Niemann–Pick type C (NPC) disease is a fatal complex neurodegenerative lysosomal storage disorder caused by genetic mutations in the proteins NPC1 (95% of patients) or NPC2 that decrease intracellular cholesterol trafficking, resulting in accumulation of unesterified cholesterol and sphingolipids in lysosomal storage organelles. Unfortunately, treatment options for NPC disease are still very limited, although miglustat, which inhibits glucosylceramide synthase, thus limiting ganglioside accumulation, has been approved for treatment of NPC disease. Here we discuss advances in the understanding of NPC1 and its functions, and several new strategies for interfering with cholesterol and sphingolipid accumulation in NPC1-null mice. We also describe several recent studies demonstrating that histone deacetylase inhibitors may correct cholesterol-storage defects in human NPC1 mutant fibroblasts by increasing expression of the low-transport-activity NPC1 mutant protein. These studies may lead to development of new therapeutic approaches for treatment of NPC disease.
Niemann–Pick disease type C (NPC) is a fatal, autosomally recessive, neurodegenerative lysosomal storage disease characterized by the accumulation of many lipids, particularly unesterified cholesterol and sphingolipids, in multiple organs [1-5]. The disease is caused by mutations in either NPC1 (95% of cases) or NPC2 (5%), and hundreds of disease-causing mutations have been found. Patients typically present in early childhood with ataxia and progressive impairment of motor and intellectual function, and death from complications of the disease occurs in the teenage years or early adulthood . Late-occurring cases typically have a less dramatic onset of symptoms and have begun to be diagnosed in the adult population with cognitive deficits .
Both the NPC1 and NPC2 proteins are lysosomally localized, with NPC1 being an integral membrane protein and NPC2 being intraluminal. NPC1 is a large polytopic protein (1278 amino acids) containing 13 putative transmembrane segments that span the entire lysosome membrane. Its cholesterol-binding site faces the lumen and is located in the N–terminal region . There are two large lumenal loops in the tail region, the latter of which is cysteine-rich and is the site of the most common mutations that cause NPC1 disease . Mature human NPC2 has 132 amino acids, with an additional signal peptide of 19 amino acids. This small lysosomal glycoprotein is soluble, ubiquitously expressed, and specifically binds unesterified cholesterol with sub-micromolar affinity . NPC2 acts in concert with NPC1 and plays an important role in the egress of cholesterol from the endosomal/lysosomal compartment. NPC1 and NPC2 function as a cellular ‘tag team duo’ to mobilize cholesterol within the multivesicular environment of the late endosome and to effect its egress through the limiting bilayer [9-12]. NPC2 binds unesterified cholesterol that has been released from low-density lipoprotein in the lumen of the late endosomes/lysosomes , and transfers it to the cholesterol-binding pocket of the N–terminal domain of NPC1, which then either directly or indirectly facilitates its transport from the late endosome/lysosome compartment [10-12]. A recent study demonstrated that oxysterols may act as pharmacological chaperones and bind to a second site on NPC1 that leads to correction of the localization/maturation defect of the mutant protein . Intriguingly, NPC1 is a receptor for filoviruses, such as Ebola , and is the first known viral receptor that recognizes its ligand within an intracellular compartment and not at the plasma membrane.
Lipid accumulation in NPC disease
Biochemically, NPC is characterized by the accumulation in late endosomes of many types of lipids, including unesterified cholesterol, sphingomyelins, various glycosphingolipids, and the sphingolipid metabolite sphingosine [4, 5, 16]. NPC patients accumulate lipids due to the dramatic block in vesicular traffic of late endosomes to lysosomes [3, 17]. The accumulation of so many different lipids and the dysregulation of their metabolism raise the question of which lipid(s), through their excess or absence, are responsible for the NPC phenotypes. This information is vital to the development of targeted therapies. The obvious suspect is cholesterol. Indeed, the cholesterol chelator 2–hydroxypropyl-β–cyclodextrin overcomes the cholesterol-transport defect in cells and ameliorates disease progression in mice [11, 18-21] and cats . However, disappointingly, even though statins and cholesterol restriction can lower cholesterol levels significantly, they do not alter disease progression in humans or in NPC mutant mice or cats [23-25].
