Phospholipase PLA2G6, a Parkinsonism-Associated

Article

Phospholipase PLA2G6, a Parkinsonism-Associated Gene, Affects Vps26 and Vps35, Retromer Function, and Ceramide Levels, Similar to a-Synuclein Gain Graphical Abstract

Authors Guang Lin, Pei-Tseng Lee, Kuchuan Chen, ..., Wen-Wen Lin, Liping Wang, Hugo J. Bellen

Correspondence [email protected]

In Brief Lin et al. show that loss of the fly homolog of the neurodegeneration gene PLA2G6 (PARK14) impairs retromer function, causes ceramide accumulation, and leads to lysosomal dysfunction, which causes neuronal dysfunction. Their further data suggest that the uncovered retromer/ceramide/lysosome pathway may be common in Parkinson disease.

Highlights d

d

d

d

iPLA2-VIA regulates ceramide and sphingolipid metabolism Loss of iPLA2-VIA progressively impairs retromer and lysosomal function in neurons iPLA2-VIA, vps35, vps26, and a-Syn affect retromer function and ceramide levels Reducing ceramides or enhancing retromer function alleviates neuronal insults

Lin et al., 2018, Cell Metabolism 28, 605–618 October 2, 2018 ª 2018 Elsevier Inc. https://doi.org/10.1016/j.cmet.2018.05.019

Cell Metabolism

Article Phospholipase PLA2G6, a Parkinsonism-Associated Gene, Affects Vps26 and Vps35, Retromer Function, and Ceramide Levels, Similar to a-Synuclein Gain Guang Lin,1,5 Pei-Tseng Lee,1,5 Kuchuan Chen,2,5 Dongxue Mao,2,5 Kai Li Tan,2,5 Zhongyuan Zuo,1,5 Wen-Wen Lin,1,5 Liping Wang,2,5 and Hugo J. Bellen1,2,3,4,5,6,* 1Department

of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA in Developmental Biology, Baylor College of Medicine, Houston, TX 77030, USA 3Howard Hughes Medical Institute, Baylor College of Medicine, Houston, TX 77030, USA 4Department of Neuroscience, Baylor College of Medicine, Houston, TX 77030, USA 5Jan and Dan Duncan Neurological Research Institute, Texas Children’s Hospital, Houston, TX 77030, USA 6Lead Contact *Correspondence: [email protected] https://doi.org/10.1016/j.cmet.2018.05.019 2Program

SUMMARY

Mutations in PLA2G6 (PARK14) cause neurodegenerative disorders in humans, including autosomal recessive neuroaxonal dystrophy and early-onset parkinsonism. We show that loss of iPLA2-VIA, the fly homolog of PLA2G6, reduces lifespan, impairs synaptic transmission, and causes neurodegeneration. Phospholipases typically hydrolyze glycerol phospholipids, but loss of iPLA2-VIA does not affect the phospholipid composition of brain tissue but rather causes an elevation in ceramides. Reducing ceramides with drugs, including myriocin or desipramine, alleviates lysosomal stress and suppresses neurodegeneration. iPLA2-VIA binds the retromer subunits Vps35 and Vps26 and enhances retromer function to promote protein and lipid recycling. Loss of iPLA2-VIA impairs retromer function, leading to a progressive increase in ceramide. This induces a positive feedback loop that affects membrane fluidity and impairs retromer function and neuronal function. Similar defects are observed upon loss of vps26 or vps35 or overexpression of a-synuclein, indicating that these defects may be common in Parkinson disease.

INTRODUCTION The phospholipase A2 (PLA2) protein family is a diverse group of lipid-modifying enzymes. They maintain homeostasis of cell membranes by remodeling membrane glycerol phospholipids, preventing membrane fatty acid (FA) peroxidation and promoting the release of docosahexaenoic acid (DHA) and arachidonic acid (AA) (Adibhatla and Hatcher, 2008). AA modulates cell survival (Hooks and Cummings, 2008) and initiates inflammatory responses (Murakami and Kudo, 2002). Biochemically, PLA2 hydrolyzes the second carbon chain of glycerol phospholipids

to generate a FA, usually an unsaturated FA, such as AA or DHA, and a lysophospholipid (LPL), a process referred to as the Lands cycle (Figure S1A) (Lands, 1958). LPL has an inverted-cone shape structure, which allows bending of cellular membranes during vesicle formation or tubulation (Figure S1B) (Brown et al., 2003). Indeed, addition of a catalytically active secreted PLA2 facilitates the formation of small vesicles from a giant unilamellar vesicle in in vitro assays (Staneva et al., 2004). PLA2G6 encodes iPLA2-b, a Ca2+-independent PLA2. iPLA2-b is expressed in the brain and enriched in dendrites and axon terminals (Ong et al., 2005). In cultured cells, loss of PLA2G6 impairs retrograde trafficking from Golgi to endoplasmic reticulum (ER) (de Figueiredo et al., 1998, 2000), recycling of transferrin and transferrin receptors from endosomes (de Figueiredo et al., 2001), and assembly and maintenance of tubulovesicular structures (TVS) of the Golgi complex (de Figueiredo et al., 1999). These data suggest that iPLA2-b plays a role in the dynamics of cellular membranes. Recessive mutations in PLA2G6 cause infantile neuroaxonal dystrophy (INAD) (OMIM no. 256600), atypical NAD (aNAD) (OMIM no. 610217) (Khateeb et al., 2006; Morgan et al., 2006), and PLA2G6-related dystonia-parkinsonism, also called Parkinson disease 14 (PARK14) (OMIM no. 612953) (Paisan-Ruiz et al., 2009). Iron accumulation in the basal ganglia region has been observed in some INAD and aNAD patients but not in PARK14 patients. These diseases are classified as neurodegeneration with brain iron accumulation 2 (OMIM no. 610217) (Morgan et al., 2006). PLA2G6-associated neurodegeneration (PLAN) is a collective term referred to all diseases caused by a variation in PLA2G6 (Kurian et al., 2008). Early-onset ataxia (age 1–3 years) is the earliest symptom of INAD. The patients show mental and motor deterioration, hypotonia, progressive spastic tetraparesis, visual impairments, bulbar dysfunction, and extrapyramidal signs. Affected children usually die during the first decade. aNAD and PARK14 have a later onset of symptoms compared with INAD and tend to have progressive dystonia and parkinsonism. Cerebellar atrophy is a characteristic symptom in both INAD and aNAD, but not in PARK14 (Sumi-Akamaru et al., 2015). A neuropathological hallmark of PLAN is the formation of spheroids throughout the nervous system (Hedley-Whyte et al., 1968). These spheroids comprise TVS, a diagnostic marker

Cell Metabolism 28, 605–618, October 2, 2018 ª 2018 Elsevier Inc. 605

of INAD (Sumi-Akamaru et al., 2015) that is usually positive for a-synuclein (a-Syn) and ubiquitin (Riku et al., 2013). In addition, the presence of Lewy bodies and phosphorylated tau-positive neurofibrillary tangles in neurons were also reported (Riku et al., 2013). Mice that lack PLA2G6 are viable but exhibit a slow progressive motor defect between the first and the second year (Malik et al., 2008; Shinzawa et al., 2008). In agreement with the symptoms observed in INAD patients, neuropathological features of these knockout mice include progressive axonal degeneration, cerebellar atrophy, and accumulation of a-Syn-containing spheroids in the TVS (Malik et al., 2008; Shinzawa et al., 2008). However, no iron accumulation has been documented in the basal ganglia of these mice. Transmission electron microscopy (TEM) revealed a swelling of the presynaptic membrane and synaptic degeneration (Beck et al., 2011). Interestingly, abnormal mitochondria with tubular and branching cristae were observed at a later stage, suggesting that loss of PLA2G6 causes a slow progressive degeneration of the mitochondrial inner membrane (Beck et al., 2011). The function of PLA2G6 appears to be conserved in flies. A P element insertion in iPLA2-VIA was shown to impair iPLA2-VIA expression, and homozygous animals are viable but exhibit a shorter lifespan (40 days) and locomotor defects. In addition, mitochondrial abnormalities and degeneration of mitochondrial inner membranes at the end stage were also documented (Kinghorn et al., 2015). These phenotypes recapitulate some of the defects observed in mice and human. How iPLA2-b contributes to the above defects described in flies, mice, and humans are unknown. Here, we report that mutant animals that lack iPLA2-VIA display a slow progressive neurodegeneration, accumulate ceramide intermediates, but do not show alterations in phospholipid composition or levels. We argue that loss of iPLA2-VIA disrupts retromer function and impairs retrieval of ceramide phosphoethanolamine (CPE)/ sphingomyelin (SM). This leads to an increase of ceramides and induces stress. In summary, our data indicate that the retromer and sphingolipid metabolism play a critical role in PLA2G6associated diseases. These observations may be relevant to Parkinson disease (PD) given that loss of vps35 or gain of a-Syn also affects retromer function and causes ceramide accumulation. RESULTS Loss of iPLA2-VIA Reduces Lifespan and Affects BangSensitivity and Starvation Tolerance To study the molecular mechanisms underlying PLAN, we created null alleles by imprecise excision of a P element (P[EPgy2] iPLA2-VIAEY05103) inserted in the 50 UTR of Drosophila iPLA2-VIA. We generated two iPLA2-VIA deletion alleles, iPLA2-VIAD174 (D174) and iPLA2-VIAD192 (D192), and a precise excision as control (C) (Figure 1A). The Drosophila iPLA2-VIA encodes four transcripts (PA, PB, PC, and PD). Transcripts PB, PC, and PD are translated into identical proteins, whereas PA is 13 amino acids shorter (Figure 1A). We generated a polyclonal antibody (iPLA2-VIArb) against the N-terminal Ankyrin repeat domain (see epitope in Figure 1A) that recognizes an 85-kDa protein on western blots (predicted 606 Cell Metabolism 28, 605–618, October 2, 2018

MW is 90 kDa). This protein co-migrates with a ubiquitously overexpressed (Act-GAL4) C-terminal hemagglutinin (HA)tagged iPLA2-VIA (iPLA2-VIA-HA) in adult heads (Figure 1B). In both iPLA2-VIA mutant alleles, D174 and D192, the 85-kDa band is absent, suggesting that they are null alleles (Figure 1B). Both D174 and D192 alleles are homozygous viable, but these flies exhibit a greatly reduced lifespan and die between days 14 and 30 (Figure 1C), compared with 80–90 days for wild-type flies. A 20-kb P[acman] genomic fragment containing the iPLA2-VIA locus (GR; D174 and GR; D192) (Venken et al., 2009) rescues the lifespan reduction (Figure 1C) as well as other phenotypes associated with the D174 and D192 mutations (Figure 1D). In summary, these experiments show that we have generated null alleles for iPLA2-VIA that exhibit a severely reduced lifespan. We next assessed a variety of phenotypes, including the ability to respond to starvation, bang-sensitivity, and measured ATP level. We tested 7- to 15-day-old null mutant flies that do not display motor defects. iPLA2-VIA mutant flies exhibit bangsensitivity and paralysis (Figure 1D), and ubiquitous expression (Act-GAL4 driver > iPLA2-VIA) of fly (PA or PB) or human PLA2G6 cDNA suppresses the bang-sensitivity phenotype (Figure 1E), showing functional conservation between the fly and human gene. In addition, the bang-sensitivity is not enhanced in transheterozygous mutants that carry a deficiency (D174/ Df(3L)BSC394) (Figure 1D). Moreover, loss of iPLA2-VIA impairs survival of both male and female flies to starvation, and females are more sensitive than males (Figure 1F). In addition, ATP levels are reduced in iPLA2-VIA flies compared with controls (Figure 1G). We also observed severe defects in odor avoidance behavior in 1-week-old mutant flies (Figure 1H). In summary, the data support a severe impairment of neuronal function in agreement with data documented in Kinghorn et al. (2015). iPLA2-VIA Is Present in the Cytoplasm and Mitochondria of Neurons We determined the expression of endogenous iPLA2-VIA by using transgenic line that carries an iPLA2-VIA C-terminal GFP-tagged genomic fragment that tags all isoforms (iPLA2VIA-GFP; Figure 1A) (Sarov et al., 2016). This construct rescues lifespan reduction and bang-sensitivity, and hence encodes a functional iPLA2-VIA protein. Based on whole-brain projection images upon long exposure to Triton X-100, iPLA2-VIA broadly localizes to the neuropil (marked by DLG staining) of adult brains (Figure S1C, a-a’). However, the long exposure to Triton X-100 diminishes iPLA2-VIA-GFP staining in cell bodies. A short Triton X-100 exposure staining protocol (Figure S1C, a’’) reveals that the antibodies mostly label the membranes and/or cytoplasm of neurons (Figure S1C, b-b’’). To assess where the proteins encoded by the PA and PB isoform are localized, we expressed UAS-PA or PB with the Actin-GAL4 driver. The PA-encoded protein localizes to the cytoplasm of axons and does not colocalize with the mitochondrial marker, ATP5a (Figure S1C, c-c’’). The protein encoded by the PB isoform shows a faint signal in cytoplasm (Figure S1C, d-d’’) but is mostly colocalized with ATP5a (Figure S1C, d-d’’). A similar expression pattern of PA and PB was observed in third-instar brains. Ubiquitous expression of either PA, PB, or human iPLA2-b cDNA rescues lifespan reduction of D174 and

A

B

E

G

C

D

F

H

Figure 1. Loss of iPLA2-VIA Reduces Lifespan and Causes Behavioral Defects Related to Dysfunctional Mitochondria (A) Reagents generated or used in this study. The dashed lines represent the deleted regions in each allele. (B) iPLA2-VIA protein levels in the indicated genotypes. (C) Lifespan of flies of the indicated genotypes (n = 300). (D and E) Bang-sensitivity of flies with the indicated genotypes (n = 60). (F) Starvation tolerance assays for male (left) and female (right) 7-day-old flies (n = 3 trials). (G) The ADP/ATP ratio measured in indicated genotypes (n = 3). (H) Odor avoidance index (n = 8 trials). Error bars represent SEM. *p < 0.05; **p < 0.01; ***p < 0.001. See also Figure S1.

