FTY720/fingolimod increases NPC1 and NPC2 expression and reduces cholesterol and sphingolipid accumulation in Niemann-Pick type C mutant fibroblasts

Abstract

Niemann-Pick type C (NPC) disease is a fatal neurodegenerative disorder caused by mutations in NPC1 or NPC2 with decreased functions leading to lysosomal accumulation of cholesterol and sphingolipids. FTY720/fingolimod, used for treatment of multiple sclerosis, is phosphorylated by nuclear sphingosine kinase 2, and its active phosphorylated form (FTY720-P) is an inhibitor of class I histone deacetylases. In this study, administration of clinically relevant doses of FTY720 to mice increased expression of NPC1 and -2 in brain and liver and decreased cholesterol in an SphK2-dependent manner. FTY720 greatly increased expression of NPC1 and -2 in human NPC1 mutant fibroblasts that correlated with formation of FTY720-P and significantly reduced the accumulation of cholesterol and glycosphingolipids. In agreement with this finding, FTY720 pretreatment of human NPC1 mutant fibroblasts restored transport of the cholera toxin B subunit, which binds ganglioside GM1, to the Golgi apparatus. Together, these findings suggest that FTY720 administration can ameliorate cholesterol and sphingolipid storage and trafficking defects in NPC1 mutant fibroblasts. Because neurodegeneration is the main clinical feature of NPC disease, and FTY720 accumulates in the CNS and has several advantages over available histone deacetylase inhibitors now in clinical trials, our work provides a potential opportunity for treatment of this incurable disease.—Newton, J., Hait, N. C., Maceyka, M., Colaco, A., Maczis, M., Wassif, C. A., Cougnoux, A., Porter, F. D., Milstien, S., Platt, N., Platt, F. M., Spiegel, S. FTY720/fingolimod increases NPC1 and NPC2 expression and reduces cholesterol and sphingolipid accumulation in Niemann-Pick type C mutant fibroblasts.

Niemann-Pick type disease C (NPC) is a rare, fatal, autosomal recessive lipid storage disease caused by mutations in the NPC1 (95% of cases) or NPC2 (5% of cases) gene. NPC2 binds cholesterol and transfers it to the cholesterol-binding domain of NPC1, facilitating transport of cholesterol from the late endosome/lysosome compartment (1, 2). Mutations in NPC1 or -2 typically decrease lysosomal functions of the proteins, resulting in accumulation of a variety of lipids in late endosome and lysosomal compartments, most prominently unesterified cholesterol and sphingolipids (3–7). Patients with NPC disease have profound and progressive neurologic defects that are responsible for most of the morbidity and mortality associated with the disease. Current treatment strategies for NPC disease include depletion of accumulated cholesterol with cyclodextrin (8) and decreasing glycosphingolipid (GSL) accumulation with the glucosylceramide synthase inhibitor miglustat (9). Miglustat has been approved by some regulatory agencies and has some reported efficacy at slowing the progression of the disease. 2-Hydroxypropyl-β-cyclodextrin (HPBCD) therapy has shown promise in animal studies (2, 10–16), and VTS-270, a defined preparation of HPBCD, is currently being evaluated in a phase 2b/3 clinical trial (NCT02534844). There have been challenges to treatment with HPBCD; particularly related to the drug’s inability to cross the blood–brain barrier and rapid export from the brain resulting in the need for intrathecal administration into the cerebral spinal fluid. Although manageable in the context of a lethal disease, HPBCD administration is associated with ototoxicity in both animal models (16, 17) and humans (18).

It has recently been shown that histone deacetylase (HDAC) inhibitors can reduce the cholesterol storage phenotype in fibroblasts derived from patients with NPC disease by epigenetically increasing the expression of low-activity mutant NPC proteins to high enough levels to overcome the defect in cholesterol efflux from late endosomes and lysosomes (19). Similarly, a genome-wide conditional synthetic lethality screen in yeast identified HDAC inhibition as a potential therapy for NPC disease and demonstrated that treatment of human NPC mutant fibroblasts with a HDAC inhibitor ameliorated lysosomal accumulation of both cholesterol and sphingolipids and defective esterification of low-density lipoprotein (LDL)–derived cholesterol (20).

Other studies have suggested that sphingosine accumulation is another contributing factor to the pathogenesis in NPC1 disease (6, 21, 22). In the major pathway of sphingolipid metabolism, sphingosine must exit the lysosome, be phosphorylated by sphingosine kinase isoenzymes, sphingosine kinase (SphK)-1 or -2, generating the potent bioactive sphingolipid metabolite sphingosine-1-phosphate (S1P). Although most of the known actions of S1P are mediated by 5 specific GPCRs, termed S1P receptor (S1PR)1–5, it also has intracellular actions (23). We have shown that S1P produced by nuclear SphK2 is an endogenous inhibitor of HDAC1/2 that epigenetically regulates gene expression (24). The sphingosine analog FTY720 (fingolimod), used for the treatment of multiple sclerosis, enters the nucleus where it is phosphorylated by SphK2. Its active form FTY720-phosphate (FTY720-P), which accumulates there, is also a potent class I HDAC inhibitor (25). Moreover, we found that FTY720-P accumulates in the brains of rodents, enhances histone acetylation and gene expression programs associated with memory and learning, and facilitates extinction of fearful memories (25). Therefore, in this work we examined whether administration of FTY720, similar to other HDAC inhibitors (19, 20, 26, 27), could enhance expression of NPC1 and -2 and concomitantly reduce cellular cholesterol and sphingolipids in cultured fibroblasts from patients with NPC1.

