| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 280, Issue 32, 29096-29106, August 12, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Found in Spinocerebellar Ataxia Type 14 Is Susceptible to Aggregation and Causes Cell Death*
||**
From the
Department of Molecular and Pharmacological Neuroscience, the ||Department of Ophthalmology and Visual Sciences, the **Department of Neurosurgery, and the Department of Clinical Neuroscience and Therapeutics, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima 734-8551, Japan and the Laboratory of Molecular Pharmacology and ¶Biosignal Research Center, Kobe University, Kobe 657-8501, Japan
Received for publication, February 15, 2005 , and in revised form, June 15, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
(PKC) gene have been found in different SCA14 families. To elucidate how the mutant PKC causes SCA14, we examined the molecular properties of seven mutant (H101Y, G118D, S119P, S119F, Q127R, G128D, and F643L) PKCs fused with green fluorescent protein (PKC-GFP). Wild-type PKC-GFP was expressed ubiquitously in the cytoplasm of CHO cells, whereas mutant PKC-GFP tended to aggregate in the cytoplasm. The insolubility of mutant PKC-GFP to Triton X-100 was increased and correlated with the extent of aggregation. PKC-GFP in the Triton-insoluble fraction was rarely phosphorylated at Thr514, whereas PKC-GFP in the Triton-soluble fraction was phosphorylated. Furthermore, the stimulation of the P2Y receptor triggered the rapid aggregation of mutant PKC-GFP within 10 min after transient translocation to the plasma membrane. Overexpression of the mutant PKC-GFP caused cell death that was more prominent than wild type. The cytotoxicity was exacerbated in parallel with the expression level of the mutant. These results indicate that SCA14 mutations make PKC form cytoplasmic aggregates, suggesting the involvement of this property in the etiology of SCA14. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Recently, six different missense mutations in protein kinase C (PKC) gene (PRKCG) have been found in SCA14 families (6–9). Five mutations are located in exon 4, encoding the C1B region in the regulatory domain of PKC, and one mutation is in exon 18, encoding the C terminus of the catalytic domain of PKC (Fig. 1). Furthermore, we found a novel mutation in a Japanese SCA14 family (Fig. 1, bold lined box).2 Because mutations associated with SCA14 affect highly conserved amino acids among the PKC family members, it is possible that these mutations disturb the fundamental function or conformation of PKC. However, how these mutations cause cerebellar degeneration remains controversial.
PKC is a family of serine/threonine kinases which plays important roles in signal transduction and the regulation of various cellular functions. Among PKC subtypes, PKC is specifically present in the central nervous system and is especially abundant in cerebellar Purkinje cells and hippocampal pyramidal cells (10). Therefore, PKC is thought to be involved in various neuronal functions including synaptic plasticity and memory via modulating long term potentiation and long term depression (11). PKC knock-out mice showed mildly impaired motor coordination and incomplete developmental elimination of synapses between Purkinje cell and climbing fibers (12, 13). Furthermore, in model mice of SCA1 overexpressing mutant ataxin-1 with elongated polyglutamine, PKC was down-regulated and abnormally localized to the cytoplasmic vacuoles in Purkinje cells (14). These findings suggest that PKC may be involved in SCA.
Previous live imaging studies using green fluorescent protein (GFP)-tagged PKC (PKC-GFP) demonstrated that PKCs are translocated to several cellular organelles in an isoform- and stimulation-specific manner when PKCs are activated by different stimulations. Thereafter, PKCs recognize and phosphorylate their substrates at the targeted subcellular regions and cause the subsequent cellular responses (PKC targeting). This PKC targeting is considered to be the molecular basis underlining the multiplicity of PKC-mediated functions. Using transgenic mice overexpressing PKC-GFP, we have recently reported that the translocation of PKC-GFP, which was induced by the electrical stimulation of parallel fibers, propagated along the dendritic shaft of the cerebellar Purkinje cells (15), indicating that PKC targeting is prerequisite for various PKC-involved neuronal functions in Purkinje cells.
