Article Neuronal lipolysis participates in PUFA-mediated neural function and neurodegeneration Leilei Yang1,2,†, Jingjing Liang1, Sin Man Lam3, Ahmet Yavuz4, Guanghou Shui1,2, Mei Ding1,2 \& Xun Huang1,2,* Abstract concentration of lipids. However, LDs in the nervous system are generally found in glial cells but not in neurons under normal condi- Lipid droplets (LDs) are dynamic cytoplasmic organelles present in tions in vivo (Kis et al, 2015). most eukaryotic cells. The appearance of LDs in neurons is not usually observed under physiological conditions, but is associated The origin and the role of glial LDs have been investigated only with neural diseases. It remains unclear how LD dynamics is regu- in recent years. The formation of LDs in glia, which act as a niche lated in neurons and how the appearance of LDs affects neuronal for neuroblasts, preserves Drosophila larval neuroblast proliferation functions. We discovered that mutations of two key lipolysis genes under ROS-inducing stress conditions such as hypoxia. It is postu- atgl-1 and lid-1 lead to LD appearance in neurons of Caenorhabditis lated that the incorporation of polyunsaturated fatty acids (PUFAs) elegans. This neuronal lipid accumulation protects neurons into neutral lipids, and their storage in LDs, reduces the ROS insult from hyperactivation-triggered neurodegeneration, with a mild and the toxic peroxidation of PUFAs in neuroblasts (Bailey et al, decrease in touch sensation. We also discovered that reduced 2015). Several other studies reported the neuronal origin of glial biosynthesis of polyunsaturated fatty acids (PUFAs) causes similar LDs. When Drosophila neurons are under ROS insult or have mito- effects and synergizes with decreased lipolysis. Furthermore, we chondrial dysfunction, neuronal lipid production is increased demonstrated that these changes in lipolysis and PUFA biosynthe- through SREBP-mediated lipogenesis. Interestingly, instead of form- sis increase PUFA partitioning toward triacylglycerol, and reduced ing LDs in neurons, lipids are transferred to neighboring glia incorporation of PUFAs into phospholipids increases neuronal through fatty acid transfer protein (FATP) or apolipoprotein to form protection. Together, these results suggest the crucial role of LDs (Liu et al, 2015, 2017). In cultured hippocampal neurons, neuronal lipolysis in cell-autonomous regulation of neural func- hyperactivated neurons also produce excess fatty acids, which are tions and neurodegeneration. transferred, via lipid particles associated with ApoE, to astrocytes and are incorporated into LDs. The storage of fatty acids in astrocyte Keywords Caenorhabditis elegans; lipid droplet; lipolysis; neurodegeneration; LDs and their subsequent b-oxidation in mitochondria protects polyunsaturated fatty acid neurons during periods of enhanced activity (Ioannou et al, 2019). Subject Categories Membrane \& Trafficking; Neuroscience These findings suggest that LD formation in glia plays a role in DOI 10.15252/embr.202050214 | Received 13 February 2020 | Revised 26 protecting neurons from stress insults. However, it is unknown why August 2020 | Accepted 8 September 2020 neurons do not form LDs in an autonomous fashion to protect them- EMBO Reports (2020) e50214 selves under stress conditions. Introduction Although neurons do not normally have LDs, some neuronal diseases are associated with LD biology and neuronal LDs have Lipid droplets (LDs) are dynamic cytoplasmic organelles which are been reported in some disease models. The Parkinson’s disease present in most, if not all, eukaryotic cells and many prokaryotic protein a-Synuclein is located on the surface of LDs in lipid-loaded cells. By storing excess lipids in the form of neutral lipids including primary hippocampal neurons and a-Synuclein expression is corre- triacylglycerol (TAG) and sterol ester (SE), LDs maintain cellular lated with LD accumulation in yeast (Cole et al, 2002; Outeiro \& lipid homeostasis, along with the coordinated actions of lipogenesis Lindquist, 2003). Huntington’s disease cells, including primary stri- and lipolysis (Chen et al, 2019; Olzmann \& Carvalho, 2019). The atal neurons and glia in Huntington’s disease mice, have dramati- nervous system, including neurons and glia, has a high cally increased LDs (Martinez-Vicente et al, 2010). Several hereditary spastic paraplegia (HSP) proteins affect LD dynamics, such as Spartin, spastin, atlastin-1, seipin, and REEP1 (Eastman et al, 2009; Klemm et al, 2013; Ebihara et al, 2015; Papadopoulos 1 State Key Laboratory of Molecular Developmental Biology, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, China 2 University of Chinese Academy of Sciences, Beijing, China 3 LipidAll Technologies Co., Ltd., Changzhou, China 4 Department of Molecular and Human Genetics, Huffington Center on Aging, Howard Hughes Medical Institute, Baylor College of Medicine, Houston, TX, USA *Corresponding author. Tel/Fax: +86 10 64806560; E-mail: xhuang@genetics.ac.cn †Present address: Vector Core, Chinese Institute for Brain Research, Beijing, China ª 2020 The Authors EMBO reports e50214 | 2020 1 of 15

EMBO reports Leilei Yang et al et al, 2015; Renvoise et al, 2016; Ding et al, 2018). Despite the xd314, while xd310 fails to complement xd314. Through SNP apparent association between LD and neuronal diseases, the causal mapping, we narrowed down the region of the xd288 mutation to link between neuronal LD dynamics and neuronal disorders remains chromosome I between +3.85 cM and +5.05 cM. Through fosmid largely elusive. transgenic rescue assays, we found that the WRM0632aE01 fosmid fully rescues the neuronal LD phenotype of xd288. The gene lid-1 In this study, we explored the dynamics and the physiological (lipid droplet protein 1), which encodes one of the C. elegans homo- role of LDs in neurons. Using Caenorhabditis elegans as a model, logs of mammalian CGI-58, is found in this fosmid (Lee et al, 2014). we found that ATGL-1/LID-1-mediated lipolysis autonomously regu- Importantly, we found a C332T point mutation in lid-1, and this lates neuronal LD dynamics. ATGL-1 is the C. elegans homolog of mutation leads to a missense S111L mutation of LID-1 (Fig 1C). mammalian ATGL, the rate-limiting enzyme of TAG hydrolysis, and S111 is located in the putative a/b hydrolase domain of LID-1 and is LID-1 is the C. elegans homolog of mammalian CGI-58 (also named conserved in C. elegans, Drosophila, mouse, and human as ABHD5), which is the best known co-activator of ATGL (Lass (Appendix Fig S2). Interestingly, S111 of LID-1 corresponds to S115 et al, 2006). Mutations in either CGI-58 or ATGL lead to neutral- in human, which is mutated in NLSD (Ben Selma et al, 2007). lipid storage disease (NLSD) with neurological abnormalities in human (Schweiger et al, 2009; Massa et al, 2016). Importantly, A previous report showed that LID-1 binds to the C. elegans defective neuronal lipolysis reduces PUFA-mediated touch sensation ATGL homolog, ATGL-1, and promotes ATGL-1-dependent lipolysis and protects neurons from hyperactivation-triggered neurodegenera- during fasting conditions (Lee et al, 2014). As expected, we found tion. The neuronal protective effect of atgl-1 mutants is significantly that xd310 and xd314 are mutants of C. elegans atgl-1. The xd310 enhanced by reduction of de novo PUFA synthesis and/or incorpora- mutation causes a missense G210E change, while xd314 causes a tion of PUFAs into phospholipids. Together, our results show that missense G19R change in ATGL-1 (Fig 1D). G19 of ATGL-1 is neuronal LDs participate in PUFA-mediated neural functions and conserved in C. elegans, Drosophila, mouse, and human neurodegeneration. (Appendix Fig S2B). The G19R and G210E mutations are located in an active (a/b/a) sandwich domain of ATGL-1, which is responsible Results for its enzymatic activity. Oil red O staining shows that both lid-1 (xd288) and atgl-1(xd314) have more fat than controls, especially in LDs accumulate in neurons of both atgl-1 and lid-1 mutants the head region (Fig 1E) and intestine (Appendix Fig S2). This suggests that, just like in NLSD, the overall fat content is increased To observe LDs in worm neurons, we generated a neuron-specific in lid-1 and atgl-1 mutants. The recessive nature of the xd288, GFP reporter line xdIs109[Punc-119::PLIN1::GFP/rol-6] expressing the xd310, and xd314 mutations and the lipolytic function of ATGL and LD surface protein PLIN1. PLIN1::GFP forms ring-like structures CGI-58 suggest that defective lipolysis leads to LD accumulation in when expressed in tissues with LDs (Bi et al, 2012; Liu et al, neurons. 2014). In xdIs109 animals, we mainly focused on the head region, which is the location of most neuronal soma. We found that there These mutants were isolated based on the appearance of are very few GFP rings in xdIs109 young adults (Fig 1A). This neuronal LDs in the xdIs109 background. To avoid potential inter- indicates that similar to mammals, there are few LDs in C. elegans ference from the overexpression of PLIN1::GFP, we also used neurons under normal conditions. It has been reported that both BODIPY, a neutral-lipid dye, combined with a pan-neuron marker aging and general obesity may increase LD accumulation in non- Prab-3::mCherry to observe LDs in the ventral nerve cord of lid-1 adipose tissues (Zhou et al, 2000; Shimabukuro et al, 2016; Pali- (xd288) and atgl-1(xd314) mutants that do not carry the xdIs109 karas et al, 2017). To explore the influence of aging and overall marker. BODIPY-positive LDs are found in neurons of lid-1(xd288) fat content increase on neuronal LDs, we examined the xdIs109 and atgl-1(xd314) but not N2 control (Appendix Fig S3). In addition, GFP pattern in 8-day-old wild-type adults, daf-2(e1370) mutants we used electron microscopy (EM) to observe whether there are and glp-1(e2141) mutants. It is well known that the latter two LDs in lid-1(xd288) and atgl-1(xd314) neurons without the xdIs109 mutants have an overall increase in lipid storage (O’Rourke et al, marker. LDs were easily found in the neuronal soma of xd288 and 2009). We found that compared with young xdIs109 controls, there xd314 mutants, but not in wild type (Fig 1F–H). We could even is no significant increase of LD number in 8-day-old xdIs109, daf-2 trace whole LDs in mutants from continuous EM serial sections (e1370); xdIs109 and glp-1(e2141); and xdIs109 (Fig 1B). This (Fig 1I). This demonstrates that atgl-1(xd314) and lid-1(xd288) indicates that aging and general obesity do not necessarily result mutants indeed have ectopic LDs in neurons. in neuronal LD accumulation. The above results indicate that ATGL-1/LID-1-mediated lipolysis To reveal the underlying mechanism(s) that control LD dynamics prevents the appearance of visible LDs in neurons. In contrast to in neurons, we performed an EMS screen using the xdIs109 marker lipolysis, lipogenesis promotes lipid storage. We wondered whether to search for mutants with neuronal LD accumulation. We isolated overexpression of lipogenesis-related genes in neurons could cause the xd288, xd310, and xd314 mutants, which show LDs in many LD accumulation. DGAT1 and DGAT2 are key enzymes in the neurons in the head region, ventral nerve cord and tail (Fig 1A and synthesis of TAG from diacylglycerol (DAG) (Fig 2A). DGAT1 has Appendix Fig S1). We also quantified LDs in the head region and only one homolog, MBOA-2, in C. elegans. DGAT2 has four homo- found that the number of LDs is increased significantly in these logs (DGAT-2/F59A1.10, K07B1.4, DGTR-1/W01A11.2, and mutants (Fig 1B). Y53G8B.2) in C. elegans (Fig 2B). We pan-neuronally overexpressed mboa-2, F59A1.10, and K07B1.4, respectively. In these transgenic Complementation tests divided these three mutants into two animals, there are ectopic LDs in neurons as revealed by the xdIs109 complementation groups. xd288 complements both xd310 and reporter (Fig 2C). BODIPY staining also confirms the presence of LDs in mboa-2 or F59A1.10 overexpressing neurons labeled by 2 of 15 EMBO reports e50214 | 2020 ª 2020 The Authors.

