Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
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 ISI Web of Science
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 (31)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Cowan, K.J.
Right arrow Articles by Welch, W.J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cowan, K.J.
Right arrow Articles by Welch, W.J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2003, Vol. 12, No. 12 1377-1391
DOI: 10.1093/hmg/ddg151
© 2003 Oxford University Press

Polyglutamine protein aggregation and toxicity are linked to the cellular stress response

K.J. Cowan1, M.I. Diamond2 and W.J. Welch1,*

1Department of Surgery, University of California, San Francisco, Bldg 1, Room 210, San Francisco General Hospital, 1001 Potrero Ave, San Francisco, CA 94110, USA and 2Department of Neurology, University of California, San Francisco, CA, USA

Received February 6, 2003; Accepted April 7, 2003


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Chronic exposure of cells to expanded polyglutamine proteins results in eventual cell demise. We constructed mouse cell lines expressing either the full-length androgen receptor (AR), or truncated forms of AR containing 25 or 65 glutamines to study the cellular consequences of chronic low-level exposure to these proteins. Expression of the polyglutamine-expanded truncated AR protein, but not the full-length expanded protein, resulted in the formation of cytoplasmic and nuclear aggregates and eventual cell death. Nuclear aggregates preferentially stained positive for heat shock protein (hsp)72, a sensitive indicator of a cellular stress response. Biochemical studies revealed that the presence of nuclear aggregates correlated with activation of the c-jun NH2-terminal kinase (JNK). Different metabolic insults, including heat shock treatment, and exposure to sodium arsenite or menadione, proved more toxic to those cells expressing the polyglutamine-expanded truncated protein than to cells expressing the non-expanded form. Cells containing cytoplasmic polyglutamine–protein aggregates exhibited a delayed expression of hsp72 after heat shock. Once expressed, hsp72 failed to localize normally and instead was sequestered within the protein aggregates. This was accompanied by an inability of the aggregate-containing cells to cease their stress response as evidenced by the continued presence of activated JNK. Finally, activation of the cellular stress response increased the overall extent of polyglutamine protein aggregation, especially within the nucleus. Inclusion of a JNK inhibitor reduced this stress-dependent increase in nuclear aggregates. Abnormal stress responses may contribute to enhanced cell vulnerability in cells expressing polyglutamine-expanded proteins and may increase the propensity of such cells to form cytoplasmic and nuclear inclusions.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
At least nine neurological disorders derive from glutamine expansion within a particular protein. These include Huntington's disease, spinobulbar muscular atrophy (SBMA), and the spinocerebellar ataxias (reviewed in 1,2). Each disease targets specific neurons within the central nervous system, despite the ubiquitous expression of the polyglutamine-expanded protein in many other cell types. Vulnerable neurons in patients and transgenic animal models frequently contain insoluble protein aggregates within the nucleus, which appear to be a marker of polyglutamine toxicity (3–9). Using a cell culture model we previously reported that polyglutamine-expanded protein aggregates within the nucleus were associated with increased cell toxicity (10). In SBMA, polyglutamine expansion within the N-terminus of the androgen receptor (AR) results in selective degeneration of motor neurons in the spinal cord and brainstem. As in Huntington's disease, several lines of evidence suggest that AR toxicity may derive from cleavage of the full-length protein to produce a polyglutamine-containing N-terminal fragment that is capable of aggregation within the cell nucleus (3,4,11,12).

Previous studies have described the co-localization of different heat shock proteins (hsps) with the polyglutamine protein aggregates (13–15). Such co-localization may have adverse cellular consequences since many of the hsps act as molecular chaperones, important components that facilitate protein maturation and degradation events throughout the cell (reviewed in 16,17). Indeed, support for this idea follows from several studies where overexpression of hsp70 family members reduces the extent of polyglutamine-dependent protein aggregation and cellular toxicity (13,18–22). Thus, neurodegenerative diseases characterized by the presence of insoluble protein aggregates may arise and/or progress due to a chronic disturbance of cellular homeostasis arising from a reduction in essential cellular components such as the molecular chaperones. Moreover, such a reduction in the available levels of the molecular chaperones might also lead to enhanced vulnerability of the cells to routine metabolic insults including oxidative injury or excitotoxicity that occur within the normal lifespan of the cell. Thus, in the present study we hypothesized that chronic exposure to an expanded polyglutamine protein might perturb normal physiologic stress responses, thereby increasing cell vulnerability.

To test this hypothesis we engineered mouse fibroblasts to express the AR containing either 25 or 65 continuous glutamine residues [referred to as AR(25) and AR(65), respectively]. In addition, we constructed mouse fibroblasts expressing only the first 127 acids of the androgen receptor (N127), again containing either 25 (unexpanded) or 65 (expanded) continuous glutamine residues, under the control of the metallothionein promoter [referred to as N127(25) and N127(65), respectively]. As has been reported for many polyglutamine related diseases, expression of the unexpanded or expanded full-length forms of the protein did not result in the formation of visible protein aggregates, nor any obvious signs of cell toxicity (23,24). In contrast, expression of the truncated and polyglutamine-expanded N127(65) protein resulted in the formation of intracellular aggregates. Protein aggregates were initially restricted to the cytoplasm but over time were observed within the nucleus and were associated with increased cell demise. Cells containing nuclear inclusions preferentially stained positive for heat shock protein (hsp)72 and exhibited activation of the c-jun terminal kinase (JNK), indicative that these cells were undergoing a cellular stress response.

We also examined the ability of the cells to respond to different metabolic insults. Cells expressing the full-length forms of AR, with or without the polyglutamine expansion, showed no differences in their activation of the cellular stress response and exhibited similar cell survival profiles. In contrast, analysis of the stress response in the N127(25) and N127(65) expressing cells revealed that the presence of the polyglutamine-expanded protein aggregates correlated with a delay in the activation of the cellular stress response and increased cell demise. Those cells that did survive the stress treatment were found to significantly increase their overall number of nuclear and cytoplasmic polyglutamine-protein aggregates. These studies help provide a biological framework to better understand aspects of chronic cellular dysfunction observed in polyglutamine expansion diseases. Moreover, we suggest that various environmental factors that activate cellular stress responses may accelerate the disease process.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
To examine the consequences of chronic exposure to polyglutamine-containing proteins we developed cell lines expressing either the full-length androgen receptor, or a truncated form of the protein consisting of only the first amino-terminal 127 amino acids. The truncated form of the protein has previously been shown to produce neurodegeneration in animal models (25). Figure 1A shows a schematic of the constructs. Full-length AR, with either 25 or 65 continuous glutamine residues [referred to as AR(25) or AR(65), respectively], and containing a hemagluttinin (HA) epitope at the amino-terminus, was engineered behind a cytomegalovirus (CMV) promoter. Similarly, sequences encoding the amino-terminal 127 amino acids, again containing either 25 or 65 glutamines [referred to throughout as N127(25) or N127(65), respectively], were HA-tagged and inserted behind a regulatable promoter, metallothionein. The rationale for using a regulatable promoter followed from the afore-mentioned observations that continued expression of truncated, polyglutamine-expanded proteins can lead to cell death. For both the full-length proteins as well as the truncated proteins, cells were transfected and a number of stable cell lines were obtained and characterized.