MATERIALS AND METHODS

Animal studies

Animal studies were conducted in the Animal Research Core Facility at Virginia Commonwealth University (VCU) School of Medicine in accordance with all institutional guidelines. All procedures were approved by the VCU Institutional Animal Care and Use Committee, which is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care (A3281-01). Mice were kept on a 12-h light/dark cycle with free access to food. Littermate wild-type (WT) C57BL/6 and Sphk2^−/− mice were obtained from Dr. Richard Proia [National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health (NIH), Bethesda, MD, USA]. Male mice were used in all the experiments. FTY720 (Sigma-Aldrich, St. Louis, MO, USA) in saline or saline alone was administered by gavage. Livers were removed immediately after death, and hippocampus and cerebellum were dissected from the brain after removing the cortex (25). Tissues were snap frozen in liquid nitrogen and stored at −80°C until analyzed.

Cell culture and transfection

NPC1 mutant skin fibroblast lines were derived from patients with NPC1 who were evaluated under a National Institute of Childhood Health and Human Development, NIH Institutional Review Board–approved clinical protocol and obtained with consent. The following cell lines were used: NPC-25 (fs-exon 20, c.2979dupA, and p.N701K); NPC-5 (p.I1061T and p.R1186G); NPC-26 (c.3742_3745delCTCA fs-exon24 and p.R1059Q); WT-2 and WT-A (controls). Dermal fibroblasts were cultured in minimum essential medium (MEM; Thermo Fisher Scientific, Waltham, MA, USA) containing 10% fetal bovine serum (FBS, Serum Source International, Charlotte, NC, USA), 100 U/ml penicillin, and 100 μg/ml streptomycin (Thermo Fisher Scientific). Chinese hamster fibroblasts null for Npc1 in which the Npc1 locus in CHO-JP17 cells was disrupted by retrovirus-mediated gene trap mutagenesis (28) were maintained in MEM-α (Thermo Fisher Scientific) with 10% FBS as previously described (29).

NIH 3T3 fibroblasts were obtained from the American Type Culture Collection (CRL-1658; Manassas VA, USA) and maintained in DMEM (Thermo Fisher Scientific) containing 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin. NIH 3T3 fibroblasts were transfected with vector, Sphk2, or catalytically inactive Sphk2^G212E (24). Sphk2 was downregulated by transfection with On-TargetPlus SMARTpool small interfering RNA (siRNA) or control scrambled siRNA (Dharmacon, Lafayette, CO, USA).

Cholesterol assays

Cholesterol and cholesterol esters were measured with the Amplex Red Cholesterol Assay Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. In brief, cells in 6-well plates were washed 3 times with cold PBS, scraped into 200 µl cold PBS, and pulse sonicated 3 times for 10 s on ice. Aliquots were incubated for 30 min at 37°C with horseradish peroxidase (HRP; 2 U/ml), cholesterol oxidase (2 U/ml), without or with cholesterol esterase (0.2 U/ml), in the presence of Amplex Red reagent (300 µM) in 0.1 M potassium phosphate buffer (pH 7.4) containing 50 mM NaCl, 5 mM cholic acid, and 0.1% Triton X-100. Total cholesterol and cholesterol esters (total minus free) were determined from standard curves and normalized to protein measured with the Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA, USA). Fluorescence was measured on a Victor ×4 2030 Multilabel Plate Reader (Perkin-Elmer, Waltham, MA, USA). Tissues were snap frozen in liquid nitrogen and powdered with a mortar and pestle. Lipids were extracted in chloroform:isopropanol:NP-40 (7:11:0.1 v/v/v), the solvents evaporated, the lipids solubilized in Amplex reaction buffer, and the cholesterol measured, as previously described.

Immunoblot analysis

Proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes with a Pierce G2 fast blotter (Thermo Fisher Scientific). Membranes were blocked with 5% nonfat dry milk in Tris-buffered saline containing 0.1% Tween 20. The following primary antibodies were used for immunoblot analysis: NPC1 (1:5000 dilution; Abcam, Cambridge, MA, USA); NPC2 (1:1000 dilution; Sigma-Aldrich); GAPDH and tubulin (1:5000 dilution; Cell Signaling Technology, Danvers, MA, USA); acetylated histone H3 and histone H3(1:1000 dilution; Abcam); V5 (1:5000 dilution; Thermo Fisher Scientific); and SphK2 (1:1000 dilution; from Dr. R. Proia). Immunopositive bands were visualized by incubating with HRP-conjugated secondary antibodies (1:5000 dilution; Jackson ImmunoResearch, West Grove, PA, USA) for 1 h at room temperature before adding SuperSignal West Pico (Thermo Fisher Scientific) chemiluminescence substrate (24). Blots from at least 3 independent experiments were quantified with ImageJ and normalized to their corresponding loading controls.