Leilei Yang et al EMBO reports A B CD E FI GH Figure 1. lid-1(xd288) and atgl-1(xd314) display ectopic LD accumulation in neurons. A lid-1(xd288) and atgl-1(xd314) mutants show LD accumulation in most neurons in Caenorhabditis elegans, including neurons in the head region, tail, and ventral nerve cord. xdIs109 is a stable transgenic line with pan-neuronal expression of the LD marker PLIN1::GFP. The white arrows indicate PLIN1::GFP-positive LDs in the ventral nerve cord of mutants. Scale bar: 20 lm. The right panels depict the head region. Scale bar: 5 lm. The insets in the right panels are enlarged views of PLIN1::GFP rings, representing LDs. B Quantification of the numbers of LDs in head region neurons (each dot represents one worm). The data were analyzed using one-way ANOVA with Dunnett’s multiple comparison test. Asterisks denote significant differences as compared to the control xdIs109. **P < 0.01, ***P < 0.001, ****P < 0.0001. ns: not statistically significant. Data show mean Æ SEM. n ≥ 9. C, D The genetic loci of lid-1 and atgl-1. The coding regions are in light blue boxes, and the noncoding regions are shown as lines. The UTRs are in red boxes. xd288 is a point mutant of lid-1 and causes a missense S111L mutation in the a/b hydrolase domain. xd314 and xd310 are point mutants of atgl-1. xd314 harbors a missense G19R mutation, and xd310 harbors a missense G210E mutation. E Oil red O staining show dramatically increased neutral lipids in the head region of lid-1(xd288) and atgl-1(xd314) compared with N2. Scale bar: 10 lm. F–H EM images of neurons. The red dashed line marks the outline of neurons. The red arrows mark LDs. Scale bar: 0.5 lm. The scale bar in enlarged images is 100 nm. I Serial sections of atgl-1(xd314) show two LDs (indicated by blue and red arrows) in a neuron cell body. Scale bar: 0.5 lm. ª 2020 The Authors. EMBO reports e50214 | 2020 3 of 15

EMBO reports Leilei Yang et al A C B Figure 2. LID-1/ATGL-1-mediated lipolysis and DAGT1/2-mediated lipogenesis regulate LD dynamics in neurons. A The pathways of lipolysis and lipogenesis. G-3-P: glycerol-3-phosphate. LPA: lysophosphatidic acid. PA: phosphatidic acid. DAG: diacylglycerol. TAG: triacylglycerol. MAG: monoacylglycerol. FA: fatty acid. B The evolutionary phylogenetic trees of DGAT1 and DGAT2. C Overexpressing the homologs of DGAT1/2 in neurons of Caenorhabditis elegans leads to LD accumulation in neurons. Each dot represents one worm. The data were analyzed using one-way ANOVA with Dunnett’s multiple comparison test. Asterisks denote significant differences as compared to the control xdIs109. ***P < 0.001, ****P < 0.0001. Data show mean Æ SEM. n ≥ 8. Prab-3::mCherry marker (Appendix Fig S3). Therefore, either and are able to associate spontaneously to form fluorescent GFP elevating lipogenesis or reducing lipolysis leads to LD accumulation (Cabantous et al, 2005). As controls, transgenic animals expressing in neurons. Together, these results indicate that neurons have the either Patgl-1::GFP1-10 or pan-neuronal Prab-3::GFP11 do not exhibit ability to form and hydrolyze LDs. The lack of visible LDs in GFP fluorescence (Fig 3B). The GFP fluorescence is clearly seen in neurons reflects the dominance of lipolysis over lipogenesis under neurons marked by pan-neuronal Prab-3::mCherry, including head normal conditions. region, ventral nerve cord, and tail neurons, in animals carrying both Patgl-1::GFP1-10 and Prab-3::GFP11 (Fig 3B and C). This result provides atgl-1 autonomously regulates neuronal LDs further evidence for the neuronal expression of ATGL-1. ATGL-1 is expressed strongly in worm intestine (Lee et al, 2014), To investigate whether atgl-1 functions autonomously or non- but its expression in neurons has not been reported in detail. To autonomously, we performed tissue-specific rescue and tissue- observe the expression pattern of ATGL-1, we first mapped the specific knockout experiments. We found that neuronal Punc-119:: promotor of atgl-1. The excess neuronal LD phenotype of atgl-1 atgl-1 expression, but not glial Phlh-17::atgl-1, hypodermal Pajm-1:: (xd314) was fully rescued by several overlapping atgl-1(+) atgl-1 or intestinal Pvha-6::atgl-1 expression, rescues the neuronal LD fosmids. The overlapping region of the rescuing fosmids highlights phenotype of atgl-1(xd314); xdIs109 (Fig 3D and E). We also gener- a 3.4 Kb promoter region which may be important for the rescuing ated a neuron-specific knockout of atgl-1 using the Cre-loxP system. activity. We further found that an atgl-1 genomic fragment with a We first used CRISPR-Cas9 to insert two loxP sites flanking the atgl- 3 Kb promoter fully rescues the neuronal LD phenotype of atgl-1 1 coding region, and then we expressed Cre in neurons to specifi- (xd314) (Appendix Fig S4). Using this 3 Kb promoter region to cally knock out atgl-1 in neurons (Appendix Fig S4). We found that drive a GFP reporter, we examined the expression pattern of atgl-1 similar to atgl-1(xd314) mutants, neuronal-specific deletion of atgl-1 in detail. The GFP fluorescence is bright in the intestine, consistent [atgl-1-loxP(xd426); xdIs182 (Prab-3::Cre/Podr-1::RFP)] causes LD with a previous report (Lee et al, 2014). We also found weak accumulation in neurons (Fig 3F and G). Together, these results expression in neurons in the nerve ring, ventral nerve cord, and demonstrate that ATGL-1 acts autonomously in neurons to prevent tail (Fig 3A). neuronal LD accumulation. To specifically analyze ATGL-1 expression in neurons, we used Paralogs of ATGL-1 and LID-1 do not affect neuronal LD dynamics the self-complementing split GFP system. Split GFP is composed of two separate GFP fragments (a 15-amino acid fragment, GFP11, and There are two LID-1 paralogs (C37H5.2 and C37H5.3) and two a 215-amino acid fragment, GFP1-10), which are expressed separately ATGL-1 paralogs (B0524.2 and D1054.1) in C. elegans (Fig EV1A 4 of 15 EMBO reports e50214 | 2020 ª 2020 The Authors.

Leilei Yang et al EMBO reports and B). Although LID-1 is reported to regulate lipolysis by interact- phosphorylation by AMPK (Narbonne \& Roy, 2009; Xie \& Roy, ing with ATGL-1 (Lee et al, 2014), C37H5.3, also named as cgi-58, 2015a,b). To test whether ATGL-1 and LID-1 paralogs affect LD had been shown to promote ATGL-1 activity at the dauer stage homeostasis in neurons, we examined the neuronal LD phenotypes when the protein stability of ATGL-1 is negatively regulated via of C37H5.2(ok3722), C37H5.3(ok3245), D1054.1(tm3111), and AB C DE F G Figure 3. EMBO reports e50214 | 2020 5 of 15 ª 2020 The Authors.

