|Home | About | Journals | Submit | Contact Us | Français|
Autophagy is a cellular catabolic process in which various cytosolic components are degraded. For example, autophagy can mediate lipolysis of neutral lipid droplets. In contrast, we here report that autophagy is required to facilitate normal levels of neutral lipids in C. elegans. Specifically, by using multiple methods to detect lipid droplets including CARS microscopy, we observed that mutants in the gene bec-1 (VPS30/ATG6/BECN1), a key regulator of autophagy, failed to store substantial neutral lipids in their intestines during development. Moreover, loss of bec-1 resulted in a decline in lipid levels in daf-2 [insulin/IGF-1 receptor (IIR) ortholog] mutants and in germline-less glp-1/Notch animals, both previously recognized to accumulate neutral lipids and have increased autophagy levels. Similarly, inhibition of additional autophagy genes, including unc-51/ULK1/ATG1 and lgg-1/ATG8/MAP1LC3A/LC3 during development, led to a reduction in lipid content. Importantly, the decrease in fat accumulation observed in animals with reduced autophagy did not appear to be due to a change in food uptake or defecation. Taken together, these observations suggest a broader role for autophagy in lipid remodeling in C. elegans.
Energy in the form of triglycerides (TGs) is required for all metazoans to maintain cellular homeostasis.1 Following nutrient starvation, a primary cellular response is the induction of macroautophagy (referred to as autophagy), a process that degrades cellular components and recycles amino acids and other molecules critical for cellular survival. At the same time, nutrient starvation results in the mobilization of cellular lipid stores to supply free lipids for energy,2 thus pointing to regulatory and functional similarities between autophagy and lipid metabolism.
Recent studies have demonstrated links between autophagy and lipid metabolism in vertebrates.3,4 Originally, degradation of lipid droplets was thought to take place in the cytosol by resident lipases. However, selective macroautophagy, referred to as lipophagy, has recently been described to be required for the delivery of lipid droplets for lysosomal degradation.3 Specifically, inhibition of autophagy in cultured mouse hepatocytes, by knockdown of Atg5 or inhibition of autophagy by 3-methyladenine, results in a significant increase in TG storage in lipid droplets. Similar results have been reported using an in vivo mouse model, as autophagy mutant animals have fatty livers.3 In contrast to the role of autophagy in hepatocytes, knockdown of Atg5 or Atg7 in 3T3 preadipocytes, results in decreased levels of adipocyte differentiation factors, failure of white adipocytes to differentiate, and consequently an inhibition of TG accumulation in these undifferentiated cells.4 A decrease in adipose tissue mass is similarly observed in an in vivo mouse model containing an adipose-specific deletion of Atg7.5 Therefore, autophagy can facilitate lipid breakdown in certain cell types such as the liver but may also enable lipid storage by affecting cell differentiation in adipose tissues. While autophagy has been proposed to mediate lipolysis via a mechanism involving lipases localized to the autophagosome6,7 or the lysosome,8 the link between autophagy and lipid storage has so far only been described to occur with cell differentiation.4
The genetics of lipid storage have recently become an area of intense study in the nematode Caenorhabditis elegans.9-11 Recent studies indicate that modulating lipid metabolism is crucial for long-term survival in C. elegans.12,13 C. elegans do not have a specialized tissue to store fat, however the intestine and the hypodermis (epidermis) are both tissues where many aspects of the regulation of lipid storage appear to be conserved.10 Young larvae respond to starvation by entering an alternative developmental program resulting in a dauer animal, until favorable conditions are resumed.14 Entry into the dauer stage is associated with reduced insulin/IGF-1-like signaling, an increase in lipid storage, and metabolic changes that enable long-term survival in harsh environmental conditions. C. elegans larvae with compromised insulin/IGF-1-like signaling (e.g., daf-2/IIR mutants) constitutively enter dauer stage, whereas mutant adults are long-lived.14 We have previously shown that many of the metabolic and physiological changes associated with the constitutive dauer phenotype of daf-2 mutants do not occur in animals that lack autophagic activity.15 In particular, dauers, in which autophagic flux has been compromised, show reduced levels of stored neutral lipids raising the possibility that autophagy can directly facilitate lipid storage, at least in this specialized larva.15 We, and others, also have shown that autophagy is induced and required for mediating the increased life span of adults with compromised insulin/IGF-1-like signaling (IIS), suggesting that autophagy is part of the anti-aging effects observed in compromised IIS mutants.15-18
Another signaling pathway recently linked to lipid metabolism is that of germline-less glp-1/Notch mutants.19 Similar to daf-2 mutants, glp-1 mutant animals also display an increase in lipid contents20 and increased autophagy levels.6 While lipid breakdown and autophagy have been linked in germline-less glp-1 animals via the lipase LIPL-4,6 it is unknown whether autophagy plays a direct role in fat storage in C. elegans raised under favorable food conditions.
