Bcl-2 homologue Debcl enhances α-synuclein-induced phenotypes in Drosophila

Introduction

Parkinson disease (PD) is a human movement disorder that is strongly associated with the selective and profound degeneration and loss of dopaminergic (DA) neurons to result in a set of marked clinical features (Forno, 1996). The neuropathological hallmarks exhibited by PD patients include the presence of Lewy Bodies (LB) which are intracytoplasmic proteinaceous inclusions composed of α-synuclein and ubiquitin among other proteins (Forno, 1996; Leroy et al., 1998; Polymeropoulos et al., 1997). This atypical protein aggregation and accumulation is believed to lead to cellular toxicity and contribute to the pathogenesis of PD. Additional pathological mechanisms that are associated with PD include aberrant protein aggregation and mitochondrial damage (Gupta, Dawson & Dawson, 2008; Schulz, 2007; Whitworth, 2011). Familial forms of PD have highlighted the genetic basis of PD and the study of the associated gene loci in model organisms offers great understanding of the disease aetiology and pathology (Ambegaokar, Roy & Jackson, 2010; Gasser, 2009; Guo, 2012). The gene encoding α-synuclein, a small soluble protein of largely unknown function predominantly found in neural tissues, was first to be identified as responsible for inherited PD (Polymeropoulos et al., 1997). Mitochondrial dysfunction due to the accumulation of α-synuclein has been implicated as one of the mechanisms leading to PD (Chinta et al., 2010; Choubey et al., 2011; Esteves et al., 2011; Zhu et al., 2011). The association of α-synuclein with components of the mitochondria is thought to lead to oxidative stress, apoptosis, autophagy and eventually, neurodegeneration. The first Drosophila model of PD utilized a human α-synuclein transgene to induce the PD-like symptoms (Feany & Bender, 2000). This model system is very successful and widely applied, as it displays the age-dependent loss of locomotor function, the degeneration of DA neurons and LB-like inclusions, features that are present in human PD (Auluck et al., 2002; Botella et al., 2009; Büttner et al., 2014; Feany & Bender, 2000; Kong et al., 2015; Staveley, 2014; Webb et al., 2003; Zhu et al., 2016). Drosophila has available tissue specific gene enhancers such as TH-Gal4, elav-Gal4 and Ddc-Gal4, which are used to model PD in flies in combination with the powerful bipartite UAS/Gal4 (Brand & Perrimon, 1993) system. Of importance is the correlation between DA neuron loss and the age-dependent loss of locomotor function (Park, Schulz & Lee, 2007; Staveley, 2014) which validates the implication that age-dependent loss of locomotor function is as a result of DA neuron degeneration.

The Bcl-2 family of genes are crucial controllers of apoptosis in animals and are functionally composed of proapoptotic and antiapoptotic members (Adams & Cory, 1998; Cory & Adams, 2002; Fu & Fan, 2002; Siddiqui, Ahad & Ahsan, 2015). In mammals, this multigene family has about 20 members, the antiapoptotic proteins protect the mitochondria from disruption by the proapoptotic proteins (Colin et al., 2009; Cory & Adams, 2002; Martinou & Youle, 2011; Suen, Norris & Youle, 2008; Tsujimoto, 2002). The antiapoptotic members possess four Bcl-2 homology (BH) domains while the proapoptotic members have three to four BH domains. The proapoptotic proteins initiate apoptosis by the permeabilization of the outer mitochondrial membrane which results in the release of apoptogenic factors into the cytosol (Delbridge & Strasser, 2015; Doerflinger, Glab & Puthalakath, 2015; Li & Dewson, 2015; Lopez & Tait, 2015). The antiapoptotic members protect the mitochondria from permeabilization by the proapoptotic members and block the release of apoptogenic factors such as cytochrome c, apoptosis inducing factor (AIF) among others from being released from the inner mitochondrial membrane into the cytosol.

Drosophila melanogaster possesses many of the apoptotic pathway proteins that participate in the intrinsic and extrinsic cell death pathways (Kornbluth & White, 2005; Richardson & Kumar, 2002). The Bcl-2 family member homologues in Drosophila are limited to the single antiapoptotic Buffy (Quinn et al., 2003), and the sole proapoptotic death executioner Bcl-2 homologue, Debcl (Brachmann et al., 2000; Colussi et al., 2000; Igaki et al., 2000; Quinn et al., 2003; Zhang et al., 2000). Debcl has a strong similarity with the mammalian mitochondria outer membrane permeabilization protein Bok/Mtd.

