Inhibition of Apoptosis by Z-VAD-fmk in SMN-depleted S2 Cells*

,

The autosomal recessive disease, spinal muscular atrophy (SMA) 1 occurs in 1:6,000 live births and is one of the most common genetic causes of infant death, resulting from death of motor neurons in anterior horn cells. SMA is clinically categorized as types I, II, and III based on age of onset and clinical progression. Type I (severe) infants have onset of weakness within the first six months of life and typically die within the first two years. Type II (intermediate) children experience symptoms before age two and never walk unassisted. Type III (mild) patients exhibit less severe clinical symptoms with a highly variable age of onset. It was found that the survival of motor neuron (SMN) gene is the SMA-determining gene and is deleted, converted, or mutated in Ͼ95% of SMA patients. There are two human SMN genes, SMN1 and SMN2 (1). However, only loss of the SMN1 gene causes the disease. SMN2 is maintained in most of the SMA patients, in many cases with multiple copies. However, 80% of SMN2 products are the alternatively spliced SMN⌬7 that lacks exon 7 and is unstable, whereas the primary product of SMN1 is full-length SMN mRNA. It has been demonstrated that clinical severity of SMA disease is inversely correlated with the number of SMN2 copies in SMA patients (2), suggesting that disease progression is inversely dosage-dependent on full-length SMN proteins produced.
Mice have only a single copy of the SMN gene that does not undergo alternative splicing of exon 7 and therefore is a functional SMN1 homologue. Homozygous knockout of the mouse SMN gene is embryonic lethal at a very early developmental stage (3). By crossing mice transgenic for human SMN2 into a SMNϩ/Ϫ heterozygous mouse, animals with an hSMN2/Ϫ genotype were born. These mice did not die in utero, implying SMN2 can rescue the lethality of homozygous knockout. A single copy of human SMN2 was sufficient to restore mouse viability from SMNϪ/Ϫ lethality; however, severe weakness developed and the mice died within few days after birth. Mice with eight copies of hSMN2, on the other hand, appear normal, providing additional evidence that SMN2 functions as a disease modifying gene by producing low levels of full-length SMN, explaining the correlation between disease severity and SMN2 copy numbers (4). SMN homologues have been also isolated from Caenorhabditis elegans (5), Danio rerio (zebrafish) (6), Schizosaccharomyces pombe (7)(8)(9), and recently from Drosophila (10). Biochemically, the SMN proteins from these species retain a number of properties identified in the human SMN, including RNA binding activity and self-association. Defects of SMN proteins in C. elegans, D. rerio, and S. pombe affect cell viability and growth.
Drosophila models have been generated by ectopic over-expression of human SMN protein domains that may function as dominant negatives to disrupt the endogenous dSMN protein, resulting in abnormal positioned wings and legs or in many cases, pupal lethality (10). However, inactivation of function of a protein using dominant negative mutants can yield misleading results if the expressed protein has additional functions or if it interacts with its endogenous counterpart (11). During the last few years, RNA interference has emerged as a promising and powerful technology to investigate the loss-of-function of a protein by specifically suppressing expression of a gene in a population of cells. Thus, RNAi was used to investigate the mechanisms of SMN-deficiency-induced cell death in a cell culture model, the insect S2 cells. Our results showed that cells with SMN expression suppressed by RNAi underwent apoptosis. Apoptosis occurred via a caspase pathway-dependent process by activation of DRONC, an apical caspase in Drosophila, which further activated its effector caspase, DRICE. Furthermore, Z-VAD-fmk (N-benzyloxycarbonyl-Val-Ala-Aspfluoromethylketone), which is a cell-permeable caspase inhibitor, efficiently blocked cell death induced by SMN deficiency. Taken together, these results strongly suggest that investigation of factors that mediate SMN and caspase-dependent apoptosis may provide clues on functions of the SMN protein and targets for identification of small molecules/drugs for SMA therapy.

EXPERIMENTAL PROCEDURES
Antibodies-To generate antibodies specific for Drosophila SMN, a plasmid construct containing full-length of dSMN was fused to glutathione-S-transferase in pGEX-4T1 (Amersham Biosciences). Fusion protein was purified according to the manufacturer's instructions. Anti-dSMN antibodies (dSMNAb) were raised against the fusion protein in rabbits and affinity purified by the GST or GST-fusion proteins that were covalently linked to glutathione Sepharose-4B resin (12). Antibodies specific for the human SMN are from Santa Cruz Biotechnology (Santa Cruz, CA) or Transduction Laboratory (Lexington, KY). Anti-HA antibody 12CA5 is from Roche Molecular Diagnostics. The monoclonal anti-tubulin antibody is from Oncogene (San Diego, CA). The polyclonal rabbit anti-DRONC, rabbit anti-DRICE, and monoclonal mouse anti-DIAP1 (Drosophila inhibitor of apoptosis protein 1) antibodies were a generous gift from Dr. Bruce A. Hay (CalTech, Pasadena, CA).