View larger version (91K):
[in this window]
[in a new window]
  		Figure 1. Construction and analysis of 3T3 cells expressing a truncated form of the androgen receptor. (A) Expression plasmids containing the entire coding sequence for the androgen receptor, with either 25 or 65 contiguous glutamine residues (Q25 or Q65), were engineered behind a CMV promoter [top, referred to as AR(25) and AR(65)]. In parallel, plasmids expressing the first amino-terminal 127 amino acids of the androgen receptor, containing either 25 or 65 glutamine residues, were engineered behind a metallothionein promoter [bottom, referred to throughout as N127(25) or N127(65), respectively; note, truncation not drawn to scale]. A sequence encoding the HA epitope was inserted at the extreme amino-terminus of each protein to facilitate detection. DBD denotes DNA-binding domain; LBD denotes ligand-binding domain. (B) 3T3 cells were transfected with plasmids expressing the truncated androgen receptor proteins, N127(25) or N127(65), and stable cell lines selected as described in the Methods. ZnSO4 (50 µM) was added to the cell media to elicit expression of the N127(25) and N127(65) proteins. After 3 days of expression the cells were fixed with paraformaldehyde and analyzed for the presence of the N127(25) and N127(65) proteins by indirect immunofluorescence. Shown are the phase contrast and fluorescence micrographs. Shown in blue is DAPI staining of the nuclei and, in red, the N127(25) or N127(65) proteins. (C) Lysates were prepared from cells expressing either N127(25) or N127(65) for 7 days and analyzed by SDS–PAGE and western blotting using an anti-HA antibody. Shown is the Coomassie blue stained gel (lanes 1–3) and the corresponding western blot (lanes 4 and 5). Lanes 1 and 4: N127(25) cells. Lanes 2 and 5: N127(65) cells. Lane 3 contains molecular mass markers.

 
Analysis by immunofluorescence revealed that both AR(25) and AR(65) localized within the cytoplasm and the nucleus. Addition of a hormone agonist (dihydrotestosterone, DHT) resulted in a nuclear locale for both proteins (data presented later). Both forms of the protein displayed diffuse staining and there was no evidence of protein aggregation over the course of 2 weeks of expression (data not shown). In the case of the truncated proteins, the unexpanded N127(25) protein exhibited diffuse fluorescence throughout the cytoplasm. In contrast, the polyglutamine-expanded N127(65) protein readily formed cytoplasmic inclusions that were evident within 3 days after induction of the transgene by addition of ZnSO4 to the media (Fig. 1B). Analysis of the cell lysates by western blotting using the HA antibody revealed expression of either the N127(25) or the N127(65) proteins (Fig. 1C). For both of the truncated proteins two species were detected, the faster migrating forms probably representing a partial degradation fragment. Note that neither protein was obvious within the total cell lysates after Coomassie blue staining of the gel. Thus, the stable cell lines express the proteins of interest at modest levels.

N127(25) and N127(65) protein expression over the course of 2 weeks was analyzed by both indirect immunofluorescence and western blotting. Again only diffuse cytoplasmic staining of the unexpanded N127(25) protein was observed (data not shown). In the case of the polyglutamine-expanded N127(65) protein, cytoplasmic protein aggregates were observed after only 3 days of expression (Fig. 2A). After 14 days of expression, the cells now exhibited either cytoplasmic or nuclear N127(65) protein aggregates (Fig. 2A). Analysis of representative fields of the cells expressing the N127(65) protein revealed that approximately 60% of the cells contained cytoplasmic aggregates after 3 days (Fig. 2B). After 2 weeks of expression, however, approximately two-thirds of the aggregate-containing cells displayed the protein aggregates within the nucleus (Fig. 2B), similar to the situation in both neurons and transgenic mice expressing the polyglutamine-expanded protein (10,25).



View larger version (47K):
[in this window]
[in a new window]
  		Figure 2. Polyglutamine–protein aggregates are found preferentially within the cytoplasm but over time also become distributed within the nucleus. (A) Cells expressing the N127(65) protein were grown in the presence of 50 µM ZnSO4 for 3, 7 or 14 days. At each time point the cells were fixed with methanol and analyzed for their distribution of the polyglutamine-expanded protein by indirect immunofluorescence. Shown in blue is the DAPI staining and in red the N127(65) protein. (B) The intracellular distribution of N127(65) protein aggregates after 3 days and 14 days (e.g. data shown in A) was quantified as described in the methods. The percentage of cells containing nuclear aggregates was observed to increase over time (n=3±SEM; *P<0.005). (C) Cells expressing the N127(25) or N127(65) proteins for 3, 7 or 14 days were harvested and nuclei were isolated as described in the Methods. Aliquots of total cell lysates, as well as nuclear isolates, were then analyzed by western blotting for the presence of the HA-tagged N127(25) and N127(65) proteins. Equal amounts of total protein were applied to the gel. Lanes 1–3, N127(25) expressing cells after 3, 7 and 14 days, respectively; lanes 4–6, N127(65) expressing cells after 3, 7 and 14 days, respectively.

 
These results were confirmed by nuclear fractionation combined with western blotting. The total levels of the N127(25) protein changed little over the 2 week time course and none of the protein was observed within isolated nuclei (Fig. 2C). In the case of N127(65) protein, however, expression of the protein peaked at 7 days, and by 14 days we observed less of the protein in the total cell lysates (Fig. 2C, total cell lysates, lanes 4–6). Analysis of isolated nuclei revealed a steady increase in the nuclear content of the polyglutamine-expanded protein (Fig. 2C, nuclear lysates).

Previous studies have reported a co-localization of different heat shock proteins with the polyglutamine–protein aggregates (13–15). Consistent with these previous reports we observed that two constitutively expressed heat shock proteins, hsp73 and hsp40 co-localized with the protein aggregates (data not shown). Sequestration of heat shock proteins with unfolded polypeptides, as occurs after heat shock treatment or exposure to other proteotoxic conditions, is believed to be one activator of the cellular stress response (reviewed in 26). Hence, we examined whether any of the cells containing the polyglutamine–protein aggregates had activated a stress response, assayed by the induction of hsp72. While not expressed in cells maintained under normal growth conditions, induction and accumulation of hsp72 serves as a useful indicator that a cellular stress response has been activated (27). After 3 days of expression, N127(65) protein aggregates present within the cytoplasm were observed, yet expression of hsp72 was not evident (Fig. 3A, top panels). After 7 days of expression, and with the cells now displaying both nuclear and cytoplasmic protein aggregates, hsp72 expression was observed (Fig. 3A, middle panels). By 14 days, the majority of cells contained nuclear aggregates and approximately 10% of the expressing cells stained positive for hsp72 (Fig. 3A and B).



View larger version (66K):
[in this window]
[in a new window]
  		Figure 3. Polyglutamine–protein aggregates within the nucleus result in the expression of hsp72, a marker of the cellular stress response. (A) N127(65) cells were grown in the presence of ZnSO4 for either 3 days (top panels), 7 days (middle panels) or 14 days (bottom panels) and then fixed and analyzed for the distribution of the N127(65) protein (red) and hsp72 (green) by double-label indirect immunofluorescence. Merging of the two images is shown at the right. Arrows in the panels indicate the same cells. Arrowheads in the middle panels indicate a cytoplasmic inclusion and longer arrows indicate nuclear inclusions. (B) The number of cells exhibiting hsp72 expression after 3 or 14 days was quantified as described in the Methods. After 14 days approximately 10% of the cells displayed hsp72 staining (n=3±SEM; *P<0.005).