Cell Metabolism 28, 605–618, October 2, 2018 607

D192 (Figure S2A), suggesting that mitochondrial localization is not required for survival and that fly and human gene are functionally and evolutionarily conserved. However, neuron-specific expression of iPLA2-VIA PA, PB, or human iPLA2-b is not sufficient to rescue lifespan (Figure S2B), indicating that iPLA2-VIA is required in cell types other than neurons.

rated by omega-like structures (Figure S3C), indicative of a defect in endocytosis (Verstreken et al., 2002). Lastly, by day 30 some mitochondria in synapses display aberrant morphology (Figure S4B) characterized by the presence of aberrant inner membranes. These data suggest that iPLA2-VIA plays an important role in maintaining proper membrane homeostasis and shape.

iPLA2-VIA Mutants Display Age-Dependent Loss of Neuronal Activity To assess neuronal function, we recorded electroretinograms (ERGs) in young and aged flies raised in constant darkness. Ten-day-old mutant flies exhibit normal ERG and on-transient amplitudes. The amplitudes of ERGs and on-transients progressively decrease in aged flies maintained in constant darkness over a 30-day period, but not in controls (Figures 2A and S2C), documenting a slow progressive neurodegenerative phenotype independent of neural activity. Moreover, exposure to a 12-hr light (3,600 lux)/dark cycle did not affect the loss of the amplitude of ERG depolarization (data not shown), but slightly enhanced the loss of on-transient amplitudes (Figure S2D0 ). Consistent with this phenotype, we observed accumulation of rhodopsin 1 (Rh1) in the cytoplasm of the photoreceptors (PRs) (Figure S2D00 ), suggesting a defect in recycling that has been previously shown to be associated with retromer dysfunction (Wang et al., 2014). Lastly, the ERG defects caused by loss of iPLA2-VIA were rescued by neuronal expression (C155-GAL4) of the fly (PA and PB) and human iPLA2-b cDNAs (Figure S2E), demonstrating that the ERG defects are due to the cell autonomous loss of iPLA2-VIA.

Loss of iPLA2-VIA Leads to an Accumulation of Sphingolipid Intermediates To test if loss of iPLA2-VIA causes a disruption of the Lands cycle (Figure S1A), we performed a phospholipidomic analysis and compared the phospholipid profile of control and iPLA2-VIA mutants. Surprisingly, none of the phospholipids detected, including PE, PC, PS, PI, PG, and PA (Figures 3A and S5A), or the two major LPL, lyso-PE and lyso-PC (Figures 3B and 3C), showed any significant difference in relative abundance between controls and iPLA2-VIA mutants. Moreover, loss of iPLA2-VIA does not affect total PLA2 activity in adult heads (Figure S5B). In addition, expressing a catalytically impaired human iPLA2-b (S519A) rescues phenotypes associated with iPLA2-VIA mutants, including bang-sensitivity (Figure S5C), ERG defects (Figure S5D), loss of PR number (Figure S5E), and lifespan reduction (Figure S5F). These data indicate that loss of iPLA2-VIA does not disrupt the Lands cycle but still leads to demise of neurons. Interestingly, loss of iPLA2-VIA reduces the levels of the 14C18 ceramide phosphoethanolamines (CPEs) (Figure 3D). Drosophila lacks sphingomyelin (SM) but generates CPE, an analog of SM (Acharya and Acharya, 2005; Vacaru et al., 2013). CPE/SM are components of cellular membranes required for proper proteins trafficking between membrane compartments (van Meer et al., 2008). To further quantify the levels of CPE in iPLA2-VIA mutants, we performed a systematic analysis of most CPE species in lysates of 1,000 wild-type or mutant fly heads. We found that the levels of 19 out of 22 CPEs show a decrease compared with controls (Figure 3E). The sphingolipid pathways are shown in Figure 3F. In brief, CPE/SM, the higher-order forms of sphingolipids, are synthesized from ceramide by CPE/SM synthase (Vacaru et al., 2013). In contrast, lysosomal acidic sphingomyelinase (SMase) breaks down these higher-order sphingolipids to ceramides (Jenkins et al., 2009). Ceramides are a pathway hub of the sphingolipid metabolism network. They are made through the de novo synthesis pathway and the salvage pathway. Serine palmitoyltransferase (SPT) (Lace in Drosophila) catalyzes the first and rate-limiting step of the de novo synthesis pathway to generate 3-ketosphinganine (Miyake et al., 1995). 3-Ketosphinganine is then converted to other sphingolipid intermediates including dihydrosphingosine and dihydroceramide in the de novo synthesis pathway. In the salvage pathway, ceramide is converted to sphingosine and then sphingosine-1-phosphate. Given the decrease in CPE observed in iPLA2-VIA mutants, we measured the levels of ceramide and other metabolic sphingolipids, sphingolipid intermediates. Out of 26 ceramides and other sphingolipid intermediates (including dihydroceramides, dihydrosphingosine, and sphingosine) detected by mass spectrometry, 18 are upregulated (Figure 4A). Taken together, our data show that CPEs are downregulated, whereas ceramides and sphingolipid intermediates are upregulated in iPLA2-VIA mutants. The data suggest that the loss of iPLA2-VIA disrupts

iPLA2-VIA Mutants Progressively Lose Photoreceptors and Develop Aberrant Synapses To assess ultrastructural defects in retina and lamina, we performed TEM in 15- and 30-day-old flies. Figure 2B, a, illustrates the structure of a fly PR. Drosophila compound eyes are composed of well-organized repeating units called ommatidia. In each ommatidium, R1-6 surround R7/8. R1-6 project axons to the lamina where they form cartridges. These cartridges contain the terminals of R1-6 in the periphery and the terminals of postsynaptic cells in the center. In iPLA2-VIA mutants, PR number and morphology appear normal in newly eclosed flies (Figure 2C). However, PRs with aberrant morphology and loss of some PRs are observed at day 15, and these phenotypes progressively worsen by day 30 (Figures 2B and C). Interestingly, mutant PRs contain TVS-like structures in 28% of the ommatidia (Figure 2D), suggesting that iPLA2-VIA mutants develop membrane defects that resemble those observed in INAD patients or PLA2G6 knockout mice (Beck et al., 2011). TEM images of the PR terminals in the lamina revealed numerous defects in mutant iPLA2-VIA flies. First, the postsynaptic dendrites (the blue-shaded region) are swollen by day 15 (Figure S3A, a and b) and become dense black masses that resemble the high-density TVS by day 30 (Figure S3A, a’ and b’). Second, the size of the presynaptic PR terminals (R1-R6) is significantly expanded at day 30 (Figure S3A, a’ and b’). Third, these PR terminals exhibit a reduction in the density of synaptic vesicles that is observed at day 2–3 and worsens over time (Figures S3A and S3B). Fourth, we observe a decrease in capitate projections (Figure S4A), suggestive of a loss in synaptic transmission. Fifth, the membrane of the presynaptic terminals are deco608 Cell Metabolism 28, 605–618, October 2, 2018

Figure 2. iPLA2-VIA Mutants Display an Age-Dependent Loss of Neuronal Activity (A) ERG recordings of 10- and 25-day-old flies of the indicated genotypes. The ERG depolarization and ERG on-transient amplitudes were quantified. Error bars represent SEM (n > 15); ***p < 0.001. (B and C) (B) (a) An image illustrating the structure of fly photoreceptors. (b–b’) TEM images of 15-day-old fly retina. Open arrow indicates morphologically normal PR (n = 20). Filled arrow indicates abnormal PR. *, mitochondria; **, rhabdomere. The degeneration of PR in mutants is quantified in (C). (D) Enlarged image of a TVS-like structure observed in the retinal section of PRs (n = 20). See also Figures S2–S4.

the homeostasis between higher-order sphingolipids (CPE/SM) and ceramides, as well as sphingolipid intermediates. CPE/SM and ceramides are important membrane components that are enriched in the plasma membrane (van Meer

et al., 2008) where they form subdomains called lipid rafts that play an important role in cell signaling (Castro et al., 2014). In vitro studies have shown that the addition of acidic SMase to SM-containing vesicles increases ceramide-enriched domains, Cell Metabolism 28, 605–618, October 2, 2018 609

A

C

Figure 3. Loss of iPLA2-VIA Leads to an Accumulation of Sphingolipid Intermediates

B

(A) The percentage of glycerophospholipid composition of the indicated genotypes. PA, phosphatidic acid; PG, phosphatidylglycerol; PI, phosphatidylinositol; PS, phosphatidylserine; PC, phosphatidylcholine; PE, phosphatidylethanolamine. (B and C) The percentage of lysophospholipid composition of the indicated genotypes. Lyso-PE (B) and lyso-PC (C) (n = 3). NS, not significant. (D) Fold change of 14C18 CPE levels in the mutants compared with the controls (n = 3). (E) Fold change of CPE levels in iPLA2-VIA mutants. The dashed line represents the relative CPE levels in the controls. (F) The CPE metabolism pathways. Error bars represent SEM; ***p < 0.001. See also Figure S5.

D

E

F

reduces the membrane’s fluidity and disrupts vesicle stability (Goni and Alonso, 2009; Lopez-Montero et al., 2007). Hence, loss of iPLA2-VIA may lead to an accumulation of ceramides, which causes cell membrane defects leading to the development of TVS-like structures (Figures 2D and S3A, b’), disruption of the endocytosis of synaptic vesicles (Figures S3B and S3C), and a reduction in capitate projections numbers (Figure S4A). To assess if decreasing ceramides and sphingolipid intermediates may suppress the neurotransmitter defects and bangsensitivity, we tested desipramine (DES), a tricylic antidepressant that is an acidic SMase inhibitor (Figure 3F) that increases CPE and decreases ceramide levels (Figure 3F) (Jenkins et al., 2011). DES significantly suppresses the ERG on-transient defects as well as the bang-sensitivity phenotype associated with iPLA2-VIA mutants (Figure 4B). To further assess if the increase in ceramides and sphingolipid intermediates or the decrease in CPE is at the root of these defects, we fed flies myriocin, an inhibitor of SPT (Lace in Drosophila) that suppresses de novo sphingolipid synthesis. As shown in Figures 4C and 4D, myriocin suppresses the bang-sensitivity phenotype (Figure 4C) and re610 Cell Metabolism 28, 605–618, October 2, 2018

duces the loss of ERG amplitude (Figure 4D) observed in iPLA2-VIA mutants. Similarly, reducing the expression of lace using an RNAi reduces bang-sensitivity and ERG defects in iPLA2-VIA mutants as well (Figures 4C and 4D). Finally, expression of Lace RNAi also suppresses PR loss from 5.3 ± 0.3 per ommatidia in iPLA2-VIA mutants to 7 ± 0 per PR in the rescues. Taken together, these data strongly indicate that elevation of ceramides and sphingolipid intermediates in iPLA2-VIA mutants plays a critical role in the functional demise of neurons. Based on previous data, this enhances the stiffness of cell membranes and affects membrane bending (Goni and Alonso, 2009; Lopez-Montero et al., 2007). We observed a reduction in mitochondrial complex I (CI) activity in iPLA2-VIA mutant (Figure S6B), consistent with an observation made using an iPLA2-VIA hypomorphic allele (Kinghorn et al., 2015). The ubiquitous expression of Ndi1p, a yeast NADH dehydrogenase that has been shown to bypass defects in CI (Vilain et al., 2012), somewhat suppresses the ERG defects in iPLA2-VIA mutants (Figure S6A, left). However, addition of myriocin and DES do not augment the Ndi1p induced improvement in iPLA2-VIA mutants, arguing that the mitochondrial defect is also dependent on the ceramide pathway (Figure S6A, right). Indeed, suppression of the sphingolipid de novo synthesis also partially restores CI activity (Figure S6B). These data indicate that the disruption of sphingolipids plays a role in mitochondrial dysfunction and promotes the demise of neurons in iPLA2-VIA mutants. Loss of iPLA2-VIA Impairs Retromer Function and Promotes Lysosomal Accumulation of Ceramides Ceramides are synthesized in the ER and transferred to the Golgi by ceramide transfer protein (CERT) (Rao et al., 2007) via a vesicle-independent pathway. In the Golgi, ceramides are

A

Dihydroceramide Dihydrosphingosine

Ceramide Fold change

5.0 4.0 3.0 2.0

Sphingosine

Loss of iPLA2-VIA leads to an increase in ceramide and its derivatives in adult heads (Day 20)

Control ∆174

1.0

14

C Co -1 n 1 4 8 : 1 tro C - l 14 14 Ce - r 14 C16 Ce r 14C1 -Ce 14 C 8-C r 2 C 0 er 2 1 0: Ce 14 4C 1-C r C 22 er 2 1 2: C 14 4C 1-C er C 24 er 1 24 -C 1 4 4 C : 1 - Ce r 14 C2 26- er d 6 C 14 hC :1- er C 14dhC14- er 14 dh 16 Ce dh C -C r 1 1 C 8 er 14 4d 20: -Ce dh hC 1-C r 1 C2 20 er 14 4dh 0: -Ce dh C 1-C r 2 14 C2 2-Cer 14 dh 2:1 e dh C -C r 2 1 C2 4- er 14 4dh 4: Ce dh C 1-C r C 26 er 26 -C 14 :1- er dh C e S r 1 4 ph Sp h

0

Bang-sensitivity:

ERG on-transients: 3.0 * 2.5 2.0 1.5 1.0 0.5 0

Recovery time (Sec.)