Here we showed that bec-1/VPS30/ATG6/BECN1 and other autophagy genes were required for lipid storage in normally developing C. elegans. Specifically, we found that mutants of the autophagy gene bec-1 failed to accumulate substantial neutral lipids during development, producing extremely lean adult animals. Similarly, RNA inactivation of bec-1, vps-34, lgg-1 and unc-51 also significantly reduced fat content in adult animals. Moreover, both daf-2 receptor mutants and germline-less glp-1 animals failed to increase lipid levels during development when autophagy was impaired, consistent with a critical role for autophagy in lipid storage. Notably, mutants with compromised autophagy had normal pharyngeal pumping, food uptake and defecation during development. In addition, mutants with impaired retrograde transport contained normal levels of lipids, implying that the cellular process of autophagy plays a specific role in lipid remodeling. Taken together, these data suggest that the autophagy process plays a role in lipid storage in C. elegans under favorable food conditions. As the C. elegans intestine is a terminally differentiated tissue, we propose that the link between autophagy and fat storage might be of a more direct nature and may not inherently involve cell differentiation.
To determine whether lipid storage is modulated by autophagy in C. elegans, we evaluated the effect of impairing autophagy on lipid content using a mutant of the autophagy gene bec-1. bec-1 is the C. elegans ortholog of yeast VPS30/ATG6 and mammalian BECN1/Beclin 1. We assayed homozygous bec-1(ok691) mutants, which have impaired autophagy.21,22 The bec-1 allele ok691 is a molecular null mutation, yet bec-1 homozygous mutants are rescued maternally and may reach adulthood, but are sterile and die as young adults.22 Considering a recent review on the limitations of various lipid analysis methods in C. elegans,23 we used Oil-Red-O staining for our initial analysis.24 Oil-Red-O staining is a fixative-based dye that stains lipids and correlates with TG mass measurements obtained using quantitative biochemical methods.20 When using this method, we found a profound decrease in the amount of Oil-Red-O stained neutral lipids in day 1 adult bec-1(ok691) homozygous mutants compared with wild-type animals (Fig. 1A).
To complement this analysis, we also used the more recent coherent anti-Stokes Raman (CARS) microscopy method to assess lipid levels in C. elegans.25 CARS microscopy allows for direct quantitative analysis of lipid storage without the use of labels. Using CARS microscopy, loss of bec-1 resulted in a very prominent decrease in lipid droplet number and overall lipid content (Fig. 1B and C). Looking at the distribution of lipid droplets in the intestine, a shift toward smaller lipid droplets was also apparent in the absence of bec-1 (Fig. S1). This shift in lipid droplet size was also evident when we stained the bec-1(ok691) homozygous animals with Oil-Red-O (Fig. 1A). The decrease in lipid levels observed in bec-1(ok691) mutants was not simply a result of reduced feeding, as the rate of pharyngeal pumping was only slightly decreased in bec-1 mutant L4 larvae compared with wild-type animals and defecation rates or BODIPY uptake remained unaffected (Fig. S2). Similarly, bec-1(ok691) mutant L4 larvae did not appear to be leaner due to increased energy expenditure, as they did not show noticeably increased activity levels compared with wild-type animals (data not shown). Collectively, these results imply that bec-1/BECN1 may play a direct role in facilitating lipid accumulation in developing C. elegans. While our assays showed a decrease in lipid levels after inactivation of bec-1/BECN1, they did not allow us to determine whether the decrease in lipid content results from an increase in lipid breakdown, a lack of lipid biosynthesis, or defective recycling and storage of lipids.
To further investigate a role for autophagy in lipid accumulation, we next examined two mutants known to display increased lipid storage capacity: the insulin/IGF-1-like receptor daf-210,26 and germline-less glp-1/Notch mutants.20 These mutants also show increased autophagy levels raising the possibility that the autophagy and lipid phenotypes may be interconnected in these animals.6,15
Consistent with previous reports,we found that day 1 adult animals carrying either daf-2(e1370) or glp-1(e2141) loss-of-function mutations, displayed an increase in lipid content as measured by Oil-Red-O staining, compared with wild-type animals (Fig. 1A). Further analysis of daf-2(e1370) or glp-1(e2141) mutants by CARS microscopy confirmed that lipid content in the intestinal cells was significantly increased in either single mutant (Fig. 1B and C). Consistent with these observations, intestinal lipid droplets in either glp-1 or daf-2 mutant worms tended to be larger than those found in wild-type animals (Fig. S1). Overall, the distribution of lipid droplets was predominantly intestinal, but a layer of small lipid droplets was detectable in the hypodermis of these animals (data not shown).
We next introduced the bec-1(ok691) mutation into daf-2 and glp-1 loss-of-function mutants to investigate the role of bec-1/BECN1 in lipid accumulation in these mutant backgrounds. We found that daf-2(e1370); bec-1(ok691) and glp-1(e2141); bec-1(ok691) double mutants no longer showed an increase in Oil-Red-O staining compared with single daf-2(e1370) and glp-1(e2141) mutants, respectively (Fig. 1A). Similarly, we found that day 1 adult daf-2(e1370); bec-1(ok691) and glp-1(e2141); bec-1(ok691) double mutants displayed a reduction in the number of lipid droplets, the size of the droplets, as well as a decrease in the lipid contents when analyzed by CARS microscopy (Fig. 1B and C).
Gene inactivation by feeding bacteria expressing dsRNA against bec-1/BECN1 also decreased normal fat contents in wild-type animals and drastically reduced the accumulation of lipids normally observed in daf-2(e1370) and glp-1(e2141) mutants (Figs. 2 and and3).3). Similar results were observed with a different glp-1 loss-of-function allele, bn18 (Fig. S6A), as well as an additional daf-2 allele, m41 (Fig. S7), which also induces accumulation of lipids.26 We found that food uptake or defecation in glp-1 or daf-2 mutants remained unchanged after RNAi against bec-1/BECN1 (Fig. S4–S6). Taken together, our analysis of daf-2 and glp-1 animals confirmed that bec-1/BECN1 plays a critical role in fat homeostasis, and supported a link between lipid metabolism and autophagy in these mutants, as observed in wild-type animals.
In addition to its role in autophagy, BEC-1/BECN1 has been shown to function in a complex with the class III phosphatidylinositol-3-kinase (PtdIns3K) VPS-34, the functional ortholog of yeast Vps34, in C. elegans retrograde transport from endosome to Golgi.22 To determine whether the autophagy process could be a determining factor in the reduced lipid levels observed in bec-1/BECN1 mutants, we inhibited other genes with effects in different steps of the autophagy process. To do this, we fed wild-type N2 animals, from hatching and throughout development, bacteria expressing dsRNA against bec-1/BECN1, vps-34/VPS34/PIK3C3, unc-51/ATG1/ULK1, and lgg-1/ATG8/LC3. Analysis in yeast and in mammalian cells has shown that the Atg1/ULK1 protein kinase acts in the induction of autophagy, Vps34/PIK3C3 acts together with Vps30/Atg6/BECN1 at the nucleation step, and Atg8/LC3 is part of the protein conjugation system for vesicle completion. Knockdown of autophagy genes was previously shown to either cause abnormal daf-2 dauers or to limit the formation of GFP::LGG-1 puncta, a commonly used assay to assess autophagy in C. elegans.27,28 Indeed, elevated GFP::LGG-1 puncta was associated with the lack of daf-2/IIR receptor activity and the lack of a germline.6 Like bec-1 deletion mutants, RNAi against vps-34/VPS34/PIK3C3, unc-51/ATG1/ULK1 or lgg-1/ATG8/LC3 resulted in a decrease in lipid contents in the intestine of 1 d-old adult wild-type animals (Fig. 2A and C; Fig. 3A). More importantly, wild-type N2 animals treated with RNAi against bec-1/BECN1, vps-34/VPS34/PIK3C3, unc-51/ATG1/ULK1, and lgg-1/ATG8/LC3, showed no decrease in food uptake or defecation, and yet these animals also displayed a decrease in lipid levels (Figs. 2 and and33; Fig. S3). Similarly, a decrease in lipid levels was observed in day 1 adult daf-2(e1370) mutants (Fig. 2A and D) following RNAi-mediated knockdown of bec-1/BECN1 and vps-34/VPS34/PIK3C3 and in glp-1(e2141) animals, following RNAi against bec-1/BECN1, vps-34/VPS34/PIK3C3, unc-51/ATG1/ULK1 and lgg-1/ATG8/LC3 (Fig. 3A and D). A decrease in lipid content was also observed in glp-1(bn18) mutants after RNAi against bec-1/BECN1, vps-34/VPS34/PIK3C3, unc-51/ATG1/ULK1, and lgg-1/ATG8/LC3 (Fig. S6). Importantly, no significant change was observed in food uptake or defecation for daf-2 or glp-1 mutant animals after RNAi treatment against bec-1/BECN1, vps-34/VPS34/PIK3C3, unc-51/ATG1/ULK1 and lgg-1/ATG8/LC3 (Figs. S3–S6). Quantification of Oil-Red-O staining confirmed a significant reduction in overall lipid content as a result of RNAi against several autophagy genes in daf-2(e1370) or glp-1(e2141) animals (Figs. 2D and and3D3D). Taken together, these observations indicated that autophagy is required for adequate lipid content in C. elegans, suggesting a more dynamic role for autophagy in lipid homeostasis.
As we have previously shown that bec-1/BECN1 and vps-34/VPS34/PIK3C3 act in the retrograde transport from endosome to Golgi,22 we decided to further investigate the role of retromer genes in lipid accumulation. Since we did not observe a decrease when we stained wild-type N2, daf-2(e1370) or glp-1(e2141) animals that had been treated with RNAi against rme-8/RME-8 or vps-35/VPS35 (Figs. 2B and and3B),3B), we also analyzed rme-8(b1023) mutants by CARS microscopy (Fig. S8). Interestingly, although rme-8 and vps-35 mutants strongly affect retromer function,29-31 these strains, unlike autophagy mutant unc-51, did not show a reduction in lipid content when compared with control animals (Fig. S8), and none of these strains showed significant change in food uptake or defecation (Fig. S9). Thus, these data suggest that lipid accumulation does not depend on retrograde transport. Instead, our findings point to a novel requirement for the autophagy process regulating lipid levels, emphasizing the possibility that lipid homeostasis is tightly linked to the autophagy/lysosomal pathway.
Autophagy has recently been linked to lipid metabolism, for instance, by mediating cell differentiation of adipocytes to ensure lipid storage.4,5 Here, our studies suggest that the regulation of lipid homeostasis in C. elegans requires functional autophagy. Specifically, we observed that the loss of bec-1/BECN1 was sufficient to significantly impair lipid accummulation in the animal’s terminally differentiated intestinal cells. Likewise, knockdown of other autophagy genes working in different steps of the autophagy process,27 such as vps-34/VPS34/PIK3C3, unc-51/ATG1/ULK1 and lgg-1/ATG8/LC3, led to a decrease in lipid content. Importantly, the reduced lipid content was not due to reduced pharyngeal pumping, nutrient absorption or defecation. It remained unclear, however, whether the decrease in lipid content resulted from an increase in lipid breakdown, a lack in lipid biosynthesis, defective recycling or storage of lipids. While future experiments are needed to differentiate between these possibilities, one hint that the decrease in lipid accumulation may not be due to an increase in the breakdown of lipids, is that glp-1 animals that have been fed RNAi against autophagy genes display reduced, not increased, lipase activity.6 Another possibility is that there is an increase in oocyte lipid uptake, however, knockdown of bec-1/BECN1 was previously shown to reduce the oocyte uptake of the yolk protein VIT-2::GFP.21 Thus the decrease in lipid levels, in the intestine, is not likely due to an increased rate of uptake by the growing oocytes. As some autophagy genes, including bec-1/BECN1, play additional roles in endocytosis,22 and this process has been linked to lipid storage,33 it is possible that impairing both autophagy and endocytosis contributes to the decrease in lipid levels. However, we found that inactivation of several retromer genes did not decrease the accumulation of lipids. Since inhibition of the autophagy process at different steps gives similar results, and inhibition of retromer genes did not have a significant effect, we suggest that autophagy is critical for lipid accumulation in C. elegans, although we have not formally excluded that bec-1/BECN1, and the other tested genes, serve nonautophagic functions in lipid homeostasis.
These novel observations add insights into earlier research on the link between autophagy and lipid homeostasis. Previous in vitro studies, using a preadipocyte mouse cell line, have shown that inhibition of autophagy results in a decrease in the accumulation of triglycerides (TGs) that occurs during the differentiation process into adipocytes. In Atg7 knockout mice, the effects of the loss of autophagy on adipocyte differentiation are more complex, in addition to the decrease in white adipose mass, the tissue displayed features of brown adipocytes.4,5 These two types of adipocytes have different functions; white adipocytes store large amount of TG for breakdown, whereas brown adipocytes have a higher number of mitochondria and an increase in the rate of fatty acid β-oxidation. While showing a role for autophagy in lipid accumulation, these mammalian studies did not address whether the decrease in lipid storage observed in autophagy-deficient animals could be secondary to a failure to differentiate adipocytes, and instead represent a direct effect on lipid storage. Since C. elegans do not formally have a white or brown adipocyte equivalent, we could not address whether autophagy has a similar effect in the intestine of these invertebrate animals. However, we observed that inhibition of autophagy, in a fully differentiated tissue, impaired fat levels while not affecting the food intake or energy expenditure of the animal. Future experiments should address how autophagy mediates these effects on lipid metabolism, for instance by regulating mitochondria function and the rate of fatty acid β-oxidation, as observed in mammalian systems. Similarly, it will be interesting to address whether autophagy regulates lipid metabolism in differentiated adult adipocytes.
While our experiments provided evidence that autophagy was required for lipid accumulation, autophagy is typically viewed as a catabolic process. Thus, an apparent question, brought up by these results, is whether active degradation of cellular components by autophagy is necessary for lipid storage, biosynthesis or remodeling via the same process. Our data are consistent with a possible, albeit controversial, mechanism in which lipid storage may be driven by the ability of cells to rapidly recycle material, such as certain fatty acids via autophagy. One alternate possibility is that by-products of lipid breakdown resulting from autophagy may be essential for storing lipids by providing raw materials needed (e.g., building blocks for membrane formation) or by indirectly stimulating lipogenesis via signaling. Another possibility is that proper mitochondrial function in lipid metabolism relies on a continual influx of fatty acids from autophagy-related lipolysis, or autophagy flux could be important for the function of a regulator of lipid storage or biosynthesis. Future experiments are needed to address how autophagy may modulate lipid accumulation in C. elegans.
The apparent benefit for cells to dynamically recycle fatty acids between processes of lipogenesis and lipolysis is still to be fully investigated. However, we noted that several healthy, long-lived longevity mutants, such as daf-2 and glp-1 mutants, display autophagy-dependent increases in both intestinal lipolysis activity6 and lipid contents.10,20,26 This is also the case in dauer larvae, in which fine-tuning of lipid partitioning suggests that particular types of lipid breakdown or levels influence the capacity of worms to survive under unfavorable conditions.13 To provide a benefit to the organism, however, it appears that concomitant lipogenesis may be required to prevent lipid store depletion, as seen in several daf-2 strains.26 We previously published that the lack of bec-1/BECN1 activity decreased the accumulation of lipids that occurs in daf-2 dauers.15 Whereas it is unclear whether the effects on lipid homeostasis in daf-2 mutant dauers and daf-2 mutant long-lived adults are mediated by identical mechanisms, our studies show that autophagy plays a role in both. In conclusion, since lipid breakdown and lipogenesis may coexist in the same tissue in C. elegans, we propose that dynamic lipid remodeling via autophagy plays an important role not only during animal development but also to ensure survival during adulthood by guaranteeing optimal lipid partitioning and energy metabolism.
The following strains were used in this study: N2 (wild-type/WT), CF1903: glp-1(e2141), GC888: glp-1(bn18), and the following strains were obtained from the Caenorhabditis Genetics Center: CB1370: daf-2(e1370), DR1564: daf-2(m41), and VC517: bec-1(ok691), CB369: unc-51(e369), KN555: vps-35(hu68) and DH1206: rme-8(b1023).
The following strains were made for this study: QU20: glp-1(e2141); bec-1(ok691)/nT1, QU36: daf-2(e1370); bec-1(ok691)/nT1.
Oil-Red-O (Sigma Aldrich, O0625) used to detect lipids after fixation. C12 BODIPY (Invitrogen, D3823) used to determine food uptake in vivo. Sodium Azide (Fisher Scientific, S227) used to paralyze animals for imaging
Animals were raised on NGM agar plates containing standard E. coli food source OP50 at permissive temperature 15°C and changed to the restrictive temperature of 25°C, as L4 larvae. Animals were assayed as 1 d-old adults. The same population was assayed for pharyngeal pumping, defecation rates, BODIPY uptake and stained with Oil-Red-O.