EMBO reports Leilei Yang et al ◀ Figure 3. atgl-1 autonomously regulates neuronal LD dynamics. A The expression pattern of atgl-1 viewed by Patgl-1::GFP transcriptional fusion reporter. Patgl-1::GFP expression is high in intestine. Relatively low fluorescent signals are found in nerve ring, motor neuron commissure, ventral nerve cord, and tail. Scale bar: 20 lm. B atgl-1 is expressed in neurons. There is no fluorescent signal when Patgl-1::GFP1-10 and Prab-3::GFP11 are expressed separately. When Patgl-1::GFP1-10 and Prab-3::GFP11 are co-expressed, the GFP signals are found in the head region, tail, and ventral nerve cord, where most neuronal cell bodies are located. The white lines mark the outline of worms. Scale bar: 50 lm. C atgl-1 is expressed widely in neurons. The left panels depict the whole worm. Scale bar: 50 lm. The insets in the right panels are enlarged views of the ventral nerve cord in the white dotted boxes of left panels. The signals of mCherry and GFP merge very well. Scale bar: 20 lm. D Visualizing LDs in head neurons with the xdIs109 reporter in different genetic backgrounds. The accumulation of LDs in head neurons of atgl-1(xd314) mutants can be rescued by expressing wild-type atgl-1 in neurons (Punc-119::atgl-1), but not in intestine (Pvha-6::atgl-1), glia (Phlh-17::atgl-1), or hypodermis (Pajm-1::atgl-1). Scale bar: 5 lm. E Quantification of the number of LDs in head neurons in different genetic backgrounds. Each dot represents one worm. The data were analyzed using one-way ANOVA with Dunnett’s multiple comparison test. Asterisks denote significant differences as compared to the control xdIs109. ns: not statistically significant. ****P < 0.0001. Data show mean Æ SEM. n ≥ 7. F Neuron-specific knockout of atgl-1 causes LD accumulation in neurons, similar to atgl-1(xd314) mutants. Scale bar: 5 lm. G Quantification of the number of LDs in head neurons in different genetic backgrounds. Each dot represents one worm. The data were analyzed using one-way ANOVA with Dunnett’s multiple comparison test. Asterisks denote significant differences as compared to the control xdIs109. t-Test was used between the groups under the crossbars. ns: not statistically significant. ****P < 0.0001. Data show mean Æ SEM. n ≥ 9. B0524.2(tm6739) mutants with the xdIs109 marker. Similar to wild Stimulated Raman Scattering (SRS) microscopy (Fig 4A). To exam- type, there are almost no LDs in neurons in these mutants ine whether the gentle touch sensation is affected in atgl-1(xd314) (Fig EV1C). These mutants are all deletion alleles (Fig EV1D), and and lid-1(xd288) mutants, we measured the touch sensitivity by a even though we are not sure whether they are null alleles, their ten-trial touch assay (Hart, 2006). mec-4(u253) and N2 were used phenotypes indicate that the ATGL-1/LID-1 pair plays the main role as the positive control and negative control, respectively. Compared in regulating neuronal lipolysis and neuronal LD dynamics. with the strong touch sensation defect in mec-4(u253) mutants, touch sensation is partially impaired in atgl-1(xd314) and lid-1 Next, we explored whether these paralogs can substitute for (xd288) mutants. Importantly, expression of neuronal-specific Punc- ATGL-1 or LID-1 in neurons. We overexpressed these paralogs in 119::atgl-1 and Punc-119::lid-1 fully rescued the touch sensation defect neurons specifically to examine whether they could rescue the in atgl-1(xd314) and lid-1(xd288), respectively (Fig 4B). This neuronal LD phenotype of atgl-1(xd314); xdIs109 or lid-1(xd288); suggests that neuronal lipolysis is required for normal gentle touch xdIs109. Because these paralogs may function in pairs, like ATGL-1 sensation. and LID-1, we overexpressed all four paralogs together in neurons. We found that they could not rescue the neuronal LD phenotype of In the de novo PUFA biosynthetic pathway, fat-4 is responsible either atgl-1(xd314); xdIs109 or lid-1(xd288); xdIs109 (Fig EV1E). for the synthesis of arachidonic acid (AA, C20:4 (20 carbons with Together, these results indicate the importance and the specificity of four double bonds)) and eicosapentaenoic acid (EPA, C20:5) the ATGL-1/LID-1 partnership in regulating neuronal LD dynamics. (Fig 4C; Watts \& Browse, 2002). Mutation of fat-4 leads to mild gentle touch sensation defects similar to those in atgl-1(xd314) and atgl-1 and lid-1 affect gentle touch sensation and genetically lid-1(xd288) mutants (Fig 4B; Vasquez et al, 2014). We then inves- interact with the de novo PUFA biosynthesis pathway tigated the genetic relationship between ATGL-1/LID-1-mediated lipolysis and de novo synthesis of PUFAs by double mutant analy- The above results show that lipolysis restricts the appearance of sis. fat-3 or fat-4 mutants do not affect neuronal LDs and neuronal visible LDs in neurons. We next examined the neuronal conse- LDs are either unchanged or slightly reduced in atgl-1 or lid-1 and quence of defective lipolysis. LDs are temporary storage places for fat-3 or fat-4 double mutants compared with atgl-1 or lid-1 single excess lipids. Since lipids are important building blocks for mutants (Fig EV2). Interestingly, the fat-4(wa14) mutation membranes, we initially focused on examining whether atgl-1 enhances the touch sensation defect of both atgl-1(xd314) and lid-1 affects neuronal morphology. The PVD neuron has very complex (xd288) single mutants (Fig 4B), which suggests that ATGL/LID-1- neurites, especially dendrites, which may require lots of lipids for regulated neuronal lipolysis participates in PUFA-mediated touch membrane biogenesis. We did not find any morphological difference sensation. between wild type and atgl-1(xd314) or lid-1(xd288) mutants based on observations with the PVD marker wdIs52(PF49H12.4::GFP) atgl-1(xd314) and lid-1(xd288) mutants have reduced neuron (Appendix Fig S5). This suggested that the appearance of LDs may hyperactivation-triggered neurodegeneration not affect the gross morphology of neurons. Glial LDs are reported to be involved in neurodegeneration in PUFAs affect the gentle touch sensation and mechanoelectrical Drosophila (Liu et al, 2015, 2017). In particular, to avoid fatty transduction in C. elegans touch receptor neurons (Kahn-Kirby acid toxicity in hyperactivated neurons, fatty acids are transported et al, 2004; Vasquez et al, 2014). A previous report showed that from neurons into glia and stored in LDs before detoxification when PUFAs are diverted away from membranes to the core of through mitochondrial oxidation (Ioannou et al, 2019). These LDs, the expansion of LDs inhibits the oxidation of PUFAs and findings prompted us to investigate whether the appearance of reduces the toxic effect of PUFAs (Bailey et al, 2015). Therefore, neuronal LDs in atgl-1 and lid-1 mutants alleviates neurodegener- the presence of LDs in neurons may limit the availability of PUFAs ation. mec-4 encodes the ion-channel protein MEC-4 and domi- and affect PUFA-associated neuronal events. There are ectopic LDs nant negative mutants of mec-4, often referred to as mec-4(d), in touch neurons in atgl-1(xd314) as revealed by non-invasive 6 of 15 EMBO reports e50214 | 2020 ª 2020 The Authors.