Amplitude (mV)

25

**

20

Blocking sphingolipid synthesis suppresses iPLA2-VIA induced Bang-sensitivity (Day 15)


B Blocking acidic SMase rescues iPLA2-VIA defects C

Figure 4. Elevation of Sphingolipids Is a Cause of Neurodegeneration in iPLA2-VIA Mutants (A) The relative change of sphingolipids levels in iPLA2-VIA mutants compared with the controls. The dashed line represents the relative sphingolipid intermediates levels in controls. (B) DES suppresses the loss of ERG on-transient amplitude (left) (n > 10) and bang-sensitivity (right) (n = 60). (C) Bang-sensitivity in iPLA2-VIA mutants with the indicated manipulations (n = 60). (D) ERG assays in iPLA2-VIA mutants with the indicated manipulations (n > 10). Error bars represent SEM; *p < 0.05; **p < 0.01; ***p < 0.001.

*

25 20 15 10 5 0

***

l

nt

ro

R ;∆ 17 4 G R + M ;∆ yr 17 io 4 ci n ∆ A 174 R ct> N A L i; ac ∆1 e 7 + ∆ 4 M 17 yr 4 io C cin o

es

+D

74

G

es

74

∆1

∆1

4

17

4+

;∆

74

∆ M 17 yr 4 io ci A n R ct> N A L i; ac ∆1 e 74

+

∆1

4 G R M ;∆ yr 17 io 4 ci n +

17

;∆

R

G

M

+

74

∆1

∆1 yr 74 io ci A n R ct> N A L i; ac ∆1 e 74

+

G

R

;∆

17

4

G R M ;∆ yr 17 io 4 ci n

Amplitude (mV)

G

G

R

R

;∆

G

R

74

∆1

D

17

es

74

∆1

+D

D

4+

17

;∆

G

R

;∆

17

es

4

The latter may increase the stiffness of the retromer membrane leading to a pos10 itive feedback loop. 5 The above hypothesis predicts a 0 disruption of the retromer complex, an increase in lysosomal stress and suggests that promoting retromer function may D Blocking the de novo sphingolipid synthesis pathway rescues iPLA2-VIA ERG defects (Day20) reduce ceramide levels and suppress degeneration. Indeed, we observed a ERG amplitude: ERG on-transient: *** *** decrease in Vps26 punctae in iPLA2-VIA 14 3.0 ** *** *** *** mutant PRs (Figure 5A), a phenotype 12 2.5 10 that is indicative of impaired retromer 2.0 8 1.5 function (Wang et al., 2014), suggesting 6 1.0 4 a decrease in retromer-derived retrieval 0.5 2 from endosomes (Wang and Bellen, 0 0 2015). To assess if lysosomal function is altered in iPLA2-VIA mutants, we stained Drosophila fat bodies with Lysotracker. As shown in Figure 5B, there is a robust increase in both number and size of lysotransformed into CPE or SM. These phospholipids are then somes in iPLA2-VIA mutants. Similarly, TEM analysis of mutant embedded in vesicles and transported to the plasma membrane PRs show a strong and progressive increase in the number where they serve as key components of lipid rafts (Taniguchi and and size of lysosomes and multivesicular bodies (Figure 5C). Okazaki, 2014). CPE/SM, as integral components of the cell These data indicate that retromer function is impaired and that membrane, are typically endocytosed where they undergo it induces lysosomal stress. If iPLA2-VIA regulates the retromer, loss of vps26 or vps35 retromer-dependent retrograde transport to Golgi or plasma membrane (Choudhury et al., 2002). The CPE/SM that are not should also cause an increase in ceramide. Given that Vps26 or retrieved via the retromer, traffic to lysosomes where they are hy- Vps35 are key players in the recycling process, their loss may drolyzed to ceramides and sphingolipid intermediates by acidic create more sever phenotypes. As shown in Figure 5D, we SMase (Frohlich et al., 2015). observed a dramatic ceramide accumulation in Drosophila The retromer complex consists of two major parts: a trimeric 5-day-old mutant eye clones lacking vps26 or vps35. These cargo recognition complex (Vps26/Vps29/Vps35), which binds data again indicate that loss of iPLA2-VIA impairs retromer functo a cargo receptor, and a membrane-deforming complex (sort- tion that leads to an accumulation of ceramides. However, to ing nexins) (Burd and Cullen, 2014). The trimeric Vps26/Vps29/ attain the levels of ceramide observed in vps26 and vps35 mutant Vps35 complex engages endosomal cargo and recruits the cells, loss of iPLA2-VIA typically takes 15–20 days, suggesting a WASH complex to promote actin polymerization, which creates milder impairment in retromer function when iPLA2-VIA is lost. a small membrane subdomain from the endosome promoting We further assessed whether iPLA2-VIA forms a complex with sorting nexin-mediated tubulation to transport cargo to Golgi the key retromer proteins, Vps26 or Vps35. We tested the local(Wang and Bellen, 2015). Hence, increased ceramide levels (Fig- ization of the endogenous C-terminal GFP-tagged iPLA2-VIA ure 4A) may be caused by retromer-mediated transport, which, (iPLA2-VIA-GFP; Figure 1A) in fat bodies of the third-instar larva in turn, promotes CPS/SM trafficking to lysosomes to generate and adult PRs. As shown in Figure S6C, endogenous iPLA2-VIA elevated levels of ceramides and sphingolipid intermediates. is homogenously distributed in the cytoplasm in both tissues. 15


Control

Δ174

5 μm

Vps26 Phalloidin

Δ174

Control

* 40 30 20 10 0

50 μm ControlΔ174

Lysotracker Control ∆174

Lysosome

800 nm MVB

Lysosomes per PR

Loss of iPLA2-VIA leads to a progressive increase in lysosomal number 400 nm Control ∆174

Day 15

C

of iPLA2-VIA enhances lysosome B Loss size and number in Drosophila fat body

Loss of iPLA2-VIA reduces Vps26 punctae Vps26 punctae/ Field

A

2.0 *** 1.8 1.6 1.4 1.2 1.0 *** 0.8 0.6 0.4 0.2 0 Day 2-3 15 30

D Loss of Vps26 or Vps35 leads to F

4 ∆1 ∆1 74 74 +R 55 G R ;∆ 17 A ∆ 4 ct 1 >V 7 ∆ p 4 A 174 s26 ct ; > ∆1 Vps 74 35 ;

** **

(A) Vps26 punctae (green) marks retromer (indicated by arrows). Phalloidin (red) labels rhabdomere (left). The number of Vps26-positive punctae is quantified (right) (n = 10). (B) Lysotracker staining of fat bodies of the indicated genotypes in third-instar larva. (C) TEM images of PR at the retinal section. Open arrows, lysosomes (enlarged at right); gray arrow, MVBs (enlarged at right). Right panel, quantification of lysosome numbers (n = 30 per genotypes). (D) Immunofluorescence staining of fly eyes. Phalloidin, red (rhabdomere); anti-ceramide antibody, green. (E) Immunoprecipitation assays. Lysate from the S2 cells transfected with the indicated constructs were immunoprecipitated by the indicated antibodies. The untagged human iPLA2-b was used to antagonize the interaction (right). (F) ERG assays in iPLA2-VIA mutants with the indicated manipulations (n > 10). Error bars represent SEM; *p < 0.05; **p < 0.01; ***p < 0.001.

17 ;∆

;∆

17

4 ∆1 ∆1 74 74 +R 55 G R ;∆ 17 4 A ct ∆1 >V 7 ∆1 ps 4 A 74 26 ct ; > ∆1 Vps 74 35 ;

Amplitude (mV)

Amplitude (mV)

Promoting retromer function suppresses loss of iPLA2-VIA-induced ERG defects ceramide accumulation 5 μm ERG on-transient: Ceramide ERG amplitude: *** 3.5 Phalloidin 3.5 14 * 16 *** 3 3.0 12 12 2.5 10 * 2 2.0 8 8 Control ∆174 1.5 6 1 1.0 4 4 0.5 2 0 0 0 0

Figure 5. Loss of iPLA2-VIA Impairs Retromer Function and Promotes Ceramide Accumulation

612 Cell Metabolism 28, 605–618, October 2, 2018

R

G

vps35 ∆MH20

G

2

R

mutant flies with R55, a pharmacological chaperone enhancing retromer stability E iPLA2-VIA interacts with the retromer proteins, Vps26 and Vps35 IP: Flag and function (Mecozzi et al., 2014). As Lysate Lysate IP: IP: iPLA2-VIA-PA-V5 + + + (1:10) (1:10) shown in Figures 5F and S6D, R55 signifFlag HA Vps26-Flag + + + Empty vector + Empty vector + iPLA2-β - 0.5 1.25 icantly suppresses the bang-sensitivity + + + iPLA2-VIA-PA-V5 + iPLA2-VIA-PA-V5 Empty vector 1.25 0.75 Vps35-HA Vps26-Flag and ERG defects. Similarly, ubiquitous + + + + WB: V5 expression of Vps26 or Vps35 supWB: V5 WB: V5 WB: iPLA2-β presses both bang-sensitivity and ERG WB: Flag WB: HA defects. Importantly, TEM reveals that Lysate WB: Actin WB: Actin expression of Vps26 suppresses the WB: Flag loss of PR and the number of aberrant WB: Actin PR in the retina at day 20 (Figure 6A). In addition, the lysosomal accumulation observed in iPLA2-VIA mutants is supHowever, it also forms punctae that partially colocalize with pressed (Figure 6B). Moreover, sections through the lamina Vps26 (Figure S6C). Given the much broader distribution of show (1) a suppression of the loss of the postsynaptic dendrites iPLA2-VIA compared with Vps26, we determined if iPLA2-VIA (the blue-shaded region), (2) a suppression of the expansion of is able to interact with Vps26 and Vps35. As shown in Figure 5E, the size of the presynaptic PR terminals, (3) a suppression of iPLA2-VIA is immunoprecipitated by either Vps26 or Vps35, the reduction in density of synaptic vesicles, and (4) a suppressuggesting that these three proteins can interact. Importantly, sion of the decrease in capitate projections (Figures 6C and overexpression of untagged human iPLA2-b antagonizes these 6D). Hence, Vps26 expression potently suppresses iPLA2-VIAinteractions, suggesting that they are evolutionarily conserved. induced degenerative features, further arguing that loss of In summary, our data indicate that iPLA2-VIA forms a complex iPLA2-VIA affects retromer function. To access the interaction between sphingolipid metabolism with retromer components and that it functions to improve retromer function and retrieval of CPE. and retromer function, we treated iPLA2-VIA mutants with both DES and R55. As shown in Figure 6E, DES and R55 do not synergize, suggesting that they function in the same pathway. InterUbiquitous Expression of Vps26 or Vps35 Alleviates estingly, ubiquitous expression of Vps26 suppresses ceramide Lysosomal Stress and Suppresses iPLA2-VIAAssociated Neuronal Defects accumulation in iPLA2-VIA mutants (Figure 6F). These data Based on the data above, we surmised that promoting Vps26/ indicate that enhancing retromer function in iPLA2-VIA Vps29/Vps35 complex activity may facilitate the recycling of mutants reduces ceramide levels, lysosomal stress, and alleviCPE/SM from endosomes, reduce the accumulation of ceram- ates numerous other phenotypes associated with the demise ides in endolysosomal compartments, and relieve the degenera- of neurons, including synaptic function, bang sensitivity, and tive phenotypes in iPLA2-VIA mutants. We therefore treated ultrastructural features. vps26

A Ubiquitous expression of Vps26 suppresses PR loss Act>Vps26, w+; Δ174

*

*

3

F Ubiquitous expression of Vps26

suppresses ceramide accumulation in iPLA2-VIA mutants

*** ***

5 μm

2 Control

Δ174

# of aberrant PR

in the retina (Day 20)

1

G R ;Δ

2 μm

*

Ceramide Phalloidin

(A) TEM images of 20-day-old fly retina of the indicated genotypes. Aberrant PR is quantified (n = 3). (B) The number of lysosomes per PR is highlighted and quantified (n = 3 per genotypes). Open arrows: lysosomes; *pigment bodies in w+. Yellow dashed line defines a PR cell. (C) TEM images of 20-day-old fly lamina of indicated genotypes. R, PR presynaptic R1-6 axon projections in the lamina. L, postsynaptic projection neurons. Yellow dashed line defines a cartridge. Degenerated postsynaptic region is highlighted in blue. Open arrow, capitate projections (CPs); #mitochondria; **synaptic vesicles. (D) Quantification of presynaptic vesicle density (left) and capitate projections (CP) (right) in (C) (n = 30 per genotypes). (E) Bang-sensitivity assays for flies with the indicated genotypes treated with R55, DES, or a combination of R55/DES. (F) Immunofluorescence staining of fly eyes of the indicated genotypes. Phalloidin, red (rhabdomere); anti-ceramide antibody, green. Arrows, ceramide staining; *pigment autofluorescence. Error bars represent SEM; *p < 0.05; **p < 0.01; ***p < 0.001; NS, not significant. See also Figure S6.