Real-time PCR

Total RNA was isolated from tissues and cells using Trizol and reverse transcribed with the High Capacity cDNA Reverse Transcription Kit (both from Thermo Fisher Scientific). Premixed primer-probe sets and TaqMan Universal PCR Master Mix (Thermo Fisher Scientific) were used to determine mRNA levels. cDNAs were diluted 10-fold (for the target genes) or 100-fold (for GAPDH) and amplified using the ABI7900HT cycler (Thermo Fisher Scientific). In some experiments SYBR Green quantitative PCR (qPCR) was performed with the CFX Connect cycler (Bio-Rad Laboratories). Gene expression levels were calculated by the ΔΔC_t method and normalized to GAPDH expression.

Quantification of FTY720 and FTY720-P

FTY720 and FTY720-P were measured by liquid chromatography, electrospray ionization–tandem mass spectrometry (LC-ESI-MS/MS;5500 QTrap; AB Sciex, Framingham, MA, USA) (25).

Measurement of GSLs

GSLs were quantified as described in Neville et al. (30). In brief, lipids from cell homogenates were extracted with chloroform and methanol and then isolated with solid-phase C18 columns (Telos; Kinesis, St. Neots, United Kingdom). Purified fractions were evaporated under nitrogen and treated with ceramide glycanase (prepared in house from the medicinal leech Hirudo medicinalis/verbena). Free glycans were then fluorescently labeled with anthranilic acid (2AA), and separated on 6S SPE columns (Supelco, Bellefonte, PA, USA) to remove excess free label. 2AA-labeled oligosaccharides were separated by normal-phase HPLC on a 4.6 × 250 mm TSK gel-Amid 80 column (Anachem, Luton, United Kingdom). The HPLC system consisted of an Alliance 2695 separations module and an in-line model 2475 multifluorescence detector (Waters Corp., Milford, MA, USA) set at 360- and 425-nm excitation and emission wavelengths, respectively.

Cholera toxin B subunit staining

Cells cultured on 22-mm² coverslips were pretreated for 24 h with either vehicle or 2 μM FTY720 (Cayman Chemicals, Ann Arbor, MI, USA), then treated with 1 µg/ml Alexa-Fluor-594-cholera toxin B subunit (CTB; Thermo Fisher Scientific) for 30 min, washed with complete medium, and incubated in medium containing 1% BSA for 90 min. Cells were then washed twice with 1 ml PBS, fixed with 4% paraformaldehyde for 15 min, and washed 3 times with PBS containing 10% glycine; the nuclei were stained with Hoechst 33342 before imaging. Alternatively, in some experiments, fixed CTB-labeled cells were permeabilized with 0.5% Triton X-100 for 3 min, washed extensively with 10 mM glycine in PBS, and incubated with antibodies to giantin (1:100; Cell Signaling Technology) and either calnexin and calreticulin (1:100; Santa Cruz Biotechnology, Dallas, TX, USA) or Rab7 antibodies (1:150; Santa Cruz) directly conjugated to Alexa 647, in PBS containing 10 mM glycine, 0.1% Triton X-100, and 1% IgG-free BSA. The cells were then washed and incubated with fluorescently labeled secondary antibodies (1:400; Thermo Fisher Scientific) for 20 min (24). Images of glycerol-mounted cells were collected with Zen software with a 700 LSM confocal microscope (Zeiss, Thornwood, NY, USA) equipped with a ×63, 1.4-NA objective. Images were collected by using identical settings across all samples within each experiment, and raw images were exported as .tif files without manipulation.

Statistical analyses

All experiments were repeated at least 3 times with consistent results. For mouse studies, 3–6 randomly chosen mice were used for each experimental group. All cell culture data were from biologic triplicates, unless otherwise indicated. Statistical analysis was performed with unpaired 2-tailed Student’s t test, for comparison of 2 groups, and ANOVA followed by post hoc tests for multiple comparisons (Prism; GraphPad, San Diego, CA, USA). A value of P < 0.05 indicates significance.

