Huntington's disease (HD) is an autosomal inherited neurological disease caused by a CAG-repeat expansion in the first exon of huntingtin gene encoding for the huntingtin protein (Htt). In HD, there is an accumulation of intracellular aggregates of mutant Htt that negatively influence cellular functions. The aggregates contain ubiquitin, and part of the HD pathophysiology could result from an imbalance in cellular ubiquitin levels. Deubiquitinating enzymes are important for replenishing the ubiquitin pool, but less is known about their roles in brain diseases. We show here that overexpression of the ubiquitin-specific protease-14 (Usp14) reduces cellular aggregates in mutant Htt-expressing cells mainly via the ubiquitin proteasome system. We also observed that the serine–threonine kinase IRE1 involved in endoplasmic reticulum (ER) stress responses is activated in mutant Htt-expressing cells in culture as well as in the striatum of mutant Htt transgenic (BACHD) mice. Usp14 interacted with IRE1 in control cells but less in mutant Htt-expressing cells. Overexpression of Usp14 in turn was able to inhibit phosphorylation of IRE1α in mutant Htt-overexpressing cells and to protect against cell degeneration and caspase-3 activation. These results show that ER stress-mediated IRE1 activation is part of mutant Htt toxicity and that this is counteracted by Usp14 expression. Usp14 effectively reduced cellular aggregates and counteracted cell degeneration indicating an important role of this protein in mutant Htt-induced cell toxicity.

INTRODUCTION

Huntington's disease (HD) is a fatal neurodegenerative disorder that leads to progressive motor, cognitive and psychiatric symptoms (1,2). The disease is caused by an extended CAG repeat in the HTT (IT15) gene, which gives rise to an elongated polyglutamine (PolyQ) tract at the first exon of the translated Htt protein (3). Mutant huntingtin (mtHtt) having more than ∼36 glutamines show a tendency to form large protein aggregates in the cells that are present both in the nucleus and in the cytosol (46). Available data from cellular models of HD indicate that the mtHtt-containing aggregates play a role in cellular toxicity (2,7). In addition, studies of transgenic mice overexpressing polyQ peptides caused neurodegeneration indicating that the polyQ stretch itself induces toxicity (8). It is considered that reducing the amount of mtHtt aggregates present in cells either by decreasing their formation or enhancing their clearance could delay the pathological process in HD (2,7,9).

Autophagy and the ubiquitin-proteasome system (UPS) are the two main degradation pathways in the cell. Proteins destined for degradation by the UPS are first labeled by ubiquitin in an energy-requiring process and then transferred to the large complex made up of the 26S proteasome for subsequent digestion by various enzymes (10). Autophagy is divided into different types that are distinguished by the mechanism and regulatory events involved (11,12). In macroautophagy, often referred to as classical autophagy, cytoplasmic components are engulfed by double-membrane structure, phagophore, which expand to form a vesicle called autophagosome (11,12). Autophagosomes further fuse with lysosomes in which the content is degraded by hydrolases (11,12). The process of autophagy occurs in the cytoplasm whereas UPS is active in all cellular compartments including the cell nucleus, and therefore, both pathways could in principle take part in the clearance of the mtHtt aggregates (13,14).

When studying cellular pathways altered in HD, we and others observed that overexpression of mtHtt induces endoplasmic reticulum (ER) stress and activation of ER signaling pathways including those regulated by the Inositol requiring enzyme-1α (IRE1α) (1518). Mutant huntingtin also impairs the ER-associated degradation (ERAD) as shown recently in yeast (19,20). Furthermore, recent data suggest that ubiquitin-specific protease 14 (Usp14) belonging to the family of deubiquitinating enzymes (DUBs) can interact with IRE1α and influence ERAD (21). The DUBs are crucial for proper cell function, and disturbances in these enzymes may lead to various diseases (22). Previous studies have shown that Usp14 associates with the 26S proteasome as studied in cells (23,24). Usp14 is able to cleave ubiquitin chains in different proteins and trim them at the proteasome thereby replenishing the cellular pool of ubiquitin (25,26). In ataxia (axJ) mice, there is a deletion of Usp14 that causes progressive neurological symptoms in the form of ataxia, tremor, hind limb paralysis and muscular weakness (27,28). The mechanisms underlying these changes are thought to involve altered synaptic functions and neurotransmitter release in the axJ mice. However, little is so far known about the possible role of Usp14 in control of protein aggregation or in neurological disorders such as HD.

In the present study, we show that overexpression of Usp14 protects neuronal cells against mHtt-mediated cell toxicity and cell death. Usp14 acts by reducing the amount of cell aggregates and by inhibiting the activation of IRE1 that is induced by mtHtt.

RESULTS

Usp14 levels are unchanged but Usp14 is redistributed in mutant Htt-expressing cells

Usp14 is one of the important DUBs in the nervous system, and we reasoned that it could be involved in neurodegenerative diseases caused by disturbed protein metabolism. Usp14 is also expressed in neuronal PC6.3 cells as shown by immunocytochemistry (Fig. 1A). We then analyzed Usp14 levels in the PC6.3 cells after the expression of Htt fragment proteins having different polyQ repeats (18Q, 39Q, 53Q and 120Q). Data showed that there was no overt change in the total cellular levels of Usp14 observed either by immunostaining or using immunoblotting (Fig. 1A and B). To study whether the distribution of Usp14 in the cells might change, we performed cell fractionation studies of control and 120Q-Htt-expressing PC6.3 cells. Data showed that the amount of Usp14 associated with the ER membrane fraction substantially decreased in the mtHtt-expressing cells whereas the level of Usp14 increased in the cytosolic fraction (Fig. 1D and E). This indicates that the subcellular localization of Usp14 was altered in mtHtt-expressing cells as shown by cell fractionation.

Figure 1.

Usp14 in control and mutant Htt-expressing neuronal PC6.3 cells. PC6.3 cells were transfected as described in Materials and methods using control EGFP vector (C) and plasmids encoding N-terminal EGFP-linked huntingtin fragment proteins containing different polyglutamine repeats (18Q–120Q). Cells were incubated for 24 h and analyzed further as indicated below. (A) Immunocytochemisty. Immunostaining using anti-Usp14 antibody (red fluorescence) was done as described in Materials and methods and shows a cytosolic localization of the protein. The expression of Htt fragment protein was shown by GFP staining (green). Note cellular aggregates in 120Q-huntingtin-expressing cells. Co-expression of Usp14 and Htt fragment proteins in the cells are shown in the merged pictures. Scale bar: 10 μm. (B) Immunoblots. The levels of Usp14 in control and huntingtin fragment protein expressing cells. β-actin was used as control. There was no significant change in overall Usp14 levels. Typical immunoblot is shown, and the experiment was repeated. (C) Immunoblots. Subcellular fractions were prepared from transfected PC6.3 cells as described in Materials and methods. The levels of Usp14 were analyzed in the cytosolic, and in the membrane (containing the endoplasmic reticulum, ER) fractions. Note a decreased amount of Usp14 in the membrane fraction in 120Q-Htt-expressing cells. Antibodies against β-actin were used as a marker for cytosol, and the antibody against the KDEL sequence in the glucose-regulated protein-94 was used as an ER marker. (D) Quantification. Ratio of Usp14 levels in the cytosol and ER. Quantification was done using ImageJ. Values represent mean ± SD, n = 3. **P < 0.01 for 120Q versus control. (E) Immunoprecipitation (IP). IP was done using anti-Usp14 antibodies and lysates from 25Q-Htt- and 103Q-Htt-stably-expressing Hela cells as described in Materials and methods. The expression of Usp14 and the 19S proteasome subunit-2 protein (19S S2) were analyzed by immunoblotting. Upper panel, interaction of Usp14 with 19S S2 protein was similar in control 25Q-Htt- and mutant 103Q-Htt-expressing cells. Lower panel, Usp14 and 19S S2 levels in cell lysates. β-actin was used as a loading control. Typical experiment is shown and was repeated three times.

Figure 1.

