Niemann–Pick disease type C is a hereditary disorder that leads to progressive neurodegeneration and death during early childhood. More than 95% of patients have defects in the NPC1 gene, whereas a small subset of patients have defects in a second gene, NPC2 . But in both cases, cholesterol accumulates aberrantly in the lysosomes. Two reports in Science now bring us a step closer to understanding the processes behind both forms of this debilitating condition.

Yiannis Ioannou and colleagues investigated the function of the protein that is mutated in NPC1 disease by looking for conserved motifs in its sequence. They found that it contains six copies of a lipid-attachment site also found in members of the resistance-nodulation-division (RND) family of prokaryotic permeases. These proteins are involved in pumping hydrophobic compounds such as antibiotic drugs, detergents and fatty acids from the cytosol of Gram-negative bacteria.

Two members of the RND family — Escherichia coli AcrB and Pseudomonas aeruginosa MexD — show weak sequence homology to the NPC1 protein throughout their entire sequences. Moreover, NPC1 and MexD have similar membrane topology and secondary structures. So could human NPC1 be an RND permease? Ioannou and colleagues tested this by looking at accumulation of a fluorescent dye called acriflavine in normal and NPC1-deficient fibroblasts. They found that efflux of acriflavine from the endosomal/lysosomal system is blocked in the absence of NPC1, and that this efflux is an active process, requiring a proton-motive force.

Next the authors engineered E. coli to express human NPC1. Then, given that the defect in NPC1 disease is the accumulation of cholesterol, they wondered whether this might be a substrate for NPC1. Ioannou and colleagues could observe no build-up of cholesterol in NPC1-expressing E. coli cells, but they did see an accumulation of oleic acid indicating that, like its prokaryotic homologues, NPC1 might transport fatty acids across a membrane.

In the second paper, Naureckiene et al. describe the protein responsible for NPC2. While trying to characterize the lysosome proteome they identified a human protein called HE1. The pig homologue of this protein is known to bind cholesterol so, given its lysosomal location, the authors wondered whether HE1 might be involved in NPC2 disease. HE1 could not be detected by Western blotting in fibroblasts from two patients with the condition, and sequence analysis revealed that both patients had mutations in the HE1 gene.

Studies with a cholesterol-binding antibiotic, filipin, indicated the abnormal accumulation of cholesterol in fibroblasts from the NPC2 patients. But when Naureckiene and co-workers treated these fibroblasts with recombinant HE1, this build-up was reduced. The HE1-conditioned medium had no effect on cholesterol accumulation in NPC1-deficient fibroblasts, however, confirming that the defects in the two forms of NPC disease are different.

These results do not rule out the possibility that NPC1 and HE1 might interact, nor that they could act sequentially in a common metabolic pathway. They also do not tell us how mutations in either protein lead to the characteristic accumulation of cholesterol. The answers must await the next threads to emerge as NPC is unpicked, piece by piece, at the cellular level.