 
Biochemical fractionation and comparative western blot analyses confirmed and extended our immunofluorescence results (Fig. 4A). Cells expressing either N127(25) or N127(65) were grown for 3, 7 and 14 days. At each time point the cells were harvested and nuclei isolated. Equal amounts of total nuclear protein were applied to the gels and hsp72 levels determined by western blot. As an alternative method to determine whether the cells had activated a stress response, we examined the status of the stress-activated protein kinase, JNK, and its downstream target c-jun. Induction of the stress response results in the activation of the JNK kinase pathway, with both the jun-kinase and c-jun undergoing increased phosphorylation (reviewed in 28). Consistent with our immunofluorescence studies, hsp72 was evident within the nuclei of those cells expressing the N127(65) protein for a period of 14 days. In addition, the levels of both phospho-JNK as well as phospho-jun were increased in the cells expressing the polyglutamine-expanded N127(65) protein. Thus, while expression of the unexpanded protein does not lead to insoluble aggregates nor induction of the stress response, expression of the polyglutamine expanded protein results in the formation of nuclear inclusions and the activation of a cellular stress response.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 4. Expression of the polyglutamine-expanded protein over time leads to activation of the cellular stress response and increased cell demise. (A) Cells expressing either N127(25) (lanes 1–3) or N127(65) (lanes 4–6) for 3, 7 or 14 days, respectively, were harvested and nuclei isolated. The nuclear material was analyzed by western blotting for the presence of hsp72, JNK and phospho-JNK, c-jun and phospho-c-jun. Equal amounts of total protein were analyzed in each case. Lanes 1 and 4, nuclear lysates from cells harvested after 3 days; lanes 2 and 5, nuclear lysates from cells harvested after 7 days; lanes 3 and 6, nuclear lysates from cells harvested after 14 days. (B) Cells expressing N127(25) or N127(65) for either 3 days or 14 days were analyzed for total cell number by propidium iodide staining as described in the Methods. Values obtained were normalized to the number of cells expressing N127(25) after 3 days (n=4±SEM; *P<0.005).

 
These results prompted us to examine the viability of the cells over a 2 week period (Fig. 4B). Cells were plated at approximately 80–90% confluence and expression of the transgenes initiated by the addition of ZnSO4. After 3 or 14 days the number of viable cells remaining was determined via propidium iodide staining. After 3 days no obvious cell demise was observed. By 14 days, however, only about 60% of those cells expressing the polyglutamine expanded protein remained.

Collectively the different heat shock proteins are known to be integral for the ability of cells to survive a variety of metabolic insults. Because the polyglutamine protein aggregates sequestered hsp40, hsp73 and in some cases hsp72, we examined whether the cells might exhibit perturbations of their normal stress response. Cells expressing the N127(25) protein or the expanded N127(65) protein were subjected to a 43°C/60 min heat shock treatment. Cells were then returned to 37°C for 24 h and relative cell survival determined using FACS analysis (Fig. 5). Cells expressing the N127(25) protein exhibited little or no cell death in response to this particular heat shock treatment. In contrast, only 50% of the cells expressing the N127(65) protein survived the heat shock treatment (Fig. 5A). Similar results were obtained when the two cell types were challenged with sodium arsenite (Fig. 5B) or menadione (Fig. 5C), two other sources of metabolic stress. In each case, cells expressing the polyglutamine-expanded N127(65) protein appeared more vulnerable to the particular stress treatment as compared with cells expressing the unexpanded N127(25) protein.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 5. Cells expressing polyglutamine–protein aggregates show enhanced vulnerability to different metabolic insults. (A) After 2 weeks of induction with ZnSO4, N127(25) and N127(65) expressing cell lines were assessed for survival 24 h after a 43°C/90 min heat shock treatment (n=3±SEM; **P<0.05). (B) Following 2 weeks of induction with ZnSO4, N127(25) and N127(65) expressing cell lines were assessed for survival after 24 h of exposure to increasing concentrations of sodium arsenite (n=3±SEM; **P<0.05). (C) After 2 weeks of induction with ZnSO4, N127(25) and N127(65) expressing cell lines were assessed for survival after 24 h of exposure to increasing concentrations of menadione (n=3±SEM; *P<0.005).

 
In an effort to understand their increased vulnerability, we examined in more detail the response of the cells to heat shock treatment. As was mentioned earlier, activation of the stress response can be easily monitored by the expression and accumulation of hsp72 (27). Early after its expression, hsp72 moves into the nucleus and nucleolus. Later, during recovery from the stress event, the protein returns to the cytoplasm and over time is degraded (29). Both the N127(25) and N127(65) expressing cells, after 5 days in culture in the presence of ZnSO4, were subjected to a 43°C/45 min heat shock treatment, and then the cells were returned to 37°C for various times before being analyzed for hsp72 expression (Fig. 6). Note that after 5 days of expression, most of the N127(65) protein aggregates were present within the cytoplasm. After 5.5 h recovery from the hyperthermic treatment cells expressing the N127(25) protein exhibited a nuclear and nucleolar deposition of hsp72, a staining pattern typical of most cell types after heat shock (Fig. 6A). In cells expressing the polyglutamine-expanded N127(65) protein, however, hsp72 staining was not observed. While hsp72 staining was evident in those cells not containing protein aggregates (indicated by vertical arrows in Fig. 6A), cells containing the polyglutamine–protein cytoplasmic aggregates (indicated by horizontal arrows in Fig. 6A) showed little or no staining of hsp72.



View larger version (104K):
[in this window]
[in a new window]
  		Figure 6. Cells containing polyglutamine–protein aggregates exhibit a delayed stress response. N127(25) and N127(65) cells, 5 days after exposure to ZnSO4, were subjected to a 43°C/45 min heat shock treatment. After 5.5 or 8.5 h of recovery at 37°C, the cells were fixed with methanol and analyzed for the presence of the highly stress-inducible hsp72 protein (green) along with either the N127(25) or N127(65) proteins (red). Arrows indicate the same cells in the two images. Note that the staining of the N127(25) protein is weak owing to fixation of the cells with methanol (see Methods), the preferred fixative for analysis of hsp72 by indirect immunofluorescence (27). Formalin fixation of parallel cells revealed diffuse cytoplasmic staining of the N127(25) protein similar to that shown in Figure 1B. (A) N127(25) and N127(65) expressing cells 5.5 h after heat shock. Vertical and horizontal arrows indicate the same cells in the three panels. Horizontal arrows indicate cells with polyglutamine–protein aggregates while vertical arrows indicate cells with no obvious protein aggregates. (B) N127(25) and N127(65) expressing cells 8.5 h after stress. Vertical and horizontal arrows indicate the same cells in the three panels. Horizontal arrows indicate cells with polyglutamine–protein aggregates while vertical arrows indicate a cell with no obvious protein aggregates.

 
Following 8.5 h of recovery from the heat shock treatment, hsp72 staining now was observed primarily within the cytoplasm of the N127(25)-expressing cells (Fig. 6B), a result similar to what we have observed for other rodent cell lines. In those cells containing the N127(65) cytoplasmic protein aggregates, however, hsp72 staining still was rarely observed (horizontal arrows in Fig. 6B). Note that many of the cells that did not contain obvious protein aggregates (vertical arrows) again showed robust hsp72 expression.

Only after 24 h of recovery from the heat shock treatment did the aggregate-containing cells begin to show expression of hsp72 (Fig. 7A). Now both cytoplasmic and nuclear staining of 72 could be detected. Note as well that the polyglutamine aggregates themselves stained intensely for hsp72. Those surrounding cells not containing protein aggregates showed staining of hsp72, but most of the staining was diffuse and restricted to the cytoplasm. Cells expressing the unexpanded N127(25) protein 24 h after heat shock exhibited an hsp72 staining pattern similar to that found for the non-aggregate-containing N127(65) cells: diffuse staining throughout the cytoplasm (data not shown). Following 48 h of recovery little or no hsp72 staining was evident in the N127(25) expressing cells, consistent with the known turnover of the protein during the recovery phase after stress (data not shown). In contrast, after 48 h of recovery hsp72 levels remained high in the N127(65) cells, with the vast majority of the protein co-localizing with the polyglutamine–protein aggregates (Fig. 7B).



View larger version (76K):
[in this window]
[in a new window]
  		Figure 7. Hsp72 locale and turnover are altered in cells containing polyglutamine protein aggregates. N127(65) expressing cells were heat shock treated at 43°C/45 min, the cells returned to 37°C, and the distribution of hsp72 (green) and N127(65) (red) analyzed 24 or 48 h later by double-label indirect immunofluorescence as described in Figure 6. (A) N127(65) expressing cells 24 h after heat shock. Horizontal arrows indicate cells with polyglutamine–protein aggregates. (B) N127(65) expressing cells 48 h after heat shock. Horizontal arrows indicate cells with polyglutamine–protein aggregates.

 
During the course of our immunofluorescence analyses we noticed that the overall number of cells containing polyglutamine–protein aggregates, especially cells with nuclear protein aggregates, appeared to increase during the recovery period following heat shock (e.g. compare HA staining in Fig. 7A and B). Analysis of representative fields of the cells confirmed such an increase. An approximate 10-fold increase in the number of cells with nuclear polyglutamine–protein aggregates was observed 24–48 h following exposure to the heat shock treatment (Fig. 8A).



View larger version (25K):
[in this window]
[in a new window]
  		Figure 8. Heat shock treatment of cells expressing N127(65) results in an increase in polyglutamine-containing protein aggregates and a sustained activation of the stress response. (A) N127(65) expressing cells, 5 days after addition of ZnSO4, were heat shock treated at 43°C for 45 min and then returned to 37°C for either 8.5 or 24 h. At either of the two time points the cells were fixed and analyzed for the distribution of the HA-N127(65) protein within the nucleus by immunofluorescence. Representative fields (five) of approximately 100 cells were analyzed as described in the Methods and presented as a bar graph (n=5±SEM; *P<0.005). (B) N127(25) expressing cells (lanes 1–6) and N127(65) expressing cells (lanes 7–12), 3 days after addition of ZnSO4, were subjected to a 43°C/45 min heat shock treatment and the cells then returned to 37°C. After 1, 2, 4, 24 or 48 h of recovery the cells were harvested, nuclei isolated and the nuclear contents analyzed by western blotting. As indicated in the figure, antibodies specific for the HA-tagged N127(25) or HA-tagged N127(65) protein (HA), hsp72, phospho-JNK (P-JNK), total JNK (JNK), phospho-c-jun (P-c-jun), total jun (c-jun) were used. Lanes 1 and 7, cells at 37°C. Lanes 2 and 8, heat shock and recovery for 1 h. Lanes 3 and 9, heat shock and recovery for 2 h. Lanes 4 and 10, heat shock and recovery for 4 h. Lanes 5 and 11, heat shock and recovery for 24 h. Lanes 6 and 12, heat shock and recovery for 48 h. (C) Cells expressing the N127(65) protein for only 10 h after the addition of ZnSO4 were left at 37°C or were subjected to a 43°C heat shock treatment for 45 min. The heat shock treated cells were returned to 37°C and at various times thereafter the cells examined for their distribution of polyglutamine–protein aggregates via indirect immunofluorescence. Representative fields (five) of approximately 100 cells were analyzed as described in the Methods. Shown at the left is a bar graph showing the percentage of cells containing total aggregates, and at the right a bar graph of the percentage of cells displaying nuclear aggregates (n=5±SEM; *P<0.005).

 
Biochemical studies confirmed and extended our immunofluorescence data. The N127(25) and N127(65) expressing cells were subjected to heat shock treatment, allowed to recover at 37°C for various times, and then nuclei were isolated by low-speed centrifugation through 0.8 M sucrose. Proteins present within the nuclear fraction were analyzed for their relative content of the unexpanded and polyglutamine-expanded proteins, as well as for hsp72. In addition, components of the stress-activated JNK pathway were analyzed for their relative extent of activation. Nuclei isolated from the N127(25) expressing cells, either before or after heat shock treatment, did not contain any appreciable amounts of the unexpanded N127(25) protein product (Fig. 8B, lanes 1–6). In the N127(65) expressing cells, however, the levels of the N127(65) protein within the nuclear fraction increased throughout the recovery periods after heat shock (Fig. 8B, lanes 7–12).

Analysis of hsp72 present within the nuclear fraction also revealed a steady increase after recovery from the heat shock treatment in both cell types. Note, however, that in those cells expressing N127(25), nuclear hsp72 levels began to diminish by 24 h (Fig. 8B, lane 5) and by 48 h of recovery nuclear hsp72 was no longer detected (Fig. 8B, lane 6). In contrast, cells expressing the polyglutamine-expanded N127(65) protein still showed significant levels of hsp72 within the nuclear fraction 48 h after heat shock (Fig. 8B, lane 12). In the case of the JNK pathway, the levels of both phospho-JNK and phospho-c-jun levels increased in both cell types early after heat shock. During the recovery periods the levels of activated JNK in the N127(25) cells began to diminish (Fig. 8B, lanes 5 and 6). In contrast, in those cells expressing the polyglutamine-expanded N127(65) protein both phosphorylated JNK and c-jun were still apparent 48 h following the heat shock treatment (Fig. 8B, lane 12).

Data presented in Figures 7 and 8A and B revealed that heat shock treatment increased the overall number of cells displaying nuclear polyglutamine–protein aggregates. Specifically, cells expressing the N127(65) protein for 5 days showed an almost 10-fold increase in their number of nuclear protein aggregates following heat shock and recovery at 37°C. We repeated these experiments in cells where we had initiated expression of the polyglutamine-expanded protein (via addition of ZnSO4) for only 10 h. At this early stage of expression we observed only a few polyglutamine–protein aggregates, with most of the aggregates being present within the cytoplasm. The cells were subjected to a 43°C/45 min heat shock treatment and then returned to 37°C. Within 24 h after the heat shock treatment an approximate 5-fold increase in the number of cells displaying polyglutamine–protein aggregates (both cytoplasmic and nuclear) was observed (Fig. 8C, total aggregates). Analysis of the number of cells displaying nuclear protein aggregates revealed an almost 10-fold increase (Fig. 8C, nuclear aggregates). These results demonstrate that heat shock treatment increases the rate of polyglutamine-expanded protein aggregation and the overall proportion of nuclear inclusions.

A recent study by Meriin and colleagues (30) showed that an active form of MEKK1, a protein kinase that regulates various stress kinsase pathways, including JNK, augmented the formation of intracellular inclusion bodies containing the amino-terminal fragment of the polyglutamine-expanded huntingtin protein. Because heat shock treatment also stimulates a number of stress kinase pathways including JNK (reviewed in 28), we examined whether pharmacological inhibition of the JNK pathway before and after heat shock treatment might affect the observed increase in the overall amount of the N127(65) protein aggregates. Cells were plated on coverslips and expression of the transgene was initiated by the addition of ZnSO4. Forty-eight hours later, the cells were either left untreated or exposed to a selective inhibitor of JNK (SP600125). Thirty minutes later, both the untreated cells as well as the cells exposed to the JNK inhibitor were subjected to a 43°C/30 min heat shock treatment and then returned to 37°C. After either 18 or 40 h of recovery from the heat shock treatment the cells were analyzed for their distribution of the N127(65) protein by indirect immunofluorescence (Fig. 9A). Heat shock treatment again resulted in a rapid increase in the overall amount of protein aggregates within the cytoplasm and nucleus. Inclusion of the JNK inhibitor before and during the heat shock treatment and recovery period reduced this heat-dependent increase in N127(65) protein aggregates.



