Fig 1.
dFatp is required for lipid storage in Drosophila retina.
(A) LD labeled with BODIPY500/510-C12 (green) were revealed using confocal microscopy in horizontal sections of whole mount retinas from one-day-old flies expressing UAS-LacZ (control), UAS-dFatp and/or UAS-Bmm-lipase under the control of a pan-retinal (GMR-Gal4) (a–d) or a dRPC-specific (54C-Gal4) (e–h) driver. Photoreceptors were counterstained with phalloidin-rhodamine (red). (B) BODIPY500/510-C12 (green) uptake in horizontal sections of whole eye wild-type (FRT82B-wild-type) and mutant (FRT82B-Aats-metFB) clone generated with the GMR-hid/FLP-FRT technique [49] from one-day-old flies in the absence (a, b) or presence of dRPC-specific (driven by 54C-Gal4) UAS-dFatp-RNAi[GD16442] (c, d). Photoreceptors were counterstained with phalloidin-rhodamine (red). Scale bar, 25 μm. (C) Quantification of BODIPY500/510-C12 uptake into lipid droplets from the images shown in (A) and (B). Data are presented as the fold change in fluorescence intensity (dots/μm2) compared with the FRT82B-wild-type flies. The boxes represent the median and lower and upper quartiles, and the whiskers represent the 1.5 interquartile range. N = 6–41 retinas per condition. Blue and red statistical stars indicate significant differences between wild type and retina overexpressing dFatp or Bmm with GMR-Gal4 or 54C-Gal4, respectively. Log adjusted values: *p<0.05, **p<0.01, ***p<0.001 by Tukey’s HSD paired sample comparison test.
Fig 2.
Lipid droplets are mainly localized in dRPCs and are increased by dFatp overexpression in Drosophila retinas.
(A) TEM of ommatidia from one-day-old flies expressing UAS-LacZ (control) or UAS-dFatp under control of the pan-retinal (GMR-Gal4) or dRPC-specific (54C-Gal4) drivers. One ommatidium in each panel shows seven photoreceptors (false colored green) with central rhabdomeres surrounded by dRPCs (false colored pink). Lipid droplets are mainly located in dRPCs (black arrowheads) but can also be observed in photoreceptors (black arrows). Lipid droplet size was increased by dFatp overexpression (b, d, d’) compared to control conditions (a, c, c’). Scale bars, 2 μm (a–d), 1 μm (c’, d’). m, mitochondria; R rhabdomeres. (B) Quantification of lipid droplet density (number per surface area) in dRPCs and photoreceptors in flies with dRPC-specific expression (54C-Gal4) of UAS-LacZ (control), UAS-dFatp, or UAS-Bmm-lipase. Means ± SD of n = 4 eyes. (C) Quantification of lipid droplet size (area in m2) in dRPCs of flies with dRPC-specific (54C-Gal4) expression of UAS-LacZ (control) or UAS-dFatp. (D) Mitochondrial size (area in m2) in photoreceptors of flies with dRPC-specific expression of UAS-LacZ (control) or UAS-dFatp. The boxes represent the median and lower and upper quartiles, and the whiskers represent the 1.5 interquartile range. N = 4 to 5 flies, from which we analyzed >150 fields of view. Log adjusted values: *p<0.05, ***p<0.001 by Tukey’s HSD paired sample comparison tests.
Fig 3.
Transfer of vesicles from dRPCs to photoreceptors.
Low (left) and high (right) magnification of retina tangential views visualized by TEM in one-day-old flies expressing (A and A’) UAS-LacZ or (B and B’) UAS-dFatp under the control of 54C-Gal4 driver. A dRPC (false colored pink) is visible sandwiched between two photoreceptors (false colored green). Arrowheads indicate lipid droplet-like vesicles entering an endocytotic invagination in the photoreceptor plasma membrane. Arrows indicate vesicles within photoreceptors surrounded by a double membrane. (A’) shows vesicles in close association with mitochondria (m) in a photoreceptor. (B’) shows various stages of the vesicle transfer from dRPC to photoreceptor. A vesicle within the dRPC (i) is taken up by the photoreceptor (ii), and surrounded by a double membrane once internalized within the photoreceptor (iii). Note the position of mitochondria (m) close to the vesicle. Scale bars, 1 μm (A and A’) and 0.5 μm (B and B’).