Ceramide

Δ174

17 4 A ct Δ1 >V 7 Δ p 4 17 s2 4 6;

0

*

*

**

**

G R ;Δ

17 4 Δ A ct 17 >V 4 Δ p 17 s2 4 6;

*

1.2 1.0 0.8 0.6 0.4 0.2 0

800 nm

C Ubiquitous expression of Vps26 suppresses loss of PR * * *

*

*

Δ174

synapses in the lamina (Day 20)

GR; Δ174

*

Act> Vps26 ; Δ174

Act>Vps26, w+; Δ174

Lysosome per PR

Δ174

Act> Lace RNAi; Δ174

B Ubiquitous expression of Vps26 suppresses lysosome increase in the retina (Day 20)

Act>Vps26; Δ174

**

Figure 6. Ubiquitous Expression of Vps26 or Vps35 Alleviates Lysosomal Stress and Suppresses iPLA2-VIA-Associated Neuronal Defects

#* L * * ***

R 1 μm

1 μm

1 μm

*

# of CP /Axon terminal

Vesicle density (%)

100 80 60 40 20 0

GR; Δ174 Δ174 Act>Vps26 ; Δ174

5 4 3 2 1 0

***

GR; Δ174 Δ174 Act>Vps26 ; Δ174

R55 with DES does not further E Combining improve bang-sensitivity in iPLA2-VIA mutants


Ubiquitous expression of Vps26 suppresses loss of presynaptic vesicles and capitate projections Vesicle density: Capitate projection (CP): * *** 120 6

D

45 40 35 30 25 20 15 10 5 0

Golgi marker, in mammalian cells. In control cells, C6-NBD-ceramide binds NS to the plasma membrane and is internalized and traffics to the Golgi complex (Figure S7C, upper panels). Interestingly, BEL induces mistrafficking to and redistribution of C6-NBD-ceramides in Δ174 Δ174 Δ174 lysosomes (Figure S7C, bottom panels). +R55 +DES +R55+DES These data suggest that the suppression of iPLA2-b activity disrupts ceramide trafficking and causes its accumulation in lysosomes. We further investigated the role of ceramide accumulation in inducing lysosomal stress. As shown in Figure S7B, BELtreated HeLa cells show an increase in Lamp1-positive signal, and myriocin suppresses BEL-induced lysosomal accumulation. We further tested the response of a neuroblastoma cell line, Neuro-2A (N2A), upon suppression of iPLA2-b. BEL treatment also leads to an increase in lysosome activity in N2A cells (Figure S7D). Hence, loss of iPLA2-b activity in two different vertebrate cell types disrupts retromer function, causes an accumulation of ceramides, and promotes lysosomal stress. To further elucidate the role of PLA2G6 in mammalian cells, we selected two small interfering RNAs (siRNAs) targeting PLA2G6. When transfected into N2A cells, these siRNAs suppress iPLA2b expression (Figure 7A) and cause a reduction in Vps26 and Vps35 (Figure 7A), similar to flies. In addition, suppressing iPLA2-b expression using two independent RNAi also leads to an elevation of ceramides and other sphingolipid intermediates in N2A cells (Figure 7B). Moreover, loss of iPLA2-b in N2A cells *

**

GR; Δ174 Δ174

Loss of iPLA2-VIA in Vertebrate Neuroblastoma Cells Affects the Abundance of Vps26 and Vps35 and Causes an Increase in Ceramide and an Expansion of Lysosomes To test if sphingolipid homeostasis and retromer function are similarly affected in vertebrate cells, we first used a wellestablished inhibitor of iPLA2-b activity, bromoenol lactone (BEL) (Balsinde and Dennis, 1996; Hazen et al., 1991). As shown in Figure S7A, BEL treatment leads to a significant decrease of Vps26 and Vps35 levels, consistent with our observations in Drosophila PRs. These data indicate a role for iPLA2-b in the regulation of retromer integrity or function. The suppression of iPLA2-b activity by BEL also leads to a significant increase in Lamp1-positive lysosomes in HeLa cells (Figure S7B), indicating that loss of iPLA2-VIA causes endolysosomal stress, similar to that observed with mutants that cause lysosomal storage diseases (Jansen et al., 2017; Kinghorn et al., 2016). To test if ceramide trafficking is impaired upon suppression of iPLA2-b activity, we tracked the subcellular localization of C6-NBD-ceramide, a well characterized

*

NS


PLA2G6-targeting siRNAs reduce Vps26/Vps35 retromer levels

siRNA Noncoding #134 #135 iPLA2-β

B Fold change

A

Vps26 Vps35

1.6 1.4 1.2 1.0 0.8 0.6 0.4 2.8

siRNA#135:

2.2 1.6 1.0 0.4 C o C ntr 1 o C 4-C l 16 e C - r C 18 C e 18 -C r : C 1- er C 20 Ce 20 -C r : C 1-Cer C 22 er 22 -C :1 er C -C C 24- er 24 C : e C 1- r C 26 Cer 2 dh 6 : 1 C e r dhC14-Ce C - r d 1 C dh hC 6-Cer C 18 er d 1 8 -C dh hC :1-Cer C 20- er d 20 Ce dh hC :1-C r C 22- er 2 C dh 2:1 er dh C - C C 24 er dh24: -Ce dh C 1-C r C 26- er 26 C :1 er dh-Ce Sp r h Sp Sp h- h 1P

Actin

Fold change

Actin

Suppression of PLA2G6 expression increases levels of ceramides and its derivatives siRNA#134: 1.8

C

Suppression of PLA2G6 expression promotes lysosomal stress and causes cell loss that is alleviated by myriocin, DES and R55 20 μm

NC

#134

#135

D Expression of a-Syn reduces retromer levels GFP

a-SynGFP

GFP

a-SynGFP

SNCAGFP

Vps26 Vps35

GFP

Actin Lamp1 #135+Myriocin

E Expression of a-Syn increases lysosomal size #135+R55

#135+DES

GFP

Lamp2 DAPI

Expression of a-Syn elevates ceramide levels 6.8

Fold change

F

SNCA-GFP

20 μm

G 5.2 4.8

Myriocin suppresses a-Syn toxicity in Drosophila eyes

Figure 7. The Mechanism that Leads to Loss of iPLA2-VIA-Induced Neurodegeneration Is Evolutionarily Conserved in Vertebrate Cells (A) The expression levels of the indicated proteins in N2A cells transfected with control or PLA2G6 siRNAs are shown. (B) The relative changes of the indicated sphingolipid levels in PLA2G6-targeting siRNA-transfected N2A cells. The dashed line represents the relative sphingolipid intermediates levels in the controls. (C) Lamp1 staining in N2A cells treated with indicated siRNAs in the presence or absence of myriocin, DES, or R55. (D) The expression levels of the indicated proteins in a human medulloblastoma cell line, Daoy cells, expressing GFP-tagged a-Syn or GFP control. (E) Lamp2 staining Daoy cells expressing GFPtagged a-Syn or GFP control. Arrows indicate lysosomes. (F) The relative changes of the indicated sphingolipid levels in GFP-tagged a-Syn-expressing cells compared with the GFP-expressing controls. The dashed line represents the relative sphingolipid intermediates levels in the controls. (G) Toxicity caused by the photoreceptor-specific (Rh1-GAL4) expression of a-Syn in retina is suppressed by myriocin. Arrows indicate degenerated tissue. See also Figures S7 and S8.

3

2

useful in isolating a suppressor of a-Syn toxicity that was subsequently validated in flies and mice (Rousseaux et al., 2016). We found that expression of a-Syn, but not a GFP control, significantly reduces the levels of Vps26 and Vps35 (Figure 7D), promotes lysosomal Rh1>a-Syn+ Myriocin expansion (Figure 7E), and causes ceramides to accumulate (Figure 7F). Hence, these data indicate that the phenotypes associated with loss of PLA2G6 may also play a role in many forms of PD. To determine that the toxicity associated with a-Syn overexpression is due to ceramide accumulation, we expressed a-Syn in Drosophila eyes using the Rh1-GAL4 driver, a driver that is only expressed in mature PRs. This causes a progressive and retinal degeneration that is quite severe in 15-day-old animals (Figure 7G) (Chouhan et al., 2016). Feeding these flies with myriocin significantly suppresses retina degeneration (Figure 7G), again indicating that ceramide accumulation plays a role in a-Syn-mediated neurodegeneration. Rh1>a-Syn

1

C on C tro 14 l C -Ce 16 r C -Ce C 18- r 18 C :1 er C -Ce C 20- r 20 C e C :1- r 20 C :4 er C -Ce C 22- r 22 C :1 er C -Ce C 24- r 24 C :1 e r C -C C 26- er 26 C dh :1 er C -C dh 14 er C -C dh 16 er dh C1 -Ce C 8- r 18 C dh :1 er dh C2 -Ce C 0- r 20 C dh :1- er dh C2 Cer C 2-C 2 dh 2:1 er dh C2 -Ce C 4- r 24 C dh :1- er dh C2 Ce C 6- r 26 C :1 er -C dh dhS er Sp ph h1P S Sp ph h1P

0

also promotes lysosome accumulation (Figures 7C and S7E– S7E00 ) and reduces cell viability (Figure S7E%). Finally, addition of myriocin, DES, or R55 suppresses lysosome accumulation and improves cell survival (Figures 7C and S7E%). These data strongly suggest that the mechanism that leads to loss of iPLA2-VIA-induced neurodegeneration in Drosophila is evolutionarily conserved in vertebrate cells. Overexpression of a-Syn in Vertebrate Medulloblastoma Cells Affects the Abundance of Vps26 and Vps35 and Causes an Increase in Ceramide and an Expansion of Lysosomes Given that mutations in vps35 have been associated with a-Syn aggregation and PD (Follett et al., 2016), and given that a-Syn accumulation is a pathological hallmark of PD, we tested if a-Syn overexpression affects the same parameters as those that we observe in vps35, vps26, and PLA2G6 mutant cells or animals. We used Daoy cells that constitutively express GFP-tagged a-Syn as these cells have been proved 614 Cell Metabolism 28, 605–618, October 2, 2018

DISCUSSION We show that loss of iPLA2-VIA reduces lifespan, starvation tolerance, and ATP levels and impairs bang-sensitivity (Figures 1C–1H), consistent with previous findings (Beck et al., 2011; Kinghorn et al., 2015). We find that iPLA2-VIA is localized to the neuronal and axonal cytoplasm and is enriched in the

neuropil as well as in punctae containing Vps26 (Figures S1C and S6C). The gene encodes two proteins (PA and PB), and expression of either isoform rescues the phenotypes associated with the loss of iPLA2-VIA. ERG recordings and TEM at different ages reveal a progressive demise and loss of neurons and severe ultrastructural defects at synapses of PR. TEM analyses further reveal an accumulation of TVS- and omega-like structures and a disruption of mitochondrial inner membranes. These observations are consistent with phenotypes present in patients and PLA2G6 knockout mice (Beck et al., 2011; Kinghorn et al., 2015; Malik et al., 2008; Shinzawa et al., 2008; Sumi-Akamaru et al., 2015) and highlight the conservation of the underlying pathways in Drosophila. An established function of PLA2s is to hydrolyze glycerol phospholipids to generate an FA and an LPL, whereas LPL acyltransferases catalyze the reverse reaction (Figure S1A). A loss of PLA2 should cause a reduction of LPL. Since LPL promotes and initiates outward membrane bending and facilitates membrane tubulation in vitro, a reduction in LPL disrupts membrane curvature and tubulation (Figure S1B) (Brown et al., 2003). In contrast to Beck et al. (2011), who found variable decreases and increases in PC and PE in a single mass spectrometry analysis of spinal cord of PLA2G6 knockout mice, we find that levels of phospholipids are not affected in iPLA2-VIA fly mutants. Moreover, total PLA2 enzymatic activity is not affected in iPLA2-VIA mutants, and expression of a catalytically impaired iPLA2-b rescues reduced lifespan. Hence, we conclude that the Lands cycle is not, or very mildly, affected in iPLA2-VIA mutants. In contrast, our data clearly implicate sphingolipid metabolism: CPE/SM are downregulated, whereas sphingolipid intermediates, including ceramides, are upregulated. Consistent with these observations, genetically or pharmacologically suppressing ceramide synthesis reduces the defects in synaptic transmission and bang-sensitivity associated with loss of iPLA2-VIA, providing compelling in vivo evidence of the involvement of ceramide accumulation in iPLA2-VIA-mediated neuronal defects. The small head group of ceramide increases its hydrophobicity and stabilizes membranes. In fact, loss of CERT in Drosophila reduces ceramide levels and has been shown to increase membrane fluidity (Rao et al., 2007). In contrast, when present in excess amounts, ceramides aggregate with other phospholipids to form gel-like membrane domains (Goni and Alonso, 2009; Wang and Silvius, 2003), decreasing membrane fluidity and curvature (Chiantia et al., 2006), destabilizing vesicles, and blocking membrane tubulation (Lopez-Montero et al., 2007). Hence, an excess of membrane ceramides may cause defects in endocytosis, affect capitate projection required for endocytosis, and promote TVS-like structures. In addition to the above-mentioned membrane defects that seem to be affected by ceramide accumulation, our data indicate an important role for PLA2G6 in regulating the function of the retromer. The following observations argue that PLA2G6 is primarily required at the retromer. First, we observed reduced levels of Vps26 or Vps35 in iPLA2-VIA mutants. Second, loss of iPLA2-VIA affects Rh1 recycling in PRs, a process that is retromer dependent (Wang et al., 2014). Similarly, loss of PLA2G6 in vertebrate cells impairs the recycling of transferrin receptor (de Figueiredo et al., 2001), which has been shown to rely on retromer function (Chen et al., 2013). Third, restoring retromer

