeat in Huntington's disease: relation to parental transmission and age of onset. J. Med. Genet., 31, 377–382.[Abstract]

  31. Dorsman, J.C., Pepers, B., Langenberg, D., Kerkdijk, H., Ijszenga, M., den Dunnen, J.T., Roos, R.A. and van Ommen, G.J. (2002) Strong aggregation and increased toxicity of polyleucine over polyglutamine stretches in mammalian cells. Hum. Mol. Genet., 11, 1487–1496.[Abstract/Free Full Text]

  32. Trottier, Y., Devys, D., Imbert, G., Saudou, F., An, I., Lutz, Y., Weber, C., Agid, Y., Hirsch, E.C. and Mandel, J.L. (1995) Cellular localization of the Huntington's disease protein and discrimination of the normal and mutated form [see comments]. Nat. Genet., 10, 104–110.[CrossRef][ISI][Medline]

  33. Darzynkiewicz, Z., Juan, G., Li, X., Gorczyca, W., Murakami, T. and Traganos, F. (1997) Cytometry in cell necrobiology: analysis of apoptosis and accidental cell death (necrosis). Cytometry, 27, 1–20.[CrossRef][ISI][Medline]

  34. Dive, C., Gregory, C.D., Phipps, D.J., Evans, D.L., Milner, A.E. and Wyllie, A.H. (1992) Analysis and discrimination of necrosis and apoptosis (programmed cell death) by multiparameter flow cytometry. Biochim. Biophys. Acta, 1133, 275–285.[Medline]

  35. Weber, G.F., Daley, J., Kraeft, S.K., Chen, L.B. and Cantor, H. (1997) Measurement of apoptosis in heterogenous cell populations. Cytometry, 27, 136–144.[CrossRef][ISI][Medline]

  36. Yoshizawa, T., Yamagishi, Y., Koseki, N., Goto, J., Yoshida, H., Shibasaki, F., Shoji, S. and Kanazawa, I. (2000) Cell cycle arrest enhances the in vitro cellular toxicity of the truncated Machado–Joseph disease gene product with an expanded polyglutamine stretch. Hum. Mol. Genet., 9, 69–78.[Abstract/Free Full Text]

  37. Giuliano, P., De Cristofaro, T., Affaitati, A., Pizzulo, G.M., Feliciello, A., Criscuolo, C., De Michele, G., Filla, A., Avvedimento, E.V. and Varrone, S. (2003) DNA damage induced by polyglutamine-expanded proteins. Hum. Mol. Genet., 12, 2301–2309.[Abstract/Free Full Text]

  38. Farabaugh, P.J. (1996) Programmed translational frameshifting. Annu. Rev. Genet., 30, 507–528.[CrossRef][ISI][Medline]

  39. Farabaugh, P.J. and Bjork, G.R. (1999) How translational accuracy influences reading frame maintenance. EMBO J., 18, 1427–1434.[CrossRef][ISI][Medline]

  40. Harger, J.W., Meskauskas, A. and Dinman, J.D. (2002) An ‘integrated model’ of programmed ribosomal frameshifting. Trends Biochem. Sci., 27, 448–454.[CrossRef][ISI][Medline]

  41. Mathews, D.H., Sabina, J., Zuker, M. and Turner, D.H. (1999) Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. J. Mol. Biol., 288, 911–940.[CrossRef][ISI][Medline]

  42. Michlewski, G. and Krzyzosiak, W.J. (2004) Molecular architecture of CAG repeats in human disease related transcripts. J. Mol. Biol., 340, 665–679.[CrossRef][ISI][Medline]

  43. Dinman, J.D., Ruiz-Echevarria, M.J., Czaplinski, K. and Peltz, S.W. (1997) Peptidyl-transferase inhibitors have antiviral properties by altering programmed –1 ribosomal frameshifting efficiencies: development of model systems. Proc. Natl Acad. Sci. USA, 94, 6606–6611.[Abstract/Free Full Text]

  44. Ioannou, M., Coutsogeorgopoulos, C. and Synetos, D. (1998) Kinetics of inhibition of rabbit reticulocyte peptidyltransferase by anisomycin and sparsomycin. Mol. Pharmacol., 53, 1089–1096.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles: