PDZD8-deficient mice accumulate cholesteryl esters in the brain as a result of impaired lipophagy

Abstract

Dyslipidemia including the accumulation of cholesteryl esters (CEs) in the brain is associated with neurological disorders, although the underlying mechanism has been unclear. PDZD8, a Rab7 effector protein, transfers lipids between endoplasmic reticulum (ER) and Rab7-positive organelles and thereby promotes endolysosome maturation and contributes to the maintenance of neuronal integrity. Here we show that CEs accumulate in the brain of PDZD8-deficient mice as a result of impaired lipophagy. This CE accumulation was not affected by diet, implicating a defect in intracellular lipid metabolism. Whereas cholesterol synthesis appeared normal, degradation of lipid droplets (LDs) was defective, in the brain of PDZD8-deficient mice. PDZD8 may mediate the exchange of cholesterol and phosphatidylserine between ER and Rab7-positive organelles to promote the fusion of CE-containing LDs with lysosomes for their degradation. Our results thus suggest that PDZD8 promotes clearance of CEs from the brain by lipophagy, with this role of PDZD8 likely contributing to brain function.

Keywords: Biological sciences; Molecular biology; Neuroscience.

Graphical abstract

Figure 1

Abnormal CE accumulation in the brain of PDZD8-deficient mice (A) Schematic representation of mouse brain regions subjected to lipidome analysis. (B–D) Heat maps of lipid amount ratios for PDZD8-KO relative to WT mice as determined by lipidome analysis. The ratios are shown according to the indicated color scale (B, left). The tissues compared in each analysis were obtained from the same mice, but the mice fed the ND or the HFD in (D) were different. (B) BG and Cx for three mice at 7 months (M) of age. (C) BG and liver from two mice at 5 months of age, with the ratios for each type of CE being shown to the right of the heat map. (D) BG and liver for three ND- or HFD-fed mice at 3 months of age (HFD feeding for 1 month). (E) Heat maps of lipid amount ratios for HFD-fed mice relative to ND-fed mice determined by lipidome analysis. The ratios are shown according to the indicated color scale (left) and were determined for the mice analyzed in (D). Lipid abbreviations: CE, cholesteryl ester; TG, triacylglycerol; DG, diacylglycerol; MG, monoacylglycerol; Chol, cholesterol; FA, fatty acid; Cer, ceramide; PA, phosphatidic acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphatidylinositol; PS, phosphatidylserine; HexCer, hexosylceramide; LPC, lyso-PC; LPE, lyso-PE; SM, sphingomyelin; OHCE, oxidized CE; OHC, oxysterol. See also Figures S1 and S2.

Figure 2

Diet-independent dyslipidemia in the brain of PDZD8-deficient mice (A–C) Lipid amount ratios (PDZD8-KO/WT) for lipid classes in the BG and Cx as in Figure 1B, in the BG and liver as in Figure 1C, or in the BG and liver of ND- or HFD-fed mice as in Figure 1D, respectively. The ratios were determined as the mean + SD in each case. (D and E) Amount of each type of CE in the BG and liver of WT and PDZD8-KO mice as in Figure 1C or in Figure 1D, respectively. Data (pmol/mg protein) are the mean + SD in each case. See also Figures S1 and S2.