transport by overexpressing Vps26 or Vps35 in iPLA2-VIA mutants suppresses ceramide accumulation and neurodegenerative phenotypes. Fourth, feeding iPLA2-VIA mutants a retromer chaperone that promotes the interaction between Vps29 and Vps35, R55, suppresses neurodegeneration. Fifth, loss of iPLA2-VIA, as well as Vps26 and Vps35, causes ceramide accumulation, and a progressive increase in lysosome number and size in flies (Wang et al., 2014). Sixth, iPLA2-VIA interacts with Vps26 and Vps35. Note that iPLA2-VIA contains eight N-terminal Ankyrin repeat domains (Hsu et al., 2009), and that Vps26 and Vps35 have previously been reported to interact with Ankyrin repeat domain-containing proteins, such as VARP (Hesketh et al., 2014) and Rank5 (Zhang et al., 2012). Finally, suppressing sphingolipid synthesis reduces lysosomal expansion caused by the loss of iPLA2-VIA, suggesting a positive feedback loop. The above-mentioned features that were tested in vertebrate cells are also conserved, arguing that vertebrate PLA2G6 regulates the stability/function of the retromer complex. This mechanism further modulates sphingolipid homeostasis and affects membrane dynamics. We propose that loss of iPLA2-VIA impairs the proper function of the retromer. This in turn interferes with CPE/SM recycling and promotes CPE/SM hydrolysis in the endolysosomal compartment (Frohlich et al., 2015), causing an accumulation of ceramides and sphingolipid intermediates. Ceramide accumulation stiffens the membrane and further impairs the retromer. This positive feedback loop eventually promotes the demise of neurons. Hence, a reduction in the ceramide levels or an improvement in retromer function suppresses the iPLA2VIA/PLA2G6-associated phenotypes. Variants in vps35 have been shown to cause PD (PARK17) (Vilarino-Guell et al., 2011; Zimprich et al., 2011). Loss of vps35 in flies, primary cultured neurons, or the expression of Vps35 D620N in cultured neurons cause endolysosomal dysfunction (MacLeod et al., 2013; Wang et al., 2014). Moreover, mutations in GBA, a gene that encodes glucocerebrosidase and regulates sphingolipid metabolism, causes lysosomal storage disorder and is a risk factor for PD (Kinghorn et al., 2017). These studies support the involvement of endolysosomal dysfunction and sphingolipid imbalance in PD. Interestingly, endolysosomal dysfunction is also observed in other PD models, including a-synuclein (PARK1) (Manzoni and Lewis, 2013), Parkin (PARK2) (Dehay et al., 2013), and LRRK2 (PARK8) (Henry et al., 2015). In Drosophila, overexpression of Vps35 reduces locomotor defects and prolongs lifespan in LRRK2 and parkin mutants (Linhart et al., 2014; Malik et al., 2015), suggesting that they affect the same pathway. In addition, Vps35 depletion by RNAi exacerbates a-Syn-induced locomotor impairment and eye disorganization (Miura et al., 2014). We show that loss of vps26 and vps35 not only causes endolysosomal dysfunction but also leads to ceramide accumulation. More importantly, a-Syn overexpression also disrupts retromer components by decreasing Vps26 and Vps35 levels, leads to ceramide accumulation, and expands lysosomal size. These data suggest that the endolysosomal-mediated regulation of sphingolipid homeostasis may play a broader role in the pathogenesis of PD. Finally, it is interesting to note that PD patients show elevated levels of ceramide in the plasma (Mielke et al., 2013). In summary, our data show a hereto unknown role of PLA2G6/ PARK14 in sphingolipid homeostasis and endolysosomal function Cell Metabolism 28, 605–618, October 2, 2018 615

(Figure S7F). Loss of PLA2G6/PARK14 causes a loss of retromer integrity, which promotes CPE/SM mistrafficking and endolysosomal dysfunction. The misregulated sphingolipid metabolism increases the levels of ceramides and sphingolipid intermediates, which further disrupts retromer function and augments lysosomal stress and contributes to the demise of neurons. Suppressing sphingolipid synthesis or enhancing retromer function alleviates the ceramide accumulation and suppresses the neurodegenerative phenotypes in iPLA2-VIA mutants and vertebrate cells. These findings suggest a novel therapeutic potential for manipulating sphingolipid levels or retromer function using DES or R55 in PARK14, INAD, and aNAD patients. Given that we also observe similar defects in vps26 and vps35 mutants, as well as in a-Syn-expressing cells, it is possible that the endolysosomal and sphingolipid metabolism pathway may be affected in many forms of PD. Limitations of Study Some of the limitations of this study relate to the observed mitochondrial defects and the possible link with other loci/genes that are causing PD or parkinsonism. Indeed, mitochondrial dysfunction has been identified in many models for PD or parkinsonism, including models related to Parkin (PARK2), PINK1 (PARK6), DJ-1 (PARK7), LRRK2 (PARK8), ATP13A2 (PARK9), and FBX07 (PARK15) (Haelterman et al., 2014). In addition, mitochondrial CI deficiencies and mitochondrial electron transport chain defects have been reported in the substantia nigra of PD patients (Keeney et al., 2006). Although we show that aged iPLA2-VIA mutant PRs display aberrant mitochondria characterized by the presence of disrupted inner membranes and we report a reduction in ATP levels, as well as a reduced mitochondrial CI activity, we did not investigate the underlying mechanisms. However, suppressing ceramide levels by expressing the Lace RNAi significantly restored CI activity and alleviated neurodegeneration. This raises some important questions: Do the defects in the endolysosomal pathway affect mitochondrial function? Do the accumulated ceramides affect mitochondrial functions? Do the defects in the endolysosomal pathway and mitochondria synergize to cause the demise of neurons or do they merely cause additive effects? Finally, our data suggest that accumulated ceramides may play an important role in other forms of PD or parkinsonism given the observed ceramide accumulation when a-Syn is overexpressed. STAR+METHODS Detailed methods are provided in the online version of this paper and include the following: d d d

d

KEY RESOURCES TABLE CONTACT FOR REAGENT AND RESOURCE SHARING EXPERIMENTAL MODEL AND SUBJECT DETAILS B Drosophila B Mammalian Cell Culture METHOD DETAILS B Fly Strains and Genetics B Drosophila Behavioral Assay B ERG Recording B Western Blotting


B

Drug Administration in Fly Food Molecular Cloning B Immunofluorescence Staining B Mitochondrial Functional Assay B Total PLA2 Activity Measurement B Transmission Electron Microscopy (TEM) B Sample Preparation for Phospholipidomics Analysis B Immunoprecipitation B Mammalian Tissue Culture Assay B C6-NBD Ceramide Localization Assay B Transfection of siRNA in Neuro-2A Cells B Cell Survival Assay QUANTIFICATION AND STATISTICAL ANALYSIS B

d

SUPPLEMENTAL INFORMATION Supplemental Information includes eight figures and can be found with this article online at https://doi.org/10.1016/j.cmet.2018.05.019. ACKNOWLEDGMENTS We thank Drs. Megan Campbell, Hsiao-Tuan Chao, Nele Haelterman, Paul Marcogliese, Karen Schulze, Joshua Shulman, Michael Wangler, and Shinya Yamamoto for insightful comments. We thank Lita Duraine for TEM imaging, Hui Ye and Yarong Li for Drosophila histology assays, and Hongling Pan and Yuchun He for injections to create transgenic flies. We thank the Bloomington Drosophila Stock Center (NIH P40OD018537) and Vienna Drosophila Resource Center for providing stocks and reagents and the Lipidomics Core at the Medical University of South Carolina and the Kansas Lipidomics Research Center at Kansas State University for glycerol phospholipid and sphingolipid analysis. This project was supported in part by IDDRC grant number 1U54 HD083092 from the Eunice Kennedy Shriver National Institute of Child Health & Human Development. IDDRC Neurovisualization Core was used for this project. We acknowledge support from Friedreich’s Ataxia Research Alliance, Target ALS, the Huffington Foundation, and the Robert A. and Renee E. Belfer Family Foundation to H.J.B. H.J.B. is an Investigator of the Howard Hughes Medical Institute. AUTHOR CONTRIBUTIONS G.L. conceived and designed the study, acquired the data, analyzed and interpreted the data, and drafted and revised the article. K.C. conceived and designed the study, analyzed and interpreted the data, and drafted and revised the article. P.-T.L., D.M., and K.L.T. acquired the data, analyzed and interpreted the data, and drafted and revised the article. Z.Z., W.-W.L., and L.W. acquired the data and drafted and revised the article. H.J.B. conceived and designed the study, analyzed and interpreted the data, and drafted and revised the article. DECLARATION OF INTERESTS The authors declare no competing financial interests. Received: October 18, 2017 Revised: March 23, 2018 Accepted: May 22, 2018 Published: June 14, 2018 REFERENCES Acharya, U., and Acharya, J.K. (2005). Enzymes of sphingolipid metabolism in Drosophila melanogaster. Cell Mol. Life Sci. 62, 128–142. Adibhatla, R.M., and Hatcher, J.F. (2008). Phospholipase A(2), reactive oxygen species, and lipid peroxidation in CNS pathologies. BMB Rep. 41, 560–567.

Balsinde, J., and Dennis, E.A. (1996). Distinct roles in signal transduction for each of the phospholipase A2 enzymes present in P388D1 macrophages. J. Biol. Chem. 271, 6758–6765. Beck, G., Sugiura, Y., Shinzawa, K., Kato, S., Setou, M., Tsujimoto, Y., Sakoda, S., and Sumi-Akamaru, H. (2011). Neuroaxonal dystrophy in calcium-independent phospholipase A2beta deficiency results from insufficient remodeling and degeneration of mitochondrial and presynaptic membranes. J. Neurosci. 31, 11411–11420. Brown, W.J., Chambers, K., and Doody, A. (2003). Phospholipase A2 (PLA2) enzymes in membrane trafficking: mediators of membrane shape and function. Traffic 4, 214–221. Burd, C., and Cullen, P.J. (2014). Retromer: a master conductor of endosome sorting. Cold Spring Harb. Perspect. Biol 6. Castro, B.M., Prieto, M., and Silva, L.C. (2014). Ceramide: a simple sphingolipid with unique biophysical properties. Prog. Lipid Res. 54, 53–67. Chen, C., Garcia-Santos, D., Ishikawa, Y., Seguin, A., Li, L., Fegan, K.H., Hildick-Smith, G.J., Shah, D.I., Cooney, J.D., Chen, W., et al. (2013). Snx3 regulates recycling of the transferrin receptor and iron assimilation. Cell Metab. 17, 343–352. Chiantia, S., Kahya, N., Ries, J., and Schwille, P. (2006). Effects of ceramide on liquid-ordered domains investigated by simultaneous AFM and FCS. Biophys. J. 90, 4500–4508. Choudhury, A., Dominguez, M., Puri, V., Sharma, D.K., Narita, K., Wheatley, C.L., Marks, D.L., and Pagano, R.E. (2002). Rab proteins mediate Golgi transport of caveola-internalized glycosphingolipids and correct lipid trafficking in Niemann-Pick C cells. J. Clin. Invest. 109, 1541–1550. Chouhan, A.K., Guo, C., Hsieh, Y.C., Ye, H., Senturk, M., Zuo, Z., Li, Y., Chatterjee, S., Botas, J., Jackson, G.R., et al. (2016). Uncoupling neuronal death and dysfunction in Drosophila models of neurodegenerative disease. Acta Neuropathol. Commun. 4, 62. de Figueiredo, P., Doody, A., Polizotto, R.S., Drecktrah, D., Wood, S., Banta, M., Strang, M.S., and Brown, W.J. (2001). Inhibition of transferrin recycling and endosome tubulation by phospholipase A2 antagonists. J. Biol. Chem. 276, 47361–47370. de Figueiredo, P., Drecktrah, D., Katzenellenbogen, J.A., Strang, M., and Brown, W.J. (1998). Evidence that phospholipase A2 activity is required for Golgi complex and trans Golgi network membrane tubulation. Proc. Natl. Acad. Sci. USA 95, 8642–8647. de Figueiredo, P., Polizotto, R.S., Drecktrah, D., and Brown, W.J. (1999). Membrane tubule-mediated reassembly and maintenance of the Golgi complex is disrupted by phospholipase A2 antagonists. Mol. Biol. Cell 10, 1763–1782. de Figueiredo, P., Drecktrah, D., Polizotto, R.S., Cole, N.B., LippincottSchwartz, J., and Brown, W.J. (2000). Phospholipase A2 antagonists inhibit constitutive retrograde membrane traffic to the endoplasmic reticulum. Traffic 1, 504–511. Dehay, B., Martinez-Vicente, M., Caldwell, G.A., Caldwell, K.A., Yue, Z., Cookson, M.R., Klein, C., Vila, M., and Bezard, E. (2013). Lysosomal impairment in Parkinson’s disease. Mov. Disord. 28, 725–732. Follett, J., Bugarcic, A., Yang, Z., Ariotti, N., Norwood, S.J., Collins, B.M., Parton, R.G., and Teasdale, R.D. (2016). Parkinson disease-linked Vps35 R524W mutation impairs the endosomal association of retromer and induces alpha-synuclein aggregation. J. Biol. Chem. 291, 18283–18298. Frohlich, F., Petit, C., Kory, N., Christiano, R., Hannibal-Bach, H.K., Graham, M., Liu, X., Ejsing, C.S., Farese, R.V., and Walther, T.C. (2015). The GARP complex is required for cellular sphingolipid homeostasis. Elife 4, https://doi.org/ 10.7554/eLife.08712. Goni, F.M., and Alonso, A. (2009). Effects of ceramide and other simple sphingolipids on membrane lateral structure. Biochim. Biophys. Acta 1788, 169–177. Haelterman, N.A., Yoon, W.H., Sandoval, H., Jaiswal, M., Shulman, J.M., and Bellen, H.J. (2014). A mitocentric view of Parkinson’s disease. Annu. Rev. Neurosci. 37, 137–159.