PKC is a member of the classical PKCs (cPKCs), which are activated by diacylglycerol (DG) and Ca2+ in the presence of phosphatidylserine (16). PKC has C1 and C2 domains, which bind DG and Ca2+, respectively (17), in its regulatory domain (Fig. 1). The C1 domain of PKC is subdivided into two cysteine-rich repeats (C1A and C1B), both of which bind with DG and phorbol ester with high affinity (18, 19). The C1 and C2 domains have crucial roles in PKC targeting through binding to these PKC activators (20). As described above, 6 of 7 missense mutations found in SCA14 families are located in the C1B domain of PKC (Fig. 1). Therefore, it is possible that these missense mutations influence the targeting of PKC. In the present study, to elucidate how mutant PKCs induced the neuronal degeneration and the pathology of SCA14, we focused on PKC targeting. We expressed mutant PKC-GFP in culture cells and compared its localization and receptor-mediated translocation with those of wild type.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
PKC (Thr514), anti-phospho-PKC(Thr655) and anti-phospho-PKC(Thr674) polyclonal antibodies were from BIOSOURCE International (Camarillo, CA). Horseradish peroxidase-conjugated goat anti-rabbit IgG antibody was from Jackson ImmunoResearch Laboratories (West Grove, PA). Anti-PKC polyclonal antibody was from Santa Cruz Biotechnology (Santa Cruz, CA).
Plasmid Construction—Human PKC cDNA was cloned from a human cDNA library by PCR and subcloned into pBluescript II KS(+) vector (Stratagene, La Jolla, CA). Mutant human PKC cDNAs were constructed by using QuickChange multisite-directed mutagenesis kit (Stratagene). To construct the plasmids encoding wild-type or mutant PKC-GFP, PKC and GFP cDNAs were together subcloned into the expression vector, pcDNA3 (Invitrogen). The GFP cDNA followed the PKC cDNA so that GFP protein was fused with the C terminus of PKC. All wild-type and mutant PKC cDNAs were verified by sequencing.
Cell Culture—The CHO-K1 cell strain was a gift from Dr. Nishijima (National Institute of Health, Tokyo, Japan). CHO cells were cultured in Ham's F12 medium (Sigma), supplemented with 10% fetal bovine serum, 100 units/ml of penicillin, and 100 µg/ml of streptomycin in a humidified atmosphere containing 5% CO2 at 37 °C.
Immunoblotting—Plasmids (5 µg) were transfected into CHO cells (2 x 105 cells) by lipofection using the FugeneTM6 transfection reagent (Roche Applied Science) according to the manufacturer's directions. Transfected CHO cells were spread onto 6-cm diameter dishes and cultured for 2 days. Cells were harvested by 500 x g centrifugation, followed by washing with 1 ml of homogenate buffer (250 mM sucrose, 10 mM EGTA, 2 mM EDTA, and 50 mM Tris-HCl, pH 7.4). For preparing total cell fractions, cells were resuspended in 100 µl of RIPA buffer (1% Nonidet P40, 0.1% sodium deoxycholate, 0.1% SDS, 150 mM NaCl, 1 mM EDTA, 20 µg/ml of leupeptin, 1 mM phenylmethylsulfonyl fluoride, 1 mM sodium orthovanadate, 1 mM NaF, and 100 nM calyculin A, and 10 mM Tris-HCl, pH 7.4) and sonicated (UR-20P, TOMY SEIKO, Tokyo, Japan) (output, 4; duty, 50%) for 15 times at 4 °C. For immunoblotting, the same amounts (10–20 µg) of samples were subjected to 7.5% SDS-PAGE, and the separated proteins were electrophoretically transferred onto polyvinylidine difluoride (PVDF) filters (Millipore, Bedford, MA). Nonspecific binding sites on PVDF filters were blocked by incubation with 5% skim milk in PBS-T (0.01 M phosphate-buffered saline containing 0.03% Triton X-100) for >1 h at room temperature. After washing with PBS-T, the PVDF filters were incubated with anti-GFP polyclonal antibody (diluted 1:2000) or anti-phospho-PKC (Thr514) polyclonal antibody (diluted 1:1000) for >1 h at room temperature. After further washing, the filters were incubated with horseradish peroxidase-conjugated anti-rabbit IgG antibody (diluted 1:10,000) for >30 min at room temperature. After three more washes, the immunoreactive bands were visualized with a chemiluminescence detection kit (ECLTM Western blotting detection reagents, Amersham Biosciences) The band densities were quantified with Fluor-S MultiImager (Bio-Rad).
For preparing Triton-soluble (S) and -insoluble (I) fractions, cells were suspended in lysis buffer (homogenate buffer containing 1% Triton X-100, 20 µg/ml of leupeptin, 1 mM phenylmethylsulfonyl fluoride, 1 mM sodium orthovanadate, 1 mM NaF, and 100 nM calyculin A) and sonicated. Samples were centrifuged at 15,000 x g for 15 min at 4 °C, and the supernatants were collected as the S fraction. The pellets were resuspended with 50 µl of RIPA buffer, sonicated, and used as the I fraction. One-twentieth volume of each fraction was subjected to 7.5% SDS-PAGE and immunoblotted by the same method as described above.