Usp14 in control and mutant Htt-expressing neuronal PC6.3 cells. PC6.3 cells were transfected as described in Materials and methods using control EGFP vector (C) and plasmids encoding N-terminal EGFP-linked huntingtin fragment proteins containing different polyglutamine repeats (18Q–120Q). Cells were incubated for 24 h and analyzed further as indicated below. (A) Immunocytochemisty. Immunostaining using anti-Usp14 antibody (red fluorescence) was done as described in Materials and methods and shows a cytosolic localization of the protein. The expression of Htt fragment protein was shown by GFP staining (green). Note cellular aggregates in 120Q-huntingtin-expressing cells. Co-expression of Usp14 and Htt fragment proteins in the cells are shown in the merged pictures. Scale bar: 10 μm. (B) Immunoblots. The levels of Usp14 in control and huntingtin fragment protein expressing cells. β-actin was used as control. There was no significant change in overall Usp14 levels. Typical immunoblot is shown, and the experiment was repeated. (C) Immunoblots. Subcellular fractions were prepared from transfected PC6.3 cells as described in Materials and methods. The levels of Usp14 were analyzed in the cytosolic, and in the membrane (containing the endoplasmic reticulum, ER) fractions. Note a decreased amount of Usp14 in the membrane fraction in 120Q-Htt-expressing cells. Antibodies against β-actin were used as a marker for cytosol, and the antibody against the KDEL sequence in the glucose-regulated protein-94 was used as an ER marker. (D) Quantification. Ratio of Usp14 levels in the cytosol and ER. Quantification was done using ImageJ. Values represent mean ± SD, n = 3. **P < 0.01 for 120Q versus control. (E) Immunoprecipitation (IP). IP was done using anti-Usp14 antibodies and lysates from 25Q-Htt- and 103Q-Htt-stably-expressing Hela cells as described in Materials and methods. The expression of Usp14 and the 19S proteasome subunit-2 protein (19S S2) were analyzed by immunoblotting. Upper panel, interaction of Usp14 with 19S S2 protein was similar in control 25Q-Htt- and mutant 103Q-Htt-expressing cells. Lower panel, Usp14 and 19S S2 levels in cell lysates. β-actin was used as a loading control. Typical experiment is shown and was repeated three times.

Apart from its association with the ER, USP14 is localized to proteasomes and specifically binds the protein Rpn1 (23,24). We performed an immunoprecipitation assay for Usp14 using control and mtHtt-stably-expressing Hela cells to study the interaction of Usp14 with Rpn1. Data showed that the binding of Usp14 to Rpn1 was roughly the same in both cell types suggesting that Usp14 binds the proteasome also in mtHtt-expressing cells.

Usp 14 expression reduces cell aggregates in mutant Htt-expressing cells

We have previously shown that the expression of mtHtt induces ER stress and leads to cell degeneration associated with formation of cellular aggregates containing mtHtt proteins (16,29,30). In view of the interaction of Usp14 with the ER (21), we wanted to study whether enhanced expression of Usp14 can influence ER stress and cellular aggregates in mtHtt-expressing cells. Figure 2A shows that cellular aggregates are present in 120Q-mtHtt-expressing PC6.3 cells but not in 18Q-Htt-expressing cells. Most importantly, co-expression of Usp14 led to a decrease in the number of cells present with aggregates (Fig. 2A). Quantification of the data showed that cells with aggregates decreased from ∼20% to <8% after Usp14 expression (Fig. 2B). We then performed a solubility assay to study the influence of Usp14 on the amount of insoluble and soluble mtHtt in these cells.

Figure 2.

Usp14 reduces cell aggregates in mutant Htt-expressing cells. Cells were transfected with plasmids expressing wild-type and mtHtt proteins alone or together with the Usp14-expressing plasmid. Cells were analyzed further as indicated. (A) Fluorescence images of PC6.3 cells expressing 18Q-Htt and 120Q-Htt constructs. Note the presence of aggregates in 120Q-huntingtin-expressing cells that are decreased by co-expression of Usp14. Scale bar: 10 μm. (B) Quantification. The number of aggregates was counted using confocal images of 18Q- or 120Q-huntingtin-expressing cells alone or after co-expression of Usp14. More than 100 cells per cover slip were counted in triplicate samples, and experiments were repeated. Values represent mean ± SD, n = 3. **P < 0.01 for 120Q versus C and for 120Q + Usp14 versus 120Q. (C) Solubility assays. PC6.3 cells were transfected as indicated, and insoluble and soluble huntingtin fragment proteins were separated by immunoblotting. The levels of Usp14 protein were determined using anti-Usp14 antibody. β-actin used as a control. Note the presence of insoluble Htt in 120Q-huntingtin-expressing cells that were reduced by co-expression of Usp14. Typical experiment is shown and was repeated three times. (D) PC6.3 cells were transfected with 120Q-expressing Htt plasmid in conjunction with Usp14 plasmid as indicated. Cells were incubated for 24 h and further treated with 10 μm MG132 (a proteasome inhibitor) or with 5 mm 3-methyladenine (an autophagy inhibitor) for 4 h. The number of aggregates was counted from confocal images of 120Q-huntingtin-expressing cells in the absence or presence of Usp14 co-expression. Values represent means ± SD, n = 3. **P < 0.01 for 120Q + Usp14 versus 120Q. *P < 0.05 for 120Q + MG321 + Usp14 versus 120Q + Mg321. (E) PC6.3 cells were transfected with the mutant Usp14 (mUsp14) plasmid lacking enzymatic activity in conjunction with the 120Q-mtHtt plasmid. The number of cells with aggregates was counted as mentioned earlier. There was no significant effect of mutant Usp14 in these experiments. (F) PC6.3 cells were transfected with plasmids expressing the FL normal 17Q-containing Htt or the disease-causing 75Q-containg mtHtt in conjunction with GFP-linked ubiquitin construct. Cells were observed under a fluorescent microscope to visualize the distribution of GFP-ubiquitin. Note the presence of ubiquitin in aggregates in mutant 75Q-Htt-expressing cells. Typical image is shown and experiment was repeated. Scale bar: 10 μm. (G) PC6.3 cells were treated with 50 ng/NGF to induce neuronal differentiation. Cells were further transfected with wild-type and 18Q- and 120Q-Htt together with Usp14 as indicated and the amount of soluble and insoluble Htt species resolved by immunoblotting. Upper panel, the level of Usp14 was not influenced by the expression of Htt fragment proteins in NGF-treated PC6.3 cells. Lower panel, Usp14 reduced the amount of SDS-insoluble aggregates in 120Q-Htt-expressing cells as shown by immunoblot. Typical experiment using NGF-treated PC6.3 cells is shown and was repeated with same results. (H) Wild-type and mutant Usp14 (mUsp14) were expressed for 24 h in conjunction with 120Q-Htt in PC6.3 cells and the amount of soluble and insoluble Htt species resolved by immunoblotting, 10 μm MG132 was added to some cells for 6 h. mUsp14 did not influence mtHtt levels whereas MG132 increased both soluble and insoluble mtHtt levels in 120Q-Htt-expressing cells. Typical experiment is shown and was repeated with similar results.

Figure 2.

Usp14 reduces cell aggregates in mutant Htt-expressing cells. Cells were transfected with plasmids expressing wild-type and mtHtt proteins alone or together with the Usp14-expressing plasmid. Cells were analyzed further as indicated. (A) Fluorescence images of PC6.3 cells expressing 18Q-Htt and 120Q-Htt constructs. Note the presence of aggregates in 120Q-huntingtin-expressing cells that are decreased by co-expression of Usp14. Scale bar: 10 μm. (B) Quantification. The number of aggregates was counted using confocal images of 18Q- or 120Q-huntingtin-expressing cells alone or after co-expression of Usp14. More than 100 cells per cover slip were counted in triplicate samples, and experiments were repeated. Values represent mean ± SD, n = 3. **P < 0.01 for 120Q versus C and for 120Q + Usp14 versus 120Q. (C) Solubility assays. PC6.3 cells were transfected as indicated, and insoluble and soluble huntingtin fragment proteins were separated by immunoblotting. The levels of Usp14 protein were determined using anti-Usp14 antibody. β-actin used as a control. Note the presence of insoluble Htt in 120Q-huntingtin-expressing cells that were reduced by co-expression of Usp14. Typical experiment is shown and was repeated three times. (D) PC6.3 cells were transfected with 120Q-expressing Htt plasmid in conjunction with Usp14 plasmid as indicated. Cells were incubated for 24 h and further treated with 10 μm MG132 (a proteasome inhibitor) or with 5 mm 3-methyladenine (an autophagy inhibitor) for 4 h. The number of aggregates was counted from confocal images of 120Q-huntingtin-expressing cells in the absence or presence of Usp14 co-expression. Values represent means ± SD, n = 3. **P < 0.01 for 120Q + Usp14 versus 120Q. *P < 0.05 for 120Q + MG321 + Usp14 versus 120Q + Mg321. (E) PC6.3 cells were transfected with the mutant Usp14 (mUsp14) plasmid lacking enzymatic activity in conjunction with the 120Q-mtHtt plasmid. The number of cells with aggregates was counted as mentioned earlier. There was no significant effect of mutant Usp14 in these experiments. (F) PC6.3 cells were transfected with plasmids expressing the FL normal 17Q-containing Htt or the disease-causing 75Q-containg mtHtt in conjunction with GFP-linked ubiquitin construct. Cells were observed under a fluorescent microscope to visualize the distribution of GFP-ubiquitin. Note the presence of ubiquitin in aggregates in mutant 75Q-Htt-expressing cells. Typical image is shown and experiment was repeated. Scale bar: 10 μm. (G) PC6.3 cells were treated with 50 ng/NGF to induce neuronal differentiation. Cells were further transfected with wild-type and 18Q- and 120Q-Htt together with Usp14 as indicated and the amount of soluble and insoluble Htt species resolved by immunoblotting. Upper panel, the level of Usp14 was not influenced by the expression of Htt fragment proteins in NGF-treated PC6.3 cells. Lower panel, Usp14 reduced the amount of SDS-insoluble aggregates in 120Q-Htt-expressing cells as shown by immunoblot. Typical experiment using NGF-treated PC6.3 cells is shown and was repeated with same results. (H) Wild-type and mutant Usp14 (mUsp14) were expressed for 24 h in conjunction with 120Q-Htt in PC6.3 cells and the amount of soluble and insoluble Htt species resolved by immunoblotting, 10 μm MG132 was added to some cells for 6 h. mUsp14 did not influence mtHtt levels whereas MG132 increased both soluble and insoluble mtHtt levels in 120Q-Htt-expressing cells. Typical experiment is shown and was repeated with similar results.