The promoter region of Debcl contains four dNF-Y-binding consensus sequences which play positive roles in promoter activity and indicate that dNF-Y regulates Debcl gene expression (Ly et al., 2013). The tumour suppressor gene Retinoblastoma (Rbf1 in Drosophila) induces a Debcl-and Drp1-dependent mitochondrial cell death (Clavier et al., 2015). Rbf1 induces cell death by reducing the expression of the sole Debcl antagonist Buffy (Clavier et al., 2014). The Rbf1-induced apoptosis is dependent on Debcl-dependent mitochondrial ROS production and essentially Debcl is required downstream of Buffy for apoptosis to occur. The Debcl-induced ROS production appears to be through Glycerophosphate oxidase 1 participation to increase mitochondria ROS accumulation (Colin et al., 2015). The organic solute carrier partner 1/oxidored nitrodomain-containing protein 1 (OSCP1/NOR1), a known tumour suppressor induces apoptosis by the down-regulation of the Buffy gene and the up-regulation of the Debcl gene (Huu, Yoshida & Yamaguchi, 2015). Debcl is not required for most developmental cell death, but has been shown to play a role in embryonic cell death (Galindo et al., 2009) and stress-induced apoptosis (Sevrioukov et al., 2007). Antiapoptotic Buffy antagonizes Debcl-induced apoptosis by physical interaction (Quinn et al., 2003), probably at the mitochondria where Debcl localizes (Doumanis, Dorstyn & Kumar, 2007). The presence of a mitochondrial outer membrane (MOM)-targeting motif in Debcl indicates it possibly has a role in mitochondrial cell death pathway.

The role of the mitochondria in PD pathogenesis makes the α-synuclein-induced model of PD (Feany & Bender, 2000) a very attractive model for the investigation of the role of Bcl-2 proteins. Here, we investigate the potential enhancement or suppression of the α-synuclein-induced PD phenotypes by the inhibition and overexpression of the pro-apoptotic Bcl-2 homologue Debcl.

Materials and Methods

Results

Debcl is similar to the human proapoptotic Bcl-2 ovarian killer (Bok)

Bioinformatic analysis of the protein sequences encoded by the Debcl and Bok genes reveal 37% identity and 55% similarity. The Debcl protein consists of 300 amino acids and indicates the existence of the BH1, BH2, BH3, BH4 and TM domains, similar to the 212 amino acids human Bok (Fig. 1). An ELM resource search for functional sites (Dinkel et al., 2016) indicates the presence of a transmembrane domain (membrane anchor region), an inhibitor of apoptosis binding motif (IBM) at amino acids 1–5, a PDZ domain at amino acids 295–300, an ER retention motif at amino acids 109–115 and between amino acids 258–262, an Atg8 binding motif at amino acids 36–42, a nuclear receptor box motif at amino acids 295–300, and a ubiquitination motif of the SPOP-binding consensus at amino acids 2–6 and another one at position 74–79. There is a number of BH3-homology region binding sites in the central region of the protein as determined by an NCBI conserved domain search (Marchler-Bauer et al., 2015). Although the two proteins Bok and Debcl have been determined to be antiapoptotic, both show the presence of a BH4 domain, the homology domain that is most often associated with pro-survival proteins.

Directed misexpression of Debcl in DA neurons alters lifespan and locomotor ability

The inhibition of Debcl in the DA neurons by RNA interference results in a lifespan with a median survival of 64 days that is similar to 62 days for the controls expressing the benign lacZ transgene as determined by a Log-rank (Mantel-Cox) test (Fig. 2A). The locomotor ability showed a slight improvement when nonlinear fitting of the climbing curves was performed, with significant differences at 95% confidence intervals (Fig. 2B). This suggests that the inhibition of the proapoptotic Debcl confers a small advantage for the normal functioning of DA neurons.

When Debcl is overexpressed in DA neurons, the survival criteria of these flies differ greatly (Fig. 2A), with Debcl-overexpressing flies having a median lifespan of 48 days compared to 62 days for the controls expressing the benign lacZ transgene as indicated by a Log-rank (Mantel-Cox) test. The overexpression of Debcl in DA neurons severely impairs climbing ability as determined by the nonlinear fitting of the curve with 95% CI (Fig. 2B). This suggests that the overexpression of Debcl in DA neurons interferes with the normal functioning of these flies and results in compromised “healthspan.”