Leilei Yang et al B EMBO reports A C Figure 4. atgl-1(xd314) and lid-1(xd288) show a gentle touch sensation defect. A LDs accumulate in touch neurons in atgl-1(xd314) visualized by SRS (Stimulated Raman Scattering). The zdIs5 reporter marks touch neurons. The white arrows indicate LDs in touch neurons in atgl-1(xd314). Scale bar: 10 lm. B atgl-1(xd314) and lid-1(xd288) show a gentle touch sensation defect, which is rescued by the neuron-specific expression of atgl-1 and lid-1, respectively. The fat-4 mutation enhances the touch sensation defect in atgl-1(xd314) and lid-1(xd288). Wild type and mec-4(u253) act as the negative control and positive control, respectively. The data were analyzed using Kruskal–Wallis test with Dunn’s test. Asterisks denote significant differences as compared to wild type. ns: not statistically significant. **P < 0.01, ***P < 0.001, ****P < 0.0001. Data show mean Æ SEM. # signify significant differences between the groups under the crossbars. ##P < 0.01. Data show mean Æ SEM. Number of worms analyzed for each strain n = 25. C The pathway of PUFA synthesis in Caenorhabditis elegans. LA: linoleic acid. ALA: alpha-linolenic acid. GLA: gamma-linoleic acid. STA: stearidonic acid. DGLA: dihomo- gamma-linolenic acid. ETA: eicosatetraenoic acid. AA: arachidonic acid. EPA: eicosapentaenoic acid. Fatty acids are also indicated by chemical abbreviations. For example, the abbreviation C18:2nÀ6 means 18 carbons with two double bonds and the first double bond is located at xÀ6. have been widely used as models of neuron hyperactivation-trig- 2002; Watts \& Ristow, 2017). Consistent with the potential involve- gered neurodegeneration (Driscoll \& Chalfie, 1991; Calixto et al, ment of PUFAs in neuron hyperactivation-triggered neurodegenera- 2012). The neurodegeneration in mec-4(d) occurs as early as the tion, we found that mec-4(d)-triggered neurodegeneration was embryonic stage, and by the L4 stage, the majority (63 Æ 2%) of significantly reduced in fat-4(wa14) and fat-3(wa22) mutants but mec-4(d) animals only have two touch neurons left as viewed by not fat-1(wa9) (Fig 5D). The proportion of animals with three or the zdIs5 marker, which labels six touch sensory neurons in wild more surviving touch neurons increased from 6 Æ 1% in mec-4(d) type (Fig 5A). We found that atgl-1 mutation reduces the to 35 Æ 1% in fat-4(wa14); mec-4(d) and 43 Æ 3% in fat-3(wa22); neuronal degeneration of mec-4(d) (Fig 5A and B). Specifically, mec-4(d) worms. These results indicate that reducing the biosynthe- the proportion of animals with three or more surviving touch sis of PUFAs alleviates the neuronal loss of mec-4(d). neurons increased from 6 Æ 1% in mec-4(d) to 28 Æ 1% in atgl-1 (xd314); mec-4(d) worms (Fig 5C). Similarly, lid-1 mutation also Since ATGL-1/LID-1-mediated neuronal lipolysis is involved in reduced the neuronal loss of mec-4(d). The proportion of animals PUFA-mediated touch sensation (Fig 4A and B), and both atgl-1 with three or more surviving touch neurons increased to and lid-1 mutants have reduced mec-4(d)-induced neurodegenera- 23 Æ 2% in lid-1(xd288); mec-4(d) worms (Fig 5C). Moreover, tion (Fig 5B), we next explored the neuroprotective effect of the reduction of mec-4(d)-triggered neurodegeneration by atgl-1 double mutations in neuronal lipolysis and de novo synthesis of and lid-1 mutations can be reversed by neuronal-specific expres- PUFAs. Interestingly, atgl-1(xd314); fat-1(wa9), atgl-1(xd314); fat- sion of wild-type atgl-1 and lid-1, respectively (Fig 5C). These 3(wa22) and atgl-1(xd314); fat-4(wa14) double mutants all have data indicate that defective neuronal lipolysis reduces neuron significantly reduced mec-4(d)-triggered neurodegeneration hyperactivation-triggered neurodegeneration. compared with single mutants (Fig 5D). The proportion of animals with three or more surviving touch neurons increased from Defects in neuronal lipolysis and de novo PUFA biosynthesis 6 Æ 1% in mec-4(d) to 52 Æ 3% in fat-1(wa9); atgl-1(xd314); synergistically decrease the neuronal loss caused by mec-4(d) mec-4(d), 78 Æ 2% in fat-4(wa14); atgl-1(xd314); mec-4(d) and 77 Æ 1% in fat-3(wa22); atgl-1(xd314); mec-4(d) worms. mboa-2 Hyperactivated neurons are vulnerable because the peroxidation of overexpression also reduced mec-4(d) neurodegeneration, although fatty acids, in particular PUFAs, can result in neurodegeneration it did not exhibit a synergistic effect with fat-4(wa14) (Ioannou et al, 2019). fat-1, fat-3, and fat-4 encode key enzymes for (Appendix Fig S6). Therefore, blocking de novo PUFA synthesis de novo synthesis of different PUFAs in C. elegans (Watts \& Browse, and neuronal lipolysis together greatly protects neurons from degeneration caused by mec-4(d). ª 2020 The Authors. EMBO reports e50214 | 2020 7 of 15

EMBO reports B Leilei Yang et al A C D E Figure 5. PUFA synthesis defect significantly enhances the beneficial effect of atgl-1 mutants in preventing the neurodegeneration of mec-4(d). A Image of touch neurons viewed by zdIs5 in different genetic backgrounds. Scale bar: 20 lm. B Quantification of the percentage of animals with 0, 1, 2, 3, 4, 5, and 6 surviving touch neurons in different genetic backgrounds. Data show mean Æ SEM. At least 50 animals per strain were analyzed in each experiment. Number of experiments n = 5. C The percentage of worms that have three or more surviving touch neurons in different genetic backgrounds in (B). The lid-1(xd288) and atgl-1(xd314) mutations significantly increase the number of surviving neurons. Neuron-specific expression of atgl-1 or lid-1 reverses the protective effect of atgl-1 or lid-1 mutation. The data were analyzed using one-way ANOVA with Turkey’s multiple comparison test. Asterisks signify significant differences between the groups under the crossbars. ***P < 0.001, ****P < 0.0001. Data show mean Æ SEM. Number of experiments n = 5, with at least 50 animals per strain analyzed in each experiment. D The percentage of worms that have three or more surviving touch neurons in different genetic backgrounds. Mutations of fat-1, fat-3 and fat-4 significantly enhance the suppression effect of atgl-1(xd314) in preventing the neurodegeneration of mec-4(d). The data were analyzed using one-way ANOVA with Bonferroni’s multiple comparison test. ### and #### denote significant differences as compared to atgl-1(xd314); mec-4(d). Asterisks signify significant differences as compared to mec-4 (d). ns: not statistically significant. ****P < 0.0001. ###P < 0.001, ####P < 0.0001. Data show mean Æ SEM. Number of experiments n ≥ 3, with at least 50 animals per strain analyzed in each experiment. E The percentage of worms that have three or more surviving touch neurons in different genetic backgrounds supplemented with different fatty acids. AA and EPA but not LA significantly enhance the neurodegeneration triggered by mec-4(d). The data were analyzed using one-way ANOVA with Bonferroni’s multiple comparison test. Asterisks signify significant differences between the groups under the crossbars. ns: not statistically significant. ***P < 0.001, ****P < 0.0001. Data show mean Æ SEM. Number of experiments n ≥ 3, with at least 50 animals per strain analyzed in each experiment. The PUFAs AA and EPA promote the neurodegeneration nÀ6) and EPA (C20:5, nÀ3) or their derivatives, may promote mec- of mec-4(d) 4(d)-induced neurodegeneration. We fed worms on NGM medium supplemented with different fatty acids. fat-4(wa14); mec-4(d) and Since fat-3(wa22) and fat-4(wa14) show similar neural protection fat-4(wa14); atgl-1(xd314); mec-4(d) animals show a significant phenotypes (Fig 5D), we reasoned that PUFAs, such as AA (C20:4, increase of neurodegeneration when grown in AA-supplemented 8 of 15 EMBO reports e50214 | 2020 ª 2020 The Authors.

Leilei Yang et al EMBO reports NGM compared with normal NGM (Fig 5E). Feeding worms with hyperactivation in both C. elegans and mammals (Calixto et al, EPA, but not linoleic acid (LA, C18:2), had a similar effect (Fig 5E). 2012; Sangaletti et al, 2017; Ioannou et al, 2019). We then tested In contrast, feeding worms with a mixture containing two saturated whether ROS contributes to PUFA-mediated neurodegeneration. We fatty acids (palmitic acid (PA, C16:0) and stearic acid (SA, C18:0)) fed mec-4(d), fat-4(wa14); mec-4(d) or atgl-1(xd314); mec-4(d) and one monounsaturated fatty acid (oleic acid (OA, C18:1)) did not mutants with the antioxidants N-acetylcysteine (NAC), vitamin C affect the neurodegeneration phenotype in fat-4(wa14); mec-4(d) or vitamin E. The survival of neurons was slightly increased in and fat-4(wa14); atgl-1(xd314); mec-4(d) (Appendix Fig S6). These NAC-treated mec-4(d) worms and vitamin C-treated fat-4(wa14); results indicate that the PUFAs AA and EPA promote the neurode- mec-4(d) worms, but not in worms subjected to other treatments generation of mec-4(d). (Fig EV3A). We introduced mutation of sod-2, which encodes a ROS scavenger, and found that it did not significantly decrease the PUFAs promote neurodegeneration through incorporation number of surviving neurons in atgl-1(xd314); mec-4(d), fat-4 into phospholipids (wa14); mec-4(d), or fat-4(wa14); atgl-1(xd314); mec-4(d) (Fig EV3B). Furthermore, overexpression of SOD-1, SOD-2, SOD-3, Elevated ROS and subsequent lipid peroxidation are proposed to be or SOD-4 in touch neurons did not affect mec-4(d)-triggered a key event for neurodegeneration triggered by neuron neurodegeneration (Fig EV3C). Lipid peroxides produced by PUFAs AB C DE FG Figure 6. EMBO reports e50214 | 2020 9 of 15 ª 2020 The Authors.