View larger version (48K):
[in this window]
[in a new window]
  		Figure 9. JNK inhibitor reduces polyglutamine–protein aggregation. (A) N127(65) expressing cells growing on glass coverslips were treated with ZnSO4 for 48 h. A cell-permeable inhibitor of JNK was added to some of the cells (final concentration 20 µM). Thirty minutes later, the cells were left at 37°C (control), while the others were heat shock-treated at 43°C for 30 min. The cells then were returned to 37°C, and the distribution of the N127(65) protein (red) was analyzed 18 and 40 h later by indirect immunofluorescence. Dapi staining of nuclei is shown in blue. Top panels show cells with no added JNK inhibitor and the bottom panels cells treated with the JNK inhibitor. (B) The percentage of cells displaying visible N127(65) protein aggregates in (A) was quantified. Representative fields (five) of approximately 100 cells were analyzed as described in the Methods. Shown at the left is the percentage of cells containing total aggregates, and at the right the percentage of cells displaying nuclear aggregates (n=5±SEM; *P<0.05). (C) N127(65) cells were plated on 6 cm dishes and treated with or without JNK inhibitor exactly as described in (A). Cells were left either at 37°C or subjected to a 43°C/30 min heat shock treatment and then returned to 37°C for 40 h. JNK inhibitor was present throughout the heat shock and recovery period. Nuclei were isolated as described in the methods and the content of N127(65) determined via western blotting using the HA antibody (left-hand panel). The analysis was performed in triplicate plates of cells. Lanes 1–3, cells left at 37°C, no JNK inhibitor; lanes 4–6, cells left at 37°C in the presence of JNK inhibitor; lanes 7–9, cells heat shock-treated in the absence of JNK inhibitor; lanes 9–12, cells heat shock-treated in the presence of JNK inhibitor. Quantitation of the western blot is shown on the right-hand side of the panel.

 
Representative fields of the cells analyzed by immunofluorescence in Figure 9A were chosen and the number of cells displaying polyglutamine–protein aggregates determined (Fig. 9B, left panel). In addition, we estimated the number of cells displaying only nuclear aggregates (Fig. 9B, right panel). This analysis revealed a reduction in both the number of overall aggregates, as well as the number of nuclear aggregates in cells exposed to the JNK inhibitor prior to and during the heat shock treatment. Interestingly, this JNK-inhibitor-dependent reduction in the number of aggregates was also observed for those cells not heated. Isolation of nuclei and analysis of the nuclear content of the N127(65) protein by western blotting corroborated the immunofluorescence results (Fig. 9C, each condition analyzed in triplicate). The levels of the polyglutamine-expanded protein within the nuclear fraction of cells maintained at 37°C in the presence of the JNK inhibitor (Fig. 9C, lanes 4–6) were less than those observed for the untreated cells (Fig. 9C, lanes 1–3). Heat shock treatment increased nuclear N127(65) content (Fig. 9C, lanes 7–9) and this increase again was blunted in those cells heated in the presence of the JNK inhibitor (Fig. 9C, lanes 10–12). Thus, blocking the JNK pathway appears to reduce the extent of polyglutamine-dependent protein aggregation.

We repeated many of these experiments in our 3T3 cell lines expressing the full-length forms of the androgen receptor containing either 25 or 65 glutamine residues [AR(25) and AR(65), respectively]. Immunofluorescence analysis revealed a diffuse nuclear locale for both proteins in cells treated with an androgen receptor agonist, DHT (Fig. 10A, top panels, 37°C). We did not observe the formation of intracellular inclusions containing either form of the full-length protein over a 7 day period, whether in the presence or absence of agonist (data not shown). Western blotting of the AR(25) and AR(65) expressing cells, treated with or without agonist, revealed modest expression of the full-length proteins (Fig. 10B). Analysis of the heat shock response showed no significant differences between the two cell types. Both exhibited a nuclear/nucleolar locale of hsp72 within a few hours after heat shock treatment (Fig. 10A, lower panels, HS). Analysis of isolated nuclei from both the AR(25) and AR(65) cells before and after heat shock revealed similar levels of the respective AR protein (Fig. 10C). In addition, hsp72 movement into the nuclear fraction and the rate of its eventual turnover also appeared similar. In both cell types activation and subsequent inactivation of JNK following heat shock treatment appeared similar (Fig. 10C). Finally, no significant differences were observed between the two cell types in their survival curves following heat shock treatment, or exposure to either menadione or sodium arsenite (Fig. 10D).



View larger version (33K):
[in this window]
[in a new window]
  		Figure 10. Analysis of 3T3 cells expressing AR(25) and AR(65). (A) Stable cell lines expressing either AR(25) or AR(65) were treated with agonist (DHT) and left at 37°C, or subjected to a 43°C/45 min heat shock treatment followed by recovery at 37°C for 6 h. In the top panels the 37°C cells were fixed and analyzed by indirect immunofluorescence for their distribution of either the AR(25) (left) or AR(65) (right) proteins using the HA antibody (red). In the bottom panels the cells after heat shock treatment were analyzed for their distribution of hsp72 (green). Dapi staining is included to identify nuclei. (B) Lysates were prepared from cells expressing either AR(25) or AR(65), in the presence or absence of DHT, and analyzed by SDS–PAGE and western blotting using the anti-HA antibody. Shown in lanes 1–4 is the Coomassie blue-stained gel and in lanes 5–8 the corresponding western blot. Lanes 1 and 5, AR(25) cells, no DHT; lanes 2 and 6, AR(25) cells grown in the presence of DHT; lanes 3 and 7, AR(65) cells, no DHT; lanes 4 and 8, AR(65) cells grown in the presence of DHT. The lane at the far left of the panel shows molecular mass markers. (C) AR(25) and AR(65) expressing cells were subjected to a 43°C/45 min heat shock treatment and the cells then returned to 37°C. After 4, 12, 24 or 48 h of recovery the cells were harvested and nuclei prepared. The nuclear isolates were analyzed by western blotting for the presence of AR(25) and AR(65) (using the HA antibody) along with antibodies for hsp72, JNK, and phospho-JNK. Equal amounts of total protein were analyzed in each case. Lane 1, cells at 37°C; lane 2, heat shock and recovery for 4 h; lane 3, heat shock and recovery for 12 h; lane 4, heat shock and recovery for 24 h; lane 5, heat shock and recovery for 48 h. (D) AR(25) and AR(65) cells were plated at 90% confluency. Three days later DHT was added and then the cells were left untreated, or subjected to different types of metabolic stress including heat shock treatment, exposure to menadione or exposure to sodium arsenite. Twenty-four hours later cell survival was determined by FACS analysis as described in the methods. 1, Cells maintained at 37°C; 2, exposure to a 43°C/90 min heat shock treatment; 3, exposure to 10 µM menadione; 4, exposure to 20 µM menadione; 5, exposure to 300 µM sodium arsenite; 6, exposure to 500 µM sodium arsenite.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Mouse fibroblasts were constructed to express the full-length androgen receptor, as well as amino-terminal derived truncated forms of the androgen receptor, containing either 25 or 65 continuous glutamine residues. With expression of both forms of full-length AR proteins we did not observe any adverse cellular consequences nor the formation of intracellular inclusions. The truncated AR protein containing 25 glutamines N127(25) exhibited diffuse staining, remained detergent soluble, and did not lead to any deleterious cellular effects. In contrast, the polyglutamine-expanded AR truncation product N127(65) readily formed cytoplasmic aggregates and was found preferentially within the detergent insoluble fraction within 2–3 days after initiating its expression.

After 2 weeks of expression, approximately 25% of the cells now exhibited polyglutamine-expanded protein aggregates within the nucleus. Double-label indirect immunofluorescence revealed that cells containing nuclear protein aggregates often showed intense nuclear staining of hsp72, a member of the stress protein family whose expression is usually restricted to cells undergoing a stress response (27). In some instances, cells with cytoplasmic protein aggregates also exhibited hsp72 staining, but much less frequently as compared with those cells with nuclear aggregates. These observations are in line with previous studies which similarly concluded that polyglutamine-expanded protein aggregates present within the nucleus (but in some cases cytoplasmic aggregates as well) elicited induction of hsp72 (14,15). Our biochemical fractionation and western blot analyses revealed steady increases in the amount of both N127(65) and hsp72 within the nucleus over time. Analysis of the stress-activated JNK kinase pathway also revealed modest increases in the overall levels of both phospho-JNK and phospho-c-jun, results consistent with previous studies examining the cellular impact of polyglutamine–protein expansion (31,32). Taken together, these results demonstrate that nuclear localized polyglutamine–protein aggregates are linked to the activation of a cellular stress response. We suspect that this chronic activation of the cellular stress response may account for the increased cell demise we observed for the cells expressing the polyglutamine-expanded protein over the 2 week time period. It is important to point out that our results are derived from cultured fibroblasts that continue to proliferate in vitro. Hence, it is difficult to predict how our results may extend to the situation of post-mitotic motor neurons in the spinal cord, the major target of the disease.