RESULTS

Oral administration of FTY720 to mice increased expression of NPC1 and -2 in a SphK2-dependent manner

It has been shown that HDAC inhibitors enhance expression of NPC1 (19), and we have found that the phosphorylated form of FTY720, which accumulates in mouse hippocampus, is a potent class I HDAC inhibitor (25). Therefore, we sought to investigate the effects of FTY720 administration on NPC1 and -2 expression in mice. In clinical trials with FTY720 in humans, renal transplant recipients received FTY720 at doses of 2.5 or 5.0 mg/d (31) and similar doses were also used in the first phase 2 trial for relapsing multiple sclerosis (32). Mice were treated orally with increasing concentrations of FTY720 up to 0.5 mg/kg, equivalent to a 2.5 mg dose in humans based on the dose conversion between mice and humans (33). FTY720 significantly increased hippocampal mRNA levels of both Npc1 and -2 in a dose-dependent manner, with a maximum effect at 0.5 mg/kg in WT C57BL/6 mice (Fig. 1A). Because we have shown that the prodrug FTY720 is mainly phosphorylated by SphK2 in vivo and less FTY720-P was observed in the hippocampus of SphK2^−/− mice (25), we compared the effects of FTY720 on NPC1 and -2 expression in SphK2^+/+ and SphK2^−/− mice. Daily administration of FTY720 to SphK2^−/− mice did not significantly increase Npc1 and -2 mRNA expression, in contrast to its effects in SphK2^+/+ mice (Fig. 1B). Similarly, Npc1 and -2 protein levels determined by immunoblot analysis with specific antibodies were increased only in WT mice, but not in SphK2-knockout mice (Fig. 1C). We next examined changes in expression of Npc1 and -2 in the cerebellum, the region of the brain that is significantly affected in NPC disease. NPC1 and -2 expression was significantly increased in WT but not in SphK2-knockout mice by administration of FTY720 (Fig. 2A). NPC1 and -2 protein levels were increased to a much larger extent in the cerebellum than in the hippocampus (Fig. 2B vs. Fig. 1C), which correlated with much larger increases in levels of FTY720-P in the cerebellum (Fig. 2C). Likewise, administration of increasing doses of FTY720 increased NPC1 and -2 mRNA expression in the liver (Fig. 3A), where cholesterol is known to accumulate in NPC1 disease (13, 34). However, NPC1 and -2 mRNA and protein were not significantly increased by FTY720 in SphK2^−/− mice (Fig. 3B, C), concomitant with the significantly decreased formation of FTY720-P in these mice (Fig. 3D). FTY720 administration significantly decreased liver cholesterol levels in WT mice, but not in SphK2-null mice (Fig. 3E). These results show that FTY720 increases NPC1 and -2 mRNA and protein and their activity in cholesterol clearance.

Figure 1.

Treatment of mice with FTY720 increased hippocampal NPC1 and -2 expression in an SphK2-dependent manner. A) C57BL/6 mice were treated without or with increasing doses of FTY720 (0.01, 0.1, 0.5 mg/kg) for 4 d by oral gavage (n = 3 mice per group). B, C) Sphk2-knockout mice and WT controls were treated without or with 0.5 mg/kg FTY720 for 4 d by oral gavage. A, B) Levels of Npc1 and -2 mRNA were determined in hippocampus by qPCR and normalized to Gapdh. C) NPC1, NPC2, and SphK2 proteins in hippocampi were analyzed by immunoblot analysis with the indicated antibodies. Blots were stripped and reprobed with anti-GAPDH antibody to show equal loading and transfer. Triplicate blots were quantitated by densitometry. Data are means ± sd. *P < 0.05 vs. vehicle-treated mice; ^#P < 0.05 vs. WT-treated mice.

Figure 2.

FTY720-induced increases in NPC1 and -2 in cerebellum requires SphK2. A, B) Sphk2-knockout mice and WT controls were treated without or with 0.5 mg/kg FTY720 for 4 d by oral gavage (n = 3 mice per group). A) Levels of Npc1 and -2 mRNA were determined in cerebellum by qPCR and normalized to Gapdh. Data are means ± sd. *P < 0.01 vs. vehicle-treated mice; ^#P < 0.05 vs. WT-treated mice. B) NPC1 and -2 proteins in cerebellum were determined by immunoblot analysis with anti-NPC1 and -2. Blots were stripped and reprobed with anti-GAPDH antibody to show equal loading and transfer. Triplicate blots were quantitated by densitometry. Data are means ± sd. *P < 0.01 vs. vehicle-treated mice; ^#P < 0.01 vs. WT-treated mice. C) In a duplicate experiment, accumulation of FTY720-P in hippocampus and cerebellum from WT control and Sphk2-knockout mice was measured by LC-ESI-MS/MS. Data are means ± sd. *P < 0.01 vs. hippocampus.

Figure 3.

FTY720 administration increased NPC1 and -2 expression and reduced cholesterol in liver of WT but not SphK2-knockout mice. A) C57BL/6 mice were treated without or with FTY720 (0.01, 0.1, 0.5 mg/kg) for 4 d by oral gavage. B–E) SphK2-knockout and WT mice were treated without or with 0.5 mg/kg FTY720 for 4 d by oral gavage (n = 3 mice per group). A, C) Levels of Npc1 and -2 mRNA were determined in liver by qPCR and normalized to Gapdh. *P < 0.01 vs. vehicle-treated mice; ^#P < 0.01 vs. WT-treated mice. B) NPC1, NPC2, and SphK2 proteins in liver were analyzed by immunoblot analysis with the indicated antibodies. Blots were stripped and reprobed with anti-GAPDH antibody. Triplicate blots were quantitated by densitometry. *P < 0.05 vs. vehicle-treated mice; ^#P < 0.05 vs. WT-treated mice. D) Accumulation of FTY720-P in liver from WT control and Sphk2-knockout mice was measured by LC-ESI-MS/MS. ^#P < 0.01 vs. WT-treated mice. E) Total liver cholesterol was determined by Amplex Red fluorescence assays. Data are means ± sd. *P < 0.01 vs. vehicle-treated mice.

