95% of NPC patients have a mutation at chromosome 18q11 (locus of NPC1 gene), with the mutations in the remaining patients in the NPC2 gene at chromosome 14q24.3 (Patterson et al, 2010). Mutations in NPC1 and NPC2 give a clinically identical phenotype, with the gene products serving non-redundant functions in the same cellular pathway (Sleat et al, 2004). NPC2 encodes a small (132 amino acids) soluble glycoprotein present in the lysosomal lumen (Babalola et al, 2007). NPC2 has been shown to bind cholesterol with a high affinity (Friedland et al, 2003) and is thought to shuttle cholesterol from inner lysosomal vesicles to the limiting membrane (Babalola et al, 2007).
The NPC1 gene is constitutively expressed in all mammalian cells, and encodes a large (1278 amino acids) integral transmembrane glycoprotein of the late endosomal/lysosomal membrane (Loftus et al, 1997). NPC1 has 13 Transmembrane domains, a small cytoplasmic tail and three loops present in the lysosomal lumen. NPC has partial homology (between residues 615-797) to the sterol-sensing domain (SSD) present in other membrane proteins. Over 200 mutations in the NPC1 gene have so far been identified in patients (Garver et al, 2007). Mutations are scattered throughout the gene, affecting all protein domains except the leucine zipper motif, located in a lumenal N-terminal domain loop (The NPC1 domain). A significant proportion of mutations are localised to a portion of the cysteine -rich lumenal loop (residues 927-958) (Vanier and Millat, 2003). The most common mutation in NPC1 is I106T, whose occurrence leads to the classic disease phenotype. The variant phenotype is attributed to P1007A or G992W, and these cells have a milder trafficking defect.
Get your grade
or your money back
using our Essay Writing Service!
The function of NPC1 is currently uncertain, with recent genetic, biochemical and evolutionary evidence indicating a role in sphingolipid recycling (Malathi et al, 2004).
NPC-/- and NPC+/- brains are characterised by the progressive death of Purkinje neurons (a class of neurons in the cerebellum). Experiments using a long-lived NPC+/- mouse model demonstrated progressive neurodegeneration, with Purkinje cell loss becoming significant (relative to NPC+/+) at around 104 weeks. This is accompanied by an increase in hyperphosphorylated tau protein, and a rise in activity of the Mitogen-Activated Protein Kinase (MAPK), the enzyme responsible for specific tau phosphorylation (Yu et al, 2005).
It is suggested that altered cholesterol metabolism in NPC cells may be the cause of neuronal cell death. Levels of cellular cholesterol are proportional to MAPK activity (and therefore the amount of hyperphosphorylated tau) (Sakamura et al, 2003), and NPC-/- have increased mitochondrial membrane cholesterol leading to mitochondrial dysfunction and a fall in cellular ATP levels (Yu et al, 2005). These two paths have been independently shown to lead to cell death, and may function synergistically in NPC-/- cells (Mizukami et al, 2004).
Purkinje Neurons death occurs selectively. In both global and conditional NPC-/- mice Purkinje cell death occurs in an anterior-to-posterior direction (Sarna et al, 2003, Elrick et al, 2010). This cell death is cell-autonomous, and does not depend on events in events in embryonic or early postnatal development. Purkinje cells in lobules IX and X are resistant to degeneration , despite lacking NPC1 and having increased filipin staining. Conditional NPC-/- mice suffer from ataxia, but do not lose weight or die early (Elrick et al, 2010). It is hypothesised that these NPC phenotypes may instead be due to defects in the hypothalamus or brain stem degeneration (Luan et al, 2008).
NPC in heterozygotes
Current Model for disease
Extracellular cholesterol is endocytosed and converted to the unesterified form in the lysosome. Re-esterification occurs in the endoplasmic reticulum (ER) (Brown et al, 1976). The trafficking defect in NPC leads to the accumulation of unesterified cholesterol in the acidic compartment (Kruth et al, 1986).
There is an increasing body of evidence against the importance of cholesterol in NPC pathogenesis. The reduction of liver cholesterol content by both pharmacological and genetic methods has no beneficial effect on CNS disease progression in NPC-/- mice (Erickson et al, 2000), with similar findings in a feline model of disease (Davies et al, 2000). Although the isolated NTD of NPC1 is able to bind cholesterol (Infante et al, 2008) this has not been demonstrated in the full-length protein - indeed, NPC1 expressed in E.coli was shown to be unable to transport cholesterol across membranes (Davies et al, 2000). Cholesterol efflux from the acidic compartment in NPC-/- cells has been demonstrated - in one such pathway MLN64 transports cholesterol to the inner mitochondrial membrane (IMM) (Charman et al, 2009).