Pharyngeal pumping rates were measured in 1 d-old adults under a dissecting scope in 20 sec intervals and pumping rate (pumps/minute) were determined in at least 10 animals per trial, for each of two trials analyzed. Prior to measurements being taken, animals were incubated at room temperature (22°C) for 1 h. ANOVA comparisons (performed with the software GraphPad Prism 5.0) were made using control animals as stated.
Defecation rates were determined by measuring the average of 2 consecutive cycles.34 Prior to measurements taken, animals were incubated at room temperature (22°C) for 1 h. At least 10 animals were analyzed for each trial and the values represent the combined average of 2 trials. ANOVA comparisons (performed with the software GraphPad Prism 5.0) were made using control animals as stated.
One day-old adults were placed on 2 cm NGM OP50 E. coli seed with 100 ul of 5 uM C1BODIPY-C12 (Invitrogen, D3823) diluted in phosphate buffer saline (PBS).33 Animals were fed BODIPY for 1 h at room temperature (22°C), collected and washed with 1× phosphate buffered saline (PBS), mounted and immobilized on 2% agarose pads in 5 ul M9 buffer containing 25 mM sodium azide. Using a magnification of 160×, fluorescent images were taken using AxioCam Zeiss Imager A1. Fluoresence was quantified using Image J software. On average, 30 worms per trial were quantified (the values represent the combined average of 3 trials). ANOVA comparisons (performed with the software GraphPad Prism 5.0) were made using control animals as stated.
Lipid levels were analyzed using Oil-Red-O staining of fixed animals as previously described.24 Quantification of Oil-Red-O staining was performed as previously described.25 On average 50 animals were quantified for each trial (values represent the combined data from 3 trials), using Image J software, representing a total population of 100 animals.
Coherent anti-Stokes Raman Scattering (CARS) microscopy was performed using the following parameters: the signal (924.2 nm) and idler (1255 nm) outputs of an optical parametric oscillator (OPO, Levante Emerald) were used as the pump and Stokes beams, respectively, to produce a frequency difference of 2851 cm−1. The OPO was synchronously pumped by the second harmonic output (532 nm) of a mode-locked Nd:YVO4 laser (HighQ Laser). The pump and Stokes beams were inherently synchronized and collinearly overlapped at the exit port of the OPO. The laser beams were passed through a laser scanner (C1plus, Nikon) and focused with a 60× IR objective into the sample. The combined laser power was attenuated with neutral density filters to 66 mW at sample. Epi-reflected signal was directed into a multichannel detector, spectrally separated with dichroic mirrors, selected with bandpass filters (Semrock), and detected with red-sensitive photomultiplier tubes (Hamamatsu, R10699). Bandpass filters for autofluorescence and CARS signal were 510/42 nm and 736/128 nm, respectively.
Images were acquired at 10 sec per frame and presented as 3-D stacks of approximately 30 frames taken at 1-micron increment along the vertical axis. Image analysis was performed post-acquisition using NIH ImageJ software. Quantification was performed on 4 stacks (3 µm apart) per animal encompassing the intestine cells. ANOVA statistical analyses of lipid droplet number, size and lipid content were performed using the software GraphPad Prism 5.0. Wild-type and daf-2(e1370) mutants were raised at the permissive temperature, while glp-1(e2141) animals were raised at the restrictive temperature of 25°C.
To circumvent defects that occur during embryonic development, L1 larvae were collected as they hatched and transferred to plates seeded with HT115 E. coli that express dsRNA for the corresponding target gene. Bacteria expressing specific dsRNA were obtained from the Ahringer35 and the Vidal libraries.36 Only healthy individuals were assayed following RNAi treatment. L4440 RNAi bacteria were used as control in all experiments.
We thank all Hansen and Meléndez lab members for helpful discussions and comments on the manuscript. We thank Lana Tolen and Zahava Rubel in the Meléndez lab for their initial studies on the role of bec-1 in lipid storage. We thank Drs. Cathy Savage Dunn and Hannes Bülow for comments on the manuscript, and Dr. Nathalia Holtzman for advice on image processing and quantification.
This work was also supported by an NIH/UCSD grant to LL (grant number P50 AG005131-29), a grant from the Nevada INBRE Program of the National Center for Research Resources to TTL, an R01 grant from the National Institute of Aging to MH (grant number R01 AG038664), a National Science Foundation Research Initiation Award (grant number 0818802) and a National institute of Aging supplement award to A.M. M.H. and A.M. are both Ellison Medical Foundation New Scholars in Aging.
No potential conflicts of interest were disclosed.
Previously published online: www.landesbioscience.com/journals/autophagy/article/22930