Other studies have suggested that the sphingolipid backbone sphingosine may also be a causative agent of NPC. First, exogenous sphinganine induces lysosomal cholesterol accumulation in cells . Second, in a drug-induced NPC disease cellular model, sphingosine storage in the acidic compartment led to calcium depletion in these organelles, which then resulted in cholesterol, sphingomyelin and glycosphingolipid storage . This led to the suggestion that NPC1 is a sphingosine transporter that, when non-functional, leads to defective endocytosis and lipid storage, and that sphingosine storage is an initiating factor in NPC1 disease pathogenesis that causes altered calcium homeostasis, leading to secondary storage of sphingolipids and cholesterol . Further studies are required to verify the function of accumulated sphingosine in NPC disease.
Potential treatments in development
Another class of lipids that has been implicated in NPC disease is glycosphingolipids [28, 29]. These are a widely diverse set of sphingolipids, including gangliosides, that are abundantly expressed in brain and consist of ceramides with sugar chains attached. The best case for their involvement comes from the observation that inhibiting glucosylceramide synthase reduced pathology and increased lifespan in animal models [28, 29]. Indeed, an inhibitor known as miglustat (N–butyldeoxynojirimycin) prolongs survival in NPC1 mouse and cat models [22, 28, 30]. Miglustat delayed disease progression in patients, and is now a clinically approved drug for the treatment of NPC [31, 32]. There are a number of lipid storage diseases associated with servere neuropathology and the inability to degrade individual glycosphingolipids or classes of glycosphingolipids, although which glycosphingolipids are responsible in NPC is still unclear. When NPC mutant mice were crossed with mice that are unable to make complex gangliosides, the central nervous system, but not the peripheral organs, showed a marked improvement in storage pathology . However, no clinical improvement was observed, suggesting that it is either the less complex gangliosides that are responsible for the pathology or that storage of complex gangliosides and cholesterol may not be the sole generators of clinical disease. Unfortunately, miglustat only slows NPC progression, so the search is on for more effective therapies.
Another possibility is the cholesterol chelator 2–hydroxypropyl-β–cyclodextrin. While questions remain about its exact mechanism of action [11, 34], several studies have reported that it reduces the severity of symptoms and increases lifespan in NPC1 null mice and cats [11, 18-22]. These encouraging preclinical studies led to the treatment of two human patients with cyclodextrin, and, although there were some improvements in hepatosplenomegaly and central nervous system dysfunction, especially during the first 6 months of treatment, neurological deficits remained , probably due to its inability to cross the blood–brain barrier. It was suggested that drug delivery into the central nervous system must be enhanced for more effective treatment of NPC patients with cyclodextrin , as achieved in cats with NPC disease . A phase I clinical trial of cyclodextrin treatment for NPC at the US National Institutes of Health has been proposed that utilizes cyclodextrin immobilized in Ommaya reservoirs that are surgically implanted under the scalp in an attempt to overcome the problem of delivery to the central nervous system (http://clinicalstudies.info.nih.gov/cgi/wais/bold032001.pI?).
Other studies have examined the role of agonists for liver X receptor, a member of the nuclear receptor family of transcription factors, as these are up-regulated in response to excess cholesterol and promote its transport across the plasma membrane . NPC-null mice treated with the liver X receptor agonist T1317 showed increased liver function but only a modest increase in lifespan , although agonists such as T1317 may be useful in combination therapies, especially with cyclodextrin .
Many NPC mutations lead to poorly functional proteins. One such mutation is NPC1I1061T, which accounts for up to 20% of all disease alleles . This mutation encodes a functional protein that fails to exit the endoplasmic reticulum and is targeted as mis-folded and degraded . More recently, it was shown that oxysterols, such as 25–hydroxycholesterol, act as molecular chaperones that diffuse into the endoplasmic reticulum, where they bind to the mutant folding intermediate and induce it to complete the correct folding process . This results in correction of both the NPC1 protein localization/maturation defect and the intracellular cholesterol accumulation in fibroblasts derived from patients expressing the NPC1I1061T mutation . It was also discovered that NPC1 contains a second sterol-binding site . These findings have important implications for understanding of the mechanism of NPC1-mediated cholesterol efflux from endo-lysosomes. Importantly, this also suggests that screening chemical libraries for compounds with better pharmacological chaperone activity that bind to this second site may be a worthwhile approach for developing effective treatments of NPC disease.