Observation of PKC-GFP Localization—CHO cells (1x105 cells) were spread onto poly-D-lysine-coated glass bottom culture dishes (Mat-Tek Corp., Ashland, MA) and were transfected with 2.5 µg of plasmid by lipofection. Transfected cells were cultured for 2 days until the observation. After the culture medium was replaced with 1 ml of HEPES buffer (135 mM NaCl, 5.4 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, 5 mM HEPES, and 10 mM glucose, pH 7.3), the fluorescence of GFP was monitored with a confocal laser scanning fluorescent microscope (LSM510META, Carl Zeiss, Esslingen, Germany) at 488-nm argon laser excitation using a 505–530-nm band pass barrier filter.
To analyze the expression level of wild-type or mutant PKC-GFP in individual cells, fluorescence images of randomly selected CHO cells expressing PKC-GFPs were obtained. For this purpose, parameters of confocal laser scanning fluorescent microscope (e.g. pinhole, laser intensity, and sensitivity of fluorescence) were adjusted to the same level. The fluorescence intensity and the area of the whole cell were measured using LSM510META software. The fluorescence intensity per area (FI/A) was used as an index for estimating the expression level of PKC-GFPs in each cell.
Fluorescent Recovery after Photobleaching (FRAP) Analysis—Circular regions in the cytoplasm of CHO cells expressing mutant PKC-GFP were photobleached by scanning for 15 s with an argon laser of the highest power. Before and after photobleaching, the bleached cells were monitored for 30 min.
Observation of PKC-GFP Translocation—Wild-type or mutant PKC-GFP-transfected CHO cells were cultured in glass bottom dishes for 2 days until observation. After the culture medium was replaced with 0.9 ml of HEPES buffer, the GFP fluorescence was monitored with a confocal laser scanning fluorescent microscope. Translocations of GFP-fused proteins were triggered by a direct application of 0.1 ml of ATP solution at 10x higher concentration into HEPES buffer to obtain the appropriate final concentration. Images were recorded every 5 s for 5–10 min before and after the stimulation. All experiments were performed at room temperature.
Immunostaining and Staining with Golgi Complex Marker—Two days after transfection, CHO cells were fixed with 4% paraformaldehyde and 0.2% picric acid in 0.1 M phosphate buffer, pH 7.4, for more than 30 min. After washing twice with PBS-T, the cells were treated with PBS containing 0.3% Triton X-100 and 5% normal goat serum (NGS) for 5 min at room temperature. For immunostaining, the cells were then incubated with the anti-PKC polyclonal antibody (1:1000) and 5% NGS in PBS-T for 1 h at room temperature. After three times washing with PBS-T, the cells were incubated with Alexa546-conjugated goat anti-rabbit IgG antibody (1:500) and 5% NGS in PBS-T for 1 h at room temperature, followed by three washes with PBS-T. For staining with Golgi complex marker, the cells were incubated with 1 µg/ml Alexa633-conjugated WGA and 5% NGS in PBS-T for 40 min at room temperature, followed by three washes with PBS-T. The fluorescence of Alexa546 and Alexa633 was observed with a confocal scanning fluorescent microscope at 543-nm and 633-nm HeNe laser excitation using a 560-nm and 650 nm-long pass barrier filter, respectively.