Cell Culture and Apoptosis Induction-S2 cells were grown in Schneider's Drosophila media (Invitrogen) supplemented with 10% fetal bovine serum, 20 units/ml penicillin, 20 g/ml streptomycin, and 0.434 mg/ml of GlutaMax at 27°C. For apoptotic induction as a positive control, cells were treated with 1 M actinomycin D for 7 h. For all the analyses, three independent experiments were conducted.
Plasmid Constructs-The 0.7-kilobase fragment of dSMN coding region with HA-tag was cloned into KpnI and SalI sites of a Drosophila expression vector, pRmHa-3, containing the metallothionein promoter (13) or the pcDNA3 vector. Expression of the HA-dSMN protein in pcDNA3-HA-dSMN transfected mammalian C33A cells or in the pRmHa-3-HA-dSMN transfected S2 cells that were induced by 500 M copper sulfate for 6 -8 h was detected by Western blot analysis with anti-HA antibody or anti-dSMN antibodies. Similarly, a construct for human SMN was generated in pRmHa-3. Expression of hSMN in S2 cells was detected by immunofluorescence with anti-hSMN antibodies as described below.
RNAi in S2 Cells and Z-VAD-fmk Treatment-PCR products containing coding sequences for the dSMN (forward primer: 5Ј-TAA TAC GAC TCA CTA TAG GG AAG ACG TAC GAC GAG TCG-3Ј; and reverse primer: 5Ј-TAA TAC GAC TCA CTA TAG GG GTG GTG CTG GCT TCT TTC-3Ј; product length, 601bps; bold and italics letters represent T7 promoter sequences) and control Drosophila Presenilin (dPsn) gene (forward primer: 5Ј-TAA TAC GAC TCA CTA TAG GG TG GCT GCT GTC AAT CTC-3Ј; and reverse primer: 5Ј-TAA TAC GAC TCA CTA TAG GG CGA TAG CAA CGC TTC TTG-3Ј; product length: 543bps) were obtained and gel-purified. Double-stranded RNAs (dsRNA) were generated by in vitro transcription with Ribomax T7 Transcription kit (Promega) and digested with Rnase-free DNase. The dsRNA products were ethanol precipitated and annealed by incubation at 65°C for 30 min and then slowly allowed to cool at room temperature. The annealed dsRNA products were analyzed on a 1% agaorse gel to ensure the majority of dsRNA existed as a single band. The dsRNA (2 g) and/or plasmid DNAs (2 g) were introduced into cells by using Cellfectin (Invitrogen) according to the manufacturer's instructions. Caspase inhibition was achieved by using 50 M of Z-VAD-fmk (Alexis Biochemicals, San Diego, CA) in the culture medium.
Semi-quantitative Multiplex RT-PCR-Total RNA was isolated using the Trizol reagent (Invitrogen) from cells transfected with dSMN dsR-NAs, yeast tRNA, dPsn dsRNAs, or mock-transfected. For first-strand cDNA synthesis, 0.5 g of total RNA was used. To amplify the endogenous dSMN and dPsn transcripts, a multiplex RT-PCR was performed using primers from dSMN and dPsn under the following PCR condition (94°C for 45 s, 55°C for 1 min, and 72°C for 1 min for 28 cycles and then 72°C 10 min extension). The reaction gave rise to two bands (dSMN and dPsn), each of which was used as an internal control for their relative expression levels, and they were quantified on an agarose gel by gel-documentation software (UVP, Upland, CA).
