Here I’ll sum up a recent paper in Developmental Cell by E. Thomas Danielson and Morten E. Moeller. The paper, entitled “A Drosophila Genome-Wide Screen Identifies Regulators of Steroid Hormone Production and Developmental Timing,” utilized the Vienna Drosophila RNAi Center’s RNAi collection. This is a set of thousands of fly strains, where each enables the expression of a unique dsRNA. In principal, any gene in the genome can be knocked down with this collection. The authors knocked down 12,504 genes (90% of protein coding genes) in the Drosophila endocrine gland, the prothoracic gland (PG). They surmised that any gene impacting developmental progression would likely be involved in ensuring that the major developmental hormone, ecdysone, is synthesized and delivered properly to the developing animal. They found 1,906 genes affecting developmental progression. These fell into six classes:
Of the ~1,800 genes that hadn’t previously been implicated in steroidogenesis, they chose to focus on “stuck in traffic (sit).” It’s a fatty acid elongase. These enzymes create the long-chain fatty acid substrate needed to synthesize ceramides, which are themselves substrates for sphingolipid synthesis. Sphingolipids are essential components of cell membranes.
When sit was knocked down in the PG, the larvae arrested at the 3rd instar stage and overgrew, consistent with sit being implicated in ecdysone synthesis or release. Using labeling-free coherent anti-Stokes Raman scattering (CARS) microscopy to visualize lipids, they found a significant increase in lipid droplets in the cells of the PG when sit was knocked down (see below). Those lipid droplets are filled with cholesterol. However, when knocked down in the fat body, no lipid droplets were observed. That result suggests a specific endocrine cell function for sit in lipid metabolism, and that lipid droplet trapped cholesterol, a precursor to ecdysone, precludes an ability to synthesize sufficient ecdysone. Knocking down a ceramide synthase also caused the 3rd instar arrest and lipid droplet accumulation. Thus ceramides, via sit, are essential for lipid homeostasis in the endocrine gland. When ceramides are low, cholesterol is packaged into lipid droplets and ecdysone pools are limiting.
Knockdown of npc1a, also led to an elevation in lipid droplets (see below). Npc1a is the ortholog of npc1, which is defective in most cases of the disease Niemann Pick C (NPC). Remarkably, over expression of sit cleared the lipid droplets when npc1a was knocked down. It may not be a coincidence that sit, which is essential to provide ceramides and sphingolipids, was implicated in a genetic interaction with npc1a. Glycosphingolipids are elevated in NPC cells, and they have long been implicated in the disease. The drug Miglustat inhibits glucosylceramide synthase, which is the first committed step to glycosphingolipid synthesis. Miglustat is approved for treating NPC in Europe.
Interested in signaling pathways that coordinate organismal growth, they blocked Tor signaling in the PG by knocking down Tsc1/Tsc2. The basal level of lipid droplets were cleared when Tor was inhibited:
Since Tor inhibition leads to an activation of autophagy, they suspected that autophagy induction might liberate cholesterol from lipid droplets. That seems to be the case, because inhibition of autophagy increased lipid droplets (below, A), and induction of autophagy via over expression of Atg1/13 rescued the developmental delay of larvae where npc1A had been knocked down (below, B).
Since npc1A deficiency leads to a failure to traffic cholesterol, resulting in reduced ecdysone levels, it’s likely that autophagy mobilized cholesterol pools in those cells and enabled ecdysone production. These data suggest that activation of autophagy could be a therapeutic avenue for NPC. That hypothesis had been made previously. Confirmation of that idea in an animal model by Danielson and Moeller makes a model whereby autophagy can mobilize cholesterol pools in NPC more realistic. In fact, 2-hydroxypropyl-β-cyclodextrin (HP-β-CD), which is now in a phase IIB/III trial for NPC, has been shown to induce autophagy in tissue culture models. Might an aspect of HP-β-CD-mediated efficacy be due to its ability to regulate autophagy? The conventional view is that HP-β-CD works in NPC by a well-documented ability to mobilize trapped lysosomal cholesterol. This enables cholesterol delivery to cholesterol-deficient cellular regions, and restores cellular cholesterol to an equilibrium. An intriguing possibility is that HP-β-CD-mediated efficacy is driven by both autophagic and cholesterol homeostatic mechanisms.