We previously reported that co-expression of a mutant form of the glucocorticoid receptor significantly enhanced the formation of polyglutamine–protein aggregates within the nucleus. Again, many of the nuclear-aggregate-containing cells (in contrast to cells with cytoplasmic aggregates) showed robust activation of hsp72 and increased cell death (10). More recent studies by Yang et al. (33) addressed this question by constructing small expanded-polyglutamine peptides (Q42) either lacking or containing a nuclear localization signal sequence. Following their uptake by the cells the peptides were observed to aggregate within the cytoplasm or nucleus, respectively. Over time those cells that displayed nuclear aggregates showed significantly higher cell death. Hence, aggregation events occurring within the nucleus are linked to cell injury and eventual cell demise.

While chronic cell dysfunction is a well-known feature of polyglutamine diseases, the work presented here shows that exposure to these toxic proteins may also compromise cell responses to stress. Specifically, the presence of polyglutamine protein aggregates perturbed normal hsp induction and localization, and correlated with increased cell death in response to a variety of stressful stimuli. Those cells that did survive the stress response failed to degrade hsp72 during the recovery period and exhibited continued activation of the stress-activated JNK kinase pathway.

Failure to down-regulate the stress response in a timely fashion following heat shock may result in further deleterious consequences for the aggregate-containing cells. For example, Volloch et al. (34) previously demonstrated that cells unable to express hsp72 following heat shock treatment exhibited a sustained activation of the JNK pathway and over time these cells underwent apoptotic death. If, however, hsp72 expression was restored to the cells, for example via forced expression using adenovirus, the cells now could effectively attenuate JNK activation after heat shock and survive. Hence, sequestration of hsp72 by the polyglutamine–protein aggregates after heat shock treatment may interfere with the normal down-regulation of the stress response, including attenuation of the JNK pathway. This in turn may lead to a state of ‘chronic stress’ and an inability of the cells to maintain normal homeostasis.

Our data have also linked metabolic stress to the formation of polyglutamine–protein aggregates, especially nuclear aggregates. Recently, Ding and colleagues (35) similarly concluded that heat shock treatment increased the aggregation of a GFP–polyglutamine chimera. In addition, Nishitoh (36) reported that thapsigargin or tunicamycin, two inducers of the stress response in the endoplasmic reticulum, also increased polyglutamine-dependent protein aggregation. Finally, direct inhibition of the proteosome pathway, yet another activator of the stress response, increased the levels of polyglutamine–protein aggregates (37,38). Increases in the overall amounts of polyglutamine–protein aggregates after metabolic stress might arise due to a number of factors. First, the increased demand on the existing pool of molecular chaperones after heat shock (e.g. to deal with the global increase in damaged proteins) may result in a compromise in ‘quality control pathways’ normally involved in the recognition and disposal of abnormally folded proteins, including polyglutamine-expanded proteins. Support for this idea follows from a number of studies showing that overexpression of one or more molecular chaperones of the hsp70 family (e.g. hsp70 and its co-factors such as hsp40 and other DnaJ homologs) suppresses aggregate formation and/or leads to a reduction in cellular toxicity associated with the polyglutamine–protein aggregates (13,18–22). Second, induction of cellular stress responses is accompanied by the activation of one or more stress kinase pathways, most notably JNK and p38 (reviewed in 28). Activation of these pathways, even in the absence of other stress-related events, appears sufficient to increase polyglutamine protein aggregation and toxicity. For example, Meriin et al. (30) showed that the transfection of cells with an active form of MEKK1 (an upstream activator of the JNK pathway) was sufficient to increase the overall levels of polyglutamine–protein aggregates. Conversely, suppression of stress kinase pathways, as we showed here using the JNK inhibitor, appeared to slow the formation of the polyglutamine–protein aggregates.

Thus, expression of polyglutamine-expanded proteins may disrupt the normal expression and locale of various heat shock proteins that participate in cellular quality control pathways. This in turn probably increases the probability of nuclear aggregation events which appear to be deleterious to the cell. In addition, the presence of both cytoplasmic and nuclear aggregates may also interfere with the normal cellular response to injury, thereby further increasing the risk of cell damage from other insults. Polyglutamine–protein toxicity would thus have at least two major elements: an initial effect (possibly by soluble protein forms) to disturb overall cell homeostasis, and a later effect due to protein aggregation. Thus, in patients with polyglutamine-related neurodegenerative diseases CNS injuries (e.g. ischemia/reperfusion, oxidative stress, excitotoxicity) or even normal aging may contribute to disease progression by initiating cellular stress responses that cannot be properly executed and/or attenuated. As a consequence, cells may enter into a state of chronic stress that either results in premature cell death or, alternatively, interferes with their ability to carry out normal cellular functions. Based on our results we predict that relevant CNS injuries may contribute to either the onset or progression of neurodegenerative diseases involving abnormal protein folding. Having defined such alterations in the stress response in vitro, our focus now will be to determine whether animals expressing polyglutamine-expanded proteins in the brain similarly exhibit dysfunctional stress responses.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Preparation of plasmids
HA-tagged fragments of N-terminal 127 amino acids of human AR with either 25 or 65 glutamines [referred to as N127(25) and N127(65)], previously described in (10) were subcloned into pMEP4 (Invitrogen, Carlsbad, CA, USA) containing the regulatable metallothionein promoter. HA-tagged full-length AR with either 25 or 65 glutamines were subcloned into the pIRES.Neo vector (Clontech).

Cell culture
NIH3T3 cells were grown in DME-21 medium with 10% FBS. To create stable cell lines, cells were transfected on 10 cm dishes with 5 µg of the N127(25) or N127(65) plasmids using Effectene transfection reagents (Qiagen). After 48 h the resultant transfectants were selected by inclusion of 400 U/ml hygromycin in DME-21 medium containing 10% FBS. Note that the selection media did not contain any ZnSO in order to minimize expression of the transgenes until the selection process was complete. After 2–3 weeks surviving cells were characterized for expression of the N127(25) and N127(65) proteins via western blot and immunofluorescence analyses. In all experiments, expression of the N127(25) and N127(65) transgenes was initiated by the addition of 50 µM ZnSO4 to the cells. This level of ZnSO4 did not elicit expression of the stress-inducible hsp72 protein and did not produce any noticeable effects on cell growth or viability (data not shown).

Similarly, 3T3 cells were transfected with plasmids containing either AR(25) and AR(65) as described above and 48 h later the cells were selected by inclusion of 500 µg/ml G418 in DME-21 plus 10% FBS. After 3 weeks, surviving cells were characterized for their expression of AR(25) and AR(65) by western blotting and immunofluorescence. Subsequent propagation of the cells was done in the same media containing 300 µg/ml G418.