FTY720 conversion to FTY720-P by SphK2 increased NPC1 and -2 expression and decreased cholesterol content in mouse fibroblasts

We next examined the effects of FTY720 on expression of NPC1 and -2 in NIH 3T3 mouse fibroblasts, as treatment of fibroblasts with FTY720 has been shown to elevate nuclear FTY720-P accompanied by decreased HDAC activity and increased histone acetylation (35). In agreement, after FTY720 treatment, FTY720-P accumulated in the nucleus and enhanced histone H3K9 acetylation, with a corresponding increase in expression of NPC1 and -2 at both the protein and the mRNA levels (Fig. 4A–C). Moreover, overexpression of SphK2, which increased formation of nuclear FTY720-P, further increased expression of NPC1 and -2 compared with vector transfectants. However, in contrast, overexpression of catalytically inactive SphK2 had no (Fig. 4A) or only minor effects compared with vector-transfected cells (Fig. 4B, C). Consistent with increased levels of NPC1 and -2, FTY720 reduced total cholesterol in the cells and even more so in cells overexpressing SphK2, but not in cells expressing catalytically inactive SphK2 (Fig. 4D).

Figure 4.

FTY720 treatment enhanced NPC1 and -2 expression and decreased cholesterol in NIH 3T3 fibroblasts in an SphK2-dependent manner. NIH 3T3 fibroblasts transfected with vector, V5-SphK2, or catalytically inactive V5-SphK2 (ci-SphK2) were treated with vehicle or FTY720 (2 µM) for 8 h. A) Proteins were analyzed by immunoblot analysis with the indicated antibodies. H3 and tubulin were used as loading controls for the nucleus and whole-cell extracts, respectively. Triplicate blots were quantitated by densitometry. *P < 0.05 vs. the respective vehicle control. B) Npc1 and -2 mRNAs were quantified by qPCR and normalized to Gapdh. C) Nuclear levels of FTY720-P were determined by LC-ESI-MS/MS. D) Cellular cholesterol content was determined by Amplex Red fluorometric assay. Data are means ± sd. *P < 0.01 vs. the respective vehicle control.

SphK2 was downregulated to further confirm that the effects of FTY720 treatment on NPC1 and -2 expression and cholesterol levels are dependent on SphK2. Reminiscent of the results in SphK2-null mice (Figs. 1–3), downregulation of SphK2 attenuated by more than 60% the FTY720-induced expression of NPC1 and -2 (Fig. 5A), as well as its effects on their protein levels (Fig. 5B). Moreover, transfection with siSphK2 completely prevented the reduction of total cellular cholesterol levels by FTY720 treatment (Fig. 5C).

Figure 5.

Downregulation of SphK2 suppressed FTY720-induced NPC1 and -2 expression and cholesterol reduction. NIH 3T3 fibroblasts transfected with siControl or siSphK2 were treated with vehicle or with the indicated concentrations of FTY720 for 8 h. A) Npc1, Npc2, and Sphk2 mRNAs were quantified by qPCR and normalized to Gapdh. B) Proteins were analyzed by immunoblot with the indicated antibodies. Blots were stripped and reprobed with anti-tubulin antibody. Triplicate blots were quantitated by densitometry. *P < 0.01 vs. vehicle-treated cells. ^#P < 0.05 vs. siControl cells-treated with FTY720. C) Cellular cholesterol content was determined by Amplex Red fluorometric assay. *P < 0.01 vs. vehicle-treated cells; ^#P < 0.01 vs. siControl cells treated with FTY720.

FTY720 treatment increased NPC1, -2, and ABCA1 expression in human NPC1 mutant fibroblasts

We next examined the effects FTY720 treatment of NPC1 mutant primary fibroblasts from well-characterized patients with a wide disease spectrum followed at NIH (https://clinicaltrials.gov/; no. NCT00344331), and control human fibroblasts (36, 37). Age-adjusted NIH neurologic severity scores (AANSSs) (38) ranged from 0 to 6; NPC-26 (c.3742_3745delCTCA fs-exon24, p.R1059Q) from a patient with adult-onset NPC, AANSS, 0.7; NPC-5 (p.I1061T, p.R1186G), AANSS 1.4; and NPC-25 (fs-exon 20, c.2979dupA, p.N701K), AANSS, 6. Treatment of these NPC1 mutant and WT fibroblasts with FTY720 significantly increased expression of NPC1 and -2 mRNA (Fig. 6A). Although the well-known pan-HDAC inhibitor suberoylanilide hydroxamic acid (SAHA; vorinostat) increased mRNA of NPC1 to a greater extent than did FTY720 treatment, up-regulation of NPC2 by SAHA and FTY720 was not significantly different. In agreement with these findings, treatment with FTY720 also increased NPC1 and -2 protein levels in NPC1 mutant fibroblasts (Fig. 6B, C). Given that the NPC1 mutant fibroblasts have reduced levels of NPC1 protein, the effects of FTY720 appeared to be more pronounced than in the control fibroblasts.