Evaluating and Counting Cells with Aggregation—CHO cells transfected with PKC-GFP were cultured for 2 days and fixed as described above. After two washes with PBS, cells were observed using fluorescent microscopy. We classified cells expressing PKC-GFP into three types: cells without aggregation, with massive aggregations, and with dot-like aggregations (Fig. 2, A–C). We evaluated the cell type and counted the number of each cell type in 50–60 GFP-positive cells
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
PKC-GFPs Tended to Aggregate in CHO Cells—Six different missense mutations (5 around the C1 domain and 1 in the catalytic domain) have been reported in the PKC gene in different SCA14 families (Fig. 1). Recently, we found a novel mutation around the C1 domain (Ser119 replaced with Phe) of PKC in a Japanese SCA family (Fig, 1, bold lined box).2 To clarify how these mutations affect the molecular properties of PKC and trigger neurodegeneration, we introduced these missense mutations into PKC-GFP and investigated molecular properties of 7 mutant PKC-GFPs (H101Y, G118D, S119P, S119F, Q127R, G128D, and F643L) expressed in CHO cells. Most of the CHO cells expressing wild-type PKC-GFP had a ubiquitous GFP fluorescence in the cytoplasm, but not in the nucleus (Fig. 2A). On the other hand, cells expressing mutant PKC-GFP frequently had aggregated GFP fluorescence in the cytoplasm (Fig. 2, B and C). Precise observation revealed that there were two patterns of mutant PKC-GFP aggregation, massive aggregation and dot-like aggregation as described in Fig. 2, B and C, respectively. Massive aggregation was seen in the vicinity of the nucleus (Fig. 2B), whereas dot-like ones were seen in cytoplasm (Fig. 2C). Therefore, we classified aggregation patterns into these two types. We evaluated the extent of PKC-GFP aggregations in 50–60 GFP-positive CHO cells expressing wild-type or mutant PKC-GFP. Results are shown in Fig. 2D. Although wild-type PKC-GFP aggregations were observed in 20.9 ± 2.7% of expressed cells, all seven mutants were preferably aggregated, and the percentage of cells having aggregation was over 30%. In six mutants (G118D, S119P, Q119F, Q127R, G128D, and F643L), the percentages were significantly greater than that in wild-type PKC-GFP. In cells expressing H101Y, S119F, and F643L mutant PKC-GFPs, massive aggregations were more often observed than in cells expressing wild type, whereas G118D, S119P, Q127R, G128D, and F643L mutants more frequently formed dot-like aggregations than wild type (Table I). Similar massive and dot-like aggregations were observed when PKC-GFPs were expressed in other cell lines such as COS-7 and SH-SY5Y cells (data not shown).
|
PKC-GFPs depends on their excessive expression, but not on the properties of mutated proteins, we investigated the relationship between the expression level and the aggregate formation of PKC-GFP. In CHO cells expressing wild type and S119P and G128D PKC-GFPs, fluorescent images of randomly selected 30–40 cells were obtained using the same parameters of confocal laser scanning microscope. The fluorescence intensity per area (FI/A), which represents the mean of the fluorescence intensity throughout the cell, was used as an index exhibiting the expression level of PKC-GFPs in each cell. In Fig. 3A, histograms show the distribution of cells with or without aggregation classified by PKC-GFP expression level. In wild-type PKC-GFP, all cells with FI/A <100 did not have any aggregation (0/23 cells). Aggregations were observed only in cells with FI/A >100 (7/12 cells). In contrast, in S119P and G128D mutant PKC-GFPs, aggregations were frequently observed in cells whose FI/A were 20–100 (22/25 cells in S119P and 19/26 cells in G128D). Fig. 3B shows the representative images of cells expressing wild-type and mutant (S119P and G128D) PKC-GFPs with FI/A 40–60, indicating that the expression levels of PKC-GFPs were similar among these three cell types. Although wild-type PKC-GFP was uniformly expressed in the cytoplasm, mutant PKC-GFPs formed dot-like aggregation in the cytoplasm. These results indicate that the susceptibility to aggregate formation of mutant PKC-GFPs is determined by their properties, but not by their expression level.
|
PKC-GFP were colocalized with cellular organelles. The massive aggregation of wild-type PKC-GFP was strongly colocalized with the wheat germ agglutinin (WGA), a Golgi complex marker (Fig. 4A). However, the massive aggregation of mutant PKC-GFP was partially colocalized with WGA (Fig. 4B). This result indicates that massive aggregation of mutant PKC-GFP was qualitatively different from that of wild type. In the case of dot-like aggregations, we could not establish colocalization of these aggregations with markers for Golgi complex, lysosome, or early or late endosomes (data not shown).
To elucidate whether PKC-GFP irreversibly formed dot-like aggregates, we performed a FRAP study. A PKC-GFP aggregation was photobleached with an argon laser at 488 nm, followed by observing fluorescence recovery. As shown in Fig. 5, an application of photobleaching into a circular area around dot-like aggregations (a and b) abolished the fluorescence of PKC-GFP aggregation, and the GFP fluorescence was not recovered, at least within 30 min. In contrast, the fluorescence of non-aggregated PKC-GFP (Fig, 5c) was recovered to a level similar to that in the unbleached cytoplasm within 1 min after photobleaching (Fig. 5d). This result indicates that mutant PKC-GFP of dot-like aggregation tightly associates each other and that the aggregates were not exchangeable with free PKC-GFP in the cytoplasm.
