Who's in, who's out? Re‐evaluation of lipid raft residents

Lipid rafts, membrane microdomains enriched with (glyco)sphingolipids, cholesterol, and select proteins, act as cellular signalosomes. Various methods have been used to separate lipid rafts from bulk (non‐raft) membranes, but most often, non‐ionic detergent Triton X‐100 has been used in their isolation. However, Triton X‐100 is a reported disruptor of lipid rafts. Histological evidence confirmed raft disruption by Triton X‐100, but remarkably revealed raft stability to treatment with a related polyethylene oxide detergent, Brij O20. We report isolation of detergent‐resistant membranes from mouse brain using Brij O20 and its use to determine the distribution of major mammalian brain gangliosides, GM1, GD1a, GD1b and GT1b. A different distribution of gangliosides—classically used as a raft marker—was discovered using Brij O20 versus Triton X‐100. Immunohistochemistry and imaging mass spectrometry confirm the results. Use of Brij O20 results in a distinctive membrane distribution of gangliosides that is not all lipid raft associated, but depends on the ganglioside structure. This is the first report of a significant proportion of gangliosides outside raft domains. We also determined the distribution of proteins functionally related to neuroplasticity and known to be affected by ganglioside environment, glutamate receptor subunit 2, amyloid precursor protein and neuroplastin and report the lipid raft populations of these proteins in mouse brain tissue. This work will enable more accurate lipid raft analysis with respect to glycosphingolipid and membrane protein composition and lead to improved resolution of lipid–protein interactions within biological membranes. Using a data‐driven partitioning strategy, we report that a significant proportion of gangliosides, long believed to be lipid raft resident, reside outside raft domains. Immunohistochemistry and imaging mass spectrometry revealed that the distribution of gangliosides remains largely unchanged by treatment with Brij O20 detergent compared to the widely used Triton X‐100, which induces ganglioside redistribution. Lipid raft analyses of Brij O20 brain extracts revealed that gangliosides are not all lipid raft depending on the ganglioside structures. Likewise, different neuroplasticity protein raft distributions were revealed. This work enables more accurate lipid raft analyses for resolution of lipid–protein interactions in biological membranes.