Fig 4.
hFATP1 overexpression in mouse RPCs increases neutral lipid accumulation and mitochondrial respiration.
(A, B) Nile Red staining of neutral lipids (red) in mRPCs of a flat-mounted eye cup from wild-type (C57BL/6J) mice (A) and hFATP1 transgenic mice (B). Nuclei are stained with DAPI (blue). Scale bars, 10 μm. (C) Quantification of Nile Red staining in wild-type (WT) and hFATP1 transgenic (hFP1) mRPCs (as shown in A and B). Mean ± SD of n = 19 and 20 animals, respectively. (D) Quantification of sterol esters (SE) and triacylglycerides (TAG) in mRPCs from WT and hFATP1 transgenic mice. Mean ± SD of n = 5 and 7 animals, respectively. (E–E”) mRPCs in whole-mount retinas from WT mice double-stained with Nile Red (E) and anti-perilipin (green, E’). The merged image (E”) shows extensive overlap between Nile Red and perilipin stainings. Scale bars, 5 μm. (F, G) TEM images of mRPCs from wild-type (C57BL/6J) mice (F) and hFATP1 transgenic mice (G). am: apical membrane, Bruch mb: Bruch membrane, me: melanosome, mi: mitochondria, p: phagosome. (H–I’) mRPCs in whole-mount retinas of WT (H) and hFATP1 transgenic (I, I’) mice double-stained with Nile Red and anti-ATP synthase antibody (green, localized to mitochondria). (I’) Magnification of the box in (I) shows the juxtaposition of mitochondria and neutral lipid stores. Scale bars, 5 μm (H, I’), 10 μm (I). (J) Quantification of mitochondrial respiratory function (O2 consumption) in RPCs isolated from WT and hFATP1 transgenic mice. EIImo, basal respiratory rate; EIIImo, β-oxidation, EIIIMPGSO, global respiratory chain function of mitochondrial complexes I and II. The boxes represent the median and lower and upper quartiles, and the whiskers represent the 1.5 interquartile range. N = 21 and 10 WT and transgenic animals, respectively. (K–N) Activities of the TCA cycle enzymes, citrate synthase (K), isocitrate dehydrogenase (DH) (L), oxoglutarate dehydrogenase (OGDH) (M), and fumarase (N) in RPCs isolated from WT and hFATP1 transgenic mice. Mean ± SD of n > 10 mice for each condition. *p<0.05, **p<0.01, ***p<0.001 for WT vs transgenic groups by two-sample t-test.
Fig 5.
Cell non-autonomous effect of mRPC-specific hFATP1 overexpression on the lipid content and respiratory rate in the neural retina.
(A) Quantification of sterol esters (SE) and triacylglycerides (TAG) in the neural retina of 3-month-old WT and hFATP1 transgenic mice. (B) Quantification of total phospholipid content in the neural retina of WT and hFATP1 transgenic mice. Mean ± SD of n = 5 and 7 retinas, respectively. (C) Quantification of mitochondrial respiratory function in the neural retina of WT and hFATP1 transgenic mice. EIImo, basal respiratory rate; EIIImbgso, β-oxidation; EIIImpgso, global respiratory chain function of complexes I and II. The boxes represent the median and lower and upper quartiles, and the whiskers represent the 1.5 interquartile range. N = 12 and 14 WT and transgenic retinas, respectively. (D–G) Activities of citrate synthase (D), isocitrate dehydrogenase (DH) (E), oxoglutarate dehydrogenase (OGDH) (F), and fumarase (G) in neural retinal extracts from WT and hFATP1 transgenic mice. Mean ± SD of n > 9 retinas. *p<0.05, **p<0.01, ***p<0.001 for WT vs transgenic groups by two-sample t-test.
Fig 6.
Consequences of FATP-dependent accumulation of LDs on RPCs and photoreceptors in Drosophila and mice.
