Cholesterol Transport through Lysosome-Peroxisome Membrane Contacts: Concerns about the Unexpected Connection

Gaining a better insight into the mechanisms governing intracellular lipid transport is of fundamental importance to medical science. Recently, Song and coworkers reported in Cell the existence of dynamic membrane contact sites between lysosomes and peroxisomes that mediate cholesterol transport from the former to the latter organelle (Chu et al., 2015). In addition, the authors argue that these contact sites are formed by interaction of lysosomal synaptotagmin VII (SYT7) with phosphatidylinositol 4,5 bisphosphate (PI(4,5)P2) on the peroxisomal membrane and that cholesterol is accumulating in cultured cells from patients and mice with diverse peroxisomal disorders. The authors employed an elegant genome-wide RNAi screen in HeLa cells to detect genes that affect cholesterol trafficking from lysosomes to the plasma membrane. To validate the unexpected enrichment of various genes involved in peroxisome biogenesis and metabolism, they designed an organelle co-precipitation method to identify components of, and functionally characterize, the lysosome-peroxisome membrane contacts observed by advanced microscopy at the biochemical level. However, to our surprise, this assay was based on the affinity capturing of peroxisomes in HeLa cells expressing EGFPHis6-SKL on Ni-Sepharose. Indeed, given that SKL-tagged variants of EGFP are routinely used in the peroxisome field as bona fide marker proteins for the peroxisome lumen, EGFPHis6-SKL is physically shielded from the Ni-affinity matrix by the peroxisomal membrane and cannot be used to study direct interactions between the peroxisomal and lysosomal compartments. As such, all conclusions deriving from this particular experimental setup need to be interpreted with care, including the cholesterol transfer from lysosomes to peroxisomes via (lysosomal) Syt7-(peroxisomal)PI(4,5)P2 membrane contact sites. Also, we are very puzzled by the authors’ observations that a knockdown of PEX1, PEX3, PEX6, PEX10, or PEX26 results in fewer peroxisome-lysosome contact sites and the accumulation of cholesterol, but not in a cytosolic mislocalization of EGFP-His6-SKL. Of course, the latter finding may be explained by the fact that residual amounts of these peroxins are sufficient to maintain a functional peroxisomal protein import machinery. However, if this is indeed the case, it may well be that the in cellulo decrease in lysosome-peroxisome contact sites can be attributed to a reduction in peroxisome number, a well-known consequence of PEX gene disruption (Steinberg et al., 2006), rather than to altered peroxisome properties.