Dentatorubral Pallidoluysian Atrophy (DRPLA) Protein Is Cleaved by Caspase-3 during Apoptosis*

Dentatorubral pallidoluysian atrophy (DRPLA) is an autosomal dominant neurodegenerative disorder. It is associated with an abnormal CAG repeat expansion resulting in formation of a protein with an elongated polyglutamine stretch. However, neither the physiological roles of the DRPLA gene product nor molecular mechanisms of its pathogenesis have yet been elucidated. Here we report that the DRPLA protein is cleaved at a site near the N terminus during apoptosis induced by VP-16, staurosporine, or glucocorticoid. Moreover, thein vitro translated DRPLA protein is cleaved by recombinant caspase-3, a member of the cysteine protease family, which is thought to be a main executioner of apoptosis. Using mutant DRPLA proteins, the cleavage site was identified as 106DSLDG110. The cleavage, however, was not modulated by the length of the polyglutamine stretch. These findings suggest that the DRPLA protein is one of the physiological substrates of caspase-3, and its cleavage may result in structural and biochemical alterations associated with apoptosis.

Dentatorubral pallidoluysian atrophy (DRPLA) is an autosomal dominant neurodegenerative disorder. It is associated with an abnormal CAG repeat expansion resulting in formation of a protein with an elongated polyglutamine stretch. However, neither the physiological roles of the DRPLA gene product nor molecular mechanisms of its pathogenesis have yet been elucidated. Here we report that the DRPLA protein is cleaved at a site near the N terminus during apoptosis induced by VP-16, staurosporine, or glucocorticoid. Moreover, the in vitro translated DRPLA protein is cleaved by recombinant caspase-3, a member of the cysteine protease family, which is thought to be a main executioner of apoptosis. Using mutant DRPLA proteins, the cleavage site was identified as 106 DSLDG 110 . The cleavage, however, was not modulated by the length of the polyglutamine stretch. These findings suggest that the DRPLA protein is one of the physiological substrates of caspase-3, and its cleavage may result in structural and biochemical alterations associated with apoptosis.
Apoptosis is an indispensable phenomenon for normal development and maintenance of homeostasis in multicellular organisms. Although the signal transduction of apoptotic stimuli remains unclear, accumulating evidence indicates that various types of apoptotic signals ultimately converge on the activation of the same machinery. The components of this machinery are cysteine proteases belonging to a caspase family (1,2). So far, about 10 members have been identified including interleukin-1␤-converting enzyme (ICE or caspase-1) (3) and caspase-3 (CPP32/Yama/Apopain) (4 -6). Each of these enzymes is synthesized as a proenzyme, and then proteolytically activated. Like systems in blood clotting and complement activation, the protease cascade is known to exist where one protease activation can lead to processing of the same or other members of the caspase family. Poly(ADP-ribose) polymerase (PARP) 1 was identified for the first time as a substrate of the caspase family (5). Since the function of PARP is associated with genome maintenance and DNA repair, the cleavage of PARP may play an important role during apoptosis. However, it is unlikely that its cleavage is essential for apoptosis since PARP-deficient mice do not show any deregulation of apoptosis (7). Therefore, there must be other substrates that are critical for apoptosis, or more likely, the coordinated degradation of several key proteins is necessary for the execution of apoptosis. At least 10 proteins are known to be cleaved during apoptosis, including nuclear lamins, actin, a 70-kDa protein component of the U1-ribonucleoprotein (U1-70 kDa), DNA-dependent protein kinase, and DNA fragmentation factor (8 -12). To elucidate the final step of apoptosis, other candidates of the "death substrates" cleaved during apoptosis need to be identified.
Dentatorubral pallidoluysian atrophy (DRPLA) is an autosomal dominant neurodegenerative disorder characterized by progressive dementia, myoclonic epilepsy, cerebellar ataxia, and choreoathetotic movement. Pathological findings of the brain were represented as combined degeneration of dentatofugal and pallidofugal pathways. We and others previously found expansion of unstable CAG repeats of the DRPLA gene of the patients (13,14). Although the physiological functions of the DRPLA gene product or the pathogenesis of DRPLA are still unclear, it is tempting to speculate that they are related to apoptosis for a number of reasons. First, neuronal death is reported to be apoptotic in other disorders such as Huntington disease or Machado-Joseph disease, which are also caused by expansion of CAG repeats (15,16). Second, mice deficient for the Huntington disease gene show increased apoptosis and embryonic lethality (17). Here we report that the DRPLA protein is proteolytically cleaved during apoptosis by caspase-3.