EMBO reports Leilei Yang et al ◀ Figure 6. PUFA-containing phospholipids regulate neurodegeneration of mec-4(d). A Lipidomic data show that the relative content of total C20:3 is increased in atgl-1(xd314); mec-4(d), fat-4(wa14); mec-4(d) and atgl-1(xd314); fat-4(wa14); mec-4(d). The increase is greatest in mutants with the fat-4 mutation. B Lipidomic data show that the relative content of total C20:4 is increased in atgl-1(xd314); mec-4(d), fat-4(wa14); mec-4(d) and atgl-1(xd314); fat-4(wa14); mec-4(d). The increase is greatest in mutants with the fat-4 mutation. C Lipidomic data show that the relative content of total C20:5 is slightly increased in atgl-1(xd314); mec-4(d) but is dramatically decreased in fat-4(wa14); mec-4(d) and atgl-1(xd314); fat-4(wa14); mec-4(d), which matches the role of FAT-4 in synthesizing C20:5 (EPA). D The percentage of C20:3-, C20:4-, or C20:5-containing TAGs to total lipids containing the same PUFA in different genetic backgrounds. The percentage of C20:4- containing TAGs to total lipids containing C20:4 is significantly increased in atgl-1 and fat-4 single mutants. The percentage is further increased in atgl-1 and fat-4 double mutant. E The percentage of C20:3-, C20:4-, or C20:5-containing phospholipids to total lipids containing the same PUFA in different genetic backgrounds. The percentage of C20:4-containing phospholipids to total lipids containing C20:4 is significantly decreased in atgl-1 and fat-4 single mutants. The percentage is further decreased in atgl-1 and fat-4 double mutant. F The percentage of worms that have three or more surviving touch neurons in different genetic backgrounds. mboa-7 mutation significantly increases the percentage of three or more surviving touch neurons in mec-4(d), atgl-1(xd314); mec-4(d), fat-4(wa14); mec-4(d) and atgl-1(xd314); fat-4(wa14); mec-4(d). The data were analyzed using one-way ANOVA with Bonferroni’s multiple comparison test. Asterisks signify significant differences between the groups under the crossbars. **P < 0.01, ****P < 0.0001. Data show mean Æ SEM. Number of experiments n ≥ 4, with at least 50 animals per strain analyzed in each experiment. G The percentage of C20:4-containing phospholipids to total lipids that have C20:4 is dramatically decreased in mboa-7(gk399); atgl-1(xd314); fat-4(wa14); mec-4(d). The data were analyzed using one-way ANOVA with Dunnett’s multiple comparison test. Asterisks signify significant differences as compared to mec-4(d). ***P < 0.001, ****P < 0.0001. Data show mean Æ SEM. Number of experiments n = 5, with about 10,000 animals per strain analyzed in each experiment. Data information: (A–E) FFA: free fatty acids; TAG: triacylglycerol; PL: phospholipids. The data were analyzed using two-way ANOVA with Dunnett’s multiple comparison test. Asterisks denote significant differences as compared to mec-4(d). ns: not statistically significant. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Data show mean Æ SEM. Number of experiments n = 5, with about 10,000 animals per strain analyzed in each experiment. can be cleared with glutathione peroxidase. We found that overex- However, in the double mutants, we also noticed persistent and pression of GPX-1, GPX-2, and GPX-6, which are homologs of further changes in PUFA partitioning between phospholipids and human phospholipid hydroperoxide glutathione peroxidase (GPx4), TAG, in particular the partitioning of C20:4. There is a further reduc- do not affect the neurodegeneration of mec-4(d) and fat-4(wa14); tion of the percentage of C20:4-containing phospholipids in atgl-1; mec-4(d) (Fig EV3D and E). Therefore, ROS may play a role, fat-4 double mutants compared with either single mutant (Fig 6D although probably not a major role, in neuronal lipolysis and and E). The altered partitioning of PUFAs between phospholipids PUFA-mediated neuronal loss caused by mec-4(d). and TAG correlates well with the moderate neuroprotective effects in atgl-1 and fat-4 single mutants and the strong neuroprotective To further investigate how PUFAs promote neurodegeneration, effects in atgl-1; fat-4 double mutants. These results suggest that we performed a lipidomic analysis of mec-4(d), atgl-1(xd314); mec-4 PUFA-containing phospholipids may promote neurodegeneration. (d), fat-4(wa14); mec-4(d) and fat-4(wa14); atgl-1(xd314); mec-4(d) to find correlations between the level of PUFA-containing lipids and We further explored the hypothesis that PUFA-containing phos- the severity of neurodegeneration. In fat-4 mutants, C20:3- or C20:4- pholipids contribute to mec-4(d)-induced neurodegeneration. containing lipids are increased significantly while C20:5-containing MBOA-7 incorporates PUFAs into phospholipids, especially PI (Lee lipids are decreased dramatically when normalized to total polar et al, 2008). We found that mboa-7 mutation alone significantly lipids (Fig 6A–C). This is consistent with the role of FAT-4 in gener- increased the percentage of mec-4(d) animals with ≥ 3 surviving ating EPA. Furthermore, although our analysis cannot distinguish touch neurons (from ~ 5% to ~ 20%) (Fig 6F). Moreover, mboa-7 between eicosatetraenoic acid (ETA, C20:4 nÀ3) and AA (C20:4 mutation further increases neuron survival in atgl-1(xd314); mec-4 nÀ6), the increased C20:4 in fat-4 mutants is ETA, as revealed in a (d), fat-4(wa14); mec-4(d) and atgl-1(xd314); fat-4(wa14); mec-4(d) previous study (Watts \& Browse, 2002). Moreover, despite the (Figs 6F, and EV4A and B). The enhancement of the neuroprotec- increased level of C20:4-containing lipids, the percentage of C20:4- tive effect is probably because the mboa-7 mutation reduces PUFA containing phospholipids is reduced in fat-4 mutants, while the incorporation into phospholipids. Consistent with that, the percent- percentage of C20:4-containing TAG is increased significantly age of C20:4-containing phospholipids is further decreased in atgl-1 (Fig 6D and E). Therefore, fat-4 mutation reduces the levels of some (xd314); fat-4(wa14); mboa-7(gk399); mec-4(d) mutants compared PUFAs and may also alter the partitioning of PUFAs between phos- with atgl-1(xd314); fat-4(wa14); mec-4(d) mutants (Fig 6G). pholipids and TAG. Together, these results indicate that PUFAs promote neuron hyper- activation-mediated neurodegeneration of mec-4(d) through their Compared with fat-4 mutants, the levels of C20:3-, C20:4-, or incorporation into phospholipids. C20:5-containing lipids are slightly increased in atgl-1 mutants, when normalized to total polar lipids (Fig 6A–C). Interestingly, consistent Discussion with a storage role for LDs, the percentages of C20:3, C20:4, or C20:5-containing TAGs are significantly increased, while the percent- This study addresses two questions: Why are there normally no LDs ages of C20:3, C20:4, or C20:5-containing phospholipids are signifi- in neurons, and what are the consequences of LD accumulation in cantly decreased in atgl-1 mutants (Fig 6D and E). Therefore, atgl-1 neurons? We reveal that the balance of lipolysis and lipogenesis mutation alters the partitioning of PUFAs into phospholipids and regulates neuronal LD dynamics. In particular, we show that defec- TAG. tive ATGL-1/LID-1-mediated lipolysis causes LD accumulation In atgl-1; fat-4 double mutants, there were similar changes of C20:3, C20:4, and C20:5 fatty acids as in fat-4 single mutants. 10 of 15 EMBO reports e50214 | 2020 ª 2020 The Authors.

Leilei Yang et al EMBO reports A B Figure 7. Neuronal lipolysis participates in the homeostasis of PUFAs, which regulates neuronal functions and neurodegeneration through incorporation of PUFAs into phospholipids. A, B The homeostasis of PUFAs is determined by PUFA de novo synthesis and the balance of neuronal lipolysis and lipogenesis. The incorporation of PUFAs into phospholipids regulates normal neuronal functions (such as touch sensation) and affects neurodegeneration of mec-4(d). autonomously in C. elegans neurons. Defective neuronal lipolysis why they do not form LDs themselves. Could the formation of LDs affects normal gentle touch sensation, while importantly, it allevi- in neurons be worm-specific because most neurons, including touch ates neurodegeneration triggered by neuron hyperactivation. The neurons, in C. elegans are not surrounded by glial cells? The appear- neuroprotective effect is synergistically enhanced when de novo ance of LDs in DDHD2À/À mice, which have a defect in a lipase, biosynthesis of PUFAs is blocked. Lastly, the incorporation of strongly argues against this possibility (Inloes et al, 2014). Alterna- PUFAs into phospholipids likely underlies neuronal lipolysis and tively, it is possible that under conditions of intact neuronal lipoly- PUFA-mediated neurodegeneration. Therefore, both neuronal LD sis, excess lipids can still engage in generating toxic hyperoxidated dynamics and de novo PUFA synthesis regulate the availability of PUFAs in neurons. Therefore, to avoid neuronal toxicity, exporting PUFAs, which may be incorporated into phospholipids to maintain excess lipids to glia is a better choice. Examining the flux of lipid the proper function of neurons under normal conditions and to export and lipid incorporation into neutral lipids may provide a promote neurodegeneration under stress conditions (Fig 7). further explanation. LDs in neuron: a tug-of-war between lipogenesis and lipolysis Interestingly, mice with deficiencies of ATGL or CGI-58 accumu- late massive amounts of neutral lipids in many tissues/organs Neurons mainly use glucose to generate energy instead of lipids to including brain (Radner et al, 2010; Etschmaier et al, 2011). Simi- avoid extra ROS production by lipid b-oxidation. Although normally larly, lipid accumulation in brain was clearly demonstrated in CGI- there are no LDs in neurons, our results demonstrate that neurons 58 human patients (Huigen et al, 2015). However, in both mouse have the ability to form and hydrolyze LDs. Both overexpression of and human studies, there is no direct evidence of LD accumulation lipogenesis genes and loss of the key lipolysis genes atgl-1 and lid-1 in neurons. Compared with LDs in glia or blood vessels, neuronal lead to LD accumulation in neurons. It is possible that the lipogene- LDs may be too small to observe and may therefore be neglected. sis in neurons is limited and lipolysis is kept at a relatively high Specific examination of neuronal LDs in these lipolysis-defective level, which promotes a fast turnover of LDs to meet the lipid mutants will be required to characterize the neuronal defects. requirements of neurons. Thus, there is a tug-of-war between lipo- genesis and lipolysis in neurons. Neuronal lipolysis and neuronal normal function Both atgl-1 and lid-1 are highly conserved from worm to Here, we show that intact neuronal lipolysis is important for main- mammals. Previous studies on Drosophila and cultured hippocam- taining normal neuron function in worm. Currently, there is no pal neurons show that instead of forming LDs, neurons transfer the study on neurological function of ATGL-mediated lipolysis in mice. excess lipids into glial cells to form LDs under stress conditions (Bai- In patients with mutated CGI-58, neurological symptoms, including ley et al, 2015; Liu et al, 2015, 2017; Ioannou et al, 2019). If cognitive impairment and psychiatric disorders, are a frequent char- neurons have the ability to form and hydrolyze LDs, it is puzzling acteristic (Schweiger et al, 2009). A human ATGL mutation case ª 2020 The Authors. EMBO reports e50214 | 2020 11 of 15

EMBO reports Leilei Yang et al with global cognitive impairment was also reported (Massa et al, because only limited amounts of the small antioxidant molecules 2016). Therefore, the critical role of neuronal lipolysis in neurons is actually reach the neuronal cells. Phospholipids are building blocks likely conserved. The detail neuron-specific pathological phenotype of all membranes in mammalian cells, providing the structural integ- and its underlying mechanism remain to be elucidated in ATGL or rity that is necessary for protein function. They also serve as precur- CGI-58 mouse mutant models and more importantly in human sors for various second messengers such as AA, DHA, ceramide, 1,2- patients. diacylglycerol, IP3, phosphatidic acid, and lyso-phospholipids (O’Donnell et al, 2018). Generally, phospholipids with PUFA chains Our results show that neuronal lipolysis participates in PUFA- are more flexible. The cis-double bond lowers the packing density of mediated touch sensation. Through incorporation into membrane- the acyl chains, which increases membrane fluidity, bending stiff- forming phospholipids or as signals regulating neuron activity, ness, and effective viscosity (Holthuis \& Menon, 2014; Vasquez et al, PUFAs and their derivatives play important roles in neurons (Watts 2014). One possibility is that a reduced level of PUFA-containing \& Ristow, 2017). In C. elegans mechanical and touch sensation, phospholipids impacts membrane fluidity and in turn reduces ion- PUFA-containing phospholipids modulate the activity of particular channel activity. For example, AA and DHA can regulate channels channels through membrane remodeling and changes of membrane such as K+ channels and TRPV4 (Villarroel \& Schwarz, 1996; Hori- fluidity (Matsuda et al, 2008; Vasquez et al, 2014). In addition, moto et al, 1997; Caires et al, 2017). PUFA synthesis mutations also PUFA depletion causes defects in neurotransmission (Lesa et al, decreased neuronal Ca2+ transients in C. elegans (Kahn-Kirby et al, 2003; Marza \& Lesa, 2006). It remains to be explored whether the 2004). Thus, it is possible that PUFA-containing phospholipids regu- neurological defects in ATGL or CGI-58 human patients are linked to late the constitutively active MEC-4(d) ion channel through Ca2+ altered PUFA-mediated neuronal functions. transients. Moreover, phospholipids with PUFAs can be degraded and produce the second messengers 2-AG, 1,2-diacylglycerol and Neuronal LDs and neurodegeneration IP3. Therefore, another possibility is that reduced levels of phospho- lipids with PUFAs affect downstream signals that execute neurode- In this study, we found that LDs have a neuroprotective effect in generation triggered by ion-channel hyperactivation. the neuron hyperactivation-triggered neurodegeneration model. Interestingly, neuronal LDs also show a protective effect in another Our study indicates that reducing the influx of PUFAs into neurodegenerative disease, Parkinson’s disease. The expression of neuron phospholipids protects neurons from degeneration triggered the Parkinson’s disease protein a-Synuclein alters the cellular lipid by hyperactivation. This suggests a potential strategy for treating profile, notably by elevating DAG and monounsaturated fatty acid neurodegeneration caused by ion hyperactivation. OA. By partitioning excess DAG and OA into TAG, the formation of LDs reduces neuron death triggered by expression of a-Synu- Materials and Methods clein (Fanning et al, 2019). In these two neurodegeneration models, LD acts as a harbor to store excess and/or toxic fatty acids Strains which promote neurodegeneration, although the detail mecha- nisms of neuroprotective effect of LDs are different. Therefore, the Worms were cultured on OP50-seeded nematode growth medium presence of LDs may be beneficial for neurodegenerative diseases (NGM) plates at 22°C (Brenner, 1974). Wild-type (N2), CB4856, in general. VC3025 C37H5.2(ok3722), RB2386 C37H5.3(ok3245), TU253 mec-4 (u253), CB1370 daf-2(e1370), CB4037 glp-1(e2141), CB1611 mec-4 Besides Parkinson’s disease, many other neuronal diseases, in (e1611), VC942 mboa-7(gk399), and GA184 sod-2(gk257) were particular the HSP diseases, are reported to be accompanied by obtained from the Caenorhabditis Genetics Center (CGC). B0524.2 abnormal levels of neutral lipids, and the disease genes are related (tm6739) and D1054.1(tm3111) were obtained from the NBRP to LD dynamics. Knockdown of Spartin, also known as SPG20, (Japan). Mutant strains BX24 fat-1(wa9), BX30 fat-3(wa22), and increases the number and size of LDs in cells loaded with oleic acid BX17 fat-4(wa14) were kindly provided by Dr. Bin Liang. lid-1 (Papadopoulos et al, 2015). In another HSP model, DDHD2À/À (xd288), atgl-1(xd310), and atgl-1(xd314) were generated by EMS. mice exhibit LD accumulation in neurons (Inloes et al, 2014). xdIs182 (Prab-3::Cre/Podr-1::RFP) was generated by integration of Ex Despite the alteration of LD homeostasis, the role of LD accumula- (Prab-3::Cre/Podr-1::RFP). atgl-1-loxP(xd426) was generated by tion in neurons under these disease conditions has not been inves- CRISPR-Cas9-mediated genome editing. All the transgenic worms tigated. Our findings also raise the possibility that the accumulation were generated by microinjection of the respective plasmids or of LDs is a compensatory response to relieve neuronal stress fosmids with co-injection markers. All the fosmids were kindly in some HSP diseases. Further studies will be required to explore provided by Dr. Xiaochen Wang. this possibility. Molecular biology PUFAs and neurodegeneration The promoters of Punc-119 (1,235 bp), Pvha-6 (1,808 bp), Phlh-17 We show that reducing PUFA incorporation into phospholipids has a (2,049 bp), Pajm-1 (3,500 bp), Prab-3 (1,207 bp), Patgl-1 (3,004 bp), beneficial effect in alleviating neuronal loss in a neurodegeneration and Pmec-4 (206 bp) were amplified from N2 genomic DNA. The model. The mechanism underlying this effect is not fully clear. coding regions of lid-1, atgl-1, C37H5.2, C37H5.3, B0524.2, D1054.1, Although we only observed a marginal effect of NAC and vitamin C mboa-2, F59A1.10, K07B1.4, sod-1, sod-2, sod-3, sod-4, gpx-1, gpx-2, treatment, it is still possible that lipid peroxides produced from and gpx-6 were amplified from cDNA. The promotor and target PUFAs contribute to the neuronal degeneration. It is possible that these antioxidants do not show a strong effect on neurodegeneration 12 of 15 EMBO reports e50214 | 2020 ª 2020 The Authors.