Immunofluorescence
Stable cell lines expressing N127(25) or N127(65) were grown on glass coverslips in the presence of ZnSO4 for the times indicated in the figure legends. Similarly stable cells expressing AR(25) and AR(65) were grown on glass coverslips. Cells were washed in PBS, fixed via immersion in 3.7% paraformaldehyde for 10 min and permeabilized by incubation in PBS containing 0.1% Triton-X100 for 2 min. Alternatively, cells were fixed by immersion into absolute methanol (-20°C) for 3 min. Note that fixation with paraformaldehyde was the preferred method for the analysis of the N127(25) protein as well for the analysis of the AR(25) and AR(65) proteins. Specifically, diffuse staining of N127(25), AR(25) and AR(65) protein was more easily observed whenever the cells were fixed in formalin as compared with methanol. While formalin fixation also was effective for the analysis of the N127(65) protein, methanol fixation provided for enhanced detection of the polyglutamine–protein aggregates. For detection of the hsp40, hsp73 and hsp72 proteins, methanol fixation was the preferred method. Thus, for double-label staining experiments methanol fixation was employed. Primary and secondary antibodies were diluted in 5 mg/ml bovine serum albumin (BSA) in PBS. Nuclei were detected via incubation with DAPI. Antibodies that were used included: mouse monoclonal anti-hsp72 (StressGen Biotechnologies C92, SPA-810, described in (27); rat anti-HA (Roche antibody 3F10); mouse monoclonal anti-HA antibody (Babco, Richmond, CA, USA); rabbit anti-hsp73 (39); rabbit anti-hsp40 (40). Primary antibodies were visualized using the appropriate affinity-purified goat anti-mouse, goat anti-rat, or goat anti-rabbit antibodies conjugated with either fluorescein, rhodamine, or Cy-3 (Jackson Laboratories, West Grove PA, and Cappel, Irvine, CA, USA).

The number of cells exhibiting HA staining or hsp72 staining at specific time points was quantified by counting three to five different fields containing approximately 100 cells. Results are presented as an average of the fields examined (±SEM).

Nuclear isolation
Isolation of crude nuclei was performed as described previously (27). Cells grown on 10 cm dishes were washed twice with PBS and once quickly with hypotonic buffer (25 mM Tris pH 7.4, 10 mM NaCl, 5 mM KCl, 5 mM MgCl2). A small volume of ice-cold hypotonic medium was added back to the cells and the cells allowed to swell on ice for 4–5 min. Triton X-100 and sodium deoxycholate were added to a final concentration of 0.5% each, along with protease and phosphatase inhibitors. The cells were scraped off the dish, and the cell lysates were vortexed for 10 s. The lysed cells were layered onto a 0.88 M sucrose cushion prepared in hypotonic buffer, and the nuclei were pelleted for 2 min at 14 000g at 4°C. The supernatant was removed carefully, and the pelleted nuclei were resuspended in Laemmli sample buffer, boiled and stored at -70°C until analysis by SDS–PAGE and western immunoblotting.

Western blotting
For experiments analyzing the expression of the N127(25) and N127(65) proteins, the cells were lysed in PBS containing 1% Triton X-100 and supplemented with protease inhibitors. After determination of total protein, the samples were adjusted with Laemmli sample buffer, heated at 95°C for 5 min, and the material then applied to the gel. Isolated nuclei (above) were solubilized in Laemmli sample buffer, heated at 95°C for 5 min, and the chromatin sheared by passage of the material through a 26 gauge needle. For those experiments analyzing different phospho-proteins, the cell lysis buffer contained a cocktail of phosphatase and protease inhibitors. In all cases equal amounts of total protein were applied to the gels. Proteins were separated on 12.5% SDS gels by SDS–PAGE, the proteins transferred to nitrocellulose and then analyzed by western blotting as described previously (10).

Propidium iodide assay
To assess cell viability over time, N127(25) and N127(65) stable cell lines were grown on 24-well dishes using 400 U/ml hygromycin and 50 µM ZnSO4 in DME-21 medium containing 10% FBS for 3 days and for 14 days. At both time points the medium was removed from the cells and replaced with 200 µl DME-21 medium containing 1% FBS. The plates were wrapped in parafilm and placed at -20°C for 3 h, then thawed at 50°C for 15 min. Propidium iodide was added to a final concentration of 50 µg/ml, and the plates were incubated at room temperature, wrapped in aluminum foil, for 1 h. Cell survival (cells remaining on plate) was then assayed using the Wallac Victor2 1420 Multilabel Counter and the1420 Manager Software.

FACS analysis
Cell viability of the stable cell lines in response to different metabolic insults was performed by FACS analysis using propidium iodide, which would stain those cells that are no longer viable. The cell culture medium was removed from each 6 cm dish representing a different treatment and recovery period (n=3–5 per condition) and saved. Remaining cells on the dish were washed twice in PBS, removed by the addition of trypsin, and then added to the initial culture medium. Samples were centrifuged for 2 min at 500g, and the resultant cell pellet was resuspended in PBS containing 50 µg/ml propidium iodide. The samples were incubated at room temperature for 15 min, then centrifuged again for 2 min at 500g. The resultant cell pellets were washed twice and finally resuspended in PBS. Cell survival was then assessed using the FACSorter and CellQuest software. Viable cells were distinguished from cells that did not maintain plasma membrane integrity by examining the resultant peaks from the FACSorter for percent cells not bound to propidium iodide after acquiring 10 000 events.

JNK inhibitor studies
Cell permeable JNK inhibitor (SP600125) was purchased from Calbiochem (catalog no. 420119) and resuspended in DMSO. JNK inhibitor was added to the cells plated on glass coverslips at a final concentration of 10–20 µM. Thirty minutes later, the cells were either left unheated or subjected to a 43°C/30 min heat shock treatment. Cells were allowed to recover at 37°C for either 18 or 40 h before analysis by immunofluorescence utilizing the HA antibody. A similar protocol was used for the isolation of nuclei. Cells plated on 6 cm dishes were treated exactly as described for the coverslips, nuclei isolated and amount of N127(65) determined by western blotting.

Statistical analysis
Statistical analysis of data used variance followed by the Student–Neuman–Keuls test. Differences were considered significant if P<0.05.


   ACKNOWLEDGEMENTS
 
We thank Alma Kabiling for outstanding technical assistance and W.J. Hansen for comments regarding the manuscript. This work was supported by the NIH (W.J.W., grant no. GM33551 and M.I.D., grant no. NS01976). M.I.D. also acknowledges support from the Muscular Dystrophy Association and the American Philosophical Society.


   FOOTNOTES
 
* To whom correspondence should be addressed. Tel: +1 4152066948; Fax: +1 4152066997; Email: welch{at}itsa.ucsf.edu Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Zoghbi, H.Y. and Orr, H.T. (1999) Polyglutamine diseases: protein cleavage and aggregation. Curr. Opin. Neurobiol., 9, 566–570.[CrossRef][ISI][Medline]

  2. Gusella, J.F. and MacDonald, M.E. (2000) Molecular genetics: unmasking polyglutamine triggers in neurodegenerative disease. Nat. Rev. Neurosci., 1, 109–115.[ISI][Medline]

  3. DiFiglia, M., Sapp, E., Chase, K.O., Davies, 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.[Abstract/Free Full Text]

  4. Li, M., Miwa, S., Kobayashi, Y., Merry, D.E., Yamamoto, M., Tanaka, F., Doyu, M., Hashizume, Y., Fischbeck, K.H. and Sobue, G. (1998) Nuclear inclusions of the androgen receptor protein in spinal and bulbar muscular atrophy. Ann. Neurol., 44, 249–254.[CrossRef][ISI][Medline]

  5. Duyckaerts, C., Durr, A., Cancel, G. and Brice, A. (1999) Nuclear inclusions in spinocerebellar ataxia type 1. Acta Neuropathol. (Berl.), 97, 201–207.[CrossRef][Medline]

  6. Gourfinkel-An, I., Cancel, G., Duyckaerts, C., Faucheux, B., Hauw, J.J., Trottier, Y., Brice, A., Agid, Y. and Hirsch, E.C. (1998) Neuronal distribution of intranuclear inclusions in Huntington's disease with adult onset. Neuroreport, 9, 1823–1826.[ISI][Medline]