Figure 3

Defective lipophagy in the brain of PDZD8-KO mice (A) Relative mRNA abundance for the cholesterol-related proteins SREBP2, HMGCR, LDLR, ACAT1, and CYP46A1 in the BG of WT or PDZD8-KO mice at 5 months of age (n = 2). Data are means + SD. Differences between genotypes were not significant (n.s.) by Student’s t test. (B) Schematic representation of mouse brain regions subjected to RT and real-time PCR analysis in (C) and (D). A, anterior; P, posterior; HB, hindbrain. (C) Relative mRNA abundance for dopamine receptor D2 (Drd2), dopamine transporter (DAT), norepinephrine transporter (NET), and brain-derived neurotrophic factor (BDNF) as markers for brain regions a to d indicated in (B) from 3-month-old WT mice. Data are means +SD (n = 2). (D) Relative mRNA abundance for PDZD8 in brain regions a to d from 3-month-old WT mice. Data are means +SD (n = 2). (E) Schematic representation of sectioning of the mouse brain for the examination of the striatum, medial habenula (MHb), amygdala, ventral tegmental area (VTA)/substantia nigra pars reticulata (SNr), and trigeminal mesencephalic nucleus (Vme). The red lines in the sagittal sections (upper) indicate the position of the coronal sections (lower), in which the red circles indicate the regions corresponding to the images in (F). (F) Immunofluorescence analysis of PDZD8 (green) in the WT brain. Enlarged images are also shown (lower). (G) Immunofluorescence analysis of PDZD8 (green) and AChT (red) for MHb neurons in the WT or PDZD8-KO brain. Nuclear staining with Hoechst 33342 (blue) is also shown in the merged images. (H) TEM images of lipophagy in MHb neurons of the WT or PDZD8-KO brain. White arrowheads indicate LDs fused with lysosomes and undergoing degradation. Black arrowheads indicate LDs making contact but not fusing with lysosomes and not undergoing degradation. (I and J) Number of LDs undergoing degradation per 100 μm² (I) or number of intact LDs per 100 μm² (J) as shown in (H). Data are means +SD (n = 26 and 28 images for WT and PDZD8-KO mice, respectively). ∗p < 0.05 (Student’s t test). See also Figure S3.

Figure 4

PDZD8 possesses PS and cholesterol transfer activity (A and B) Phospholipid transfer activity of His₆-PDZD8(ΔTM) in the absence (A) or presence (B) of acceptor liposomes as determined with the liposome-FRET assay shown in Figure S4A. The amount of transferred lipid (nM) is shown. (C) PS transfer activity of His₆-PDZD8 deletion mutants in the presence of acceptor liposomes. (D) Domain structure of mouse wild-type PDZD8 [PDZD8(WT)] and its deletion mutants. A summary of the PS and cholesterol transfer activities of each mutant determined as in (C) and (H), respectively, is shown on the right. ND, not determined. (E) Schematic representation of model for the accessibility of cholesterol to PDZD8-SMP in PS-poor (left) or PS-rich (right) domains of a lipid bilayer. (F) Schematic representation of the liposome-FRET assay for cholesterol transfer by His₆-PDZD8(ΔTM) as performed with donor liposomes containing rhodamine-PE, NBD-cholesterol, and DGS-NTA(Ni) and in the absence or presence of acceptor liposomes. The lipid constituents of the liposomes are shown in the boxes later in discussion. (G) Cholesterol transfer activity of PDZD8(ΔTM) in the absence or presence of acceptor liposomes. (H) Cholesterol transfer activity of the indicated PDZD8 deletion mutants in the presence of acceptor liposomes. (I) Model for the mechanism of lipid transfer by PDZD8, indicating that PDZD8 exchanges PS and cholesterol between the ER and Rab7-positive organelles in a manner dependent on its SMP domain. See also Figure S4.