(A) TEM images of horizontal sections from 3-month-old wild-type (C57BL/6J) and hFATP1 transgenic mice. Abundant vacuoles (v) and thickened Bruch’s membrane (mb, dashed lines) are evident in the aged hFATP1 mice. me: melanosome, mi: mitochondria, n: nucleus. Scale bars, 500 nm. (B–D) Quantification of Bruch’s membrane width (B), mRPC width (C), and vacuole density in mRPCs (D) from the images shown in (A). Mean ± SD of n = 14 and 18 WT and transgenic animals, respectively. (E) Cornea neutralization method of the retinas using confocal fluorescence microscopy (dRPC autofluorescence) of 30-day-old Drosophila expressing (a, c) UAS-LacZ (control) or (b, d) UAS-dFatp throughout the retina (GMR-Gal4) or in dRPCs alone (54C-Gal4). Arrows show loss of dRPCs resulting in disorganization and loss of typical hexagonal ommatidial shape (dashed outline) upon dFatp overexpression. Scale bars, 10 μm. (F) Quantification of ommatidia with missing dRPCs (affected ommatidia) in retinas shown in (E). Mean ± SD of n = 5 and 8 control and transgenic flies, respectively. Black bars, GMR-Gal4; pink bars, 54C-Gal4. (G) Visualization of retinas of 40-day-old Rh1-GFP-expressing flies by the cornea neutralization method with dRPC-specific expression of (a) UAS-LacZ (control), (b) UAS-dFatp, (c) UAS-LacZ RNAi (control-RNAi), or (d) UAS-dFatp-specific RNAi (dFatp-RNAi[GD9406]). (b) dFatp overexpression in dRPCs does not affect photoreceptor survival, as indicated by intact rhabdomeres (dashed outlines). (d) dFatp-RNAi[GD9406] expression induces the loss of rhabdomeres (arrows). Scale bars, 10 μm. (H) Quantification of ommatidia with one or several missing photoreceptors (affected ommatidia), as shown in (G). (I) Quantification of affected ommatidia with missing photoreceptors in 7, 21 and 30 day-old flies expressing UAS-LacZ, UAS-dFatp or UAS-Bmm under the control of 54C-Gal4. (J) Quantification of affected ommatidia with missing photoreceptors in 30 day-old flies expressing UAS-LacZ, or UAS-dFatp under the control of 54C-Gal4 combined with repo-Gal80 (white bar), elav-Gal80 (pink bar) or wild-type (green bar). (K) Quantification of affected ommatidia with missing photoreceptors in 30 day-old flies expressing UAS-LacZ or UAS-dFatp-RNAi under the control of 54C-Gal4 in normal (yellow) or Vitamin A deficient (blue) diet. Mean ± SD of n = 5–7 flies/condition. *p<0.05, **p<0.01, ***p<0.001 by two-sample t-test.
Fig 7.
dRPC-specific knockdown of dFatp suppresses Aats-metFB-induced photoreceptor degeneration in Drosophila retina.
(A) Epifluorescence microscopy of retinal sections on whole eye clones generated with the GMR-hid/FLP-FRT technique [49] from 5-day-old FRT82B-wild-type (a, b, control) or FRT82B-Aats-metFB mutant (c, d) flies expressing UAS-dFatp-RNAi[GD16442] (b, d) or UAS-Lac-RNAi (a, c) under the control of 54C-Gal4. In Aats-metFB mutant (c), nuclei (stained with DAPI, white) are mislocalized between the distal and proximal part of the retina (red arrows), indicative of dying photoreceptors. dRPC-specific expression of UAS-dFatp-RNAi[GD16442] markedly reduces the number of dying photoreceptors in Aats-metFB mutant (d). Scale bars, 100 μm. (B) Quantification of mislocalized nuclei, as shown in A. Mean ± SD of n > 6 retinas. **p<0.01 by two-sample t-test.
Fig 8.
Schematic of the role of FATP in the metabolism of lipid droplets and communication between RPCs and photoreceptors under physiological and pathological conditions.
Under physiological conditions, FATP is involved in the maintenance of LD and their presence in RPCs is required and beneficial for photoreceptor health. Indeed, the removal of LD by expression of dFatp-RNAi, Mdy-RNAi or Bmm induces a progressive photoreceptor degeneration in flies. Overexpression of FATP in RPCs expands the cellular LD content, which has a cell non-autonomous stimulatory effect on neighbouring photoreceptors that increases LD (in flies), neutral lipids, mitochondrial respiration and β-oxidation (in mice). LD accumulation in condition of overexpression of FATP in RPCs is not toxic for photoreceptors and could be considered beneficial through increased mitochondrial function and nutrient availability in these cells. Under conditions of oxidative stress (e.g., Aats-metFB mutant flies, in which ROS levels are elevated in photoreceptors), LD are no longer beneficial. In this scenario, LD could be considered as detrimental which could be due to their abnormal turnover under high ROS leading to enhanced lipid peroxidation.