To exclude the possibility that the addition of GFP to the PKC was critical for aggregate formation, we expressed mutant PKC alone in CHO cells. We attempted to immunostain mutant PKCs with anti-PKC antibody. To confirm whether this antibody could properly detect the aggregation of PKCs, CHO cells expressing mutant PKC-GFP were stained with this antibody. As shown in Fig. 6A, the PKC immunofluorescence detected with this antibody is consistent with the S119P PKC-GFP fluorescence, although the antibody only recognized edges, not centers, of massive aggregations of S119P PKC-GFP (upper panels, arrows). In contrast, the antibody recognized the whole of dot-like aggregations (lower panels, arrowheads). These results suggest that the anti-PKC antibody properly recognizes PKC-GFP aggregation, although it might be inaccessible to the centers of these aggregations. As shown in Fig. 6B, both massive (arrow) and dot-like (arrowheads) aggregations were also observed in CHO cells expressing S119P PKC. In contrast, wild-type PKC rarely aggregated in CHO cells (data not shown). This result indicates that the addition of GFP to PKC is not critical for the aggregate formation of mutant PKC.
Phosphorylation Level and Solubility to Triton X-100 of Mutant PKC-GFP Was Decreased—PKC has three phosphorylation sites in its kinase domain: activation loop (Thr514), turn motif (Thr655), and hydrophobic motif (Thr674) site (Fig. 1). Phosphorylation of these three sites is necessary for the full activation of PKC in response to various stimulations (21–23). We examined whether the aggregate formation affected the phosphorylation level of these three sites by immunoblotting with each phosphospecific antibody. As shown in Fig. 7 and Table I, the phosphorylation levels of mutant PKC-GFPs were significantly decreased or tended to be decreased at three phosphorylation sites, compared with those of wild type. Specifically, phosphorylation levels of three mutants (S119P, G128D, and F643L) were significantly lower at all three sites than those of wild type.
Various neurodegenerative diseases are accompanied by the formation of disease-specific inclusion bodies, for examples Lewy bodies in Parkinson disease, neurofibrillary tangles and senile plaques in Alzheimer disease, and nuclear inclusion bodies in Huntington disease (3, 24). These inclusion bodies are generated by aggregated, unfolded, or misfolded protein such as 
-synuclein, tau, amyloid 
-protein, and expanded polyglutamine, respectively. It has been reported that these unfolded or misfolded proteins became detergent-insoluble in cellular and animal models of various neurodegenerative diseases (25–27). Therefore, we examined whether mutated PKC became insoluble to the detergent. Transfected CHO cells were separated into the 1% Triton-soluble (S) and -insoluble (I) fractions, and the amounts of PKC-GFP in both fractions were quantified by immunoblotting with anti-GFP antibody. As shown in the upper panel of Fig. 8A, wild-type PKC-GFP was detected with almost equal amounts in both S and I fractions. On the other hand, mutant PKC-GFPs were mostly detected in the I fraction. Especially, in S119P, Q127R, G128D, and F643L mutants, only faint bands were detected in the S fraction and very intense bands in I fraction. The amount of PKC-GFP in the I fraction was quantified and indicated as percentage of the total fraction (Fig. 8B and Table I), which was used as an index of the insolubility to Triton X-100. Except in S119F, this value tended to increase, and significant differences were detected in S119P, Q127R, G128D, and F643L mutants compared with wild-type PKC-GFP. This result suggests that mutant PKC-GFPs became insoluble to 1% Triton X-100. The insolubilities of PKC-GFPs to Triton X-100 were positively correlated (r = 0.817) with the extent of aggregated cells (Fig. 2D). For example, both values were high in S119P, Q127R, G128D, and F643L mutants and were relatively low in H101Y and S119F mutants. This correlation would imply that the insolubility of mutant PKC-GFP to detergent was caused by its aggregated form.
|
PKC, which is essential for PKC kinase activity (28, 29), between wild-type and mutant PKC-GFP. The phosphorylation state was also compared between the S and I fractions. Phospho-Thr514-specific antibody was used for assessment of the phosphorylation level (Fig. 8A, lower panel). PKC-GFP in the S fraction was intensely phosphorylated. However, very few mutant PKC-GFPs were phosphorylated in the I fraction although sufficient levels of mutant PKC-GFP existed (Fig. 8A, upper panel). The relative phosphorylation level of Thr514 per PKC-GFP amount was shown in Fig. 8C. As for the S fraction, the phosphorylation level did not significantly differ between the wild-type and mutant PKC-GFPs, although the phosphorylation level of S119F tended to be increased, compared with wild type. As for the I fraction, the phosphorylation level was obviously decreased in all wild-type and mutant PKC-GFPs. These results suggest that mutant PKC-GFPs in the S fraction were normally phosphorylated at Thr514 to the same extent as the wild type. It is possible that PKC-GFPs in the I fraction were mostly unphosphorylated at The514 and inactive. As shown in Fig. 7, the phosphorylation levels of mutant PKC-GFPs in RIPA-soluble total fraction were mostly decreased. This may be manifested because Triton-insoluble aggregated mutants were unphosphorylated.