Western Blot Analysis-Cells were grown in 100 mm cell culture dishes and transfected with dsRNA of dSMN and controls. After 72 h cells they were harvested and lysed with 500 l of radioimmune precipitation assay buffer lysis buffer (50 mM Tris-HCl (pH 7.4), 1.0% Nonidet P-40, 0.25% sodium-deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM aprotinin, and 1 g/ml of phenylmethylsulfonyl fluoride, leupeptin, and pepstatin each), and the amount of protein was quantified by BCA assay. Equal amounts of protein from control and dsRNA of dSMN transfected cells were boiled for 5 min in Laemmli's buffer (14) and separated on 12% and/or 15% SDS-PAGE, and the resolved gels were blotted on Immobilon P membrane (Millipore, Bedford, MA). The transblots were probed with antibodies against dSMN, anti-DRONC, fulllength anti-DRICE, and anti-DIAP1 (15) followed by secondary horseradish peroxidase-conjugated antibodies. In all cases, blots were stripped with stripping buffer (62.5 mM Tris (pH 6.7), 2% SDS and 90 mM 2-mercaptoethanol) and reprobed with anti-tubulin for loading controls. The signal was detected by enhanced chemiluminescence (Pierce) Immunofluorescence-Cells were cultured on glass cover slips coated with poly L-lysine and fixed with 4% paraformaldehyde for 10 min, washed 3 times with phosphate-buffered saline, permeabilized with 0.1% Triton X-100 in phosphate-buffered saline for 10 min, and blocked with 3% IgG-free bovine serum albumin in phosphate-buffered saline for one hour. The cells were incubated with primary anti-SMN (1:100) or anti-HA (1:100) antibodies overnight at 4°C and appropriate fluorescein isothiocyanate and/or Cy3-conjugated secondary antibodies (1: 50) (Jackson Immunoresearch Laboratories, Inc., West Grove, PA) were applied for one hour. The localization of proteins was visualized by fluorescence microscopy.
Flow Cytometric Assay for Apoptosis-Cells were co-transfected with dsRNAs and GFP expression plasmid as described above and collected 72 h later for staining with Annexin V for flow cytometric analysis. Untreated and cells treated with 1 M actinomycin D and cells undergoing necrosis were collected in parallel. All cells were collected by centrifugation (500 ϫ g for 2 min), resuspended in binding buffer to which Annexin V-Cy5 was added according to the manufacturer's instructions (MBL International). The cells were immediately analyzed on a BD-LSR Facstar flow cytometer (BD Biosciences) using Argon and HeNe lasers and the recommended filters. The resulting data was analyzed using Flowjo software (version 4.1.1) (Tree Star, San Carlos, CA).
Cell Death Detection ELISA Assay for Apoptosis-To quantitate apoptotic cell death using a complementary assay, we performed ELISA assays as described in the manufacturer's manual using a Cell Death Detection kit (Roche). This assay is based on the quantitative sandwichenzyme-immunoassay principle using mouse monoclonal antibodies directed against DNA and histones. This ELISA provides a quantitative in vitro determination of histone-associated DNA fragments (mono-and oligonucleosomes) in the cytoplasmic fraction of cell lysates and determined photometrically at 405 nm. Cell lysates equivalent to 10,000 cells were used for ELISA analysis.

RESULTS
Characterization of Drosophila SMN Protein-To investigate expression of Drosophila SMN (dSMN) protein, two antibodies against GST-dSMN protein were raised in rabbits and then affinity-purified. The dSMN gene was sub-cloned into HAtagged pcDNA3 vector and transfected into mammalian C33A cells. Western blot analysis shows that both dSMN antibodies specifically recognized a single band in C33A cells transiently transfected with the HA epitope-tagged full-length Drosophila SMN cDNA (Fig. 1A, upper panel, lane 1), but not in cells transfected with empty vector (Fig. 1A, upper panel, lane 2). The band was also confirmed as HA-dSMN by Western blot analysis with anti-HA antibody (12CA5) (Fig. 1A, middle panel). The molecular mass of HA-dSMN is 28 kDa, 10 kDa smaller than that of the human SMN, consistent with the length of their open reading frame sequences. HA-tagged dSMN was also cloned into a Drosophila vector pRmHa-3 and transfected into Drosophila S2 cells. The HA-tagged dSMN protein expression was detected in transfected cell lysates probed with anti-HA antibody (Fig. 1B, middle panel, lane 2), and the band was undetected in control cells transfected with empty vector (Fig. 1B, middle panel, lane 1). By using anti-dSMN antibodies, higher expression of dSMN was detected in HA-dSMN transfected cells (Fig. 1B, upper panel, lane 2) than in cells transfected with empty vector (Fig. 1B, upper panel The amount of the endogenous dSMN protein was reduced to 90% in dSMN RNAi cell lysates (Fig. 2B, lane 2), and the level was unaffected in cells treated with yeast tRNA (control) (Fig.  2B, lane 1). There was no appreciable difference between the amounts of tubulin (loading control) in control and dsRNAtreated cells (Fig. 2C). Furthermore, we also analyzed the effect of RNAi after 24 and 48 h, and the level of dSMN expression was only slightly suppressed, whereas the level of dPsn expression was reduced to 90% after 24 h (data not shown), suggesting the relative stability of dSMN mRNA.