Western Blotting-Immunoblot assays were carried out as described previously (20). In brief, resolved proteins (30 g) in SDS-PAGE gels were transferred to a nitrocellulose membrane (Schleicher & Schuell) by electroblotting. Immunoblotting was performed using a 1:750 dilution of the primary anti-DRPLA antibody followed by a 1:5,000 dilution of horseradish peroxidase-conjugated goat anti-rabbit IgG (Sigma) as a second antibody, and the proteins were visualized by ECL (Amersham Corp.). A Vectastain ABC kit (Vector Laboratories Inc.) and 3-amino-9-ethyl carbazole were also used for some of the developments. Rabbit anti-DRPLA antisera were raised against GST-DRPLA fusion protein (see below). Prior to immunoblotting, the antisera were affinity-purified by using GST and protein A columns. To detect PARP, a 1:2,000 dilution of anti-PARP monoclonal antibody (Biomol) and a 1:1,000 dilution of rabbit horseradish peroxidase-conjugated anti-mouse immunoglobulins (DAKO) were used for the primary and secondary antibodies, respectively. * This study was supported in part by Grants for Pediatric Research, for Cancer Research, and for Surveys and Research on Specific Diseases from the Ministry of Health and Welfare, and a Grant in Aid of Scientific Researches from the Ministry of Education, Science, and Culture, Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Analysis of Caspase-3 Activities-Cell extracts were prepared by resuspending the cell pellet in 100 l of lysis buffer (50 mM Tris, pH 7.5, 0.5% Nonidet P-40, 0.5 mM EDTA, 150 mM NaCl). 30 l of cell lysate were mixed with 100 l of reaction buffer (20 mM HEPES, pH 7.5, 0.1 M NaCl, 5 mM dithiothreitol) and 50 l of 100 mM Ac-DEVD-MCA (Peptide Institute, Inc., Japan) and incubated at 37°C for 2 h. A standard curve was prepared using solutions of 7-amino-4-methylcoumarin at various concentrations in the reaction buffer. The fluorescence of the cleaved substrates was measured using a spectrofluorometer set at an excitation wavelength of 365 nm and an emission wavelength of 450 nm. One unit corresponds to the activity that cleaves 1 nmol of Ac-DEVD-MCA at 37°C in 2 h.
Cleavage of in Vitro Translated DRPLA Protein-The in vitro translated proteins were prepared using the TNT coupled reticulocyte lysate systems (Promega) and radiolabeled with [ 35 S]methionine. Two l of in vitro translation reaction mixture were incubated in cleavage buffer (50 mM PIPES/KOH (pH 6.5), 2 mM EDTA, 0.1% (w/v) Chaps, 5 mM dithiothreitol) along with either apoptotic cell extracts (about 0.5 unit) or recombinant caspase-3 (20 l final volume). Apoptotic cell extracts were prepared from 697 cells treated with 10 Ϫ7 M of dexamethasone for 24 h (23). Recombinant caspase-3 was kindly provided by M. Enari and S. Nagata (24) (Osaka University, Japan). For some of the reactions, we added various concentrations of Ac-YVAD-CHO or Ac-DEVD-CHO (Peptide Institute, Inc., Japan) which are specific inhibitors of caspase-1-or caspase-3-like proteases, respectively. The mixture was incubated at 37°C for 1 h and subjected to 6% SDS-PAGE after boiling at 95°C for 5 min with Laemmli buffer. After electrophoresis, the gels were dried and autoradiographed. Quantitative evaluation of DRPLA cleavage was performed using a BAS2000 Bio-Imaging analyzer (Fuji film).

DRPLA Protein Is
Cleaved during Apoptosis-We first confirmed the specificity of the antibody used in this study that had been raised against GST-DRPLA fusion protein. As shown in Fig. 1A, the antibody reacted specifically with in vitro translated DRPLA proteins but not with an irrelevant protein, p53. As described previouly (25), the molecular masses of the DRPLA proteins containing 14 and 17 repeats of glutamine were 160 and 170 kDa, respectively, which were much larger than the predicted sizes based on their amino acid compositions 2 K. Nakamura, manuscript in preparation. (124 and 126 kDa, respectively). A nonspecific band was also detected at 135 kDa.