Hazen, S.L., Zupan, L.A., Weiss, R.H., Getman, D.P., and Gross, R.W. (1991). Suicide inhibition of canine myocardial cytosolic calcium-independent phospholipase A2. Mechanism-based discrimination between calcium-dependent and -independent phospholipases A2. J. Biol. Chem. 266, 7227–7232. Hedley-Whyte, E.T., Gilles, F.H., and Uzman, B.G. (1968). Infantile neuroaxonal dystrophy. A disease characterized by altered terminal axons and synaptic endings. Neurology 18, 891–906. Henry, A.G., Aghamohammadzadeh, S., Samaroo, H., Chen, Y., Mou, K., Needle, E., and Hirst, W.D. (2015). Pathogenic LRRK2 mutations, through increased kinase activity, produce enlarged lysosomes with reduced degradative capacity and increase ATP13A2 expression. Hum. Mol. Genet. 24, 6013–6028. Hesketh, G.G., Perez-Dorado, I., Jackson, L.P., Wartosch, L., Schafer, I.B., Gray, S.R., McCoy, A.J., Zeldin, O.B., Garman, E.F., Harbour, M.E., et al. (2014). VARP is recruited on to endosomes by direct interaction with retromer, where together they function in export to the cell surface. Dev. Cell 29, 591–606. Hooks, S.B., and Cummings, B.S. (2008). Role of Ca2+-independent phospholipase A2 in cell growth and signaling. Biochem. Pharmacol. 76, 1059–1067. Hsu, Y.H., Burke, J.E., Li, S., Woods, V.L., Jr., and Dennis, E.A. (2009). Localizing the membrane binding region of Group VIA Ca2+-independent phospholipase A2 using peptide amide hydrogen/deuterium exchange mass spectrometry. J. Biol. Chem. 284, 23652–23661. Jansen, I.E., Ye, H., Heetveld, S., Lechler, M.C., Michels, H., Seinstra, R.I., Lubbe, S.J., Drouet, V., Lesage, S., Majounie, E., et al. (2017). Discovery and functional prioritization of Parkinson’s disease candidate genes from large-scale whole exome sequencing. Genome Biol. 18, 22. Jenkins, R.W., Canals, D., and Hannun, Y.A. (2009). Roles and regulation of secretory and lysosomal acid sphingomyelinase. Cell Signal. 21, 836–846. Jenkins, R.W., Idkowiak-Baldys, J., Simbari, F., Canals, D., Roddy, P., Riner, C.D., Clarke, C.J., and Hannun, Y.A. (2011). A novel mechanism of lysosomal acid sphingomyelinase maturation: requirement for carboxyl-terminal proteolytic processing. J. Biol. Chem. 286, 3777–3788. Keeney, P.M., Xie, J., Capaldi, R.A., and Bennett, J.P., Jr. (2006). Parkinson’s disease brain mitochondrial complex I has oxidatively damaged subunits and is functionally impaired and misassembled. J. Neurosci. 26, 5256–5264. Khateeb, S., Flusser, H., Ofir, R., Shelef, I., Narkis, G., Vardi, G., Shorer, Z., Levy, R., Galil, A., Elbedour, K., and Birk, O.S. (2006). PLA2G6 mutation underlies infantile neuroaxonal dystrophy. Am. J. Hum. Genet. 79, 942–948. Kinghorn, K.J., Asghari, A.M., and Castillo-Quan, J.I. (2017). The emerging role of autophagic-lysosomal dysfunction in Gaucher disease and Parkinson’s disease. Neural Regen. Res. 12, 380–384. Kinghorn, K.J., Castillo-Quan, J.I., Bartolome, F., Angelova, P.R., Li, L., Pope, S., Cocheme, H.M., Khan, S., Asghari, S., Bhatia, K.P., et al. (2015). Loss of PLA2G6 leads to elevated mitochondrial lipid peroxidation and mitochondrial dysfunction. Brain 138, 1801–1816. Kinghorn, K.J., Gronke, S., Castillo-Quan, J.I., Woodling, N.S., Li, L., Sirka, E., Gegg, M., Mills, K., Hardy, J., Bjedov, I., and Partridge, L. (2016). A Drosophila model of neuronopathic Gaucher disease demonstrates lysosomal-autophagic defects and altered mTOR signalling and is functionally rescued by rapamycin. J. Neurosci. 36, 11654–11670. Kurian, M.A., Morgan, N.V., MacPherson, L., Foster, K., Peake, D., Gupta, R., Philip, S.G., Hendriksz, C., Morton, J.E., Kingston, H.M., et al. (2008). Phenotypic spectrum of neurodegeneration associated with mutations in the PLA2G6 gene (PLAN). Neurology 70, 1623–1629. Lands, W.E. (1958). Metabolism of glycerolipides; a comparison of lecithin and triglyceride synthesis. J. Biol. Chem. 231, 883–888. Linhart, R., Wong, S.A., Cao, J., Tran, M., Huynh, A., Ardrey, C., Park, J.M., Hsu, C., Taha, S., Peterson, R., et al. (2014). Vacuolar protein sorting 35 (Vps35) rescues locomotor deficits and shortened lifespan in Drosophila expressing a Parkinson’s disease mutant of Leucine-Rich Repeat Kinase 2 (LRRK2). Mol. Neurodegener. 9, 23.


Lopez-Montero, I., Velez, M., and Devaux, P.F. (2007). Surface tension induced by sphingomyelin to ceramide conversion in lipid membranes. Biochim. Biophys. Acta 1768, 553–561. MacLeod, D.A., Rhinn, H., Kuwahara, T., Zolin, A., Di Paolo, G., McCabe, B.D., Marder, K.S., Honig, L.S., Clark, L.N., Small, S.A., and Abeliovich, A. (2013). RAB7L1 interacts with LRRK2 to modify intraneuronal protein sorting and Parkinson’s disease risk. Neuron 77, 425–439. Malik, B.R., Godena, V.K., and Whitworth, A.J. (2015). VPS35 pathogenic mutations confer no dominant toxicity but partial loss of function in Drosophila and genetically interact with parkin. Hum. Mol. Genet. 24, 6106–6117. Malik, I., Turk, J., Mancuso, D.J., Montier, L., Wohltmann, M., Wozniak, D.F., Schmidt, R.E., Gross, R.W., and Kotzbauer, P.T. (2008). Disrupted membrane homeostasis and accumulation of ubiquitinated proteins in a mouse model of infantile neuroaxonal dystrophy caused by PLA2G6 mutations. Am. J. Pathol. 172, 406–416. Manzoni, C., and Lewis, P.A. (2013). Dysfunction of the autophagy/lysosomal degradation pathway is a shared feature of the genetic synucleinopathies. FASEB J. 27, 3424–3429. Mecozzi, V.J., Berman, D.E., Simoes, S., Vetanovetz, C., Awal, M.R., Patel, V.M., Schneider, R.T., Petsko, G.A., Ringe, D., and Small, S.A. (2014). Pharmacological chaperones stabilize retromer to limit APP processing. Nat. Chem. Biol. 10, 443–449. Mielke, M.M., Maetzler, W., Haughey, N.J., Bandaru, V.V., Savica, R., Deuschle, C., Gasser, T., Hauser, A.K., Graber-Sultan, S., Schleicher, E., et al. (2013). Plasma ceramide and glucosylceramide metabolism is altered in sporadic Parkinson’s disease and associated with cognitive impairment: a pilot study. PLoS One 8, e73094. Miura, E., Hasegawa, T., Konno, M., Suzuki, M., Sugeno, N., Fujikake, N., Geisler, S., Tabuchi, M., Oshima, R., Kikuchi, A., et al. (2014). VPS35 dysfunction impairs lysosomal degradation of alpha-synuclein and exacerbates neurotoxicity in a Drosophila model of Parkinson’s disease. Neurobiol. Dis. 71, 1–13. Miyake, Y., Kozutsumi, Y., Nakamura, S., Fujita, T., and Kawasaki, T. (1995). Serine palmitoyltransferase is the primary target of a sphingosine-like immunosuppressant, ISP-1/myriocin. Biochem. Biophys. Res. Commun. 211, 396–403. Morgan, N.V., Westaway, S.K., Morton, J.E., Gregory, A., Gissen, P., Sonek, S., Cangul, H., Coryell, J., Canham, N., Nardocci, N., et al. (2006). PLA2G6, encoding a phospholipase A2, is mutated in neurodegenerative disorders with high brain iron. Nat. Genet. 38, 752–754. Murakami, M., and Kudo, I. (2002). Phospholipase A2. J. Biochem. 131, 285–292. Ong, W.Y., Yeo, J.F., Ling, S.F., and Farooqui, A.A. (2005). Distribution of calcium-independent phospholipase A2 (iPLA 2) in monkey brain. J. Neurocytol. 34, 447–458. Paisan-Ruiz, C., Bhatia, K.P., Li, A., Hernandez, D., Davis, M., Wood, N.W., Hardy, J., Houlden, H., Singleton, A., and Schneider, S.A. (2009). Characterization of PLA2G6 as a locus for dystonia-parkinsonism. Ann. Neurol. 65, 19–23. Rao, R.P., Yuan, C., Allegood, J.C., Rawat, S.S., Edwards, M.B., Wang, X., Merrill, A.H., Jr., Acharya, U., and Acharya, J.K. (2007). Ceramide transfer protein function is essential for normal oxidative stress response and lifespan. Proc. Natl. Acad. Sci. USA 104, 11364–11369. Riku, Y., Ikeuchi, T., Yoshino, H., Mimuro, M., Mano, K., Goto, Y., Hattori, N., Sobue, G., and Yoshida, M. (2013). Extensive aggregation of alpha-synuclein and tau in juvenile-onset neuroaxonal dystrophy: an autopsied individual with a novel mutation in the PLA2G6 gene-splicing site. Acta Neuropathol. Commun. 1, 12.


Rousseaux, M.W., de Haro, M., Lasagna-Reeves, C.A., De Maio, A., Park, J., Jafar-Nejad, P., Al-Ramahi, I., Sharma, A., See, L., Lu, N., et al. (2016). TRIM28 regulates the nuclear accumulation and toxicity of both alpha-synuclein and tau. Elife 5, https://doi.org/10.7554/eLife.19809. Sarov, M., Barz, C., Jambor, H., Hein, M.Y., Schmied, C., Suchold, D., Stender, B., Janosch, S., KJ, V.V., Krishnan, R.T., et al. (2016). A genomewide resource for the analysis of protein localisation in Drosophila. Elife 5, e12068. Shinzawa, K., Sumi, H., Ikawa, M., Matsuoka, Y., Okabe, M., Sakoda, S., and Tsujimoto, Y. (2008). Neuroaxonal dystrophy caused by group VIA phospholipase A2 deficiency in mice: a model of human neurodegenerative disease. J. Neurosci. 28, 2212–2220. Staneva, G., Angelova, M.I., and Koumanov, K. (2004). Phospholipase A2 promotes raft budding and fission from giant liposomes. Chem. Phys. Lipids 129, 53–62. Sumi-Akamaru, H., Beck, G., Kato, S., and Mochizuki, H. (2015). Neuroaxonal dystrophy in PLA2G6 knockout mice. Neuropathology 35, 289–302. Taniguchi, M., and Okazaki, T. (2014). The role of sphingomyelin and sphingomyelin synthases in cell death, proliferation and migration-from cell and animal models to human disorders. Biochim. Biophys. Acta 1841, 692–703. Vacaru, A.M., van den Dikkenberg, J., Ternes, P., and Holthuis, J.C. (2013). Ceramide phosphoethanolamine biosynthesis in Drosophila is mediated by a unique ethanolamine phosphotransferase in the Golgi lumen. J. Biol. Chem. 288, 11520–11530. van Meer, G., Voelker, D.R., and Feigenson, G.W. (2008). Membrane lipids: where they are and how they behave. Nat. Rev. Mol. Cell Biol. 9, 112–124. Venken, K.J., Carlson, J.W., Schulze, K.L., Pan, H., He, Y., Spokony, R., Wan, K.H., Koriabine, M., de Jong, P.J., et al. (2009). Versatile P[acman] BAC libraries for transgenesis studies in Drosophila melanogaster. Nat. Methods 6, 431–434. Verstreken, P., Kjaerulff, O., Lloyd, T.E., Atkinson, R., Zhou, Y., Meinertzhagen, I.A., and Bellen, H.J. (2002). Endophilin mutations block clathrin-mediated endocytosis but not neurotransmitter release. Cell 109, 101–112. Vilain, S., Esposito, G., Haddad, D., Schaap, O., Dobreva, M.P., Vos, M., Van Meensel, S., Morais, V.A., De Strooper, B., and Verstreken, P. (2012). The yeast complex I equivalent NADH dehydrogenase rescues pink1 mutants. PLoS Genet. 8, e1002456. Vilarino-Guell, C., Wider, C., Ross, O.A., Dachsel, J.C., Kachergus, J.M., Lincoln, S.J., Soto-Ortolaza, A.I., Cobb, S.A., Wilhoite, G.J., Bacon, J.A., et al. (2011). VPS35 mutations in Parkinson disease. Am. J. Hum. Genet. 89, 162–167. Wang, S., and Bellen, H.J. (2015). The retromer complex in development and disease. Development 142, 2392–2396. Wang, S., Tan, K.L., Agosto, M.A., Xiong, B., Yamamoto, S., Sandoval, H., Jaiswal, M., Bayat, V., Zhang, K., Charng, W.L., et al. (2014). The retromer complex is required for rhodopsin recycling and its loss leads to photoreceptor degeneration. PLoS Biol. 12, e1001847. Wang, T.Y., and Silvius, J.R. (2003). Sphingolipid partitioning into ordered domains in cholesterol-free and cholesterol-containing lipid bilayers. Biophys. J. 84, 367–378. Zhang, J., Reiling, C., Reinecke, J.B., Prislan, I., Marky, L.A., Sorgen, P.L., Naslavsky, N., and Caplan, S. (2012). Rabankyrin-5 interacts with EHD1 and Vps26 to regulate endocytic trafficking and retromer function. Traffic 13, 745–757. Zimprich, A., Benet-Pages, A., Struhal, W., Graf, E., Eck, S.H., Offman, M.N., Haubenberger, D., Spielberger, S., Schulte, E.C., Lichtner, P., et al. (2011). A mutation in VPS35, encoding a subunit of the retromer complex, causes late-onset Parkinson disease. Am. J. Hum. Genet. 89, 168–175.

