PMP22 Regulates Cholesterol Trafficking and ABCA1-Mediated Cholesterol Efflux

Abstract

The absence of functional peripheral myelin protein 22 (PMP22) is associated with shortened lifespan in rodents and severe peripheral nerve myelin abnormalities in several species including humans. Schwann cells and nerves from PMP22 knock-out (KO) mice show deranged cholesterol distribution and aberrant lipid raft morphology, supporting an unrecognized role for PMP22 in cellular lipid metabolism. To examine the mechanisms underlying these abnormalities, we studied Schwann cells and nerves from male and female PMP22 KO mice. Whole-cell current-clamp recordings in cultured Schwann cells revealed increased membrane capacitance and decreased membrane resistance in the absence of PMP22, which was consistent with a reduction in membrane cholesterol. Nerves from PMP22-deficient mice contained abnormal lipid droplets, with both mRNA and protein levels of apolipoprotein E (apoE) and ATP-binding cassette transporter A1 (ABCA1) being highly upregulated. Despite the upregulation of ABCA1 and apoE, the absence of PMP22 resulted in reduced localization of the transporter to the cell membrane and diminished secretion of apoE. The absence of PMP22 also impaired ABCA1-mediated cholesterol efflux capacity. In nerves from ABCA1 KO mice, the expression of PMP22 was significantly elevated and the subcellular processing of the overproduced protein was aberrant. In wild-type samples, double immunolabeling identified overlapping distribution of PMP22 and ABCA1 at the Schwann cell plasma membrane and the two proteins were coimmunoprecipitated from Schwann cell and nerve lysates. Together, these results reveal a novel role for PMP22 in regulating lipid metabolism and cholesterol trafficking through functional interaction with the cholesterol efflux regulatory protein ABCA1.SIGNIFICANCE STATEMENT Understanding the subcellular events that underlie abnormal myelin formation in hereditary neuropathies is critical for advancing therapy development. Peripheral myelin protein 22 (PMP22) is an essential peripheral myelin protein because its genetic abnormalities account for ∼80% of hereditary neuropathies. Here, we demonstrate that in the absence of PMP22, the cellular and electrophysiological properties of the Schwann cells' plasma membrane are altered and cholesterol trafficking and lipid homeostasis are perturbed. The molecular mechanisms for these abnormalities involve a functional interplay among PMP22, cholesterol, apolipoprotein E, and the major cholesterol-efflux transporter protein ATP-binding cassette transporter A1 (ABCA1). These findings establish a critical role for PMP22 in the maintenance of cholesterol homeostasis in Schwann cells.

Keywords: ATP-binding cassette transporter 1; Schwann cell; apoE; cholesterol efflux; hereditary neuropathy; peripheral myelin protein 22.

Figure 1.

Intracellular cholesterol retention and perturbed membrane electrical properties in PMP22-deficient Schwann cells. A, B, Representative images of WT and PMP22 KO mouse Schwann cells (A) and quantification of filipin intensity (B) in the Golgi and at the plasma membrane (PM) after costaining with anti-GS15 antibody (n = 80 cells each genotype). “Rest” indicates the rest of a cell minus the Golgi and PM. C, Representative images of PMP22 (red) and cholesterol (blue) localization after scrambled or PMP22 shRNA-GFP plasmid (green) transfection in rat Schwann cells. Arrows point to transfected cells. D, Images depicting the location of patch pipettes relative to the cells subjected to current-clamp recordings. E–G, Current-clamp recordings of WT (gray bars) and PMP22 KO (black bars) mouse Schwann cells showing the membrane capacitance (E), membrane resistance (F), and time constant (G) for each experimental group (n = 10 cells per group from three independent replicates). For all experiments, values represent the mean ± SEM. **p < 0.01; ***p < 0.001, two-tailed Student's t test.

Figure 2.

Lipid accumulation and altered expression of lipid-metabolism-related genes in peripheral nerves from PMP22 KO mice. A, In the absence of PMP22, genes involved with cholesterol efflux (regulated by LXR/RXR) are increased, whereas genes linked to lipid synthesis (SREBP-1c and SREBP-2 pathways) are reduced (n = 3 independent experiments). RPL32 served as the internal control. B–E, Immunoblot detection (B, C) and quantification (D, E) of ABCA1 and apoE and LDLR in nerves from WT and PMP22 KO mice. F, Oil Red O staining detects accumulated neutral lipids in the perineurium and endoneurium (arrows) on longitudinal sections of PMP22 KO nerves. Nuclei (purple) are stained with hematoxylin. G, Representative images from P18 sciatic nerves showing vacuoles (V) and lipid droplets (arrow) in PMP22 KO Schwann cells. Boxed areas are magnified on the right of the panels. H, I, Immunoblot detection (H) and quantification (I) of autophagy markers LAMP1, p62, and LC3 in nerves from WT and PMP22 KO mice. Actin and GAPDH are shown as protein-loading controls. Values represent the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001, n.s., non-significant, one-sample t test or two-tailed Student's t test. B–I, n = 3–6 mice each genotype.

Figure 3.

Nerve macrophages and fibroblasts make a limited contribution to lipid abnormalities in PMP22 KO nerves. A, B, Immunoblot detection (A) and quantification (B) of ABCA1 and apoE in cultured nerve fibroblasts from WT and PMP22 KO mice. C, Representative immunoblot showing the expression of PMP22 in WT MNFs. D, Characterization of nerve fibroblast cultures by immunostaining for Thy-1 and the Schwann cell marker P75. E, Number of CD11b-positive cells in WT and PMP22 KO nerve sections after immunostaining and counting. F, G, Immunoblot detection (F) and quantification (G) of CD11b in whole-nerve lysates from WT and PMP22 KO mice. Actin or GAPDH is shown as the protein-loading control. H, Co-staining for CD11b and apoE showing macrophages positive for apoE. I, CD11b-positive macrophages are detected in the endoneurium of PMP22 KO nerves. Nuclei are stained with Hoechst dye (blue). Data were collected from three to four independent culture preparations (A–D) or from four animals per genotype (E–I). Values represent the mean ± SEM. *p < 0.05, two-tailed Student's t test.