Always on Time
Marked to Standard
Finally, cholesterol does not accumulate in the NPC brain (Salvioli et al,2004). However, it has been argued that CNS cholesterol accumulation occurs, but is masked by a net decrease, due to neuronal death and demyelination (Xie et al, 2000). GSL levels (particularly Glucosylceramide, Lactosylceramide and GM2/GM3) are elevated in the NPC-/- brain (Salvioli et al, 2004). The possibility that GSL accumulation is an upstream event in pathogenesis is cast into doubt by experiments in mice with non-functional β-1-4GalNAc transferase. These mice are unable to synthesize GM2 and other complex gangliosides, and therefore storage in the brain decreases. However, this absence of GSL storage results in no clinical benefit (Liu et al, 2000).
NPC1 is an evolutionarily ancient protein, predating the emergence of many components of the receptor-mediated sterol uptake pathway (Malathi et al, 2004). Deletion of the yeast NPC1 homologue NCR1, or mutations in the putative SSD, do not affect cholesterol metabolism. Expression of the yeast protein in mammalian cells allows wild-type cholesterol and GSL transport (Malathi et al, 2004). It was therefore suggested that aberrant cholesterol trafficking is a downstream event in NPC, and is not directly caused by non-functional NPC1. It was observed that NPC-/- cells are deficient in amine transport (Kaufmann and Krise, 2008). NPC1 is known to interact with a range of amines, including acriflavine (Davies et al, 2000), and can be bound and inhibited by the amine U18666A (Kaufmann and Krise, 2008). This prompted investigation into sphingosine, an 18-carbon amino-alcohol whose levels show a large proportional increase in all NPC-/- tissue (Lloyd-Evans and Platt, 2010). It is also elevated in NPC2 (Lloyd-Evans and Platt, unpublished observation).
Free cellular sphingosine is manufactured in the late endosome/lysosome. In the acidic compartment sphingolipids and GSLs are degraded to give ceramide (Kitatani et al, 2007). Ceramide is unable to leave the lysosome (Chatelut et al, 1998), so must first be converted to sphingosine (Riboni et al, 1999). Sphingosine is protonated (and is the only storage lipid to be so), so would require a transport protein for movement across membranes. Once released from the lysosome sphingosine can re-enter ceramide synthesis pathways or can be phosphorylated by sphingosine kinase, giving sphingosine-1-phosphate (S1P) (Taha et al, 2006). S1P is an important secondary messenger, having mitogenic and anti-apoptotic effects (Maceyka et al, 2007), and indirectly stimulates the expression of the cholesterol biosynthetic protein CYP17 (Ozbay et al, 2006).
Many LSDs have altered Endoplasmic Reticulum (ER) Ca2+ homeostasis (Ginzburg et al, 2004). The Ca2+ levels in ER of NPC-/- is normal, but lysosomal Ca2+ levels fall relative to wild-type (Lloyd-Evans et al, 2008). Ca2+ levels in the lysosome are approximate to those in the ER (~500μM - relative to ~100nM in the cytosol). Release from the lumen occurs via a NAADP-mediated channel (Galione and Churchill, 2002) and is required, along with the formation of a trans-SNARE protein complex, for acidic compartment fusion and endocytic trafficking (Luzio et al, 2005). Drug induced chelation of lysosomal calcium leads to the rapid downstream accumulation of GSLs and cholesterol, whereas an increase in cytosolic Ca2+ levels can rescue the NPC trafficking phenotype (Lloyd-Evans et al, 2008).
Sphingosine is the only lipid stored in NPC cells whose exogenous addition is able to induce the calcium homeostatic defect in NPC+/+ macrophages. It does so in a concentration dependent fashion, with the level of sphingosine required for phenotype induction equal to levels found in the NPC-/- brain. Addition of sphingosine also gives increased lysotracker and filipin staining (Lloyd-Evans et al, 2008). Sphingosine is known to inhibit Ca2+ transport (Pandol et al, 1994). Reduction of NPC-/- sphingosine levels via inhibition of serine palmitoyl transferase corrects the NPC phenotype, with calcium homeostasis re-established before a reduction in lipid storage occurs. On the basis of this evidence, sphingosine accumulation (occurring due a mutation in NPC1) is the initiating event in disease pathogenesis, and the consequent calcium phenotype leads to defects in endocytosis and the accumulation of secondary lipids such as cholesterol, GSLs and sphingomyelin. This order of events is supported by experiments in which NPC1 is inhibited via the Class II Amphiphile U18666A. Sphingosine accumulation begins 10 minutes after the addition of drug, far in advance of the Calcium defect or other lipid accumulation.
This Essay is
a Student's Work
This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.Examples of our work
Sphingosine levels may also have a more direct effect on trafficking. The accumulation of lipids in NPC is associated with the aggregated, insoluble form of the intermediary filament Vimentin (Walter et al, 2008). It is believed that the hypophosphorylated form of vimentin is able to sequester the Rab9 GTPase. Functional Rab9 protein is required for trafficking of LDL-derived cholesterol (Sarria et al, 1992). Upregulation of the protein can reduce cholesterol storage, with Rab9-/- cells have a NPC phenotype (Walter et al, 2003). Vimentin is specifically phosphorylated by protein kinase C (PKC) (Ivaska et al, 2002). PKC is regulated in response to sphingosine levels (Hannun and Bell, 1989). The addition of exogenous sphingosine to wild-type cells leads to the disappearance of soluble phosphorylated vimentin. Also, the treatment of NPC-/-cells with activators of PKC restores levels of soluble vimentin and reduces filipin staining (Walters et al, 2008).