Leilei Yang et al EMBO reports cDNA were inserted into plasmid Dpsm. The split GFP1-10 and Fatty acid, NAC, vitamin C, and vitamin E supplementation GFP11 were inserted into pPD95.75. Patgl-1 and Prab-3 were inserted upstream of GFP1-10 and GFP11, respectively. Prab-3::Cre was gener- AA (arachidonic acid), NAC (N-Acetyl-L-cysteine), vitamin C, and ated by replacing the eft-3 promotor in pDD104, which was kindly vitamin E were acquired from Aladdin. SA (stearic acid), OA (oleic provided by Dr. Shiqing Cai. atgl-1-sgRNAs for loxP insertion were acid), PA (palmitate acid), EPA (eicosapentaenoic acid), LA (linoleic designed using the Zhang laboratory’s CRISPR design tool at http:// acid), and Tergitol (70%) were acquired from Sigma-Aldrich. FAs, crispr.mit.edu to select the target sites. The sgRNAs were inserted dissolved in ethanol with 0.1% Tergitol, were added to NGM agar to into pDD162, which was kindly provided by Dr. Guangshuo Ou. All reach a final concentration of approximately 200 lM. NAC and vita- the constructed plasmids were verified by sequencing. min C was dissolved in ddH2O and added into NGM agar to reach a final concentration of approximately 10 mM, respectively. Vitamin EMS screen E was dissolved in methanol and added into NGM agar to reach a final concentration of approximately 200 lg/ml. Plates were seeded We treated L4 and young adult xdIs109 (Punc-119::PLIN1::GFP) with Escherichia coli OP50 and kept at room temperature (Vasquez worms with ethyl methanesulfonate (EMS). About 2,500 F1 were et al, 2014). For each strain, about 10 L4 worms were placed on the screened. We observed LDs using a Zeiss compound microscope. plate. We quantified the phenotype in the next generation. Mutants were selected that showed LD accumulation in neurons. Stimulated Raman Scattering microscopy SNP mapping and fosmid rescue To quantify lipid droplet metrics in the touch neurons, 1-day adult We used rapid single nucleotide polymorphism (SNP) mapping (Davis worms were imaged using Stimulated Raman Scattering (SRS) et al, 2005). Briefly, we crossed the mutant with CB4856 and picked microscopy. Worms were mounted onto 2% agarose pads with F2 animals with LD accumulation. We amplified some SNPs using F2 0.5% NaN3 as anesthetic on glass microscope slides and imaged or F3 lysates as templates, and we digested the PCR products with using a 60× water objective (UPlanAPO/IR; 1.2 N.A.; Olympus). A enzymes. After chromosome mapping and interval mapping, we iden- femtosecond-pulsed laser and picosecond-pulsed laser were used for tified a narrow region. Then, we did the fosmid rescue assay. Fosmids simultaneously imaging label-free lipids (SRS channel) and GFP- were injected individually into mutant worms at 5–10 ng/ll together labeled neurons (fluorescence channel). Images were analyzed with 50 ng/ll Podr-1::RFP which acts as the co-injection marker. using ImageJ software (NIH). The neuronal area was selected using the fluorescent signal based on several Z-projected stacks to confirm Oil red O staining the presence of lipid droplets within the cell and to exclude extracel- lular signal. Oil red O was acquired from Sigma-Aldrich. Oil red O staining was conducted as previously reported (O’Rourke et al, 2009). EM analysis BODIPY staining Young adult worms were collected for high-pressure freezing and freeze-substitution, embedding and sectioning following a procedure Worms were fixed as for Oil red O staining (O’Rourke et al, 2009). essentially as described by (Weimer, 2006). Then, the sections were Worms were washed three times with 1× PBS to remove visualized with a Hitachi HT7700 and pictures were recorded on a paraformaldehyde. After that, worms were incubated in PBS with 4,008 × 2,672 CCD camera. 2 lg/ml BODIPY (Invitrogen) for 30 min in the dark with gentle rocking. Then worms were washed with 1× PBS and imaged directly Lipidomic analysis by confocal microscopy (Leica SP8) (Tian et al, 2011). Worms were washed 9–10 times in M9. More than 10,000 worms Behavioral assays per genotype were collected per sample and five samples were analyzed per genotype. Lipid extraction and analysis were Gentle touch sensitivity was tested and scored as described (Hart, conducted as previously reported (Lam et al, 2014). The lipid 2006; Vasquez et al, 2014). Briefly, we performed ten-trial touch content was normalized by the mole fraction of each lipid to total assays. We scored the touch response percentage by stroking an polar lipids. eyebrow hair across the anterior and posterior body. Twenty-five animals were tested in each trial, and results were compared across Statistical analyses three trials. All assays were performed by investigators blinded to genotype and/or treatment. All data are presented as mean Æ SEM. The data were first analyzed normal distribution. When the data meet the normal distribution, Quantification of neuron loss we used one-way ANOVA or two-way ANOVA for statistically significant overall results. When the data do not meet the normal Animals were mounted on thin agarose pads and immobilized by 1- distribution, we used Kruskal–Wallis test. Post hoc multiple compar- phenoxy-2-propanol (10 ll/ml). Animals were visualized under a ison tests were carried out to determine significant differences. 20× objective. GFP-expressing touch neurons were counted in Significant difference is noted with a number sign (#) or an asterisk synchronized L4 animals. (*). ns represents not statistically significant. *P < 0.05, **P < 0.01, ª 2020 The Authors. EMBO reports e50214 | 2020 13 of 15

EMBO reports Leilei Yang et al ***P < 0.001, ****P < 0.0001. #P < 0.05, ##P < 0.01, ###P < 0.001, Davis MW, Hammarlund M, Harrach T, Hullett P, Olsen S, Jorgensen EM ####P < 0.0001. (2005) Rapid single nucleotide polymorphism mapping in C. elegans. BMC Genom 6: 118 Data availability Ding L, Yang X, Tian H, Liang J, Zhang F, Wang G, Wang Y, Ding M, Shui G, This study includes no data deposited in external repositories. Huang X (2018) Seipin regulates lipid homeostasis by ensuring calcium- dependent mitochondrial metabolism. EMBO J 37: e97572 Expanded View for this article is available online. Driscoll M, Chalfie M (1991) The mec-4 gene is a member of a family of Acknowledgements Caenorhabditis elegans genes that can mutate to induce neuronal degeneration. Nature 349: 588 – 593 We thank Drs. Shiqing Cai, Bin Liang, Pingsheng Liu, Guangshuo Ou, Meng C. Wang, and Xiaochen Wang for providing reagents and helpful discussions. We Eastman SW, Yassaee M, Bieniasz PD (2009) A role for ubiquitin ligases and thank the Caenorhabditis Genetics Center (CGC) and National BioResource Spartin/SPG20 in lipid droplet turnover. J Cell Biol 184: 881 – 894 Project (NBRP) for providing strains. This research was supported by grants 31630019, 2018YFA0506902, 91954207, and 2016YFA0500100 from the Ebihara C, Ebihara K, Aizawa-Abe M, Mashimo T, Tomita T, Zhao M, Gumbilai National Natural Science Foundation of China and National Key R\&D program V, Kusakabe T, Yamamoto Y, Aotani D et al (2015) Seipin is necessary for of China. normal brain development and spermatogenesis in addition to adipogenesis. Hum Mol Genet 24: 4238 – 4249 Author contributions Etschmaier K, Becker T, Eichmann TO, Schweinzer C, Scholler M, Tam- LY directed the project and her work includes EMS screen, molecular biology, Amersdorfer C, Poeckl M, Schuligoi R, Kober A, Chirackal Manavalan AP SNP mapping and fosmid rescue, Oil red O staining, behavioral assays, et al (2011) Adipose triglyceride lipase affects triacylglycerol metabolism neuronal loss analysis, fatty acid, NAC, vitamin C and vitamin E supplementa- at brain barriers. J Neurochem 119: 1016 – 1028 tion and analyzing the results. JL and MD did the EM imaging and analysis. SML and GS did the lipidomic analysis. AY did the SRS imaging. LY and XH Fanning S, Haque A, Imberdis T, Baru V, Barrasa MI, Nuber S, Termine D, wrote the manuscript. XH guided the project and edited the manuscript. Ramalingam N, Ho GPH, Noble T et al (2019) Lipidomic analysis of alpha- synuclein neurotoxicity identifies stearoyl CoA Desaturase as a target for Conflict of interest parkinson treatment. Mol Cell 73: 1001 – 1014 All authors declare that they have no conflict of interest. Hart AC (2006) Behavior. WormBook, https://doi.org/10.1895/wormbook.1891. 1887.1891 References Holthuis JC, Menon AK (2014) Lipid landscapes and pipelines in membrane Bailey AP, Koster G, Guillermier C, Hirst EM, MacRae JI, Lechene CP, Postle AD, homeostasis. Nature 510: 48 – 57 Gould AP (2015) Antioxidant role for lipid droplets in a stem cell niche of Drosophila. Cell 163: 340 – 353 Horimoto N, Nabekura J, Ogawa T (1997) Arachidonic acid activation of potassium channels in rat visual cortex neurons. Neuroscience 77: 661 – 671 Ben Selma Z, Yilmaz S, Schischmanoff PO, Blom A, Ozogul C, Laroche L, Caux F (2007) A novel S115G mutation of CGI-58 in a Turkish patient with Huigen MC, van der Graaf M, Morava E, Dassel AC, van Steensel MA, Seyger Dorfman-Chanarin syndrome. J Invest Dermatol 127: 2273 – 2276 MM, Wevers RA, Willemsen MA (2015) Cerebral lipid accumulation in Chanarin-Dorfman Syndrome. Mol Genet Metab 114: 51 – 54 Bi J, Xiang Y, Chen H, Liu Z, Gronke S, Kuhnlein RP, Huang X (2012) Opposite and redundant roles of the two Drosophila perilipins in lipid mobilization. Inloes JM, Hsu KL, Dix MM, Viader A, Masuda K, Takei T, Wood MR, Cravatt BF J Cell Sci 125: 3568 – 3577 (2014) The hereditary spastic paraplegia-related enzyme DDHD2 is a principal brain triglyceride lipase. Proc Natl Acad Sci USA 111: 14924 – 14929 Brenner S (1974) The genetics of Caenorhabditis elegans. Genetics 77: 71 – 94 Cabantous S, Terwilliger TC, Waldo GS (2005) Protein tagging and detection Ioannou MS, Jackson J, Sheu SH, Chang CL, Weigel AV, Liu H, Pasolli HA, Xu CS, Pang S, Matthies D et al (2019) Neuron-astrocyte metabolic with engineered self-assembling fragments of green fluorescent protein. coupling protects against activity-induced fatty acid toxicity. Cell 177: Nat Biotechnol 23: 102 – 107 1522 – 1535 Caires R, Sierra-Valdez FJ, Millet JRM, Herwig JD, Roan E, Vasquez V, Cordero- Morales JF (2017) Omega-3 fatty acids modulate TRPV4 function through Kahn-Kirby AH, Dantzker JL, Apicella AJ, Schafer WR, Browse J, Bargmann CI, plasma membrane remodeling. Cell Rep 21: 246 – 258 Watts JL (2004) Specific polyunsaturated fatty acids drive TRPV-dependent Calixto A, Jara JS, Court FA (2012) Diapause formation and downregulation of sensory signaling in vivo. Cell 119: 889 – 900 insulin-like signaling via DAF-16/FOXO delays axonal degeneration and neuronal loss. PLoS Genet 8: e1003141 Kis V, Barti B, Lippai M, Sass M (2015) Specialized cortex glial cells Chen L, Chen X-W, Huang X, Song B-L, Wang Y, Wang Y (2019) Regulation of accumulate lipid droplets in Drosophila melanogaster. PLoS ONE 10: glucose and lipid metabolism in health and disease. Sci China Life Sci 62: e0131250 1420 – 1458. Cole NB, Murphy DD, Grider T, Rueter S, Brasaemle D, Nussbaum RL (2002) Klemm RW, Norton JP, Cole RA, Li CS, Park SH, Crane MM, Li L, Jin D, Boye- Lipid droplet binding and oligomerization properties of the Parkinson’s Doe A, Liu TY et al (2013) A conserved role for atlastin GTPases in disease protein alpha-synuclein. J Biol Chem 277: 6344 – 6352 regulating lipid droplet size. Cell Rep 3: 1465 – 1475 Lam SM, Wang Y, Duan X, Wenk MR, Kalaria RN, Chen CP, Lai MK, Shui G (2014) Brain lipidomes of subcortical ischemic vascular dementia and mixed dementia. Neurobiol Aging 35: 2369 – 2381 Lass A, Zimmermann R, Haemmerle G, Riederer M, Schoiswohl G, Schweiger M, Kienesberger P, Strauss JG, Gorkiewicz G, Zechner R (2006) Adipose triglyceride lipase-mediated lipolysis of cellular fat stores is activated by CGI-58 and defective in Chanarin-Dorfman Syndrome. Cell Metab 3: 309 – 319 Lee HC, Inoue T, Imae R, Kono N, Shirae S, Matsuda S, Gengyo-Ando K, Mitani S, Arai H (2008) Caenorhabditis elegans mboa-7, a member of the MBOAT 14 of 15 EMBO reports e50214 | 2020 ª 2020 The Authors.