Immunolocalization of SMN in S2 Cells-Because there is no report on expression of Drosophila SMN protein in cells, we sought to determine the localization of dSMN protein in S2 cells. Constructs of both dSMN and hSMN, tagged and/or untagged in the pRmHa-3 vector were generated and co-transfected into cells. After 72 h, protein expression was induced by 500 M of copper sulfate for 6 -8 h. Immunofluorescence assay revealed co-localization of dSMN and hSMN proteins to the nucleus (Fig. 3, F-H). Staining for human SMN was also detected as discrete, punctate sub-nuclear dots in co-transfected cells, reminiscent of "gems" in the nucleus of some of S2 cells (Fig. 3L). With purified anti-dSMN antibodies, we identified the localization of the endogenous Drosophila SMN in cells, and it was diffusely distributed throughout cytoplasm (Fig.  3A). However, enriched localization was detected in the nucleus when dSMN was over expressed (Fig. 3B). Only the endogenous level of the dSMN protein expression was detected in HA-dSMN transfected but non-induced cells (Fig. 3C). The localization could not be seen in cells treated with dsRNA of dSMN (Fig. 3D), indicating specificity of anti-dSMN antibodies. Staining to cytoplasm and nucleus in some cells transfected with both dsRNA and HA-dSMN may be due to incomplete suppression of HA-dSMN over-expression or those cells failed to conceive the dsRNA (Fig. 3E).
Deficiency of dSMN in S2 Cells Induces Apoptosis-To investigate how motor neurons die in SMA patients, it is important to understand the mechanisms of cell death induced by SMN deficiency. Accordingly, we used RNAi to suppress SMN expression in S2 cells. Morphological changes were observed in dSMN dsRNA-treated cells after 48 h (Fig. 3, I-K). The cells were stained with Annexin V-Cy5 and analyzed for apoptosis by flow cytometry. As a positive control for induction of apoptosis, cells were treated with a low concentration of actinomycin D (1 M) (Fig. 4B) that has been previously shown to effectively induce apoptosis (16), whereas untreated S2 cells were used as negative controls (Fig. 4A). The percentage of Annexin-V-positive cells indicates an increase in phosphatidylserine on the cell surface of cells in response to an apoptotic stimulus. Furthermore, there was an increase in apoptosis (25.8%) in the dSMN dsRNA-treated cells (Fig. 4D) compared with the baseline percent apoptosis (9.17%) of the cells transfected with the GFP marker and CAT dsRNA (Fig. 4C). This result is representative of three separate experiments. In these three experiments, the mean percent apoptosis in SMN dsRNA/GFP co-transfectants was 18.8 Ϯ 4.0%, whereas the mean percent apoptosis in the control-treated cells was 8.3 Ϯ shown). The necrotic cell population was identified by analysis of the flow cytometric parameters forward and side scatter as previously described (17). Necrotic cells were gated out of the entire analysis. The independent cell death ELISA confirmed increased apoptosis in these cells following treatment of dSMN dsRNA (Fig. 4E).
dSMN Deficiency-induced Apoptosis Is Caspase-dependent and Blocked by a Caspase Inhibitor-To test whether the apoptosis resulting from the depletion of SMN in S2 cells is caspasedependent or -independent, we analyzed by immunoblot to detect the activation of the endogenous caspases using anti-DRONC and anti-DRICE antibodies. It was previously demonstrated (18 -20) that these two caspases interacted with each other, and the DRONC was an upstream factor that processes the effector caspase DRICE. Our results clearly indicate that whereas there was no processing of the ϳ50 kDa DRONC precursor in the control cell lysates (CAT dsRNA) (Fig. 5A), its cleaved fragments (ϳ40 kDa (Pr1) and ϳ37 kDa (Pr2)) were evident in dSMN RNAi cell lysates (Fig. 5A). The first cleaved product (Pr1) of DRONC may be due to autocatalytic processing, which has been demonstrated earlier (19), and the second cleavage product (Pr2) is the resulting product of apoptosis, which is consistent with results obtained in S2 cells previously (21). The disappearance of full-length DRONC correlated very well with the increasing activation of caspase due to dSMN RNAi. Because DRONC interacts with DRICE, we also examined processing of DRICE using anti-DRICE antibody. Similar to what has been found for DRONC, our immunoblot results indicate that there was no processing of DRICE in control (CAT dsRNA) transfected cells, but accumulation of its processed (cleaved) products (ϳ28, ϳ26, and ϳ19 kDa) were detected in cell lysates of RNAi (Fig. 5B), and these analyses suggest that they are enzymatically activated in dSMN RNAi cell lysates.