|
|
PKC-GFP Aggregated after Receptor-mediated Transient Translocation—To examine whether receptor-mediated translocation of mutant PKC-GFPs differed from that of wild type, we observed the translocation of these PKC-GFPs triggered by the stimulation of P2Y receptors, which are endogenously expressed in CHO cells (30). As reported previously (31), the application of ATP (1 mM) induced a rapid translocation of PKC-GFP from the cytoplasm to the plasma membrane within 10 s after the stimulation, followed by a re-translocation from the membrane to cytoplasm around 2 min (Fig. 9A). Like-wise, all mutant PKC-GFPs were transiently translocated from cytoplasm to the plasma membrane in CHO cells without any aggregations (Fig. 9). The translocation of mutant PKC-GFP occurred within 10 s after the stimulation, and there were no differences in the retention period at the plasma membrane between wild-type and mutants (Table I).
|
PKC-GFP formed punctuate aggregates in the cytoplasm after the reversion from plasma membrane to cytoplasm (Fig. 9C and Supplemental Video). In these cells, mutant PKC-GFP behaved in a similar manner to wild type until 2 min after the stimulation. Thereafter, PKC-GFP started to aggregate in the cytoplasm and almost all PKC-GFP in the cytoplasm aggregated within 10 min (Fig. 9C and Supplemental Video). These phenomena were seen in 45, 67, and 78% of cells expressing G118D, Q127R, and G128D, respectively, to which translocation studies were applied (Table I). These results suggest that the aggregate formation of mutant PKC-GFP was frequently triggered by its receptor-mediated translocation.
As is shown in Fig. 2, in some cells, aggregation had already observed before the stimulation. In these cells, diffusely expressed cytoplasmic PKC-GFP was translocated, but aggregated PKC-GFP was not (data not shown).
Mutant PKC-GFP Induced Cell Death in Parallel with Its Expression Level—Overexpression of etiological gene products has been reported to cause cell death in cellular and animal models of various neurodegenerative diseases (5, 32, 33). We investigated whether these mutant PKC-GFPs caused cell death. Three days after transfection of wild-type or mutant PKC-GFPs, CHO cells were stained with 7-AAD, a fluorescent DNA dye that selectively enters dead cells (34). Stained cells were applied to flow cytometry and were classified by two parameters, GFP and 7-AAD fluorescence. Representative results using wild type and S119P are shown in Fig. 10, A and B, respectively. The GFP and 7-AAD fluorescence of each cell was positioned as a dot in the graph. Cells in the right lower part (GFP(+)/7-AAD(-)) represented viable cells expressing PKC-GFPs, and cells in the right upper part (GFP(+)/7-AAD(+)) represented dead cells expressing PKC-GFPs. The GFP-positive cells expressing S119P PKC-GFP were better stained with 7-AAD than cells expressing wild-type PKC-GFP (Fig. 10, A and B). The percentages of dead cells in total GFP-positive cells were shown in Fig. 10C. In CHO cells expressing S119P and Q127R mutant PKC-GFPs, the percentages of dead cells were significantly higher (31.5 ± 2.5 and 25.5 ± 1.8%, respectively) than that in cells expressing wild type (20.0 ± 0.8%). To evaluate whether the expression level of PKC-GFP affects the cell viability, GFP-positive cells were divided into two groups: groups with low and high GFP fluorescence (low and high GFP group, Fig. 10B), and the percentages of dead cells were calculated in each group. In the low GFP group, there were no significant differences in cell viabilities between wild-type and mutant PKC-GFP-expressing cells (Fig. 10D). However, in the high GFP group, cytotoxic effects of mutant PKC-GFPs were more prominently manifested than in total GFP-positive cells (Fig. 10E). G118D, S119P, S119F, and G128D significantly exacerbated cell death, compared with wild type. In addition, H101Y and Q127R mutants tended to cause cell death more prominently than wild type. These results suggest that mutant PKC-GFP induced cell death in parallel with its expression level.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
-sheet structures and insoluble aggregates in the neuronal cells (3, 4). Similar insoluble aggregations of mutant or misfolded proteins are frequently observed in various inherited neurodegenerative disorders including Alzheimer disease and Parkinson disease (3, 24, 35). The gain of toxic function derived from aggregated mutant proteins is considered to be the etiology of these late-onset neurodegenerative disorders. In the present study, we demonstrated that missense mutations of the PKC gene, found in SCA14, induced the formation of insoluble PKC aggregates and cell death in CHO cells.