Usp14 overexpression reduced particularly the SDS-insoluble aggregates in 120Q-mtHtt-expressing PC6.3 cells as shown by the immunoblotting. Thus, the amount of insoluble Htt resolved by the stacking gel induced by mutant 120Q-Htt was virtually disappeared after Usp14 expression (Fig. 2C). Accumulation of protein aggregates in the cell is a balance between their synthesis and degradation involving the autophagy and the proteasome system. We studied these pathways in mutant 120Q-Htt-expressing cells using 3 methyladenosine (3Ma) to inhibit autophagy and MG132 to block the proteasome. In these experiments, these inhibitors were used for a relatively short period of time (4 h) and the inhibitors per se had no significant effect on the aggregates (Fig. 2D). In contrast, in conjunction with Usp14 expressed for 24 h, MG132 inhibited the Usp14-mediated reduction of cellular aggregates in mutant 120Q-Htt-expressing cells (Fig. 2D). The concomitant use of Usp14 overexpression and MG132 increased the number of mtHtt aggregates above controls indicating a crucial role of UPS in the degradation of mtHtt aggregates under these conditions. In contrast, although the autophagy blocker 3Ma did increase the number of cellular aggregates in mutant 120Q-Htt-expressing cells, the effect was not statistically significant (Fig. 2D). This suggests that autophagy was less involved in the decrease of mtHtt aggregates induced by Usp14 expression in these cells. To study whether the intact Usp14 is required for this effect, we used the enzymatically inactive mutant Usp14 (mUsp14) that was unable to reduce cellular aggregates in 120Q-Htt-expressing cells (Fig. 2E). Collectively, these results show that the Usp14-mediated decrease in the amount of cellular aggregates in mtHtt-expressing cells largely involves the UPS and less so the autophagy system. To formally prove that the cell aggregates accumulating in mtHtt cells contain ubiquitin, we transfected PC6.3 cells with full-length (FL) control 17Q-Htt or the disease-causing 75Q-mtHtt plasmid. Data showed that GFP-ubiquitin was present throughout the cells in 17Q-Htt-expressing cells but accumulated in aggregates in 75Q-Htt-expressing cells (Fig. 2F).

To examine whether the state of neuronal differentiation may influence the data obtained with Usp14, we stimulated the PC6.3 cells with 50 ng/ml of NGF for 4 days. These cells were further transfected with 18Q- and 120Q-Htt fragment proteins in conjunction with Usp14 to study the amount of SDS-soluble and insoluble aggregates as done above (Fig. 2C). Results showed that Usp14 reduced particularly the insoluble aggregates in NGF-treated PC6.3 cells (Fig. 2H) in keeping with the data obtained using undifferentiated PC6.3 cells (Fig. 2C).

To further clarify the role of the proteasomes in the removal of soluble and insoluble mtHtt aggregates, PC6.3 cells expressing 120Q-Htt were treated with Usp14 in the absence or presence of MG132. Data showed that the levels of both soluble and insoluble mtHtt aggregates were increased in the presence of MG132 for 6 h to effectively inhibit the proteasomes (Fig. 2G). In contrast, mutant Usp14 did not influence the levels of soluble or insoluble aggregates in these cells, which is in accordance with data shown in Figure 2E.

Usp14 induces clearance of aggregates in stably mutant Htt-expressing Hela cells

To examine the action of Usp14 further, we studied HeLa cells stably expressing mtHtt with the first 17 amino acids of Htt linked to a polyQ expansion (65Q or 103Q) and to monomeric CFP (HttPolyQ-mCFP) (31). Results obtained showed that the expression of Usp14 also reduced the amount of aggregates in these mutant 103Q-expressing Hela cells (Fig. 3A and B). The expression of silencing RNA against Usp14 in turn slightly increased aggregates in mutant 103Q-expressing Hela cells, but the effect was not statistically significant (Fig. 3B). Performing the solubility assay with Hela cells showed that the decrease in aggregates by Usp14 involved particularly the insoluble mtHtt species that were significantly reduced in the 103Q-Htt-expressing cells after Usp14 expression (Fig. 3C and D).

Figure 3.

Usp14 induces clearance of mtHtt aggregates in stably expressing Hela cells. Hela cells stably expressing huntingtin proteins (Htt) with different polyQ expansions (25Q, 65Q and 103Q) and fused to mCFP were cultured as described in Materials and methods. (A) Immunocytochemistry. Cellular aggregates are shown as dots in mutant 103Q-huntingtin-expressing Hela cells. The expression of Usp14 reduced the number of aggregates in these cells. (B) Quantification using ImageJ. The number of cells with aggregates was counted for 25Q- and 103Q-Htt-expressing cells alone or after co-expression of either Usp14 siRNA (siUsp14) or Usp14-expressing plasmids. Usp14 overexpression reduced the number of cells having aggregates. Values represent means ± SD, n = 3. **P < 0.01 for 103Q versus 25Q, and *P < 0.05 for 103Q + siRNAUsp14 versus 103Q. (C) Solubility assays. Lysates of Hela cells stably expressing control and mutant huntingtin proteins were analyzed by immunoblotting to separate soluble and insoluble protein as described in Materials and methods. β-actin used as a control. Note the presence of insoluble Htt in the stacking gel in lysates from 103Q-Htt-expressing cells. Overexpression of Usp14 reduced the amount of insoluble Htt in the cell. Lower panel shows the expression levels of Usp14. (D) Quantification. Values represent mean ± SD, n = 3. **P < 0.01 for 103Q versus 25Q and for 103Q + Usp14 versus 103Q. *P < 0.05 for 65Q versus 25Q. (E) Lysates from stably 25-Q and 103Q-Htt were immunoprecipitated from stably expressing cells using GFP antibodies. Hela cells expressing GFP alone were used as controls. Upper lane, Usp14 is present in the immunoprecipitate. Middle and lower panels, GFP and Usp14 expression in the cell lysates, respectively. β-actin was used as control. Note the interaction of Usp14 with 25Q-Htt and also with 103Q-Htt. The level of 103Q-Htt revealed by immunoprecipitation was lower than that of 25Q-Htt owing to the fact that most of 103Q-Htt is present in aggregates and not accessible to precipitation.

Figure 3.

Usp14 induces clearance of mtHtt aggregates in stably expressing Hela cells. Hela cells stably expressing huntingtin proteins (Htt) with different polyQ expansions (25Q, 65Q and 103Q) and fused to mCFP were cultured as described in Materials and methods. (A) Immunocytochemistry. Cellular aggregates are shown as dots in mutant 103Q-huntingtin-expressing Hela cells. The expression of Usp14 reduced the number of aggregates in these cells. (B) Quantification using ImageJ. The number of cells with aggregates was counted for 25Q- and 103Q-Htt-expressing cells alone or after co-expression of either Usp14 siRNA (siUsp14) or Usp14-expressing plasmids. Usp14 overexpression reduced the number of cells having aggregates. Values represent means ± SD, n = 3. **P < 0.01 for 103Q versus 25Q, and *P < 0.05 for 103Q + siRNAUsp14 versus 103Q. (C) Solubility assays. Lysates of Hela cells stably expressing control and mutant huntingtin proteins were analyzed by immunoblotting to separate soluble and insoluble protein as described in Materials and methods. β-actin used as a control. Note the presence of insoluble Htt in the stacking gel in lysates from 103Q-Htt-expressing cells. Overexpression of Usp14 reduced the amount of insoluble Htt in the cell. Lower panel shows the expression levels of Usp14. (D) Quantification. Values represent mean ± SD, n = 3. **P < 0.01 for 103Q versus 25Q and for 103Q + Usp14 versus 103Q. *P < 0.05 for 65Q versus 25Q. (E) Lysates from stably 25-Q and 103Q-Htt were immunoprecipitated from stably expressing cells using GFP antibodies. Hela cells expressing GFP alone were used as controls. Upper lane, Usp14 is present in the immunoprecipitate. Middle and lower panels, GFP and Usp14 expression in the cell lysates, respectively. β-actin was used as control. Note the interaction of Usp14 with 25Q-Htt and also with 103Q-Htt. The level of 103Q-Htt revealed by immunoprecipitation was lower than that of 25Q-Htt owing to the fact that most of 103Q-Htt is present in aggregates and not accessible to precipitation.