Figure 5

PDZD8 facilitates the localization of cholesterol to Rab7-positive organelles as well as that of PS to ER (A) Confocal fluorescence images of HeLa cells transfected with control (siControl) or PDZD8 (siPDZD8) siRNAs as well as with expression vectors for mCherry-KDEL (red) and GFP-Evectin2-PH (green). (B) Pearson’s correlation coefficient for the colocalization of mCherry-KDEL and GFP-Evectin2-PH in images similar to those in (A) (n = 28 and 41 cells for siControl and siPDZD8, respectively). (C) Confocal fluorescence images of HeLa cells transfected with siControl or siPDZD8 as well as with expression vectors for mCherry-Rab7a(Q67L) (red) and GFP-Evectin2-PH (green). (D) Pearson’s correlation coefficient for the colocalization of mCherry-Rab7a(Q67L) and GFP-Evectin2-PH in images similar to those in (C) (n = 57 and 50 cells for siControl and siPDZD8, respectively). (E) Confocal fluorescence images of HeLa cells transfected with siControl or siPDZD8 as well as with an expression vector for EGFP-KDEL (green) and then metabolically labeled with filipin (shown blue in the merged images). (F) Pearson’s correlation coefficient for colocalization of EGFP-KDEL and filipin in images similar to those in (E) (n = 36 and 35 cells for siControl and siPDZD8, respectively). (G) Confocal fluorescence images of HeLa cells transfected with siControl or siPDZD8 as well as with expression vectors for EGFP-Rab7a(Q67L) (green) and mCherry-D4H (red) and then metabolically labeled with filipin (shown blue in the merged images). (H–J) Pearson’s correlation coefficient for the colocalization of EGFP-Rab7a(Q67L) and filipin (H), of EGFP-Rab7a(Q67L) and mCherry-D4H (I), or of mCherry-D4H and filipin (J) in images similar to those in (G) (n = 24 and 32 cells for siControl and siPDZD8, respectively). (K) Confocal fluorescence images of HeLa cells transfected with siControl or siPDZD8 as well as with an expression vector for mCherry-D4H (red) and then metabolically labeled with filipin (shown blue in merged images). The boxed regions in the left panels are shown enlarged in those to the right. (L) Proportion of cells transfected as in (K) that showed an abnormal distribution of mCherry-D4H as indicated in Figure S5B (n = 229 cells in nine images and 207 cells in eight images for siControl and siPDZD8, respectively). All quantitative data are presented as box-and-whisker plots, with the boxes indicating the median and upper and lower quartile values and with the whiskers representing the maximum and minimum values. ∗∗p < 0.01, ∗∗∗p < 0.001 (Student’s t test). See also Figure S5.

Figure 6

PDZD8 promotes fusion between D4H-positive and Rab7-positive organelles (A) Confocal fluorescence microscopy images of HeLa cells transfected with expression vectors for mCherry-D4H (red) and EGFP-Rab7a (green) as well as with either an expression vector for FLAG epitope-tagged mouse PDZD8(WT) or the corresponding control vector. The boxed regions in the left panels are shown enlarged in the right panels. (B) Double- or triple-layered organelles indicated by the arrows in (A) are shown at higher magnification. (C) Fluorescence intensity of mCherry-D4H (red) and EGFP-Rab7a (green) for structure 1 in (B). The bars above the graph indicate the distances between the corresponding peaks of fluorescence intensity.

Figure 7

PDZD8 promotes lipophagy (A) Confocal fluorescence microscopy images of PC12 cells transfected with siControl or siPDZD8 as well as with an expression vector for EGFP-PLIN2 (green) and then metabolically labeled with LysoTracker Red (red). The boxed regions in the left panels are shown enlarged in those to the right. (B) Proportion of cells showing LD aggregation in images similar to those in (A) (n = 175 in 11 images and 140 cells in 10 images for siControl and siPDZD8, respectively). (C) Proportion of cells examined in (D) with LD aggregates containing ≥2 LDs. (D) Number of LDs per aggregate for eight individual cells determined from images similar to those in (A). (E) Pearson’s correlation coefficient for the colocalization of EGFP-PLIN2 and LysoTracker Red in images similar to those in (A) for the cells examined in (D). (F) Diameter of LDs determined for the cells examined in (D) (n = 1097 and 665 LDs, respectively). (G) Fluorescence intensity of LysoTracker Red per cell area in PC12 cells transfected with siControl or siPDZD8 as in Figure S6B (n = 93 or 99 lysosomes for siControl and siPDZD8, respectively). (H) Schematic representation of the structure of mRFP-EGFP-PLIN2. (I) Confocal fluorescence microscopy of PC12 cells transfected with siControl or siPDZD8 as well as with expression vectors for EBFP-LAMP1 (blue) and mRFP-EGFP-PLIN2 (red and green). The boxed regions in the left panels are shown at higher magnification than those to the right, with circles indicating LAMP1-positive lysosomes. Arrowheads indicate LDs enclosed within lysosomes. (J) Ratio of EGFP/mRFP fluorescence intensity for tagged PLIN2 within lysosomes as shown as in (I) (n = 108 or 94 lysosomes for siControl and siPDZD8, respectively). Quantitative data are presented as box-and-whisker plots or as means + SD. ∗∗∗p < 0.001 (Student’s t test). See also Figure S6.