STAR+METHODS KEY RESOURCES TABLE

REAGENT or RESOURCE

SOURCE

IDENTIFIER

goat anti-GFP-FITC

Santa Cruz

AB_641121

rabbit polyclonal anti-GFP

Life Technologies

AB_221569

rat anti-Elav

DSHB

AB_528218

rabbit anti-DLG

from Kwang-Wook Choi

N/A

mouse anti-DLA

DSHB

AB_528203

mouse anti-ATP5a

Abcam

AN_301447

Antibodies

mouse monoclonal anti-actin

Sigma-Aldrich

AB_2223041

mouse monoclonal anti-V5

ThermoFisher

AB_2556564

mouse monoclonal anti-HA

Covance

AB_10064220

mouse monoclonal anti-Flag

Sigma-Aldrich

AB_262044

mouse monoclonal anti-ATP5A

ThermoFisher

AB_2533548

goat anti-Vps35

Abcam

AB_296841

rabbit anti-Vps26

ProteinTech

cat#12804-1-AP; RRID: AB_2215033

guinea pig anti-dVps26

Wang et al., 2014

N/A

rabbit anti-PLA2G6 (iPLA2-b)

Sigma-Aldrich

AB_1079145

rabbit anti-Glc-Cer

Glycobiotech

cat#RAS_0010

Phalloidin 488nm

ThermoFisher

AB_2315147

DAPI

ThermoFisher

AB_2629482

Alexa 488-conjugated secondary antibodies

Jackson ImmunoResearch Labs

AB_2338059

Alexa Cy3-conjugated secondary antibodies

Jackson ImmunoResearch Labs

AB_2338013

Alexa Cy5-conjugated secondary antibodies

Bioss Inc

AB_11117143

Rabbit polyclonal anti-iPLA2-VIA

This study

N/A

Lysotracker

ThermoFisher

cat#L7528

rabbit anti-Lamp1

Sigma-Aldrich

AB_477157

mouse anti-Lamp2

Santa Cruz

AB_626858

C6-NBD Ceramide

ThermoFisher

cat#N1154

Chemicals, Peptides, and Recombinant Proteins Vectashield

Vector Labs

cat#H-1000

Pierce Protease Inhibitor Tablets, EDTA-free

ThermoFisher

cat#88266

Schneider’s Drosophila medium

ThermoFisher

cat#21720

Fetal Bovine Serum, Heat-inactivated

Sigma-Aldrich

cat#F4135

Penicilin Streptomycin

ThermoFisher

cat#15070063

Cacodylic Acid, Trihydrate Sodium 100g

EMS

cat#12300

EM-grade glutaraldehyde, 25% Aq solution

EMS

cat#16221

Osmium tetroxide 4% Aq solution

EMS

cat#19191

Paraformaldehyde 16% Aq Solution

EMS

cat#15711

Propylene Oxide

EMS

cat#20411

Koptec 200 Proof 100% ethanol Anhydrous

VWR

cat#89125-186

Embed-812

EMS

cat#14901

NMA

EMS

cat#19001

DDSA

EMS

cat#13711 (Continued on next page)

Cell Metabolism 28, 605–618.e1–e6, October 2, 2018 e1

Continued REAGENT or RESOURCE

SOURCE

IDENTIFIER

DMP-30

EMS

cat#13600

Uranyl Acetate

EMS

cat#RT22400

Lead Nitrate

EMS

cat#RT17900-25

Western Lightning Plus-ECL

PerkinElmer

cat#NEL105001EA

Myriocin from Mycelia sterilia

Sigma-Aldrich

cat#M1177

Desipramine hydrochloride

Sigma-Aldrich

cat#D3900

Retromer Chaperone, R55

Calbiochem/Millipore

cat#531084

RapiClear

SunJin Lab Co.

N/A

Effectene Transfection Reagent

Qiagen

cat#301425

ATP/ADP ratio

Abcam

cat#ab65313

Anti-V5 Agarose Affinity Gel

Sigma-Aldrich

AB_10062721

Anti-FLAG M2 Affinity Gel

Sigma-Aldrich

AB_10063035

EZview Red Anti-HA Affinity Gel

Sigma-Aldrich

AB_10109562

EnzChek Phospholipase A2 Assay Kit

ThermoFisher

cat#F10217

Critical Commercial Assays

Lipofectamine RNAiMax

ThermoFisher

cat#13778075

QuikChange Lightning Site-Directed Mutagenesis Kit, 10 Rxn

Agilent Technologies

cat#210518

Experimental Models: Cell Lines D. melanogaster: Cell line S2: S2-DRSC

laboratory of Norbert Perrimon

CVCL_Z232; FlyBase: FBtc0000181

Neuro-2A

ATCC

cat#ATCC-CCL-131

Daoy

ATCC

cat#ATCC-HTB-186

P[EPgy2]iPLA2-VIAEY05103

Bloomington Drosophila Stock Center

BDSC_15947; FlyBase: FBti0038091

y1 w*; P[Act5C-GAL4]25FO1/CyO, y+


BDSC_4414; FlyBase: FBst0004414

y w; PBac[UAS-Vps35-HA]

Wang et al., 2014

N/A

y w; PBac[UAS-Vps26-HA]

Wang et al., 2014

N/A

y w; iPLA2-VIA-GFP

FlyFosTransgeneOme (Sarov et al., 2016)

N/A

lace RNAi: P[KK102282]VIE-260B

Vienna Drosophila RNAi Center

N/A

w; P[UAS-Ndi1p]

Vilain et al., 2012

N/A

y w;; iPLA2-VIAD174

This study

N/A

y w;; iPLA2-VIAD192

This study

N/A

y w; Vps262, FRT19A/FM7c, Kr>GFP

Wang et al., 2014

N/A

w*; P[neoFRT]42D Vps35MH20/CyO, P [GAL4-Kr.C]DC3, P[UAS-GFP.S65T]DC7


BDSC_67202; FlyBase: FBst0067202

y w; PBac[UAS-iPLA2-VIA-PA-HA]

This study

N/A

y w; PBac[UAS-iPLA2-VIA-PB-HA]

This study

N/A

y w; PBac[UAS-hPLA2G6-wt]

This study

N/A

y w; PBac[UAS-hPLA2G6-S519A]

This study

N/A

Plasmid: pUASTattb_iPLA2-VIA-PA-HA

This study

cloneID: LP03302

Plasmid: pUASTattb_iPLA2-VIA-PA-V5

This study

cloneID: LP03302

Plasmid: pUASTattb_iPLA2-VIA-PB-HA

This study

cloneID: RE23733

Plasmid: pUASTattb_iPLA2-b

This study

MGC:45156

Experimental Models: Organisms/Strains

Recombinant DNA

Plasmid: pUASTattb_dVps26-Flag

This study

N/A

Plasmid: pUASTattb_dVps35-HA

This study

N/A

P[acman] BAC CH322-150H21 (chr3L:9844579..9866739)

Venken et al., 2009

N/A (Continued on next page)

e2 Cell Metabolism 28, 605–618.e1–e6, October 2, 2018

Continued REAGENT or RESOURCE

SOURCE

IDENTIFIER

scrambled RNAi

Sigma-Aldrich

SIC001

PLA2G6 RNAi #134

ThermoFisher

cloneID: s79134

PLA2G6 RNAi #135

ThermoFisher

cloneID: s79135

Other

CONTACT FOR REAGENT AND RESOURCE SHARING Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Hugo J. Bellen ([email protected]). EXPERIMENTAL MODEL AND SUBJECT DETAILS Drosophila Mixed genders of flies (approximately 50% male and 50% female) were used for all experiments. Flies were raised on molasses based food at 25 C with constant darkness unless otherwise noted. The full list of genotypes of the flies used can be found in the Key Resources Table. Mammalian Cell Culture The cell lines used in this study are HeLa, Neuro-2A (N2A) and Daoy cells. Cells were maintained in DMEM medium supplemented with 10% FBS, 1X GlutaMAX, 1X pyruvate and 1X penicillin and streptomycin. All reagents were purchased from ThermoFisher. METHOD DETAILS Fly Strains and Genetics iPLA2-VIAD174 (D174) and iPLA2-VIAD192 (D192) alleles were generated via the imprecise remobilization of a P-element (P[EPgy2] iPLA2-VIAEY05103) inserted at the 50 UTR of iPLA2-VIA. A precise remobilization of the P-element was created as control that is used in this study. The genetically modified regions of all alleles were confirmed by Sanger DNA sequencing. To generate the iPLA2-VIA genomic rescue line, the 20 kb P[acman] BAC CH322-150H21 (chr3L:9844579..9866739) (Venken et al., 2009) was microinjected into y w FC31;VK37 embryos. Transgenic lines harboring the P[acman] construct were selected. All the fly stocks were routinely maintained at ambient temperature. For genetic interaction experiments, flies were raised at 25 C to enhance the GAL4 activity. The following strains were also used to generate fly stocks in this study: y1 w67c23;; P[EPgy2]iPLA2-VIAEY05103 (Bloomington Drosophila Stock Center) y w;; iPLA2-VIAD174 (generated in this study) y w;; iPLA2-VIAD192 (generated in this study) y w; Vps262, FRT19A/FM7c, Kr>GFP (Wang et al., 2014) w*; P[neoFRT]42D Vps35MH20/CyO, P[GAL4-Kr.C]DC3, P[UAS-GFP.S65T]DC7 (Bloomington Drosophila Stock Center) y w; PBac[UAS-iPLA2-VIA-PA-HA] (generated in this study) y w; PBac[UAS-iPLA2-VIA-PB-HA] (generated in this study) y w; PBac[UAS-hPLA2G6-wt] (generated in this study) y w; PBac[UAS-hPLA2G6-S519A] (generated in this study) y w; PBac[UAS-Vps35-HA] (Wang et al., 2014) y w; PBac[UAS-Vps26-HA] (Wang et al., 2014) y w; iPLA2-VIA-GFP (FlyFosTransgeneOme collection) (Sarov et al., 2016) lace RNAi: P[KK102282]VIE-260B (Vienna Drosophila RNAi Center) w; P[UAS-Ndi1p] (Vilain et al., 2012) (gift from Patrik Verstreken) Drosophila Behavioral Assay To measure life span, 10 newly eclosed flies were grouped per vial and incubated at 25 C. Survival was recorded every 5 days. The life span of 100 flies was recorded per data point. To perform bang-sensitive paralytic analysis, 5 adult flies were grouped per vial. The flies were vortexed at maximum speed for 15 s and the time required for flies to stand on their feet was counted. At least 50 flies were tested per data point. For the starvation assay, 7-day-old flies were incubated in a vial that only contained cotton presoaked with water. The survival of the flies was monitored every 2 hr. The odor avoidance behavior was examined in a T-maze apparatus. In each assay, 100 one-week-old adult flies were tested with either 3-octanol or 4-methylcyclohexanol by comparison Cell Metabolism 28, 605–618.e1–e6, October 2, 2018 e3