As these Hela cells express stably mtHtt, they are ideal to study also the possibility that Usp14 might interact with Htt proteins.

Immunoprecipitation experiments were done and showed that Usp14 binds both 25Q-Htt and 103Q-mtHtt proteins (Fig. 3E). These results indicate that Usp14 can interact with Htt that may be part of its action in reducing cellular aggregates.

We then also studied the possible role of Usp14 in the clearance of mtHtt aggregates that can be induced by doxycycline in these cells and which have been shown to involve mainly the macroautophagy pathway (31). The results showed that reducing Usp14 levels by silencing RNA or employing the mutant Usp14 did not interfere with the doxycycline-mediated clearance of mtHtt aggregates (data not shown). This indicates that the process of macroautophagy is not influenced by the relative lack of active Usp14 in these cells. Unfortunately, we were not able to study the opposite as both Usp14 and doxycycline alone are efficient in reducing mtHtt aggregates, and no additive effect could therefore be shown.

Usp14 counteracts caspase-3 activation and cell death induced by mtHtt

To study whether the decrease in aggregates brought about by Usp14 has an influence on cell toxicity, we analyzed the cleavage of caspase-3 that is involved in cell death regulation (16,29,30). In line with previous data, we observed that caspase-3 is activated in cells expressing 120Q-Htt and disease-causing 75Q-Htt proteins (Fig. 4A and B). Poly(ADP-ribose) polymerase (PARP) that is a downstream substrate for caspase-3 was also cleaved in these cells as shown here for 120Q-mtHtt (Fig. 4A). Most importantly, the expression of Usp14 inhibited the cleavage of caspase-3 and PARP induced by mtHtt in these cells (Fig. 4A and B). In contrast, the expression of mUsp14 was not able to reduce the activation of caspase3 induced by 120Q-Htt as shown by immunoblotting (Fig. 4C). To study whether the reduced caspase-3 activation caused by wild-type Usp14 can have functional consequences, we determined the amount of cell death in these cultures. Data showed that Usp14 decreases mtHtt-induced cell death as evident by the reduced number of cells with chromatin condensation and fragmentation (Fig. 4D). Taken together these results show that Usp14 expression can beneficially influence cell viability of mtHtt-expressing cells.

Figure 4.

Usp14 overexpression counteracts mutant Htt-induced cell toxicity. PC6.3 cells were transfected for 36 h with control and mutant huntingtin-expressing plasmids alone or together with Usp14 as indicated. (A) Immunoblots. Cells were transfected with 18Q- and 120Q-fragment Htt-expressing plasmids. Co-expression of Usp14 in 120Q-Htt-expressing cells reduced caspase-3 (17 kDa) and PARP cleavage (89 kDa). β-actin was used as control. (B) Cells were transfected with plasmids expressing normal 17Q-containing FL Htt or the disease-causing 75Q-containing mtHtt. Co-expression of Usp14 reduced caspase-3 cleavage (17 kDa) in mtHtt-expressing cells. Typical experiment is shown and was repeated. (C) Mutant Usp14 (mUsp14) was expressed in conjunction with 120Q-Htt. mUsp14 did not reduce cleavage of caspase-3 induced by 120Q-Htt. Typical experiment is shown and was repeated with similar results. (D) Cells were transfected with 120Q-containing Htt fragment protein alone or together with Usp14. Cell death was determined at 36 h by counting the number of cells having condensed and fragmented nuclei as described in Materials and methods. *P < 0.05 120Q versus 120Q + Usp14.

Figure 4.

Usp14 overexpression counteracts mutant Htt-induced cell toxicity. PC6.3 cells were transfected for 36 h with control and mutant huntingtin-expressing plasmids alone or together with Usp14 as indicated. (A) Immunoblots. Cells were transfected with 18Q- and 120Q-fragment Htt-expressing plasmids. Co-expression of Usp14 in 120Q-Htt-expressing cells reduced caspase-3 (17 kDa) and PARP cleavage (89 kDa). β-actin was used as control. (B) Cells were transfected with plasmids expressing normal 17Q-containing FL Htt or the disease-causing 75Q-containing mtHtt. Co-expression of Usp14 reduced caspase-3 cleavage (17 kDa) in mtHtt-expressing cells. Typical experiment is shown and was repeated. (C) Mutant Usp14 (mUsp14) was expressed in conjunction with 120Q-Htt. mUsp14 did not reduce cleavage of caspase-3 induced by 120Q-Htt. Typical experiment is shown and was repeated with similar results. (D) Cells were transfected with 120Q-containing Htt fragment protein alone or together with Usp14. Cell death was determined at 36 h by counting the number of cells having condensed and fragmented nuclei as described in Materials and methods. *P < 0.05 120Q versus 120Q + Usp14.

The interaction of Usp14 with IRE1α is decreased in mutant Htt-expressing cells

To study the mechanisms by which Usp14 is able to counteract cell degeneration, we analyzed ER stress that is present in mtHtt-expressing cells (1619). Previous data had also shown that Usp14 interacts with IRE1α, a transmembrane protein in the ER involved in ER signaling (21). We asked whether the reduced association of Usp14 with ER as observed in mtHtt-expressing cells (Fig. 1C and D) is reflected in a decreased binding of Usp14 to IRE1α. Immunoprecipitation experiments revealed that Usp14 avidly binds IRE1α in control, 18Q-Htt but not in 120Q-Htt-expressing cells (Fig. 5A). Immunostaining also showed that Usp14 is partly co-localized with IRE1α in control but not in 120Q-Htt-expressing cells (Fig. 5B). To study whether the decreased binding of Usp14 influences the activation of IRE1α, we employed an antibody against active, phosphorylated IRE1α. As shown by immunoblotting, the levels of p-IRE1α increased in the mtHtt-expressing cells (Fig. 5C and D). Most importantly, the concomitant expression of Usp14 was able to reduce p-IRE1α induced by 120Q-Htt (Fig. 5C and D). In contrast to IRE1α, Usp14 did not influence the Protein kinase RNA-like endoplasmic reticulum kinase (PERK) and ATF6 pathways that are also activated in 120Q-Htt-expressing cells (Fig. 5F). This was shown by immunoblotting to detect phospho-eIF2α levels and ATF6 cleavage as markers for activation of PERK and ATF6 pathways, respectively (Fig. 5F). This shows that Usp14 plays a modulatory role preferentially in IRE1α signaling that as such may mediate some of the deleterious effects of mtHtt on cellular aggregate formation and cell stress.

Figure 5.

Usp14 interacts with IRE1 and reduces its phosphorylation in mtHtt-expressing cells. PC6.3 cells were transfected and analyzed further as indicated. (A) Immunoprecipitation (IP). IP was performed with cell lysates using anti-Usp14 antibody followed by immunoblotting to show levels of Usp14 and of IRE1α. Upper panel, interaction of Usp14 with IRE1α in 18Q- but not in 120Q-huntingtin-expressing cells. Lower panel, IRE1α and Usp14 levels in cell lysates. β-actin was used as control. Typical experiment is shown and was repeated three times. (B) Confocal images. Immunostaining was done using anti-IRE1α and anti-Usp14 antibodies in 18Q- and 120Q-Htt-expressing cells. Usp14 (violet fluorescence) and IRE1α (red) are present in different compartments in 120Q-Htt-expressing cells. Scale bar: 10 μm. (C) Immunoblot. The level of p-IRE1α increased in 120Q-Htt-expressing cells and was reduced by Usp14 expression. The expression of Usp is shown by the Flag tag. (D) Relative level of p-IRE1 to IRE1. Values represent means ± SD, n = 3. **P < 0.01 for 120Q versus C, and *P < 0.05 for Usp14 + 120Q versus 120Q. (E) Mutant Usp14 (mUsp14) was expressed in conjunction with 120Q-Htt. mUsp14 did not reduce p-IRE1α levels induced by 120Q-Htt. The expression of mUsp14 is shown by the Flag tag. Typical experiment is shown and was repeated with similar results. (F) ER stress signaling pathways PERK and ATF6 in PC6.3 cells were studies using specific antibodies as described earlier. Usp14 expression did not reduce activation of PERK as shown by the intact phospho-eIF2 levels, nor did it reduce activation of ATF6 as shown by the presence of cleavage of ATF6 in 120Q-Htt-expressing cells. Typical experiment is shown and was repeated with similar results.

