Fig. 7: GLaz has functional and physical interactions with LpR1-short.

a GLaz and LpR1-short show genetic interactions in supporting LNv dendrite development. Left: representative projected confocal images (red) and 3D reconstructions (gray) of the LNv. Right: the quantifications of LNv dendrite volume. Compared to wildtype controls, the trans-heterozygous mutants (GLaz^+/−; LpR1^+/−) display a large reduction in the LNv dendrite volume, significantly lower than controls and both heterozygous LpR1 and GLaz mutants. Data are presented as mean values +/− SEM. Statistical significance was assessed by one-way ANOVA with Tukey’s post hoc test. ANOVA: p < 0.0001, F = 11.53, df = 67; WT/LpR1^+/−: p = 0.0257; WT/GLaz^+/−: p = 0.0486; WT/LpR1^+/−; GLaz^+/−: p < 0.0001; LpR1^+/−/LpR1^+/−;GLaz^+/−: p = 0.0226; GLaz ^+/−/LpR1^+/−;GLaz^+/−: p = 0.0187. n = 19, 18, 16, and 18 for WT, LpR1^−/−, GLaz^+/−, and LpR1^+/−; GLaz^+/−. n represents individual larval brain samples. *p < 0.05, ***p < 0.001. b GLaz enhancer-driven GFP-tagged GLaz (GLaz > GLaz-GFP) is secreted from astrocytes. Top: a schematic diagram illustrating the elements included in the GLaz>GLaz-GFP transgene. Bottom: Representative confocal images show that blocking astrocytes’ secretion by expressing a dominant-negative form of Rab1 (Rab1-DN) leads to the accumulation of GLaz>GLaz-GFP signals (green) in somas of astrocytes (arrows), in contrast to a diffused pattern observed in controls (observed in at least 10 brains). 24B10 staining (red) labels the LON (dashed circles). c GLaz > GLaz-GFP puncta localize near or on the surface of the LNv dendrites (observed in at least 9 brains). Left: Representative projected confocal images demonstrate the broad distribution of GLaz>GLaz-GFP (green) in the third instar larval brain. Right: Zoomed-in images of the LNv dendrites (Pdf > CD2::mCherry, red) show GLaz-GFP puncta in close proximity with LNv dendrites. d Co-IP experiment shows that GLaz binds to LpR1-short, but not LpR1-long. Both GFP-tagged LpR1-short (~140 KD) and LpR1-long (~150 KD) are expressed in neurons and detected by the anti-GFP antibody on the western blot. But only LpR1-short-GFP was found to be associated with GLaz-HA in the co-IP experiments using anti-HA beads. GLaz-HA is detected as two bands around 27KD using the anti-HA antibody. Asterisk (*) indicates a nonspecific band detected by anti-GFP antibody in the larval brain lysate. One representative result from two independent repeats is shown. e A proposed model for the distinct lipid recruitment in the Drosophila brain and peripheral tissues mediated by short vs. long isoforms of LpR1 (created with BioRender.com). LpR1-long mediates neutral lipid uptake and lipid droplet accumulation in peripheral tissues in an endocytosis-independent way, facilitated by its interactions with LTP. In contrast, LpR-short is the major isoform expressed in the fly CNS and likely recruits lipids through an endocytosis-dependent pathway, facilitated by its interactions with the glia-derived apolipoprotein GLaz.