We next assessed whether the DRPLA protein is cleaved during apoptosis. NT2 cells expressing a high level of the DRPLA protein were incubated with staurosporine, a broadspectrum protein kinase inhibitor, and VP-16 (etoposide), a topoisomerase II inhibitor, both of which induce apoptosis in a variety of cells (9,26). 697 cells were treated with dexamethasone, a synthetic glucocorticoid (23). Treated cells were harvested at various intervals, and cell lysates were subjected to immunoblotting. At zero time, the DRPLA protein was in the full-length 160-kDa form (Fig. 1, B, lanes 1 and 4, and D, lane  1). After certain hours of drug treatments, we observed a new band at 145 kDa ( Fig. 1, B, lanes 2 and 6, and D, lane 4). The appearance of the cleaved PARP, a well known substrate of caspase-3, showed the same kinetics (Fig. 1, C and E), thus strongly suggesting that the DRPLA protein was cleaved during apoptosis. To address the question if the apoptotic lysate can really cleave the DRPLA protein, the in vitro translated DRPLA protein was incubated with apoptotic lysate obtained from 697 cells treated with dexamethasone. Kinetics of the caspase activity after treatment with dexamethasone was described previously (23). As shown in Fig. 2A, we observed an additional band of 145 kDa when incubated with apoptotic lysate but not with nonapoptotic 697 lysate (lanes 1 and 2). The Cleavage Is Mediated by a Caspase-3-like Protease at a Site Near N Terminus-To determine if these cleavages are mediated by a member of the caspase family, the specific inhibitors of caspase-1-and caspase-3-like proteases, Ac-YVAD-CHO and Ac-DEVD-CHO, respectively, were added to the reaction mixtures. Cleavage of the in vitro translated DRPLA protein was almost completely inhibited with as little as 1 nM of Ac-DEVD-CHO, whereas the inhibition was negligible with the same concentration of Ac-YVAD-CHO (lanes 3 and 7) (IC 50 value of less than 1 nM for the former, and about 60 nM for the latter (Fig. 2B)). These data indicate that the cleavage is mediated by a caspase-3-like protease but not by a caspase-1-like protease.
To investigate the localization of the cleavage site, a series of C-terminal deletion mutants (Fig. 3) as well as full-length cDNA were translated in vitro and incubated with apoptotic or nonapoptotic lysate. SDS-PAGE followed by autoradiography revealed two bands from each sample incubated with the apoptotic lysate, one corresponding to an uncleaved protein and the other to a cleaved one (Fig. 4). The two bands in each lane showed the same difference in apparent molecular weights of the products. Therefore, we conclude that the cleavage site is located near the N terminus but not near the C terminus.
The DRPLA Protein Translated in Vitro Is Cleaved by Recombinant Caspase-3-To address the question if the responsible protease is caspase-3 itself or another protease(s) resembling caspase-3, we used recombinant caspase-3 prepared in Escherichia coli. As shown in Fig. 5A, the pattern of cleavage observed with comparable units of recombinant caspase-3 was the same as that for the apoptotic lysate. These results show that the DRPLA protein is a novel substrate of at least caspase-3.
Cleavage Site Is 106 DSLDG 110 -To identify the exact cleavage site of the DRPLA protein, we searched for a DXXD motif, which is the consensus cleavage site for caspase-3 (27). Three such motifs were found in the DRPLA protein, two of them were located near the N terminus ( 106 DSLD 109 and 120 DPRD 123 ). Two plasmids, pNSLA and pAPRA, were generated to produce mutant DRPLA proteins in which either 106 DSLD 109 or 120 DPRD 123 was disrupted, respectively (Fig. 3). The mutant DRPLA protein NSLA was not cleaved by caspase-3, while APRA showed the same cleavage pattern as the wild type protein (Fig. 5A, lanes 7 and 8). Therefore, we conclude that the cleavage site of the DRPLA protein is 106 DSLDG 110 . A faint cleaved product of NSLA may represent an alternative cleavage at 120 DPRD 123 when 106 DSLD 109 was disrupted (lane 8).