Altered expression of Debcl influences the α-synuclein-induced phenotypes

The inhibition of Debcl by RNAi along with the expression of α-synuclein under the direction of the Ddc-Gal4 transgene results in increased lifespan and healthier climbing ability compared to the control (Fig. 3). The Debcl-RNAi along with α-synuclein-expressing flies had a median lifespan of 67 days, while that of α-synuclein-expressing controls was 60 days as determined by a Log-rank (Mantel-Cox) test (Fig. 3A). The climbing ability of these flies was slightly improved than of the α-synuclein-expressing controls as indicated by the nonlinear fitting of the climbing curves and compared the 95% CI (Fig. 3B). These results show that the inhibition of the proapoptotic Debcl confers a significant advantage to flies under the influence of the neurotoxic effects of the human transgene α-synuclein.

The overexpression of Debcl along with α-synuclein in DA neurons results in decreased median lifespan of 44 days, compared to 60 days for the control flies as determined by a Log-rank (Mantel-Cox) test (Fig. 3A). The climbing curves indicate that there was a significant reduction in the climbing ability of the flies with overexpression of Debcl (Fig. 3B) and thus, enhancing the phenotypes observed when α-synuclein is expressed in DA neurons. This suggests that the overexpression of Debcl further increases the toxic effects of the expression of α-synuclein.

Overexpression of Debcl enhances the α-synuclein-induced developmental eye defects

The overexpression of Debcl in the Drosophila eye results in severe ablation of the eye due to apoptosis (Colussi et al., 2000; Igaki et al., 2000) while expression of α-synuclein in the eye results in developmental defects (Fig. 4D). When Debcl is overexpressed in the eye, developmental defects resulting from Gal4 (Kramer & Staveley, 2003) (Figs. 4A and 4J), inhibition of Debcl (Figs. 4B and 4J), and overexpression of Debcl (Figs. 4C and 4J) are enhanced. Biometric analysis of the ommatidia number and the percentage of eye disruption showed significant differences in the compared genotypes to the control that express the benign lacZ transgene (Fig. 4J). The inhibition of Debcl along with α-synuclein expression (Figs. 4E and 4K) and the co-expression of Debcl and α-synuclein (Figs. 4F and 4K) result in enhanced phenotypes. The disruption of the ommatidial array due to fusion of the ommatidia and smaller eye is severely enhanced by the overexpression of Debcl together with α-synuclein (Figs. 4F and 4K). The analysis of the ommatidia number and disruption of the eye reveals significant differences, the inhibition of Debcl yields “healthier” eyes and its overexpression results in worsened phenotypes (Fig. 4K). The ommatidial disarray that results from inhibition of Debcl are completely rescued by overexpression of the pro-survival Buffy (Figs. 4H and 4L), while the ablated eye that result from Debcl overexpression is partially rescued upon Buffy overexpression, this restores the eye ablation to a mildly severe rough eye phenotype (Figs. 4I and 4L). Biometric analysis showed recouped ommatidia number and a lessened disruption of the eye, though they were still significantly different from the control (Fig. 4L). These results suggest that overexpression of Debcl along with expression of α-synuclein enhances the Debcl-induced eye ablation, while the overexpression of Debcl together with Buffy partially rescues the eye phenotype.

Discussion

Since mitochondrial dysfunction is central to the pathology of both sporadic and familial forms of PD (Subramaniam & Chesselet, 2013), it was important to highlight the role and consequences of the altered expression of the proapoptotic mitochondrial gene Debcl in this process. The overexpression of Debcl in Drosophila and other systems, including mammalian, has been demonstrated to lead to apoptosis (Brachmann et al., 2000; Colussi et al., 2000; Galindo et al., 2009; Igaki et al., 2000; Senoo-Matsuda, Igaki & Miura, 2005; Sevrioukov et al., 2007; Zhang et al., 2000). The recapitulation of PD-like symptoms in Drosophila melanogaster, especially the age-dependent loss of climbing ability, has led to investigation of genes that could suppress these phenotypes (Auluck et al., 2002; Feany & Bender, 2000; Haywood & Staveley, 2004). Our results show that the overexpression of Debcl results in a severely shortened lifespan followed by premature loss in climbing ability; phenotypes that are reminiscent of PD-like symptoms in model organisms. Thus our work shows the intricate balance between life and death decisions in the sensitive dopamine producing neurons. It seems that excess amounts of Debcl protein are sufficient to upset the survival mechanisms and lead to degeneration and death of DA neurons. The importance of Debcl-induced apoptosis is exhibited by the strict control in its gene product by the tumour suppressors Rbf1 (Clavier et al., 2015), OSCP1/NOR1 (Huu, Yoshida & Yamaguchi, 2015), and NF-Y (Ly et al., 2013). Furthermore, it has a motif for ubiquitination, probably by the TrCP homologue slimb that targets it for destruction by the proteasome (Colin et al., 2014). The inhibition of Debcl had a converse result, with flies that had a longer lifespan and healthy climbing ability. It is possible that the suppression of Debcl tips the balance towards the survival pathways controlled by the antiapoptotic Buffy. Our results indicate that overexpression of Debcl appears to be a novel model of PD as a result of neuronal apoptosis.