Figure 5.

Usp14 interacts with IRE1 and reduces its phosphorylation in mtHtt-expressing cells. PC6.3 cells were transfected and analyzed further as indicated. (A) Immunoprecipitation (IP). IP was performed with cell lysates using anti-Usp14 antibody followed by immunoblotting to show levels of Usp14 and of IRE1α. Upper panel, interaction of Usp14 with IRE1α in 18Q- but not in 120Q-huntingtin-expressing cells. Lower panel, IRE1α and Usp14 levels in cell lysates. β-actin was used as control. Typical experiment is shown and was repeated three times. (B) Confocal images. Immunostaining was done using anti-IRE1α and anti-Usp14 antibodies in 18Q- and 120Q-Htt-expressing cells. Usp14 (violet fluorescence) and IRE1α (red) are present in different compartments in 120Q-Htt-expressing cells. Scale bar: 10 μm. (C) Immunoblot. The level of p-IRE1α increased in 120Q-Htt-expressing cells and was reduced by Usp14 expression. The expression of Usp is shown by the Flag tag. (D) Relative level of p-IRE1 to IRE1. Values represent means ± SD, n = 3. **P < 0.01 for 120Q versus C, and *P < 0.05 for Usp14 + 120Q versus 120Q. (E) Mutant Usp14 (mUsp14) was expressed in conjunction with 120Q-Htt. mUsp14 did not reduce p-IRE1α levels induced by 120Q-Htt. The expression of mUsp14 is shown by the Flag tag. Typical experiment is shown and was repeated with similar results. (F) ER stress signaling pathways PERK and ATF6 in PC6.3 cells were studies using specific antibodies as described earlier. Usp14 expression did not reduce activation of PERK as shown by the intact phospho-eIF2 levels, nor did it reduce activation of ATF6 as shown by the presence of cleavage of ATF6 in 120Q-Htt-expressing cells. Typical experiment is shown and was repeated with similar results.

Data showed also that mutant Usp14 was unable to reduce p-IRE1α levels in 120Q-Htt-expressing cells (Fig. 5E), further suggesting that wild-type Usp14 is required for its beneficial effects in mtHtt-expressing cells.

IRE1α is specifically activated and binds less Usp14 in the striatum of BACHD mice

To examine whether the changes in Usp14 and IRE1α were also observed in vivo, we studied brain samples obtained from the BACHD mouse model of HD. This transgenic mouse expresses FL mtHtt and recapitulates a number of clinical features of the disease (3235). Immunoblots showed that the levels of Usp14 did not change neither in the striatum nor in the cerebral cortex of BACHD mice compared with the corresponding controls (Fig. 6A). However, we observed that IRE1α was activated and p-IRE1 levels were increased in the striatum but not in the cerebral cortex of BACHD mice compared with wild-type controls (Fig. 6B and C). We also observed that the binding of Usp14 to IRE1α was reduced in samples from BACHD striatum compared with wild-type controls (Fig. 6D). The observed increase in IRE1α activation in the BACHD striatum with a reduced binding of Usp14 is in line with data obtained in culture in mutant 120Q-Htt-expressing cells.

Figure 6.

IRE1α is specifically activated in the striatum in BACHD mice. Striatal and cortical tissues were prepared from BACHD mice and their wild-type (wt) littermates. Immunoblots were made as described in Materials and methods using antibodies against Usp14 and total and phosphorylated IRE1α. β-actin was used as control. (A) Usp14 levels remained the same in the striatum of control (wt) and BACHD mice. Left, striatum. Right, cerebral cortex. Representative immunoblots are shown and were repeated five times. (B) Phosphorylated IRE1α was increased in the striatum but not in cerebral cortex of BACHD mice. Left, striatum. Right, cerebral cortex. (C) Quantification of pIRE1 levels in striatum versus IRE1. Values represent means ± SD, n = 5. **P < 0.01 for BACHD versus wild type. (D) Immunoprecipitation. IP was performed with lysates from septum of wt and BACHD mice using anti-Usp14 antibodies followed by immunoblotting with anti-IRE1α and anti-Usp14 antibodies. Usp14 interacts with IRE1α in wt striatum but not to any appreciable degree in BACHD mice. Representative IP is shown and was repeated three times.

Figure 6.

IRE1α is specifically activated in the striatum in BACHD mice. Striatal and cortical tissues were prepared from BACHD mice and their wild-type (wt) littermates. Immunoblots were made as described in Materials and methods using antibodies against Usp14 and total and phosphorylated IRE1α. β-actin was used as control. (A) Usp14 levels remained the same in the striatum of control (wt) and BACHD mice. Left, striatum. Right, cerebral cortex. Representative immunoblots are shown and were repeated five times. (B) Phosphorylated IRE1α was increased in the striatum but not in cerebral cortex of BACHD mice. Left, striatum. Right, cerebral cortex. (C) Quantification of pIRE1 levels in striatum versus IRE1. Values represent means ± SD, n = 5. **P < 0.01 for BACHD versus wild type. (D) Immunoprecipitation. IP was performed with lysates from septum of wt and BACHD mice using anti-Usp14 antibodies followed by immunoblotting with anti-IRE1α and anti-Usp14 antibodies. Usp14 interacts with IRE1α in wt striatum but not to any appreciable degree in BACHD mice. Representative IP is shown and was repeated three times.

DISCUSSION

In this work, we have shown that the expression of the DUB Usp14 has a favorable effect in mtHtt-expressing cells by decreasing the aggregate load and by enhancing cell viability. The presence of mtHtt aggregates is considered as an essential part of the cell toxicity in HD although the nature of the toxic species present or its pathophysiology is not fully understood (46). The mtHtt aggregates increase progressively in the cells if not effectively removed, and they may contribute to the adverse effects of HD manifested as alterations in cell signaling and metabolism, changes in gene transcription, vesicle transport and organelle dysfunctions with increased ROS and ER stress (1,2,7). Neurons in the striatum are particularly vulnerable in HD although some other brain areas are also affected (1,2). The molecular mechanisms determining the cell-specific effects of mtHtt are, however, so far incompletely understood.

The clinical picture of HD is rather complex, and apart from motor and cognitive deficits, psychiatric symptoms and sleep disturbances are also common (13). The BACHD mouse model recapitulates a number of clinical features of the disease and displays mild striatal pathology at a later stage (3235). Interestingly, we observed that IRE1α that acts as an ER stress sensor was preferentially activated in the striatum but not in the cerebral cortex of BACHD mice. Recently, Lee et al. also reported similar findings regarding IRE1α and suggested that the induction of ER stress leads to a decrease in autophagy and an accumulation of mtHtt in striatal cells (18). In the present work, we show further that alterations in cellular distribution of Usp14 could be the missing link between mtHtt-mediated ER stress induction and the enhanced neuronal degeneration. Along with this, we observed a reduced binding of Usp14 to IRE1α in the ER membrane in mtHtt-expressing cells in culture and in the striatum of BACHD mice, whereas a forced expression of Usp14 was able to counteract cell death and reduce cell aggregates in mtHtt-expressing cells.

Usp14 is known to bind the small 19S proteasome subunit to regulate the trimming of the ubiquitin chains in proteins prior to their degradation in the large 26S proteasome subunit (23,24). This process of deubiquitination is of utmost importance as it ensures a constant supply of ubiquitin molecules for protein ubiquitination reactions in the cell (25,26). In this work, we observed that the interaction of Usp14 with the 19S proteasome subunit-2 protein (also called Rpn11) was largely intact in mtHtt-expressing cells indicating that the deubiquitination process was not altered. Likewise, we observed that the Usp14-mediated decrease in mtHtt aggregates involved mainly the proteasome as shown by using the inhibitor MG132 with minor contribution of autophagy. Available data on the degradation of larger aggregates in HD have shown that autophagy and the formation of autophagosome are important for their clearance in the cells (13,14,18). This is based on genetic evidence and by using drugs like rapamycin to enhance autophagy to clear protein aggregates in the cells (3638). Although our results show that the UPS is important for the Usp14-mediated reduction of cellular aggregates under present conditions, they do not exclude the contribution of the process of autophagy in this process and in vivo. It is more likely that UPS and the autophagy are both involved in the removal of mtHtt aggregates depending on cell type and conditions used for the study. Along this line, previous studies have shown that the degradation of polyQ proteins by the proteasome is rather inefficient probably due to the trapping of the polyQ protein within the proteasome (39). We reasoned that part of this could be steric hindrance owing to the presence of large amounts of ubiquitin chains in the aggregates that may clog the proteasome. In line with previous data, we observed here that ubiquitin is present in the protein aggregates in mtHtt-expressing cells (Fig. 2F). We also considered that the cleavage of the ubiquitin chain by Usp14 could render the mtHtt aggregates more accessible to degradation via the proteasomes. Collectively, our data support this view as Usp14 expression effectively reduced cell aggregates both in mtHtt transiently transfected PC6.3 cells and in mtHtt-stably-expressing Hela cells. In addition, the clearance of the cellular aggregates was dependent on the proteasomes and on the intact enzymatic activity of wild-type Ups14. It remains to be studied whether the same strategy involving increased cleavage of ubiquitin chains from larger protein aggregates may be beneficial in other neurodegenerative diseases as well.