Two types of aggregation, massive and dot-like aggregations were frequently observed in CHO cells expressing mutant PKC-GFP (Fig. 2). Dot-like aggregations, frequently shown in CHO cells expressing S119P, Q127R, and G128D mutants, might be caused by the accumulation of PKC-GFP to cellular organelles such as lysosome or endosomes. However, we could not clearly find the colocalization of this aggregation with any organelles (data not shown). Moreover, FRAP analysis confirmed that PKC-GFP in these aggregations was tightly associated and not exchangeable with free PKC-GFP in the cytoplasm (Fig. 5). This result is consistent with previous reports that the GFP fluorescence of polyglutamine protein aggregates is not recovered after photobleaching (36, 37). These findings indicate that dot-like accumulations of mutant PKC-GFP are indeed aggregates, but are not the simple result of targeting to particular cell organelles.
|
|
PKC-GFPs was different from that of wild type. The long term time-lapse imaging of mutant PKC-GFP aggregation is necessary to elucidate how massive aggregations are formed. In seven missense mutations examined in the present study, six mutations were located around the C1B domain (Fig. 1), which is involved in the binding to various lipid messengers like DG (17–19). These mutations might affect the lipid binding, resulting in the aggregate formation.
Both wild-type and mutant PKC-GFPs were transiently translocated from cytoplasm to plasma membrane in CHO cells by P2Y receptor stimulation with ATP (Fig. 9). No significant differences were found in parameters of PKC translocation, such as translocation or re-translocation speed and PKC-retaining periods at plasma membrane, between wild type and mutants (Table I). It suggests that the six missense mutations around C1B domain did not robustly affect the properties of receptor-mediated PKC translocation. This is inconsistent with results reported by Verbeek et al. (38), showing that calcium ionophore-induced translocation was hastened in two mutant PKC-GFPs found in SCA14 families; one is G118D and the other is C150F, a newly found missense mutation. This discrepancy may be explained by the difference in stimulation and cell types; however, it is still controversial whether mutations affect the PKC regulation mechanism and subsequently alter its translocation.
Although the translocation process of mutant PKC is likely similar to that of wild type, three mutant (G118D, Q127R, and G128D) PKC-GFPs rapidly and irreversibly aggregated in the cytoplasm after the receptor-mediated translocation (Fig. 9C and Supplemental Video). The aggregation pattern resembled the dot-like aggregation (Fig. 2C) although we did not confirm whether these two types of aggregation were identical. When the stimulation of Gq-protein-coupled receptors translocate PKC to the plasma membrane, PKC interacts with its activator, diacylglycerol and Ca2+, which are elevated by the stimulation It is well accepted that this interaction triggers the conformational change of the PKC molecule and allows PKC to be active (21, 39). This conformational change of mutant PKC-GFP might cause the rapid aggregation of PKC-GFP. In unstimulated CHO cells, the endogenous receptor activation might trigger the translocation, activation, conformational change, and aggregation of mutant PKC-GFP during proliferation of CHO cells. The dot-like aggregation might be formed by repetitive endogenous receptor activation.
|
PKC-GFP seemed to be inactive because it was marginally phosphorylated at Thr514 in the activation loop (Fig. 8, C and D), which is phosphorylated by PDK1 and is most essential for kinase activity in the three main phosphorylation sites of PKC (Thr514, Thr655, and Thr674) (21–23, 28, 29, 40, 41) (Fig. 1). Because the insolubility of PKC-GFPs to Triton X-100 were positively correlated (r = 0.817) with the extent of its aggregation, this result would imply that aggregated PKC-GFP was unphosphorylated and inactive. This raises the possibility that the reduction of PKC function, caused by the aggregate formation, is the etiology of SCA14. Indeed, PKC knock-out mice showed mildly impaired motor coordination and mild deficits in spatial and contextual learning (12, 42), similar pathologies observed in SCA14. However, these symptoms seemed to be more severe in SCA14 patients than in knock-out mice although mutations found in SCA 14 are heterozygous, leading to the idea that the gain of toxic function, not the loss of PKC function derived from mutant PKC, would trigger various cerebellar symptoms.
