Glycolipid transfer protein knockout disrupts vesicle trafficking to the plasma membrane.

The glycolipid transfer protein (GLTP) has been linked to many cellular processes aside from its best-known in vitro function as a lipid transport protein. It has been proposed to act as a sensor and regulator of glycosphingolipid homeostasis in cells. Furthermore, through its previously determined interaction with the endoplasmic reticulum membrane protein VAP-A (vesicle-associated membrane protein-associated protein A), GLTP may also be involved in facilitating vesicular transport in cells. In this study, we characterized the phenotype of CRISPR/Cas9-mediated GLTP KO HeLa cells. We showed that motility, three-dimensional growth, and cellular metabolism were all altered by GLTP knockout. Expression of a GLTP mutant incapable of binding VAP disrupted cell spheroid formation, indicating that the GLTP-VAP interaction is linked to cellular adhesion, cohesion, and three-dimensional growth. Most notably, we found evidence that GLTP, through its interaction with VAP-A, affects vesicular trafficking, marking the first cellular process discovered to be directly impacted by a change in GLTP expression.

Glucosylceramide acyl chain length is sensed by the glycolipid transfer protein.

The glycolipid transfer protein, GLTP, can be found in the cytoplasm, and it has a FFAT-like motif (two phenylalanines in an acidic tract) that targets it to the endoplasmic reticulum (ER). We have previously shown that GLTP can bind to a transmembrane ER protein, vesicle-associated membrane protein-associated protein A (VAP-A), which is involved in a wide range of ER functions. We have addressed the mechanisms that might regulate the association between GLTP and the VAP proteins by studying the capacity of GLTP to recognize different N-linked acyl chain species of glucosylceramide. We used surface plasmon resonance and a lipid transfer competition assay to show that GLTP prefers shorter N-linked fully saturated acyl chain glucosylceramides, such as C8, C12, and C16, whereas long C18, C20, and C24-glucosylceramides are all bound more weakly and transported more slowly than their shorter counterparts. Changes in the intrinsic GLTP tryptophan fluorescence blueshifts, also indicate a break-point between C16- and C18-glucosylceramide in the GLTP sensing ability. It has long been postulated that GLTP would be a sensor in the sphingolipid synthesis machinery, but how this mechanistically occurs has not been addressed before. It is unclear what proteins the GLTP VAP association would influence. Here we found that if GLTP has a bound GlcCer the association with VAP-A is weaker. We have also used a formula for identifying putative FFAT-domains, and we identified several potential VAP-interactors within the ceramide and sphingolipid synthesis pathways that could be candidates for regulation by GLTP.

Effects of cholesterol and saturated sphingolipids on acyl chain order in 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine bilayers--a comparative study with phase-selective fluorophores.

Saturated sphingolipids have high acyl chain order. Our aim was to study how palmitoylated sphingomyelin (PSM), ceramide (PCer), glucosyl (GlcPCer)-, and galactosylceramide (GalPCer) were able to order the bulk acyl chains of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), in comparison with cholesterol. For this reason, we used lipid probes which had preferred phases that were either the disordered phase (1-oleoyl-2-propionyl[DPH-sn-glycero-3-phosphcholine (18:1-DPH-PC) or the ordered phase (trans parinaric acid (tPA). DPH was also used, although it has no clear phase preference. We measured steady-state anisotropy (all probes) and performed fluorescence lifetime analysis (tPA) as a function of composition and temperature. At concentrations where the saturated sphingolipids were not aggregated into ordered domains (and 23 °C), they did not increase POPC acyl chain order as determined from 18:1-DPH-PC anisotropy. As expected, cholesterol increased the POPC acyl chain order linearly as a function of concentration (0-28 mol %). Since PCer already forms ordered domains below 5 mol % (at 23 °C), we measured the acyl chain ordering effect of PCer at 50 °C (0-13 mol %) and observed that PCer ordered POPC acyl chains as efficiently as cholesterol. We conclude that the bulk acyl chain order of POPC was not markedly affected in bilayers where disordered and ordered domains coexist.