Figure 6.

FTY720 treatment induced expression of NPC1, NPC2, and ABCA1 in NPC1 mutant human fibroblasts. Normal human fibroblasts (WT-2 or WT-A) and the indicated NPC1 mutant fibroblasts were treated with vehicle or 2 µM FTY720 or SAHA for 8 h. A) NPC1, NPC2, and ABCA1 mRNAs were quantified by qPCR and normalized to GAPDH. B) Proteins were analyzed by immunoblot with anti-NPC1 and -NPC2 antibodies. Blots were stripped and reprobed with anti-GAPDH antibody. C) Triplicate blots were quantified by densitometry. Data are means ± sd. *P < 0.05 vs. untreated control.

Because the cholesterol trafficking defect in NPC disease results in reduced activity of the ATP-binding cassette transporter A1 (ABCA1) (39), which mediates the efflux of lipids, including cholesterol, from cells (40), we also examined the effects of FTY720 on its expression. Similar to a recent study (41), we found that HDAC inhibition by SAHA increased ABCA1 expression (Fig. 6A). FTY720 treatment also increased expression of ABCA1 to a extent similar to SAHA in both control and NPC1 mutant fibroblasts.

Cholesterol accumulation in human NPC1 mutant fibroblasts was reduced by FTY720 treatment

It was of interest to determine whether the conversion of FTY720 to FTY720-P was similar in control and NPC1 mutant fibroblasts. Similar levels of FTY720-P were produced in both control and mutant cells after treatment with 2 µM FTY720 (Fig. 7A). As expected, FTY720-P was produced in a dose-dependent manner. FTY720 treatment also induced a significant decrease in total cellular cholesterol in both control and NPC1 mutant cells at a concentration of 1–2 µM (Fig. 7B). As NPC proteins are involved in the egress of unesterified cholesterol from late endosomes/lysosomes and delivery to the ER (1–5, 7), increases in NPC may also lead to the increased cholesterol esterification that occurs in the endoplasmic reticulum (ER). In agreement with previous studies (42), NPC1 mutant fibroblasts contain a much lower percentage of cholesterol esters than WT fibroblasts (0.4% of total cholesterol compared to 20%). Treatment with FTY720 significantly increased cholesterol esters in NPC1 mutant fibroblasts, but not in the WT cells (Fig. 7C). Furthermore, treatment with FTY720 or SAHA did not decrease cholesterol levels in Npc1 null Chinese hamster ovary (CHO) cells (Fig. 7D), consistent with their effects as HDAC inhibitors on gene regulation.

Figure 7.

FTY720 treatment reduced cholesterol levels in normal and NPC1 mutant human fibroblasts. A, B) Normal human fibroblasts (WT-A, WT-2) and the indicated NPC1 mutant fibroblasts were treated for 24 h with vehicle, 2 µM FTY720 (A), or the indicated concentrations of FTY720 (B). A) FTY720-P levels were measured by LC-ESI-MS/MS. B) Cellular cholesterol levels were determined. Data are means ± sd. *P < 0.05 vs. untreated control. C) Normal human WT-2 fibroblasts and NPC1 mutant fibroblasts (NPC-5 and-25) were treated with vehicle or 2 µM FTY720 for 24 h, and levels of cholesterol esters (total minus free cholesterol) were determined by Amplex Red fluorometric assay (n = 5). Data are means ± sem. *P < 0.01 vs. untreated control. D) Npc1-null CHO cells were treated with vehicle or 2 µM FTY720 or SAHA for 24 h, and levels of cellular cholesterol were determined. Data are means ± sd.

FTY720 treatment reversed GSL storage defects in human NPC mutant fibroblasts

In addition to cholesterol, NPC disease causes the accumulation of multiple GSLs in late endosomal and lysosomal compartments (43). Thus, we checked whether FTY720 treatment also facilitates clearance of accumulated GSLs in addition to cholesterol. As anticipated, levels of GSLs, particularly globotriaosylceramide Gb3 and ganglioside GM3, were increased in NPC1 mutant fibroblasts compared with controls (Fig. 8A). Similar to its effects on cholesterol, FTY720 treatment reduced total GSL levels and those of Gb3 and GM3 in the human NPC1 mutant fibroblasts. Mutant NPC-25 and -26 fibroblasts had more pronounced accumulation of Gb3 and GM3 than did NPC-5, and thus FTY720 more drastically reduced their levels in these fibroblasts compared with NPC-5. Furthermore, FTY720 had no effect on levels of GSLs in WT fibroblasts, which in contrast to accumulation of GSLs in mutant fibroblasts, have lower levels of GSLs, including Gb3 and GM3.