Leilei Yang et al EMBO reports family, is required for selective incorporation of polyunsaturated fatty binds to lipid droplets and affects lipid metabolism. PLoS Genet 11: acids into phosphatidylinositol. Mol Biol Cell 19: 1174 – 1184 e1005149 Lee JH, Kong J, Jang JY, Han JS, Ji Y, Lee J, Kim JB (2014) Lipid droplet protein Radner FP, Streith IE, Schoiswohl G, Schweiger M, Kumari M, Eichmann TO, LID-1 mediates ATGL-1-dependent lipolysis during fasting in Rechberger G, Koefeler HC, Eder S, Schauer S et al (2010) Growth Caenorhabditis elegans. Mol Cell Biol 34: 4165 – 4176 retardation, impaired triacylglycerol catabolism, hepatic steatosis, and Lesa GM, Palfreyman M, Hall DH, Clandinin MT, Rudolph C, Jorgensen EM, lethal skin barrier defect in mice lacking comparative gene identification- Schiavo G (2003) Long chain polyunsaturated fatty acids are required for 58 (CGI-58). J Biol Chem 285: 7300 – 7311 efficient neurotransmission in C. elegans. J Cell Sci 116: 4965 – 4975 Renvoise B, Malone B, Falgairolle M, Munasinghe J, Stadler J, Sibilla C, Park Liu Z, Li X, Ge Q, Ding M, Huang X (2014) A lipid droplet-associated GFP SH, Blackstone C (2016) Reep1 null mice reveal a converging role for reporter-based screen identifies new fat storage regulators in C. elegans. J hereditary spastic paraplegia proteins in lipid droplet regulation. Hum Mol Genet Genomics 41: 305 – 313 Genet 25: 5111 – 5125 Liu L, Zhang K, Sandoval H, Yamamoto S, Jaiswal M, Sanz E, Li Z, Hui J, Sangaletti R, D’Amico M, Grant J, Della-Morte D, Bianchi L (2017) Knock-out Graham BH, Quintana A et al (2015) Glial lipid droplets and ROS induced of a mitochondrial sirtuin protects neurons from degeneration in by mitochondrial defects promote neurodegeneration. Cell 160: 177 – 190 Caenorhabditis elegans. PLoS Genet 13: e1006965 Liu L, MacKenzie KR, Putluri N, Maletic-Savatic M, Bellen HJ (2017) The glia- Schweiger M, Lass A, Zimmermann R, Eichmann TO, Zechner R (2009) neuron lactate shuttle and elevated ROS promote lipid synthesis in Neutral lipid storage disease: genetic disorders caused by mutations in neurons and lipid droplet accumulation in glia via APOE/D. Cell Metab 26: adipose triglyceride lipase/PNPLA2 or CGI-58/ABHD5. Am J Physiol 719 – 737 Endocrinol Metab 297: E289 – E296 Martinez-Vicente M, Talloczy Z, Wong E, Tang G, Koga H, Kaushik S, de Vries Shimabukuro MK, Langhi LG, Cordeiro I, Brito JM, Batista CM, Mattson MP, R, Arias E, Harris S, Sulzer D et al (2010) Cargo recognition failure is Mello Coelho V (2016) Lipid-laden cells differentially distributed in the responsible for inefficient autophagy in Huntington’s disease. Nat Neurosci aging brain are functionally active and correspond to distinct phenotypes. 13: 567 – 576 Sci Rep 6: 23795 Marza E, Lesa GM (2006) Polyunsaturated fatty acids and neurotransmission Tian Y, Bi J, Shui G, Liu Z, Xiang Y, Liu Y, Wenk MR, Yang H, Huang X (2011) in Caenorhabditis elegans. Biochem Soc Trans 34: 77 – 80 Tissue-autonomous function of Drosophila seipin in preventing ectopic Massa R, Pozzessere S, Rastelli E, Serra L, Terracciano C, Gibellini M, Bozzali M, lipid droplet formation. PLoS Genet 7: e1001364 Arca M (2016) Neutral lipid-storage disease with myopathy and extended Vasquez V, Krieg M, Lockhead D, Goodman MB (2014) Phospholipids that phenotype with novel PNPLA2 mutation. Muscle Nerve 53: 644 – 648 contain polyunsaturated fatty acids enhance neuronal cell mechanics and Matsuda S, Inoue T, Lee HC, Kono N, Tanaka F, Gengyo-Ando K, Mitani S, Arai touch sensation. Cell Rep 6: 70 – 80 H (2008) Member of the membrane-bound O-acyltransferase (MBOAT) Villarroel A, Schwarz TL (1996) Inhibition of the Kv4 (Shal) family of transient family encodes a lysophospholipid acyltransferase with broad substrate K+ currents by arachidonic acid. J Neurosci 16: 2522 – 2532 specificity. Genes Cells 13: 879 – 888 Watts JL, Browse J (2002) Genetic dissection of polyunsaturated fatty acid Narbonne P, Roy R (2009) Caenorhabditis elegans dauers need LKB1/AMPK to synthesis in Caenorhabditis elegans. Proc Natl Acad Sci USA 99: ration lipid reserves and ensure long-term survival. Nature 457: 210 – 214 5854 – 5859 O’Donnell VB, Rossjohn J, Wakelam MJ (2018) Phospholipid signaling in Watts JL, Ristow M (2017) Lipid and carbohydrate metabolism in innate immune cells. J Clin Invest 128: 2670 – 2679 Caenorhabditis elegans. Genetics 207: 413 – 446 Olzmann JA, Carvalho P (2019) Dynamics and functions of lipid droplets. Nat Weimer RM (2006) Preservation of C. elegans tissue via high-pressure Rev Mol Cell Biol 20: 137 – 155 freezing and freeze-substitution for ultrastructural analysis and O’Rourke EJ, Soukas AA, Carr CE, Ruvkun G (2009) C. elegans major fats are immunocytochemistry. Methods Mol Biol 351: 203 – 221 stored in vesicles distinct from lysosome-related organelles. Cell Metab 10: Xie M, Roy R (2015a) AMP-activated kinase regulates lipid droplet localization 430 – 435 and stability of adipose triglyceride lipase in C. elegans Dauer Larvae. PLoS Outeiro TF, Lindquist S (2003) Yeast cells provide insight into alpha-synuclein ONE 10: e0130480 biology and pathobiology. Science 302: 1772 – 1775 Xie M, Roy R (2015b) The causative gene in Chanarian Dorfman Syndrome Palikaras K, Mari M, Petanidou B, Pasparaki A, Filippidis G, Tavernarakis N regulates lipid droplet homeostasis in C. elegans. PLoS Genet 11: (2017) Ectopic fat deposition contributes to age-associated pathology in e1005284 Caenorhabditis elegans. J Lipid Res 58: 72 – 80 Zhou YT, Grayburn P, Karim A, Shimabukuro M, Higa M, Baetens D, Orci L, Papadopoulos C, Orso G, Mancuso G, Herholz M, Gumeni S, Tadepalle N, Unger RH (2000) Lipotoxic heart disease in obese rats: implications for Jungst C, Tzschichholz A, Schauss A, Honing S et al (2015) Spastin human obesity. Proc Natl Acad Sci USA 97: 1784 – 1789 ª 2020 The Authors. EMBO reports e50214 | 2020 15 of 15