Usp14 is an important Dubs that is highly expressed in the central nervous system and by neurons. Usp14 levels are reduced in the ataxia (axJ) mice owing to a recessive mutation in the Usp14 gene causing various neurological symptoms characterized by increased tremor and ataxia and by a progressive hindlimb weakness and paralysis (25,28). Recent studies have shown that many of the symptoms in the axJ mice can be reversed by the transgene expression of wild-type Usp14 protein (40). However, little is known about the possible functions of Usp14 in brain diseases or in neurodegeneration. We found that Usp14 plays an important role in the control of mtHtt aggregation and cell stress by binding to the signaling molecule IRE1α in the ER. Interestingly, the effect of Usp14 was found to be rather specific for IRE1α as both the PERK and the ATF6 pathways were not significantly inhibited by Usp14 in mtHtt-expressing cells. We have also indications that Usp14 could differentially affect the phosphorylation and the Xbp1 splicing activities of IRE1α in mtHtt-expressing cells that requires more studies in the future.

Previous studies in yeast have further shown that Usp14 can inhibit the process of ERAD by binding to the cytoplasmic region of IRE1α (21). ERAD is part of the ER quality control system for protein degradation in the cell, and it was suggested that the release of Usp14 from IRE1α under cell stress conditions might positively influence ERAD (21). As shown here, we observed a reduced binding of Usp14 to IRE1α in mtHtt-expressing cells as well as in striatal tissue in BACHD mice. We also observed that the phosphorylation of IRE1α was significantly increased in striatum and in mtHtt-expressing cells and that Usp14 could reverse this probably by binding IRE1α. It remains to be investigated which particular signaling pathways downstream of IRE1α are influenced by Usp14 in mtHtt-expressing cells. In the present work, we have so far not studied changes in ERAD that together with ER stress are early signs of mtHTT cell toxicity (19). It is likely that Usp14 by interacting with IRE1α is able to influence both ER stress and the ERAD pathway and thereby enhance the removal of accumulated toxic protein species both in cell cytosol and in the membrane.

Taken together, we show in the present study that Usp14 is cytoprotective and reduces cell aggregates in mtHtt-expressing cells. Usp14 may have a dual function in neuroprotection by trimming the ubiquitin chains in the protein aggregates thus render them accessible for the proteasome, and by reducing p-IRE1α and ER stress responses in mtHtt-expressing cells. We therefore suggest that Usp14 can be a potential target to consider in the development of novel treatments for HD and possibly other aggregate disorders.

MATERIALS AND METHODS

Animal experiments

BACHD mice expressing FL mtHtt-containing 97 polyQ repeats (32) were obtained from the Jackson Laboratories (FVB/N strain; Bar Harbor, Maine) and housed as described previously (33). Striata and cerebral cortices were dissected from 2-month-old female BACHD mice and wild-type littermates, and the tissues were immediately frozen for further studies. Tissue pieces were weighed and suspended in buffer, and an equal amount of proteins from different samples was subjected to immunoblotting as described later. All animal experiments were approved by the local ethical committee and accomplished in accordance with the European Communities Council Directive (86/609/EEC).

Reagents and plasmid constructs

Human wild-type pRK-FLAG-USP14 and mutant pRK-FLAG-USP14 C114A were kind gifts from Dr. Ye (41), and GFP-Ub was from Addgene. HTT plasmids expressing full-length or N-terminal fragment proteins have been described earlier (16,30).

Cell culture and transfections

PC6.3 cells were cultured in RPMI 1640 medium (Lonza) supplemented with 5% fetal calf serum (Chemicon) and 10% horse serum as previously described (29,30). Cells were transfected using Transfectin reagent (BioRad) for 24–48 h with expression plasmids encoding different CAG-repeat lengths of huntingtin exon-1 fused to EGFP (18Q, 39Q, 53Q and 120Q), or FL huntingtin constructs (17Q and 75Q) and using the EGFP expression plasmid as controls (Clontech). To express Usp14, pRK-FLAG-USP14 or the mutant pRK-FLAG-USP14 C114A plasmids were also used. In some experiments, 100 µm IU1 inhibiting Usp14 activity was added for 24 h (Sigma). 5 mm 3-methyladenine (3-MA, Sigma) inhibiting autophagy was used for 4 h, 30 nm Bafilomycin A for 8 h and the proteosome inhibitor 20 µm MG-132 (Calbiochem) for 8 h.

Tetracycline (tet)-regulatable HeLa cell line stably expressing the first 17 amino acids of Htt followed by a polyQ expansion (65Q or 103Q) tagged to monomeric CFP (HttPolyQ-mCFP) were used to study the clearance of aggregates (31). As a control, we used cell lines that express 25Q, which is under the aggregation threshold of 37 polyQ repeats, and which remains soluble. Hela cells were maintained in DMEM (Lonza), and 100 ng/ml doxycycline (dox) was used to shut off the production of new protein (31). Usp14 small interfering RNAs against rat Usp14 [siRNA, 20 µm, ON-TARGETplus SMARTpool, Thermo Scientific Dharmacon (21)] and siRNA universal negative control (Mission®, Sigma) were transfected by using the Transfectin reagent (BioRad) according to the manufacturer's protocol. Cells were further incubated at 37°C for 24 h.

Immunocytochemistry

PC6.3 cells plated on poly-C-lysine (Sigma)/laminin (Sigma)-coated coverslips fixed using 4% paraformaldehyde at RT for 20 min. Cells were first incubated in phosphate-based saline (PBS) containing 0.1% Triton X-100 and 5% bovine serum albumin (Sigma) for 1 h at RT, followed by an overnight incubation at +4°C in the presence of primary antibodies. The antibodies used included Anti-Usp14 (1 : 1000; Sigma) and anti-IRE1α (1 : 100; Santa Cruz) and were diluted in 3% BSA–PBST. Cells were washed with PBST and incubated in Alexa Fluor® 488 and/or 594 (1 : 500; Invitrogen Molecular Probes)-conjugated secondary antibodies for 1 h at RT. Cells were further washed with PBST, cell nuclei were stained with Hoechst 33342 blue (4 μg/ml; Sigma) for 1 min and dying cells showed condensed and/or fragmented nuclei (30,42). Cells were mounted using Mowiol (Calbiochem) containing 50 mg/ml 1,4-diazocyclo-[2,2,2]octane (Sigma–Aldrich) or using gel mounting medium (DABCO). Controls without primary antibodies showed no staining. Cells were visualized with a Leica fluorescence microscope, and images were taken using Zeiss LSM 510 META laser scanning or Zeiss LSM 780 confocal microscope at the Molecular Imaging Unit, Biomedicum Helsinki.

Subcellular fractionation

Cell fractionation was done as earlier described (29,30). Briefly, cells were lysed and homogenized in 20 mm HEPES–KOH, pH 7.5, 10 mm KCl, 1.5 mm MgCl2, 1 mm NaEDTA, 1 mm NaEGTA and 1 mm dithiothreitol containing 250 mm sucrose and a mixture of protease inhibitors (Roche). Unbroken cells and nuclei were collected by low-speed centrifugation of 800 g for 10 min, and the resulting supernatant was centrifuged for 20 min at 10 000 g to obtain mitochondria, followed by centrifugation of the supernatant for 1 h at 100 000 g. Protein fractions were analyzed with western blotting as explained later. Antibodies recognizing the KDEL sequence in the glucose-regulated protein-94 (GRP94) were used to show the presence of ER membranes, and β-actin was used as a cytosolic marker.