Figure 8.

FTY720 decreased GSL storage and corrected GM1 trafficking in NPC1 mutant human fibroblasts. Normal human fibroblasts (WT-2) and NPC1 mutant fibroblasts (NPC-5, -25, and -26) were treated with either vehicle or the indicated concentrations of FTY720 for 24 h. A) Total GSLs, Gb3, and GM3 were measured by HPLC. Data are means ± sd. *P < 0.01 vs. untreated. B) Normal human and NPC1 mutant human fibroblasts (NPC-25) were pretreated with 2 µM FTY720 or SAHA for 24 h, incubated with fluorescent labeled CTB (green), and then chased for 90 min. Cells were fixed, and counterstained with Hoechst (blue). C, D) In a duplicate experiment, cells were pulse-labeled with fluorescently labeled CTB (green) followed by immunofluorescence staining with the Golgi marker giantin (red) and the ER markers calnexin and calreticulin (blue) (C) or with the endosomal marker Rab7 (red) (D). Labeled cells were visualized by confocal laser scanning microscopy. Yellow in the merged images indicates colocalization. Representative images are shown. Scale bars, 25 µm.

It has been reported that ganglioside GM1 accumulates in early endosomes of NPC1-deficient cells (29) and in NPC1 mutant fibroblasts (44, 45). In agreement, we found that fluorescently labeled cholera toxin subunit B (CTB), which binds to GM1 at the plasma membrane, was transported to the Golgi after internalization in control fibroblasts, but accumulated in puncta observed in the cytoplasm of NPC1 mutant cells (Fig. 8B). Pretreatment of these cells with 2 µM FTY720 or SAHA before pulse-chase labeling with CTB corrected this GM1 trafficking defect and restored transport of CTB to the Golgi apparatus. Indeed, confocal microscopy showed that FTY720 treatment induced colocalization of CTB with giantin, a Golgi marker, but not with the ER markers calnexin and calreticulin (Fig. 8C) or the endosomal marker Rab7 (Fig. 8D). Altogether, these data indicate that FTY720 rescues aberrant cholesterol and sphingolipid storage and trafficking in NPC1 mutant cells.

DISCUSSION

HDAC inhibitors have long been used in psychiatry and various brain disorders and are under investigation as possible treatments for several neurologic diseases (46–48). It was shown that treatment with the class I HDAC inhibitor valproic acid up-regulated neurotrophic genes, enhanced neuronal differentiation, and improved defective cholesterol traffic in neural stem cells from NPC1-deficient mice (49). Indeed, treatment with class I HDAC inhibitors of cells with the NPC1^I1061T mutation, which has much lower activity than WT NPC1 and is the most common mutation in patients with NPC1 (50), increased NPC1 expression and reduced cellular accumulation of unesterified cholesterol (19). It was also reported that 11 HDAC genes are up-regulated in fibroblasts from patients with NPC disease and that the pan-HDAC inhibitor SAHA reversed the dysregulation of most of these genes and reduced lysosomal accumulation of cholesterol, as well as sphingolipids, and the defective esterification of LDL-derived cholesterol (20). These findings led to the initiation of a phase 1/2 clinical trial testing the efficacy of SAHA in adult patients with NPC (NCT02124083). However, SAHA is poorly soluble in aqueous solutions and does not readily cross the blood–brain barrier in mice (51). Nevertheless, the HDAC inhibitor panobinostat is orally available and crosses the blood–brain barrier (52). Moreover, although SAHA alone did not improve survival of the Npc1^nmf164 mouse model of NPC1 disease, low-dose, once-weekly intraperitoneal injections of the combination of SAHA with 2-hydroxypropyl-β-cyclodextrin and polyethylene glycol not only increased its penetration of the brain and increased histone acetylation, it also preserved neurites and Purkinje cells, delayed symptoms of neurodegeneration and systemic inflammation, and doubled their life span (27).

Recently, it has been shown that the prodrug FTY720 is phosphorylated to the active form FTY720-P in the nucleus, where it potently inhibits class I HDACs (25, 35). FTY720 also readily crosses the blood–brain barrier, and FTY720 and FTY720-P accumulate in human (53) and animal brains (25, 54). Similar to other HDAC inhibitors, oral administration of FTY720 to mice enhanced histone acetylation and gene expression programs associated with memory and learning, and rescued memory deficits of severe combined immunodeficiency (SCID) mice, independent of its immunosuppressive actions (25). Recent studies have reproduced and extended this work and showed that FTY720 treatment improved contextual fear memory (55) and increased proliferation of embryonic neural stem cells, hippocampal neurogenesis, and learning and memory abilities in adult mice (56).