In the recent study reported by Verbeek et al. (38) the increase of basal kinase activity was observed in G118D and C150F mutant PKC-GFPs. They also demonstrated that these mutant PKC-GFPs had similar phosphorylation levels at three sites to wild-type PKC-GFP. These findings appear to conflict with our findings. This discrepancy is interpreted by the difference in the timing of observation. In the present study, our experiments were conducted 2 or 3 days after transfection, whereas Verbeek et al. observed the phosphorylation and kinase activity 24 h after transfection. We confirmed that lower aggregation and lower insolubility to Triton X-100 of mutant PKC-GFPs were observed 1 day after transfection in our experiments (data not shown). Furthermore, our preliminary study demonstrated that the basal kinase activities were not increased in all of the SCA14 PKC mutants, although we confirmed the increased basal activity of G118D.2 This result suggests that the elevated basal activity found in G118D could not simply account for the whole aspect of SCA14 pathogenesis.
|
PKC-GFPs found in several SCA14 families (Figs. 2 and 7). To explore whether these variations are correlated with the clinical features of each SCA14 family, we investigated the relationship between the Triton solubility of mutant PKC-GFPs and average age at onset of ataxia in five SCA families (H101Y, G118D, S119F, Q127R, and F643L) in which more than five persons were reported to be affected. As shown in Fig. 11, the average onset age of each family is negatively correlated with the insolubility of the corresponding PKC mutant (r = -0.8115). This strongly suggests that insoluble PKC formed by the missense mutation is implicated in the pathogenesis of SCA14.
Although the formation of insoluble aggregation is thought to exert neurotoxic effects in several inherited neurodegenerative diseases (3, 24), the precise molecular mechanism is uncertain. The ubiquitin-proteasome system (UPS) is one of the major proteolytic pathways in mammalian cells that is involved in the degradation of cytosolic short-lived proteins (43). The UPS is also involved in the elimination system of inappropriate folded proteins, which prevents them from aggregating. Age-related decline of UPS function results in the increased accumulation and aggregation of misfolded proteins. Protein aggregation further exacerbates UPS dysfunction by sequestrating the 26 S proteasome complex into the aggregation (44). This is the commonly accepted mechanism of neuronal cell death in neurodegenerative diseases accompanying protein aggregation (44–46). In our preliminary experiments, increased ubiquitination was found in several mutant PKC-GFPs. Although further studies are necessary to elucidate the involvement of UPS in SCA14 pathogenesis, it is plausible that mutant PKC aggregation exerts its cytotoxic effects on neurons by disturbing the UPS.
In various neurodegenerative diseases, the disease-specific inclusion bodies, for example Lewy bodies in Parkinson disease, neurofibrillary tangles and senile plaques in Alzheimer disease, and nuclear inclusion bodies in Huntington disease (3, 24) were observed. However, the autopsy report from the SCA14 patient having the H101Y mutation revealed no aggregation of PKC in the cerebellar neurons (6). At present, we do not have adequate answers to explain why the mutant PKC does not form aggregates in the Purkinje cells of actual SCA14 patients. Further autopsy studies from specimens with mutations other than H101Y are necessary to resolve the discrepancy because the H101Y mutant had the lowest tendency to aggregate among the mutant PKC-GFP examined in the present study (Fig. 2D and Table I).
The findings in the present study provide the possibility that SCA14 is caused by a mechanism similar to other neurodegenerative diseases; that is the accumulation of aggregated protein. Although further studies are necessary to identify the precise molecular mechanism of mutant PKC to cause SCA14, the identification may lead to effective therapeutic methods not only for SCA14 but also for other neurodegenerative diseases.
|
FOOTNOTES |
|---|
The on-line version of this article (available at http://www.jbc.org) contains a Supplemental Video.
To whom correspondence should be addressed: Dept. of Molecular and Pharmacological Neuroscience, Graduate School of Biomedical Sciences, Hiroshima University, Minami-ku, 1-2-3 Kasumi, Hiroshima 734-8551, Japan. Tel.: 81-82-257-5142; Fax: 81-82-257-5144; E-mail: nsakai{at}hiroshima-u.ac.jp.
1 The abbreviations used are: SCAs, spinocerebellar ataxias; PKC, protein kinase C; CHO, Chinese hamster ovary; GFP, green fluorescent protein; RIPA, radioimmunoprecipitation assay; PVDF, polyvinylidine difluoride; NGS, normal goat serum; 7-AAD, 7-amino-actinomycin D; UPS, ubiquitin-proteasome system; PDK1, 3-phosphoinositide-dependent protein kinase-1; WGA, wheat germ agglutinin; FI/A, fluorescence intensity per area; FRAP, fluorescent recovery after photobleaching; WT, wild type; PBS, phosphate-buffered saline.
2 K. Hiramoto, H. Kawakami, K. Inoue, T. Seki, H. Maruyama, H. Morino, M. Matsumoto, K. Kurisu, and N. Sakai, submitted data.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