Furthermore, Z-VAD-fmk treatment reversed the effect of dSMN RNAi as indicated by ELISA and flow cytometric assays (Fig. 6, A-D). The percent of apoptosis in the control dsRNAtreated cells (Fig. 6C) was 11%, whereas in the dSMN dsRNAtreated cells, percent apoptosis was 19% (Fig. 6B). This result is consistent with the increased apoptosis observed with the dSMN dsRNA (Fig. 4). In three similar experiments, the mean percent apoptosis in SMNdsRNA/GFP co-transfectants was 21.5 Ϯ 6.5%, whereas the mean percent apoptosis in the Z-VAD-fmk-treated was 18.7 Ϯ 5.4%. Thus, there was little effect of Z-VAD-fmk on Annexin V binding to apoptotic S2 cells. In the Z-VAD-fmk-treated cells, the percentage of transfected cells surviving increased from 26 to 63% (Fig. 6D), a percentage similar to that of control transfected cells (Fig. 6C). This demonstrates an appreciable effect of the caspase inhibitor, Z-VADfmk upon the survival of S2 cells treated with dsRNA of dSMN. Further, cell lysates were extracted from control, dSMN RNAi, and dSMN RNAi with Z-VAD-fmk-treated cells and were immunoblotted for DRONC and DRICE protein. The results confirm that there was no accumulation of processed DRONC and DRICE products in lysates prepared from Z-VAD-fmk cotreated cells (Fig. 7). However, the endogenous expression of DIAP1, a known cellular caspase inhibitor, which belongs to the family of IAPs (22) was unaffected by dSMN RNAi (Fig.  5C), and this suggests that cell death may be independent of DIAP1 pathway.

Conservation and Diversification of Structures and Functions between dSMN and hSMN-
One of the characteristic features for mammalian SMN proteins is the formation of gems in the nucleus. However, it is unclear what roles, if there are any, the SMN protein plays in the structure. In this report, we generated specific antibodies against the Drosophila SMN protein. We demonstrated by immunofluorescence that in cotransfected insect S2 cells, hSMN and dSMN were largely co-localized in the cytoplasm. The hSMN was detected as nuclear dot-like structures in some of the S2 cells (Fig. 3L). However, only minimal dSMN in co-transfected cells but not in cells without transfection of hSMN was observed in these structures, suggesting that unlike hSMN, dSMN may not be a nuclear protein to form gems, possibly due to sequence diversification between exon 7 of dSMN and hSMN, which has been suggested to play a key role for hSMN nuclear localization (23). The minimal staining of dSMN in the nucleus from co-transfected cells may occur through a strong association between dSMN and hSMN proteins, which has been demonstrated in a separate experiment by co-immunoprecipitation assay and in vitro binding assay. 2 We recently identified an interaction between SMN and Rpp20, a subunit of the RNase P complex, and we showed that the interaction between these two proteins is interchangeable between human and Drosophila proteins. 2 These data 2 suggest that dSMN and hSMN proteins are structurally and functionally conserved in some respects but diverse in others, consistent with conservation of middle domains and diversification of C and N-terminal domains between human and Drosophila SMN sequences.
SMN Involvement in Apoptosis-The spinal muscular atrophy is caused by death of motor neurons. However, the underlying mechanisms of SMN-deficiency induced cell death are not completely understood. Disruption of SMN expression from S. pombe to mice as well as in a chicken cell line DT40 shows lethal phenotypes (3,5,(7)(8)(9)24), indicating a requirement of SMN for cell survival. It has been more recently demonstrated that over-expression of human SMN prolonged cell survival and inhibited apoptosis in PC12 cells deprived of trophic support by acting on mechanisms that mediate cell death through the mitochondrial release of cytochrome c, a key step in the activation of caspase-3 (25). Because the molecular mechanism of apoptosis is highly conserved throughout evolution, we used 2 Y. Hua and J. Zhou, manuscript in preparation. the insect S2 cells by RNA interference in this report to investigate how SMN depletion affected cell death. We demonstrated by both FACS and cell death ELISA assays that apoptotic cell death was induced by depletion of dSMN protein in the S2 cells.