In the present study, FTY720 administration increased expression of NPC1 and -2 at mRNA and protein levels in brain areas, including the hippocampus and cerebellum, and in livers of mice, in an SphK2-dependent manner. These increases were accompanied by a decrease in total liver cholesterol in WT, but not in SphK2-null mice. The large increases in NPC1 and -2 proteins in cerebellum and liver consequent to formation of FTY720-P are particularly intriguing, as they are the clinically relevant tissues in NPC disease. FTY720 and its conversion to FTY720-P also enhanced NPC1 and -2 expression and decreased cholesterol in murine NIH 3T3 fibroblasts, effects that were increased by overexpression of SphK2 and decreased by its downregulation. Consistent with our findings, it has been reported that treatment of primary macrophages with FTY720 increases NPC1, but not NPC2, and increases efflux of endosomal cholesterol to apolipoprotein A-I (57). However, the mechanism of up-regulation of NPC1 expression has not been elucidated. These effects of FTY720 are mainly related to its phosphorylation to FTY720-P in the nucleus. We also found that FTY720 treatment of several human fibroblast lines derived from patients expressing a heterogeneous set of mutations in the NPC1 gene showed increased NPC1, NPC2, and ABCA1 expression, accompanied by increased intracellular FTY720-P levels and decreased cellular cholesterol. NPC1 mutant fibroblasts from patients with disease of differing severities had responses similar to those with FTY720. FTY720 administration also significantly increased cholesterol esters in NPC1 mutant fibroblasts to the same level in WT fibroblasts. However, treatment with FTY720 or SAHA failed to decrease cholesterol levels in NPC1-null cells. These data further support the notion that the effects of FTY720 on cholesterol are related to the epigenetic increases in expression of NPC1, NPC2, and ABCA1 mediated by formation of FTY720-P, which inhibits class I HDACs (25). FTY720 treatment also ameliorated the GSL storage defects and restored transport of ganglioside GM1 to the Golgi apparatus. Taken together, these data indicate that elevation of NPC1 and -2 by FTY720 can correct the aberrant cholesterol and sphingolipid storage and trafficking observed in these NPC1 mutant fibroblasts.

We believe that FTY720 has several advantages over available HDAC inhibitors as a potential treatment for patients with NPC1 disease for several reasons: it is also an orally bioavailable drug; it has already been approved for human use for the treatment of multiple sclerosis with minor and manageable side effects (58); the action of FTY720 on HDACs is related to its phosphorylation by SphK2 in the nucleus, whereas none of the other HDAC inhibitors work in a similar manner (25); it regulates expression of only a limited number of genes (most related to cholesterol and sphingolipid metabolism), compared with other HDAC inhibitors that increase expression of thousands of genes (59); it has good pharmacokinetics and a long half-life; it can reduce de novo ceramide biosynthesis by inhibiting ceramide synthase (60, 61); it also has several other actions that may be beneficial for treatment of NPC disease, given that it is also an immunosuppressant that sequesters lymphocytes into lymph nodes by modulating S1P receptor 1, and reduces neural inflammation (58); and it is less toxic and accumulates in the brain and would therefore be able to treat the neurologic sequelae of NPC disease. Hence, our studies may pave the way for exploration of new treatments for this incurable disease. Determining whether FTY720 treatment of a mouse model of NPC disease carrying an Npc1 mutation can delay neurodegenerative and behavioral symptoms and extend life is an exciting future direction for research.

Glossary

2AA

anthranilic acid

AANSS

age-adjusted NIH neurologic severity score

ABCA1

ATP-binding cassette transporter A1

BSA

bovine serum albumin

CTB

cholera toxin B subunit

ER

endoplasmic reticulum

FBS

fetal bovine serum

FTY720-P

FTY720-phosphate

GSL

glycosphingolipid

HPBCD

2-hydroxypropyl-β-cyclodextrin

HDAC

histone deacetylase

HRP

horseradish peroxidase

LC-ESI-MS/MS

liquid chromatography-electrospray ionization tandem mass spectrometry

LDL

low-density lipoprotein

MEM

minimum essential medium

NPC

Niemann-Pick type C

qPCR

quantitative PCR

S1P

sphingosine-1-phosphate

S1PR

S1P receptor

SAHA

suberoylanilide hydroxamic acid

siRNA

small interfering RNA

SphK

sphingosine kinase

WT

wild-type

AUTHOR CONTRIBUTIONS

S. Spiegel conceived and coordinated the study and wrote the paper with assistance from J. Newton, and S. Milstien; C. Wassif, A. Cougnoux, and F. D. Porter provided NPC patient fibroblasts and technical assistance with cell culturing; J. Newton and N. C. Hait performed the experiments shown in Figs. 1 and 3–5; J. Newton and M. Maczis performed the experiments in Fig. 2; J. Newton performed the experiments in Figs. 6 and 7; M. Maceyka collected the images in Fig. 8B–D; A. Colaco, N. Platt, and F. M. Platt measured the GSLs; S. Milstien and F. M. Platt helped with data analysis; J. Newton, S. Milstien, and S. Spiegel contributed to the preparation of the figures; and all authors reviewed the results and approved the final version of the manuscript.