  7. Holmberg, M., Duyckaerts, C., Dürr, A., Cancel, G., Gourfinkel-An, I., Damier, P., Faucheux, B., Trottier, Y., Hirsch, E.C., Agid, Y. et al. (1998) Spinocerebellar ataxia type 7 (SCA7): a neurodegenerative disorder with neuronal intranuclear inclusions. Hum. Mol. Genet., 7, 913–918.[Abstract/Free Full Text]

  8. Becher, M.W., Kotzuk, J.A., Sharp, A.H., Davies, S.W., Bates, G.P., Price, D.L. and Ross, C.A. (1998) Intranuclear neuronal inclusions in Huntington's disease and dentatorubral and pallidoluysian atrophy: correlation between the density of inclusions and IT15 CAG triplet repeat length. Neurobiol. Dis., 4, 387–397.[CrossRef][ISI][Medline]

  9. Davies, S.W., Turmaine, M., Cozens, B.A., DiFiglia, M., Sharp, A.H., Ross, C.A., Scherzinger, E., Wanker, E.E., Mangiarini, L. and Bates, G.P. (1997) Formation of neuronal intranuclear inclusions underlies the neurological dysfunction in mice transgenic for the HD mutation. Cell, 90, 537–548.[CrossRef][ISI][Medline]

  10. Welch, W.J. and Diamond, M.I. (2001) Glucocorticoid modulation of androgen receptor nuclear aggregation and cellular toxicity is associated with distinct forms of soluble expanded polyglutamine protein. Hum. Mol. Genet., 10, 3063–3074.[Abstract/Free Full Text]

  11. Ellerby, L.M., Hackam, A.S., Propp, S.S., Ellerby, H.M., Rabizadeh, S., Cashman, N.R., Trifiro, M.A., Pinsky, L., Wellington, C.L., Salvesen, G.S. et al. (1999) Kennedy's disease: caspase cleavage of the androgen receptor is a crucial event in cytotoxicity. J. Neurochem., 72, 185–195.[CrossRef][ISI][Medline]

  12. Merry, D.E., Kobayashi, Y., Bailey, C.K., Taye, A.A. and Fischbeck, K.H. (1998) Cleavage, aggregation and toxicity of the expanded androgen receptor in spinal and bulbar muscular atrophy. Hum. Mol. Genet., 7, 693–701.[Abstract/Free Full Text]

  13. 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]

  14. Chai, Y., Koppenhafer, S.L., Bonini, N.M. and Paulson, H.L. (1999) Analysis of the role of heat shock protein (Hsp) molecular chaperones in polyglutamine disease. J. Neurosci., 19, 10338–10347.[Abstract/Free Full Text]

  15. Stenoien, D.L., Cummings, C.J., Adams, H.P., Mancini, M.G., Patel, K., DeMartino, G.N., Marcelli, M., Weigel, N.L. and Mancini, M.A. (1999) Polyglutamine-expanded androgen receptors form aggregates that sequester heat shock proteins, proteasome components and SRC-1, and are suppressed by the HDJ-2 chaperone. Hum. Mol. Genet., 8, 731–741.[Abstract/Free Full Text]

  16. Georgopoulos, C. and Welch, W.J. (1993) Role of the major heat shock proteins as molecular chaperones. A. Rev. Cell Biol., 9, 601–634.

  17. Hartl, F. and Hayer-Hartl, M. (2002) Molecular chaperones in the cytosol: from nascent chain to folded protein. Science, 295, 1852–1858.[Abstract/Free Full Text]

  18. Warrick, J.M., Chan, H.Y., Gray-Board, G.L., Chai, Y., Paulson, H.L. and Bonini, N.M. (1999) Suppression of polyglutamine-mediated neurodegeneration in Drosophila by the molecular chaperone HSP70. Nat. Genet., 23, 425–428.[CrossRef][ISI][Medline]

  19. Kobayashi, Y., Kume, A., Li, M., Doyu, M., Hata, M., Ohtsuka, K. and Sobue, G. (2000) Chaperones Hsp70 and Hsp40 suppress aggregate formation and apoptosis in cultured neuronal cells expressing truncated androgen receptor protein with expanded polyglutamine tract. J. Biol. Chem., 275, 8772–8778.[Abstract/Free Full Text]

  20. Kazemi-Esfarjani, P. and Benzer, S. (2000) Genetic suppression of polyglutamine toxicity in Drosophila. Science, 287, 1837–1840.[Abstract/Free Full Text]

  21. Krobitsch, S. and Lindquist, S. (2000) Aggregation of huntingtin in yeast varies with the length of the polyglutamine expansion and the expression of chaperone proteins. Proc. Natl Acad. Sci. USA, 97, 1589–1594.[Abstract/Free Full Text]

  22. Chan, H.Y.E., Warrick, J.M., Andriola, I., Merry, D. and Bonini, N.M. (2002) Genetic Modulation of polyglutamine toxicity by protein conjugation pathways in Drosophila. Hum. Mol. Genet., 11, 2895–2904.[Abstract/Free Full Text]

  23. Brooks, B.P., Paulson, H.L., Merry, D.E., Salazar-Grueso, E.F., Brinkmann, A.O., Wilson, E.M. and Fischbeck, K.H. (1997) Characterization of an expanded glutamine repeat androgen receptor in a neuronal cell culture system. Neurobiol. Dis., 3, 313–323.[CrossRef][ISI][Medline]

  24. Bingham, P.M., Scott, M.O., Wang, S., McPhaul, M.J., Wilson, E.M., Garbern, J.Y., Merry, D.E. and Fischbeck, K.H. (1995) Stability of an expanded trinucleotide repeat in the androgen receptor gene in transgenic mice. Nat. Genet., 9, 191–196.[CrossRef][ISI][Medline]

  25. Abel, A., Walcott, J., Woods, J., Duda, J. and Merry, D. (2001) Expression of expanded repeat androgen receptor produces neurologic disease in transgenic mice. Hum. Mol. Genet., 10, 107–116.[Abstract/Free Full Text]

  26. Craig, E.A. and Gross, C.A. (1991) Is hsp70 the cellular thermometer? Trends Biochem. Sci., 16, 135–140.[CrossRef][ISI][Medline]

  27. Welch, W.J. and Feramisco, J.R. (1984) Nuclear and nucleolar localization of the 72,000 dalton heat shock protein in heat shocked mammalian cells. J. Biol. Chem., 259, 4501–4513.[Abstract/Free Full Text]

  28. Gabai, V.L. and Sherman, M.Y. (2002) Interplay between molecular chaperones and signaling pathways in survival of heat shock. J. Appl. Physiol., 92, 1743–1748.[Abstract/Free Full Text]

  29. Welch, W.J. and Suhan, J. (1986) Cellular and biochemical events in mammalian cell during and after rcovery from physiological stress. J. Cell Biol., 103, 2035–2052.[Abstract/Free Full Text]

  30. Meriin, A.B., Mabuchi, K., Gabai, V.L., Yaglom, J.A., Kazantsev, A. and Sherman, M.Y. (2001) Intracellular aggregation of polypeptides with expanded polyglutamine domain is stimulated by stress-activated kinase MEKK1. J. Cell Biol., 153, 851–864.[Abstract/Free Full Text]

  31. Liu, Y.F. (1998) Expression of polyglutamine-expanded Huntingtin activates the SEK-1-JNK pathway and induces apoptosis in a hippocampal neuronal cell line. J. Biol. Chem., 273, 28873–28877.[Abstract/Free Full Text]

  32. Yasuda, S., Inoue, K., Hirabayashi, M., Higashiyama, H., Yamamoto, Y., Fuyuhiro, H., Komure, O., Tanaka, F., Sobue, G., Tsuchiya, K. et al. (1999) Triggering of neuronal cell death by accumulation of activated SEK1 on nuclear polyglutamine aggregations in PML bodies. Genes Cells, 4, 743