Potential usefulness of HDAC inhibitors
Remarkably, two recent studies showed that histone deacetylase (HDAC) inhibitors correct cholesterol storage defects in human NPC1 mutant cells by increasing expression of the low-activity mutant NPC1 protein [39, 40] (Fig. 1). HDACs remove acetyl groups from histones and play a key role in gene regulation, and HDAC inhibitors have long been used in psychiatry and various brain disorders and are being investigated as possible treatments for several neurological diseases [41-43]. The large HDAC family has been grouped into four classes: the ‘classical’ classes I, II and IV have 11 members, whereas class III (the sirtuins) has seven members. These classes differ in structure, enzymatic activity, localization and expression pattern [42, 44, 45]. Studies of mice lacking HDAC genes in various tissues have revealed highly specific functions for individual HDAC isoforms during development and adulthood [42, 44, 45].
We have previously shown that the sphingolipid metabolite sphingosine-1–phosphate (S1P) is an endogenous inhibitor of class I HDAC1 and HDAC2 . Although most of the known actions of S1P, a potent bioactive mediator formed by the sphingosine kinases SphK1 and SphK2, are mediated by five specific G protein-coupled receptors, termed S1PR1–5 , our work showed that HDAC1/2 are direct intracellular targets of S1P produced by nuclear SphK2 and link sphingolipid metabolism to epigenetic regulation of gene expression . This was the first indication that HDACs act as metabolic sensors, converting changes in sphingolipid metabolism into patterns of gene expression.
Treatment of human fibroblasts expressing the NPC1I1061T mutant (which has much lower activity than wild-type NPC1) using HDAC inhibitors increased expression of the mutant NPC1 to high enough levels to correct the defect in cholesterol efflux from endosomes and lysosomes . This is especially encouraging as this is the most common mutation in NPC1 patients. Moreover, based on extensive screening, it has been shown that class I HDAC inhibitors, particularly, LBH589 (panobinostat), an orally available HDAC inhibitor that crosses the blood–brain barrier and is currently in phase III clinical trials, are most effective in correcting the phenotype. LBH589 also enhanced NPC1 protein expression, reduced proteolytic processing of sterol regulatory element-binding protein 2 (SREBP2), and increased cholesterol esterification, consistent with improved cholesterol delivery to the endoplasmic reticulum .
Interestingly, a genome-wide, conditional synthetic lethality screen in yeast identified HDAC inhibition as a correction for cholesterol and sphingolipid transport defects in human NPC disease . This study also showed that 11 HDAC genes are up-regulated in fibroblasts from patients with NPC disease, and that the HDAC inhibitor suberoylanilide hydroxamic acid (SAHA/vorinostat) reversed the dysregulation of most of these genes and ameliorated lysosomal accumulation of both cholesterol and sphingolipids and the defective esterification of low-density lipoprotein-derived cholesterol . The fact that more efficient rescue of cholesterol accumulation was observed with global HDAC inhibitors than with isoform-specific HDAC inhibitors suggests involvement of several class I HDACs, which is consistent with the up-regulation of multiple HDAC genes observed in human NPC mutant cells . It is tempting to speculate that, due to accumulation of sphingosine in the lysosome of NPC mutants, less sphingosine may be available for phosphorylation to S1P in the nucleus, resulting in enhanced activity of HDAC1 and HDAC2 . Therefore, it is of great interest to examine mechanisms of action of HDAC inhibitors and their potential as therapies in animal models of NPC.
Biomarkers for diagnosis and treatment of NPC1 disease
One obstacle for delivery of effective treatment for NPC disease has been the lack of non-invasive diagnostic tests that can distinguish NPC from other childhood diseases. This is critical, as early intervention is most beneficial for treatment success. As a first step in this direction, a mass spectrometry-based search was performed for blood markers of NPC disease. Several cholesterol oxidation products were markedly elevated in the plasma of NPC1 subjects, but not in other neurodegenerative or lysosomal storage diseases, and correlated with the severity and age of onset of disease . Moreover, these plasma oxysterols decreased in response to therapeutic intervention in the NPC1 feline model. These biomarkers are currently being validated in a clinical trial. Confirmation of the robustness of these blood-based biomarkers may revolutionize the diagnosis of NPC disease and be useful for measuring responses to therapy.
Conclusions and future perspectives
The relationship between cholesterol and sphingolipid accumulation and the complex coordination between their metabolism remain important issues that need to be better understood. The future challenge is to translate the advances made in understanding the biochemical actions of NPC1 and NPC2 into effective treatment for NPC disease.
This work was supported by US National Institutes of Health grant R37GM043880 to S.S.