Polyglutamine Tract Does Not Modulate Caspase-3 Cleavage-Since patients with DRPLA are associated with expansion of unstable CAG repeats of the DRPLA gene, we next investigated whether the length of the polyglutamine tract affected the protein cleavage by caspase-3. Plasmids pMY1240 and pMY1247 which had 14 or 71 CAG repeats, respectively, were used for the analysis. In vitro translated proteins with different sizes of the polyglutamine tract were similarly cleaved by varying concentrations of recombinant caspase-3, indicating that the size of the CAG repeat does not modulate DRPLA cleavage by caspase-3 (Fig. 5, A and B). DISCUSSION Activation of a group of cysteine proteases recently termed caspases is commonly observed during apoptosis triggered by a variety of stimuli (1,2). The importance of the caspase family for the execution of apoptosis is underscored by several findings. First, CED-3, a caspase homolog of Caenorhabditis elegans, is required for programmed cell death during development of the nematode (3). Second, protease inhibitors specific for caspases have the ability to inhibit apoptosis (28). Third, mice deficient for a caspase such as caspase-1 or -3 show abnormal organ development or deregulations of apoptosis induced by some but not all kinds of signals (29,30). However, how the activation of these proteases can lead to the characteristic changes of apoptosis is totally unknown. Although more than 10 proteins are known to be cleaved during apoptosis, it has not yet been demonstrated that cleavage of any one of them is essential for apoptosis. Therefore, the identification of new substrates as well as novel caspases should provide an insight into the mechanism of apoptosis.
In this report, we demonstrated that the DRPLA protein is cleaved by caspase-3 during apoptosis. It is cleaved in vivo with a very similar kinetics to PARP, one of the known substrates for caspase-3. To our knowledge, the DRPLA protein is the ninth substrate of caspase-3 demonstrated experimentally to date. The cleavage site was identified as 106 DSLDG 110 , which is  4. DRPLA protein is cleaved at a site near N terminus. In vitro translated proteins derived from a series of truncated DRPLA cDNAs depicted in Fig. 3 were incubated with either control (C) or apoptotic (A) 697 cell extracts and then subjected to SDS-PAGE followed by autoradiography. similar to the consensus sequence for the caspase-3 cleavage site, D P4 XXD P1 (27). It should be noted that the DRPLA protein was cleaved at a physiological concentration of caspase-3. Excess amounts of caspase-3 could cleave proteins with less stringent requirements for the recognition sequence (data not shown). Obviously it is still possible that other members of the growing caspase family, either known or unknown, can also cleave this protein.
Since the biological function of DRPLA is still unknown, it is elusive how important role the proteolysis of this protein plays during apoptosis. Moreover, the proposed roles for the cleavage of any substrates reported previously remain largely hypothetical. Interestingly, some substrates of caspases are nuclear repair proteins such as PARP, DNA-dependent protein kinase, or U1-70 kDa whose cleavage may result in the DNA degradation characteristic of apoptosis (7,10,11). The cleavage of these proteins separates key functional domains of the molecule. In the light of these findings, it is intriguing that a bipartite nuclear localization signal (31) at the N terminus of the DRPLA protein (Fig. 3, 16 RK KEAPGPREEL RSRGR 32 ) is removed by proteolysis during apoptosis, possibly resulting in a loss of function. Although we previously detected the DRPLA gene product in the cytoplasm using immunohistochemistry (25), it is still possible that a fraction of this protein in the nucleus may enter the cytoplasm after apoptotic cleavage. Alternatively, like DNA fragmentation factor (12), it is also likely that the cleaved DRPLA protein can induce apoptosis. In this scenario, instead of the protective effect of the uncleaved protein, the cleaved fragments would be toxic to the cells or have a dominant negative effect.
Some of the other known substrates for caspases such as nuclear lamins, actin, fodrin, the actin-associated protein, or Gas2, a component of the microfilament system (8,9,32,33) have important roles for maintaining cell structure. The DRPLA protein itself or its binding protein may be a member of such proteins. These various possibilities are currently under investigation.
Recently it was shown that huntingtin, the Huntington disease gene product, is specifically cleaved by caspase-3 (34). The mutation underlying Huntington disease is also expansion of CAG repeats. They also demonstrated that the rate of cleavage increases with the length of the polyglutamine tract, raising the possibility that such a cleavage is associated with the pathogenesis of Huntington disease. Unlike in Huntington disease, the cleavage rate of the DRPLA protein was not affected by the length of polyglutamine tract and the short N-terminal polypeptide resulting from the cleavage does not contain polyglutamine. Taken together, these results suggest that the cleavage of the DRPLA protein is not directly related to the pathogenesis of the disorder, but rather more likely associated with its physiological roles during apoptosis.
Finally, caspase-3-deficient mice were reported recently. They showed specific defects in brain development such as protrusions of brain tissue or the indentation of the retinal neuroepithelium (30). Thus, although the DRPLA protein is largely ubiquitously expressed in humans (21), taking these results into consideration, it may also have a crucial role in brain development.