The α-synuclein-induced model of PD in Drosophila shows little difference in lifespan between the control and wild type, A53T and A30P α-synuclein flies (Feany & Bender, 2000). In our study, the overexpression of Debcl in the DA neurons resulted in a marked decrease in lifespan. This is in part due to toxic effects as a result of the expression of α-synuclein, and additionally, due to Debcl-induced apoptosis. The Debcl-induced apoptosis is mediated by other factors including; the mitochondrial fission protein Drp1 (Clavier et al., 2015) that interacts with Debcl to induce mitochondrial fragmentation; Glycerophosphate oxidase-1 (Colin et al., 2015) that increases mitochondrial ROS accumulation; and possibly through the initiation of autophagy, since both α-synuclein expression (Xilouri & Stefanis, 2015) and Debcl (Hou et al., 2008) overexpression are implicated in this process. This worsening of phenotypes was also observed when Debcl was overexpressed with α-synuclein in the eye. The inhibition of Debcl in the DA neurons resulted in a marked increase in survival and improved locomotor ability. This inhibition of Debcl is sufficient to negate its apoptotic role and thus promote cell survival through the opposing antiapoptotic Buffy.

Locomotor dysfunction is one of the major symptoms of PD. The demonstration of an age-dependent loss of climbing ability is pivotal to highlighting the effects of degeneration and death of DA neurons, ultimately as a consequence of altered gene expression as opposed to cellular senescence (Rodriguez et al., 2015). The overexpression of Debcl in the DA neurons produced a climbing index significantly different from that of control flies with the loss of climbing ability in an age-dependent manner and likely due to Debcl-induced neuronal degeneration. The degree of locomotor dysfunction seemed to be similar to that observed when α-synuclein is overexpressed in DA neurons. Taken together, these results would indicate a detrimental effect in overexpression of Debcl in DA neurons that result in a novel model of PD in flies.

In contrast, the inhibition of Debcl in the same neurons results in a remarkable improvement in climbing ability when compared to the controls. The inhibition of Debcl in the DA neurons of the α-synuclein-induced PD model significantly increased lifespan and climbing ability, indicating that reduced levels of Debcl are sufficient to alter the healthspan of DA neurons. The Debcl-induced apoptosis relies on downstream effectors that either induces ROS accumulation (Colin et al., 2015) or the fragmentation of the mitochondria (Clavier et al., 2015). As the down-regulation of Buffy or up-regulation of Debcl results in apoptosis (Huu, Yoshida & Yamaguchi, 2015), the cellular advantage of Debcl inhibition may be indirect through the de-repression of the Buffy gene product that confers survival advantages. The directed expression of Buffy along with Debcl results in an improved “healthspan” compared to the Debcl-induced phenotypes and corroborate other studies that show the overexpression of the pro-survival Buffy confers survival advantages through increased survival and improved climbing ability under conditions of stress (M’Angale & Staveley, 2016). Our study suggests that the overexpression of Buffy is similar to an up-regulation that ultimately blocks Debcl-induced apoptosis, similar to results obtained when its regulation by Rbf1 or dE2F2 is altered to repress it transcriptionally (Clavier et al., 2014; Clavier et al., 2015). This suppression of Buffy is sufficient to induce Debcl-dependent apoptosis, in addition to the promotion of Debcl activity by dNF-Y (Ly et al., 2013). The co-overexpression of Debcl and Buffy in the eye resulted in a partial rescue of the Debcl-induced phenotypes. Therefore, overexpression of the pro-survival Buffy suppresses the Debcl-dependent phenotypes.

Conclusions

Directed inhibition of Debcl results in improved survivorship and extended climbing ability whereas the directed expression of Debcl results in reduced lifespan and impaired locomotor function. These phenotypes are rescued upon co-expression with the pro-survival Buffy. The overexpression of Debcl enhances the effects of α-synuclein expression. Buffy counteracts Debcl-induced phenotypes, and represents a potential target to enhance neuronal survival in response to the detrimental effects of Debcl-induced apoptosis.

Supplemental Information