The execution of cell death in Drosophila requires caspase activation (26). There are three pro-apoptotic genes so far cloned in the fly (rpr (27), hid (28), and grim (29)). These genes are reported to activate at least two caspases (DRICE (30) and DCP-1 (31)). DRICE is proteolytically processed during apoptosis in S2 cells (32). In flies, a total of seven caspases have been reported: DCP-1 (31), DREDD/DCP-2 (33), DRICE (30), DRONC (34), DECAY (35), DAMM (36), and STRICA (37). Of these caspases, DRONC and DREDD (apical/initiator or class I caspases) contain extensive pro-domains. They do not require other caspases for cleavage of their zymogens but are thought to interact with other regulators through their pro-domains. However, the precise molecular mechanism for the activation of initiator caspases is still unclear. DCP-1 and DRICE (effector or class II caspases) have been shown to play a role in the execution of cell death (see the review) (38). DRONC is a homolog of caspase-9 in mammals that contains a caspase recruitment domain (CARD), suggesting that it is the functional counterpart of CED-3 in C. elegans (34). DRONC is also found in embryonic tissues like eye and brain and in adult egg chambers where apoptosis occurs naturally. Up-regulation of DRONC has been reported in salivary glands and midguts before histolysis (34,39). Recently, its interaction with DRICE, an effector caspase, has been discovered (18). Upon receipt of cell death signal, the apical or initiator caspases undergo proteolytic processing and further activate other downstream caspases that go on to cleave various target proteins causing apoptosis (40 -42). Because DRONC and DRICE activation are directly involved in execution of cell death in Drosophila cells, we analyzed by immunoblotting for caspase activation using specific antibodies against DRONC and DRICE. Our results strongly suggest that apoptotic cell death induced by SMN deficiency is caspase-dependent.
Z-VAD-fmk Inhibits Caspase Activation-DRONC and DRICE are activated after various death stimuli without any significant release of cytochrome c in the cytosol (43), and retention of cytochrome c in mitochondria during apoptosis has been demonstrated in Drosophila (44). Recently, it has also been demonstrated in Drosophila cells that the central role of ARK (apoptotic protease-activating factor-1 (Apaf-1)-related killer), also known as Dark, Hac-1, or Dapaf-1, in stress-induced apoptosis appears to be independent of cytochrome c (16). In contrast, release of cytochrome c has been reported in Drosophila cells during apoptosis (45). More studies are needed to elucidate whether any involvement of cytochrome c in SMNmediated apoptotic cell death in S2 cells. In the present investigation, cells co-treated with Z-VAD-fmk and dsRNA of dSMN result in increased survival (Fig. 6D), a decrease in nucleosomal cleavage (Fig. 6A) and decrease in caspase cleavage (Fig.  7). However, the rate of apoptosis as measured by Annexin V staining is unchanged by Z-VAD-fmk treatment. This data is consistent with that of previous investigators (46) who have shown that caspase inhibition by Z-VAD-fmk does not prevent phosphatidylserine (PS) exposure on the cell surface. In contrast to these other studies (46,47), the ultimate death of Z-VAD-fmk-treated cells was not observed here. Thus, the caspase-dependent apoptosis induced by SMN deficiency can be blocked by Z-VAD-fmk, suggesting potential targets for therapy to surpass depletion of SMN protein in SMA patients.
SMA Drosophila Models and Identification of SMA Therapeutic Targets-SMA is a disease caused by apoptotic cell death of motor neurons. However, ubiquitous expression of SMN protein in all tissues gives no explanations as to why only motor neurons are targeted in SMA patients. It has been shown previously that SMN exhibits anti-apoptotic properties and interacts with several factors including Bcl-2 (48) and p53 (49), both of which play key roles in the apoptotic process but are ubiquitously expressed. In this study we demonstrated that SMN-depletion-induced apoptosis in S2 cells is caspasedependent and could be blocked by a caspase inhibitor Z-VADfmk. We hypothesize that the specific mediators (proteins) between SMN and apoptotic pathways may exist and are downor up-regulated in motor neurons, resulting in more vulnerability of these cells to low levels of SMN protein than other cell types. More studies are needed in this direction to identify scaffold molecules involved in the activation of upstream caspase DRONC to identify the SMN-mediated apoptotic pathway in Drosophila cells. These proteins may be potentially targeted for drug development for SMA disease.