Figure 4.

Elevated expression and altered processing of PMP22 in sciatic nerves from ABCA1-deficient mice. A, B, Immunoblot detection (A) and quantification (B) of myelin proteins MAG, P0, and P75 in nerve samples from control (WT), heterozygous (HET), and homozygous (KO) ABCA1-deficient mice. C, Immunoblot detection of PHH3 from WT, ABCA1 HET, and ABCA1 KO nerves. Nerve lysate from 6-week-old PMP22 KO mice is shown as a positive control. D, E, Immunoblot detection (D) and quantification (E) of PMP22 in nerves from WT, ABCA1 HET, and ABCA1 KO mice. Actin is shown as a protein-loading control. F, G, Immunoblot detection (F) and quantification (G) of PMP22 after incubation in buffer without enzyme (C), with endo H (H), or with N-glycosidase (N). A decrease in the endo H resistant (R) PMP22 and a corresponding increase in endo H sensitive (S, arrow) PMP22 is detected in the absence of ABCA1 (G). H, Oil Red O staining detects accumulated neutral lipids in the perineurium and endoneurium (arrows) on longitudinal sections of ABCA1 KO nerves. Results are from four to five mice per genotype. Values represent the mean ± SEM. *p < 0.05, one-way ANOVA followed by the Tukey's post test.

Figure 5.

ABCA1 and PMP22 are codistributed within Schwann cells. A, B, ABCA1 and PMP22 costaining in cultured WT mouse Schwann cells after Triton X-100 permeabilization. Control for staining specificity with secondary antibody alone is shown in A. Arrows point to the cell magnified in B. B, Representative high-magnification image after Z-stack (step size = 0.15 μm). Single XY plane in the middle, and XZ and YZ plane on the side display the colocalization of the two proteins (yellow) (n = 150 cells). C, Cell surface ABCA1 and PMP22 costaining in WT mouse Schwann cells revealing colocalization at the plasma membrane (yellow on merge) (n = 100 cells). D, ABCA1 and PMP22 costaining in longitudinal sections of WT mouse sciatic nerves (arrowheads). Arrow points to Schwann cell body, with coexpression of the two proteins near the nucleus (blue). Boxed areas are magnified in the upright corner. E, F, Coimmunoprecipitation of ABCA1 and PMP22 (E) or ABCA1 and PMP22-Myc (F) from WT rat Schwann cells or cells transfected with PMP22-Myc, respectively. G, H, Coimmunoprecipitation of ABCA1 and PMP22 from WT (G) and PMP22 KO (H) mouse sciatic nerve (SN) homogenates. Dystrophin, a known ABCA1-binding partner, was precipitated as a positive control. Total protein lysate (T), preclear (PC), and IP are included for each experiment (n = 3–5 independent experiments).

Figure 6.

Decreased apoE secretion and cholesterol efflux in the absence of PMP22. A, B, Immunoblot detection (A) and quantification (B) of secreted apoE and laminin in the conditioned media of cultured mouse Schwann cells after 24 h. C, D, Immunoblot detection (C) and quantification (D) of ABCA1, apoE, and LDLR in whole WT and PMP22 KO Schwann cell lysates. Actin or Ponceau staining are shown as protein-loading controls. E, F, Visualization (E) and quantification (F) of BODIPY-LDL uptake by WT and PMP22 KO mouse Schwann cells after 3, 8, and 24 h incubation times. G, ³H-cholesterol efflux to medium without apolipoprotein induction from WT and PMP22 KO mouse Schwann cells. H, ³H-cholesterol efflux to apoA1 in normal rat Schwann cells transfected with scrambled or PMP22 shRNA. I, BODIPY–cholesterol efflux to apoA1 in WT and PMP22 KO mouse Schwann cells after treatments with 8-Br-cAMP (cAMP) or CsA of ABCA1. J, K, Quantification (J) and immunoblot detection (K) of ABCA1 protein expression after treatments with vehicle (DMSO), 8-Br-cAMP (cAMP), or CsA. Results are representative of three to five independent experiments. Values represent the mean ± SEM. *p < 0.05; **p < 0.01, two-tailed Student's t test.

Figure 7.

Reduced ABCA1 expression at the plasma membrane in the absence of PMP22. A, B, Representative images of ABCA1 immunoreactivity on the surface of WT and PMP22 KO mouse Schwann cells (A) and quantification of fluorescence intensity (B) (n = 90 cells per genotype). C, Representative images of ABCA1 immunoreactivity on the surface of normal rat Schwann cells after PMP22 shRNA-GFP transfection. Arrow points to a transfected cell as indicated by GFP expression. Asterisk marks a nontransfected cell. D, E, Representative images of ABCA1 localization in WT and PMP22 KO nerve cross-sections (D) and teased nerve fibers (E). Arrowheads point to plasma-membrane-associated ABCA1 in WT nerves, and asterisks mark Schwann cell bodies. F, G, Immunoblot detection (F) and quantification (G) of Rab8 in lysates from WT and PMP22 KO sciatic nerves. H, Native immunoblot detection of lipidated (arrowheads) and poorly lipidated (arrows) apoE in whole sciatic nerve lysates. GAPDH or actin is shown as the protein-loading control. In A–C, Results are representative of three to five independent experiments; in D–H, results are from three to four animals per genotype. Values represent the mean ± SEM. *p < 0.05; ***p < 0.001, two-tailed Student's t test.