Immunoblotting

Cells were lysed in the ice-cold radio-immunoprecipitation assay (RIPA) buffer (150 mm NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 50 mm Tris–HCl and 0.1% SDS, pH 8.0) supplemented with protease inhibitor cocktail (Roche) and phosphatase inhibitor PhosphoStop (Roche), and protein concentrations were determined (BioRad BC assay) (30,43). To detect PARP by immunoblotting, cells were lysed in separate lysis buffer (62.5 mm Tris, pH 6.8, 6 m urea, 10% glycerol, 2% SDS, 0.003% bromophenol blue and 5% 2-mercaptoethanol). Equal amount of protein (40 μg) was separated on SDS–PAGE, and proteins were blotted onto Hybond-C Extra nitrocellulose filters (Pharmacia-Amersham, Helsinki, Finland). Filters were blocked for 1 h in 5% milk–TBST or 5% BSA–TBST followed by an overnight incubation at +4°C using primary antibodies. These included antibodies against active caspase-3 (1 : 350; Cell Signaling Technology), PARP (1 : 1000; Cell Signaling Technology), Usp14 (1 : 500; Abgent), Usp14 (1 : 5000; Sigma), GFP (1 : 3000; Roche), IRE1α (1 : 1000; Abcam), p-IRE1α (1 : 1000; Abcam), p-eIF2α (1 : 2000, Cell Signaling), ATF-6 (1 : 1000, Pierce ), KDEL (1 : 250; Abcam) and β-actin (1 : 1000, Sigma). After washing, the filters were incubated with horseradish peroxidase-conjugated secondary antibodies (1 : 2500, Jackson ImmunoResearch and Pierce), followed by a detection using the enhanced chemiluminescent method Super Signal West Pico reagents (Pierce Thermo Scientific).

Immunoprecipitation

PC6.3 cells were transfected with 18Q and 120Q-huntingtin exon1 constructs for 24 h followed by lysis in 400 μl of RIPA buffer supplemented with protease inhibitors and PhosphoStop (Roche) (16,30). Protein concentrations were measured with BioRad protein assay DC (BCA protein assay kit), and equal amounts of proteins (typically 200–300 μg) were taken for immunoprecipitation. Preclearing was done by incubating the lysates with protein A- and G-agarose beads (Roche, Germany) followed by an overnight incubation with Usp14 antibody (Bethyl). Following day, 50 μl of protein G-agarose was added and incubated in a rotor at +4°C for 2 h (36,37). Agarose beads were collected by centrifugation and washed three times with lysis buffer. The agarose pellet was dissolved to 50 μl of 1× WB loading buffer and boiled for 10 min followed by centrifugation, 14 000 rpm for 5 min. Subsequently, immunoprecipitates were analyzed by western blotting with IRE1α and Usp14 antibodies as described earlier.

In immunoprecipitation experiment using 25Q-Htt- and 103Q-Htt-stably-expressing Hela cells, the NP-40 buffer (150 mm NaCl; 50 mm Tris, pH 7.5; 1 mm EDTA; 0.5% NP-40) completed with 0.2 mm Na2VO3, 50 mm NaF, protease and phosphatase inhibitors was used. Usp14 was precipitated from the cell extract using the monoclonal-anti-Usp14 antibody (Abnova) and subjected to immunobotting using the rabbit Usp14 antibody (1 : 1000, Bethyl laboratories) and the proteasome 19S subunit S2 (1 : 1000, Thermo). In some experiments, Hela cells transfected with GFP-expressing plasmid were subject to immunprecipitation and immunblotting using GFP and Usp14 antibodies, respectively.

To study interaction of Usp14 with IRE1α in vivo, lysates were made from striatum of control and BACHD mice and immunoprecipitation was done using anti-Usp14 antibodies followed by immunoblotting using anti-IRE1α antibodies. Three experiments were performed, and the typical data are shown in Figure 6.

Solubility assay

Cells were lysed in buffer containing 50 mm Tris–HCl, pH 7.5; 100 mm NaCl; 3 mm EGTA; 0.5% Triton X-100 and protease inhibitors (Roche), kept on ice for 5 min and suspended in three volumes of SDS loading buffer to obtain the total cell lysate as described previously (29,30). To detect high-molecular-weight forms of the huntingtin fragment proteins, the stacking gel was blotted after the run to reveal protein aggregates in the cells.

Quantification and statistics

Immunoblots were quantified using the ImageJ software. Results are expressed as percentage of controls (mean ± SD). Statistical analyses were done using students’ t-test with Bonferroni post hoc test. P < 0.05 was considered as significant.

FUNDING

This study was supported by Academy of Finland, Sigrid Juselius Foundation, Arvo and Lea Ylppö Foundation, Finska Läkaresällskapet, Liv and Hälsa Foundation, Magnus Ehrnrooth Foundation, Emil Aaltonen Foundation, von Frenckell Foundation, Oskar Öflund foundation Minerva Foundation and the Swedish Research Council. A. Hyrskyluoto is a PhD student in the Brain&Mind Doctoral Program.

ACKNOWLEDGEMENTS

We thank A. Norremolle for the N-terminal huntingtin plasmids, F. Saudou for the FL huntingtin and Y. Ye for Usp14 constructs. We are grateful to K. Söderholm for skillful technical assistance. Confocal imaging was done at Molecular Imaging Unit, Biomedicum Helsinki.

Conflict of Interest statement. None declared.

REFERENCES

1
Zuccato
C.
Valenza
M.
Cattaneo
E.
Molecular mechanisms and potential therapeutical targets in Huntington's disease
Physiol. Rev.
 
2010
90
905
981
2
Ross
C.A.
Tabrizi
S.J.
Huntington's disease: from molecular pathogenesis to clinical treatment
Lancet Neurol.
 
2011
10
83
98
3
The Huntington's disease Collaborative Research Group
A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington's disease chromosomes
Cell
 
1993
72
971
983
4
Davies
S.W.
Turmaine
M.
Cozens
B.A.
DiFiglia
M.
Sharp
A.H.
Ross
C.A.
Scherzinger
E.
Wanker
E.E.
Mangiarini
L.
Bates
G.P.
Formation of neuronal intranuclear inclusions underlies the neurological dysfunction in mice transgenic for the HD mutation
Cell
 
1997
90
537
548
5
Scherzinger
E.
Lurz
R.
Turmaine
M.
Mangiarini
L.
Hollenbach
B.
Hasenbank
R.
Bates
G.P.
Davies
S.W.
Lehrach
H.
Wanker
E.E.
Huntingtin-encoded polyglutamine expansions form amyloid-like protein aggregates in vitro and in vivo
Cell
 
1997
90
549
558
6
Wyttenbach
A.
Carmichael
J.
Swartz
J.
Furlong
R.A.
Narain
Y.
Rankin
J.
Rubinsztein
D.C.
Effects of heat shock, heat shock protein 40 (HDJ-2), and proteasome inhibition on protein aggregation in cellular models of Huntington's disease
Proc. Natl Acad. Sci. USA.
 
2000
97
2898
2903
7
Gil
J.M.
Rego
A.C.
Mechanisms of neurodegeneration in Huntington's disease
Eur. J. Neurosci.
 
2008
27
2803
2820
8
Ikeda
H.
Yamaguchi
M.
Sugai
S.
Aze
Y.
Narumiya
S.
Kakizuka
A.
Expanded polyglutamine in the Machado-Joseph disease protein induces cell death in vitro and in vivo
Nat. Genet.
 
1996
13
196
202
9
Krainc
D.
Clearance of mutant proteins as a therapeutic target in neurodegenerative diseases
Arch. Neurol.
 
2010
67
388
392
10
Glickman
M.H.
Ciechanover
A.
The ubiquitin-proteasome proteolytic pathway: destruction for the sake of construction
Physiol. Rev.
 
2002
82
373
428
11
Behrends
C.
Sowa
M.E.
Gygi
S.P.
Harper
J.W.
Network organization of the human autophagy system
Nature
 
2010
466
68
76
12
Chen
Y.
Klionsky
D.J.
The regulation of autophagy - unanswered questions
J. Cell. Sci.
 
2011
124
161
170
13
Ravikumar
B.
Duden
R.
Rubinsztein
D.C.
Aggregate-prone proteins with polyglutamine and polyalanine expansions are degraded by autophagy
Hum. Mol. Genet.
 
2002
11
1107
1117
14
Goldberg
A.L.
Protein degradation and protection against misfolded or damaged proteins
Nature
 
2003
426
895
899
15
Nishitoh
H.
Matsuzawa
A.
Tobiume
K.
Saegusa
K.
Takeda
K.
Inoue
K.
Hori
S.
Kakizuka
A.
Ichijo
H.
ASK1 is essential for endoplasmic reticulum stress-induced neuronal cell death triggered by expanded polyglutamine repeats
Genes Dev.
 