Investigations from this lab have revealed that in the NPC-/- mouse cerebellum sphingosine storage precludes accumulation of GSLs, whilst levels of cholesterol are not significantly different to those in the wild-type brain (Bachtar et al, Unpublished data). Quantification of lipid levels in NPC+/- tissue
The cellular NPC+/- phenotype has been well documented. Experiments with NPC+/- human fibroblasts (HFs) have revealed filipin staining and cholesterol storage levels intermediate between NPC+/+ and NPC-/-phenotypes (Kruth et al, 1986) and have shown a partial cellular phenotype in the father of an NPC-/- patient (Ceuterick and Martin, 1986). The intermediate NPC+/-phenotype was confirmed using a feline model of disease, in which CNS abnormalities in a NPC+/- organism was first demonstrated (Brown et al, 1996). Progressive neurodegeneration has been observed in the NPC+/-mouse
To date, NPC therapies have focused on treating dysfunction that occurs downstream of the initiating disease event. Better understanding of disease pathogenesis will allow for development of more targeted therapies.
The NPC-/-mouse has a significant decrease in neurosteroid content (Griffin et al, 2004). Neurosteroids are found in the blood and brain, and control neuronal growth, survival and differentiation. The neurosteroid allopregnanolone is synthesised de novo from cholesterol via the action of 3α-hydroxysteroid dehydrogenase (3αHSD), the activity of which is decreases with age in the NPC-/- brain (Griffin et al, 2004), likely due to defects in cholesterol metabolism. Addition of exogenous allopregnanolone increase Purkinje cell survival, thereby delaying neurodegeneration and doubling lifespan.
The discovery of the lysosomal calcium defect (Lloyd-Evans et al, 2008) provided a new target for therapy. The Smooth Endoplasmic Reticulum ATPase (SERCA) antagonist circumin increases levels of cytosolic Ca2+ (without affecting lysosomal levels), and thereby corrects sphingolipid trafficking and lowers GSL storage. Circumin-treated mice show increased lifespan and reduced tremor/weight loss. However, circumin has no effect on sphingosine levels (Lloyd-Evans et al, 2008), strengthening the case for the importance of sphingosine in disease pathogenesis.
N-butyldeoxynojirimycin (NB-DNJ) is an inhibitor of glucosylceramide synthase, an essential enzyme in the early stages of GSL synthesis (Zervas et al, 2001). It is able to cross the blood-brain barrier, and in mice is seen to be able to correct lipid trafficking in NPC-/- B lymphocytes (but does not affect cholesterol metabolism) and delay disease progression (Lachmann et al, 2004). Functional improvement is greater with NB-DNJ than with N-butyldeoxygalactonojirmycin (NB-DGJ), a related compound (Elliot-Smith et al, 2008) Miglustat was shown to give an improvement in clinical indications of NPC in a randomised trial (Patterson et al, 2007), and a benefit to paediatric patients in a long-term trial (Patterson et al, 2010b).
NPC1 and Tuberculosis
Tuberculosis is a potentially deadly infectious disease caused, in humans, by Mycobacterium Tuberculosis (Russell et al, 2002). Around 2 billion people are infected. The majority of infections are latent, but can become active (Jasmer et al, 2005). Macrophages can internalize the Mycobacterium. However, fusion between the phagosome and lysosome is defective, thereby preventing Mycobacterial clearance (Hart et al, 1972). M.Tuberculosis possesses a complex-lipid rich coat (Kolattukudy et al, 1997), with the role of these lipids in immunomodulation having been well documented (Barnes et al 1992, Barrow et al 1993). In particular, Trehalose dimycolate (TDM) has been shown to induce granuloma formation (Russell et al, 2002).
It was shown that Mycobacterial lipids are actively trafficked from the mycobacterial vacuole to both the extracellular space and the endosomal system (Beatty et al, 2000). Macrophages infected with live Mycobacterium Bovis show increased filipin staining distal to the Mycobacterium vacuole. Infected cells also accumulated sphingosine, SM and GSLs, and had disrupted acidic compartment Ca2+ homeostasis. A mixture of Mycobacterial lipids was able to induce an identical phenotype in the absence of live bacteria. The similarities to the NPC cell phenotype raises the possibility that Mycobacteria may disrupt endosomal transport, and thereby avoid cellular clearance, via the targeted inactivation of the NPC1 protein using lipids shed from the cell wall.
We possessed purified versions of lipids found in the mixture used above. After identifying which is responsible for induction of the NPC phenotype we will investigate the susceptibility of NPC+/- macrophages to phenotype induction. If Mycobacterial lipids function via inhibition of NPC1 we should observe a greater phenotype induction (relative to wild-type) in macrophages possessing 50% functional NPC1 protein.