to air for 2 min. The odors were diluted 1:1,000 in mineral oil (Sigma-Aldrich). The avoidance index was calculated by subtracting the number of flies tending to odorant from the number of flies tending to air, and then divided the number by the total number of flies. ERG Recording ERG recordings were performed as detailed in Wang et al., 2014. Flies were glued on a glass slide. A recording and a reference electrode filled with 0.1 M NaCl were placed on the eye or inserted into the fly head respectively. During the recording, a 1 s pulse of light stimulation was given. The ERG traces of the indicated number of flies were recorded and analyzed by AXON-pCLAMP 8 software. Western Blotting Fly heads or HeLa cells were homogenized in Modified RIPA buffer (50 mM Tris-Cl, 150 mM NaCl, 1% NP-40, 1% Sodium deoxycholate, 0.1% SDS, 50 mM NaF, 1 mM Na3VO4, 10% Glycerol and Roche protease inhibitor mix) on ice. Tissue or cell debris were removed by centrifugation. Isolated lysates were loaded into 10% gels, separated by SDS-PAGE, and transferred to nitrocellulose membranes (Bio-Rad). Primary antibodies used in this study were as follows: rabbit anti-iPLA2-VIA antiserum (this study; iPLA2-VIArb), mouse anti-Actin (ICN691001, ThermoFisher), Rabbit anti-Vps26 (GTX106297S, GeneTex) and Goat anti-Vps35 (ab10099, Abcam). Drug Administration in Fly Food The following chemical inhibitors were added freshly to the regular fly food at the indicated concentration: 100 mM Myriocin from Mycelia sterilia (M1177, Sigma-Aldrich); 0.3 mg/mL Desipramine hydrochloride (D3900, Sigma-Aldrich), and 12 mM Retromer Chaperone, R55 (531084, Calbiochem/Millipore). The flies were transferred to fresh food with or without the drugs every three days. Molecular Cloning The plasmids containing Drosophila iPLA2-VIA-PA (clone number: LP03302) and -PB cDNA (clone number: RE23733) were obtained from Drosophila Genomics Resource Center (DGRC). Human iPLA2-b cDNA was subcloned from Human clone MGC:45156 (IMAGE:5166749, Genbank:BC036742). Primers with respective cutting sites, Kozak sequence, HA tag sequence or GGS-linker sequence (listed below) were used to amplify the cDNAs by PCR. Full-length PCR products from the above described templates were subcloned into the pUAST-attB vector. The constructs were then injected into y w FC31 embryos. The transgenic flies were selected based on eye color. Primer used: Notl-Kozak-iPLA2-VIA-PA forward: cttgcggccgcgccaccatgttcaatgttctacagcgcc Notl-Kozak-iPLA2-VIA-PB forward: cttgcggccgcgccaccatggcgtggatggcgttagggg XhoI-HA-GGS-iPLA2-VIA reverse: gcctcgagctacgcgtagtcggggacgtcgtaggggtaggaacctccgcttccaccgctacctccttttaggaaattg atcatctc NotI-Kozak-PLA2G6 forward: cttgcggccgcgccaccatgcagttctttggccgcctgg XbaI-HA-GGS-PLA2G6 reverse: gctctagatcacgcgtagtcggggacgtcgtaggggtaggaacctccgcttccaccgctacctccgggtgagagcagca gctggatg Immunofluorescence Staining For whole mount adult brain staining, the dissected adult brains were fixed with 4% paraformaldehyde in 1X PBS at 4 C overnight. The fixed brains were permeabilized in two conditions, long or short permeabilization. For the long permeabilization condition, the fixed brains were permeabilized with 2% Triton X-100 in 1X PBS under vacuum at room temperature for 1 hr and then shifted to 4 C for two days. For the short permeabilization, the fixed brains were permeabilized with 0.5% Triton X-100 in 1X PBS under vacuum at room temperature for 1 hr and then shifted to 4 C overnight. For the larval brain and axon staining, the dissected third instar larvae were fixed with 3.7% formaldehyde in 1X PBS at room temperature for 20 min. The fixed larvae were permeabilized with 0.4% Triton X-100 in 1X PBS at room temperature for 1 hr. After permeabilization the adult or larval brains were used for immunofluorescence staining using the following antibodies: Rat anti-Elav (Rat-Elav-7E8A10 anti-Elav, DSHB); Rabbit anti-DLG (from Kwang-Wook Choi); Mouse anti-HA (16B12, Covance); Mouse anti-DLG (4F3 anti-Discs large, DSHB); Goat anti-GFP (sc-5385, Santa Cruz Biotechnology); Mouse anti-ATP5a (ab14748, abcam), Goat anti-Vps35 (ab10099, Abcam); Guinea pig anti-dVps26 (Wang et al., 2014); Rabbit anti-iPLA2-b (HPA001171, Sigma-Aldrich); Rabbit anti-Glc-Cer (RAS_0010, Glycobiotech) and Alexa 488-, Cy3-, or Cy5-conjugated secondary antibodies (111-545-144 , 111-585-003 and 111-175-144, Jackson ImmunoResearch Labs and bs-2673R-Cy5.5, Bioss Inc). Lysotracker (L7528, ThermoFisher), Rabbit anti-Lamp1 (L1418, Sigma-Aldrich) or mouse anti-Lamp2 (sc-18822, Santa Cruz) was used to label lysosomes. The adult brains were mounted in RapiClear (SunJin Lab Co.). The larval brains were mounted in Vectashield (H-1000, Vector Laboratories) and kept at 4 C before imaging under a confocal microscope. All the confocal images were acquired with a Model LSM 710 confocal microscope (Carl Zeiss). Confocal images were processed using LSM Image Browser (Carl Zeiss) and Photoshop (Adobe). Mitochondrial Functional Assay To extract mitochondria, Drosophila adult heads were homogenized in ice cold extraction buffer (5 mM HEPES, pH 7.5, 210 mM mannitol, 70 mM sucrose, and 1 mM EGTA) using a Dounce homogenizer. The lysate was centrifuged at 1,500 g at 4 C for 5 min e4 Cell Metabolism 28, 605–618.e1–e6, October 2, 2018

to remove undissolved debris. The supernatant was centrifuged at 8,000 g at 4 C for 15 min to pellet mitochondria. The identified mitochondria were washed once with extraction buffer. To determine ADP/ATP ratio, adult heads were collected and the ratio was determined with an ADP/ATP Ratio Assay Kit (ab65313, abcam) following the manufacturer’s instructions. Mitochondrial complex I activity (NADH/ubiquinone oxidoreductase) was measured by detecting the oxidation of NADH at 340 nm in a reaction buffer containing 25 mm potassium phosphate, pH 7.5, 0.2 mM NADH, and 1.7 mM potassium ferricyanide. Ferricyanide was used as the electron acceptor. Total PLA2 Activity Measurement Total lysate from 1,000 fly heads were homogenized in ice-cold lysis buffer (50 mM HEPES, pH 7.4, containing 1 mM EDTA, 50 mM NaF, 1 mM Na3VO4, 10% Glycerol and Roche protease inhibitor mix). PLA2 activity was determined with an EnzChek Phospholipase A2 Assay Kit (E10217, ThermoFisher) following the manufacturer’s instructions. Transmission Electron Microscopy (TEM) Drosophila eye ultrastructure was analyzed following standard electron microscopy procedures using a Ted Pella Bio Wave processing microwave with vacuum attachments. Briefly, dissected adult heads were fixed in fixative (2% paraformaldehyde, 2.5% glutaraldehyde, 0.1 M sodium cacodylate, and 0.005% CaCl2, pH 7.2) at 4 C overnight and then post-fixed in 1% OsO4. The fixed samples were dehydrated in ethanol and propylene oxide, and then embedded in Embed-812 resin (Electron Microscopy Sciences) under vacuum. Photoreceptors were then sectioned and stained in 1% uranyl acetate and saturated lead nitrate. TEM images of photoreceptor sections were taken using a JEOL JEM 1010 transmission electron microscope at 80 kV with an AMT XR-16 mid-mount 16 mega-pixel digital camera. Sample Preparation for Phospholipidomics Analysis To prepare samples for phospholipidomics analysis, 20 adult fly heads were homogenized in 500 mL Methanol. After homogenization, 1.5 mL of Methanol and 1 mL of Chloroform were added and the samples were vortexed at maximum speed for 30 s. Then, 1 mL of Chloroform and 1 mL of H2O were added to the samples and vortexed at maximum speed for 30 s. The samples were centrifuged at 1,500 rpm for 7 min. The lower layer was collected to a new tube. The collected samples were washed with Chloroform three times. In each wash, 1 mL of Chloroform was added to the samples and vortexed at maximum speed for 30 s. The samples were then centrifuged at 1,500 rpm for 7 min. The lower layer was collected to a new tube and the upper layer was subjected to another run of Chloroform wash. After Chloroform washes, all the lower layer fractions were combined and 1 mL of 1M KCl was added. The samples were vortexed at maximum speed for 30 s and then centrifuged. The lower layer was isolated and 1 mL of H2O was added to each sample and then vortexed and centrifuged. The lower layer was taken and the organic solvents were dried by introducing Nitrogen air. The prepared samples were sent for phospholipidomics analysis at the Kansas Lipidomics Research Center at Kansas State University. To measure the levels of the endogenous sphingosine bases and ceramide species (C14- and C16-Sphingoid Base) sphingolipids, 1,000 adult fly heads or 1 X 106 cultured cells were collected. The levels of the sphingolipids were measured at the Lipidomics Shared Resources at Medical University of South Carolina. Briefly, adult heads or cells were fortified with internal standards (ISs: C17 base D-erythro-sphingosine (17CSph), C17 sphingosine-1-phosphate (17CSph-1P), N-palmitoyl-D-erythroC13 sphingosine (13C16-Cer) and heptadecanoyl-D-erythro-sphingosine (C17-Cer)), and extracted with ethyl acetate/iso-propanol/water (60/30/10 v/v) solvent system. After evaporation and reconstitution in 100 mL of methanol samples were injected on the HP1100/TSQ Quantum LC/MS system and gradient eluted from the BDS Hypersil C8, 150 3 3.2 mm, 3 mm particle size column, with 1.0 mM methanolic ammonium formate/2 mM aqueous ammonium formate mobile phase system. Peaks corresponding to the target analytes and internal standards were collected and processed using the Xcalibur software system. Quantitative analysis was based on the calibration curves generated by spiking an artificial matrix with the known amounts of the target analyte synthetic standards and an equal amount of the internal standards (ISs). The target analyte/IS peak areas ratios were plotted against anlyte concentration. The target analyte/IS peak area ratios from the samples were similarly normalized to their respective ISs and compared to the calibration curves, using a linear regression model. Applying consistent mass spectral conditions of Collision Assistant Dessociation (CAD); 35 eV and Electron Spray Ionization (ESI) all sphingoid bases and related ceramides undergo uniform transition from initial molecular ion (M+1) to the respective sphingoid backbone secondary ions. Consequently, calibration curves, generated from authentic standards of the typical, 18C-sphingosine and ceramides, can be used for quantitation of other, e.g. 20C- counterparts. For the fly heads, the levels of sphingolipid species are normalized to the number of the fly heads. For the cell line assays, the levels of sphingolipid species are normalized to the phosphate amount of the samples. Although absolute concentrations determined for compounds without authentic standards (20C-LCB derivatives) may not be precise, due to possible differences in instrument response for 18C- and 20C-LCB related compounds, for comparative study, where changes in sphingolipids level rather than absolute concentration, are most important, this indirect methodology provide reliable results. Immunoprecipitation Drosophila Schneider 2 (S2) cells were maintained in Schneider’s Drosophila Medium (21720-024, ThermoFisher) supplemented with 10% FBS and 1XPenicillin-Streptomycin. 48 hr after transfection, S2 cells were lysed for 30 min on ice, using 1X NP40 lysis buffer (150 mM NaCl, 20 mM HEPES pH 7.5, 1% NP40, 50mM NaF, 1mM Na3VO4, 10% Glycerol and EDTA-free proteinase inhibitor [11836170001, Sigma-Aldrich]). The lysates were centrifuged at maximum speed for 10 min at 4 C. Supernatants were transferred Cell Metabolism 28, 605–618.e1–e6, October 2, 2018 e5

to a clean eppendorf tube and incubated overnight at 4 C with 5 mL agarose beads, which had been previously equilibrated with lysis buffer. The beads were washed 5 times in 1X NP40 lysis buffer before boiling in loading buffer. Western blotting was then performed with each sample. The following beads were used for immunoprecipitation: Monoclonal Anti-V5Agarose antibody (A7345, SigmaAldrich), monoclonal anti-Flag clone M2-agarose antibody (A2220, Sigma-Aldrich) and monoclonal anti-HA-Agarose antibody (E6779, Sigma-Aldrich). The antibodies used for western blot analyses were: mouse anti-HA (16B12, Covance), mouse anti-Flag clone M2 (F3165, Sigma-Aldrich) and mouse anti-V5 (R960-25, ThermoFisher,) and rabbit anti-PLA2G6 (HPA001171, Sigma-Aldrich). Mammalian Tissue Culture Assay HeLa or Neuro-2A (N2A) cells were seeded on 12 well plates. Bromoenol Lactone (B1552, Sigma-Aldrich) 20 mM was added and the cells were incubated at 37 C overnight. Myriocin (1 mM) or Ethanol was added and the cells were incubated at 37 C for 12 hr. Cells were fixed and stained with Lysotracker (L7528, ThermoFisher), Rabbit anti-Lamp1 (L1418, Sigma-Aldrich) or mouse anti-Lamp2 (sc-18822, Santa Cruz). All the confocal images were acquired with a Model LSM 710 confocal microscope (Carl Zeiss). Confocal images were processed using LSM Image Browser (Carl Zeiss) and Photoshop (Adobe). C6-NBD Ceramide Localization Assay To track the localization of ceramides, C6-NBD Ceramide (N1154, ThermoFisher) was added to HeLa cells at 4 C for 30 min in the presence or absence of BEL. The labeled cells were washed three times with 1X PBS to remove unlabeled C6-NBD Ceramide. After the washes, the cells were incubated at 37 C for 1 hr to allow the internalization of the C6-NBD Ceramide. Lysotracker and DAPI staining was used to label lysosomes and nuclei, respectively. Images were acquired with a Model LSM 710 confocal microscope (Carl Zeiss). Confocal images were processed using LSM Image Browser (Carl Zeiss) and Photoshop (Adobe). Transfection of siRNA in Neuro-2A Cells The negative control siRNA or siRNAs target PLA2G6 were transiently transfected to Neuro-2A cells using Lipofectamine RNAiMax (13778075, ThermoFisher) following the manufacturer’s instructions. The following siRNAs were used: Negative control siRNA (SIC001, Sigma-Aldrich); PLA2G6 siRNA #134 (4390771-s79134, ThermoFisher) and PLA2G6 siRNA #135 (4390771-s79135, ThermoFisher). 48 hr after the siRNAs were transfected to Neuro-2A cells, the indicated drugs were added to the cells for 6 hr. The cells were fixed and stained with anti-Lamp1 antibody (Sigma-Aldrich). All the confocal images were acquired with a Model LSM 710 confocal microscope (Carl Zeiss) and were processed using LSM Image Browser (Carl Zeiss) and Photoshop (Adobe). Cell Survival Assay The day before transfection, Neuro-2A cells (5 X 105 per well of 6-well plate) were seeded on a poly-D-lysine coated cover slip. The Neuro-2A cells were transfected with indicated siRNAs for 48 hr. The indicated drugs were added to the transfected cells for 6 hr. The cells on the cover slips were fixed and stained with DAPI. The images of ten random fields were taken using a Model LSM 710 confocal microscope (Carl Zeiss). Survival cells per field were counted by DAPI positive punctae using ImageJ software. QUANTIFICATION AND STATISTICAL ANALYSIS All datasets were organized and analyzed in Microsoft excel 2010 using two-tailed Student’s t test. For fly experiments, sample sizes are stated in the figure legends. All crosses or drug treatments were performed at least twice. For cell experiments, all studies were conducted in parallel with vehicle controls in the neighboring well for at least 3 biological replicates. Error bars are shown as standard error of the mean (SEM). The criteria for significance are: NS (not significant) p > 0.05; *p < 0.05; **p < 0.01 and ***p < 0.001.

e6 Cell Metabolism 28, 605–618.e1–e6, October 2, 2018