2002
16
1345
1355
16
Reijonen
S.
Putkonen
N.
Norremolle
A.
Lindholm
D.
Korhonen
L.
Inhibition of endoplasmic reticulum stress counteracts neuronal cell death and protein aggregation caused by N-terminal mutant huntingtin proteins
Exp. Cell Res.
 
2008
314
950
960
17
Vidal
R.
Caballero
B.
Couve
A.
Hetz
C.
Converging pathways in the occurrence of endoplasmic reticulum (ER) stress in Huntington's disease
Curr. Mol. Med.
 
2011
11
1
12
18
Lee
H.
Noh
J.Y.
Oh
Y.
Kim
Y.
Chang
J.W.
Chung
C.W.
Lee
S.T.
Kim
M.
Ryu
H.
Jung
Y.K.
IRE1 plays an essential role in ER stress-mediated aggregation of mutant huntingtin via the inhibition of autophagy flux
Hum. Mol. Genet.
 
2012
21
101
114
19
Duennwald
M.L.
Lindquist
S.
Impaired ERAD and ER stress are early and specific events in polyglutamine toxicity
Genes Dev.
 
2008
22
3308
3319
20
Yang
H.
Liu
C.
Zhong
Y.
Luo
S.
Monteiro
M.J.
Fang
S.
Huntingtin interacts with the cue domain of gp78 and inhibits gp78 binding to ubiquitin and p97/VCP
PLoS One
 
2010
5
e8905
21
Nagai
A.
Kadowaki
H.
Maruyama
T.
Takeda
K.
Nishitoh
H.
Ichijo
H.
USP14 inhibits ER-associated degradation via interaction with IRE1alpha
Biochem. Biophys. Res. Commun.
 
2009
379
995
1000
22
Singhal
S.
Taylor
M.C.
Baker
R.T.
Deubiquitylating enzymes and disease
BMC Biochem.
 
2008
9
Suppl 1
S3-2091-9-S1-S323
23
Borodovsky
A.
Kessler
B.M.
Casagrande
R.
Overkleeft
H.S.
Wilkinson
K.D.
Ploegh
H.L.
A novel active site-directed probe specific for deubiquitylating enzymes reveals proteasome association of USP14
EMBO J.
 
2001
20
5187
5196
24
Koulich
E.
Li
X.
DeMartino
G.N.
Relative structural and functional roles of multiple deubiquitylating proteins associated with mammalian 26S proteasome
Mol. Biol. Cell
 
2008
19
1072
1082
25
Anderson
C.
Crimmins
S.
Wilson
J.A.
Korbel
G.A.
Ploegh
H.L.
Wilson
S.M.
Loss of Usp14 results in reduced levels of ubiquitin in ataxia mice
J. Neurochem.
 
2005
95
724
731
26
Lee
M.J.
Lee
B.H.
Hanna
J.
King
R.W.
Finley
D.
Trimming of ubiquitin chains by proteasome-associated deubiquitinating enzymes
Mol. Cell. Proteomics
 
2011
10
R110.003871
27
D'Amato
C.J.
Hicks
S.P.
Neuropathologic alterations in the ataxia (paralytic) mouse
Arch. Pathol.
 
1965
80
604
261
28
Wilson
S.M.
Bhattacharyya
B.
Rachel
R.A.
Coppola
V.
Tessarollo
L.
Householder
D.B.
Fletcher
C.F.
Miller
R.J.
Copeland
N.G.
Jenkins
N.A.
Synaptic defects in ataxia mice result from a mutation in Usp14, encoding a ubiquitin-specific protease
Nat. Genet.
 
2002
32
420
425
29
Reijonen
S.
Kukkonen
J.P.
Hyrskyluoto
A.
Kivinen
J.
Kairisalo
M.
Takei
N.
Lindholm
D.
Korhonen
L.
Downregulation of NF-kappaB signaling by mutant huntingtin proteins induces oxidative stress and cell death
Cell Mol. Life Sci.
 
2010
67
1929
1941
30
Hyrskyluoto
A.
Pulli
I.
Törnqvist
K.
Ho
T.H.
Korhonen
L.
Lindholm
D.
Sigma-1 receptor agonist PRE084 is protective against mutant huntingtin-induced cell degeneration: involvement of calpastatin and the NF-κB pathway
Cell Death Dis.
 
2013
4
e646
31
Filimonenko
M.
Isakson
P.
Finley
K.D.
Anderson
M.
Jeong
H.
Melia
T.J.
Bartlett
B.J.
Myers
K.M.
Birkeland
H.C.
Lamark
T.
et al.  
The selective macroautophagic degradation of aggregated proteins requires the PI3P-binding protein Alfy
Mol. Cell
 
2010
38
265
279
32
Gray
M.
Shirasaki
D.I.
Cepeda
C.
Andre
V.M.
Wilburn
B.
Lu
X.H.
Tao
J.
Yamazaki
I.
Li
S.H.
Sun
Y.E.
et al.  
Full-length human mutant huntingtin with a stable polyglutamine repeat can elicit progressive and selective neuropathogenesis in BACHD mice
J. Neurosci.
 
2008
28
6182
6195
33
Hult Lundh
S.
Nilsson
N.
Soylu
R.
Kirik
D.
Petersén
Å.
Hypothalamic expression of mutant huntingtin contributes to the development of depressive-like behavior in the BAC transgenic mouse model of Huntington's disease
Hum. Mol. Genet.
 
2013
22
3485
3497
34
Pouladi
M.A.
Stanek
L.M.
Xie
Y.
Franciosi
S.
Southwell
A.L.
Deng
Y.
Butland
S.
Zhang
W.
Cheng
S.H.
Shihabuddin
L.S.
Hayden
M.R.
Marked differences in neurochemistry and aggregates despite similar behavioural and neuropathological features of Huntington disease in the full-length BACHD and YAC128 mice
Hum. Mol. Genet.
 
2012
21
2219
2232
35
Abada
Y.S.
Nguyen
H.P.
Schreiber
R.
Ellenbroek
B.
Assessment of motor function, sensory motor gating and recognition memory in a novel BACHD transgenic rat model for Huntington disease
PLoS One.
 
2013
8
e68584
36
Ravikumar
B.
Vacher
C.
Berger
Z.
Davies
J.E.
Luo
S.
Oroz
L.G.
Scaravilli
F.
Easton
D.F.
Duden
R.
O'Kane
C.J.
et al.  
Inhibition of mTOR induces autophagy and reduces toxicity of polyglutamine expansions in fly and mouse models of Huntington disease
Nat. Genet.
 
2004
36
585
595
37
Yamamoto
A.
Cremona
M.L.
Rothman
J.E.
Autophagy-mediated clearance of huntingtin aggregates triggered by the insulin-signaling pathway
J. Cell Biol.
 
2006
172
719
731
38
Hyrskyluoto
A.
Reijonen
S.
Kivinen
J.
Lindholm
D.
Korhonen
L.
GADD34 mediates cytoprotective autophagy in mutant huntingtin expressing cells via the mTOR pathway
Exp. Cell Res.
 
2012
318
33
42
39
Holmberg
C.I.
Staniszewski
K.E.
Mensah
K.N.
Matouschek
A.
Morimoto
R.I.
Inefficient degradation of truncated polyglutamine proteins by the proteasome
EMBO J.
 
2004
23
4307
4318
40
Crimmins
S.
Jin
Y.
Wheeler
C.
Huffman
A.K.
Chapman
C.
Dobrunz
L.E.
Levey
A.
Roth
K.A.
Wilson
J.A.
Wilson
S.M.
Transgenic rescue of ataxia mice with neuronal-specific expression of ubiquitin-specific protease 14
J. Neurosci.
 
2006
26
11423
11431
41
Wang
Q.
Li
L.
Ye
Y.
Regulation of retrotranslocation by p97-associated deubiquitinating enzyme ataxin-3
J. Cell Biol.
 
2006
174
963
971
42
Korhonen
L.
Belluardo
N.
Lindholm
D.
Regulation of X chromosome-linked inhibitor of apoptosis protein in kainic acid-induced neuronal death in the rat hippocampus
Mol. Cell Neurosci.
 
2001
17
364
372
43
Do
H.T.
Tselykh
T.V.
Mäkelä
J.
Ho
T.H.
Olkkonen
V.M.
Bornhauser
B.C.
Korhonen
L.
Zelcer
N.
Lindholm
D.
Fibroblast growth factor-21 (FGF21) regulates low-density lipoprotein receptor (LDLR) levels in cells via the E3-ubiquitin ligase Mylip/Idol and the Canopy2 (Cnpy2)/Mylip-interacting saposin-like protein (Msap)
J. Biol. Chem.
 
2012
287
12602
12611

Author notes

D.L. and L.K contributed equally to the work.