Little is known about NPC2. In particular, researchers are unsure if its ability to bind cholesterol is related to its role in the disease. Researchers at Matthew Scott's laboratory at Stanford University, California, set out to test this relationship by mutating the protein and then testing its ability to both bind the steroid and to correct the cholesterol storage problem in human NPC2 fibroblasts. They conclude that the two are inextricably linked.

By comparing NPC2 homologs from different vertebrates, first author Dennis Ko and colleagues identified four evolutionarily constrained regions that may be functionally important. Ko et al. then made 22 single amino acid replacements in these regions, and found that six of the mutants failed to disperse fibroblast intracellular cholesterol when added to culture medium. Three of these mutations, alanine for phenylalanine at position 66, phenylalanine for valine at 96, and alanine for tyrosine at 100, had less than seven percent of the activity of wild-type NPC2. Furthermore, these mutations reduced cholesterol binding below detectable levels.

In a complementary study, researchers in Ann Stock's lab at the University of Medicine and Dentistry of New Jersey, Piscataway, analyzed the structure of bovine NPC2 by x-ray crystallography. First author Natalia Friedland and colleagues found that the protein has three small hydrophobic cavities in the center of the molecule. Though these are too small to hold cholesterol, their overall shape mimics its three-dimensional structure, leading the authors to predict that this site "dilates to accommodate a cholesterol molecule."

Friedland et al. predict that a "gate" in NPC2 would first have to open for cholesterol to bind. This is interesting because the proposed gatekeepers happen to be tyrosine-100 and phenylalanine-66, two of the three amino acids Ko et al. found to be essential for NPC2 activity. As replacing these amino acids with the less bulky alanine abolishes cholesterol binding, Friedland et al. suggest that tyrosine-100 and phenylalanine-66 may play an active role in the binding process.

Questions still remain regarding the role of NPC2, perhaps the most basic being whether it acts alone or needs help to mobilize cholesterol. In this regard, it is worth noting that, on its own, the protein binds cholesterol quite slowly, leading Friedland et al. to speculate that additional cellular components might facilitate the conformational change necessary to open up the molecule and thus accelerate cholesterol binding.-Tom Fagan.

References:
Ko DC, Binkley J, Sidow A, Scott MP. The integrity of a cholesterol-binding pocket in Niemann-Pick C2 protein in necessary to control lysosome cholesterol levels. PNAS early online edition. 17-21 February 2003. Abstract

Friedland N, Liou H-L, Lobel P, Stock AM. Structure of a cholesterol-binding protein deficient in Niemann-Pick type C2 disease. PNAS early online edition. 17-21 February 2003. Abstract