Sphingomyelin homeostasis is required to form functional enzymatic domains at the trans-Golgi network.
Bottom Line: Do lipids such as sphingomyelin (SM) that are known to assemble into specific membrane domains play a role in the organization and function of transmembrane proteins?We found that TGN46, which cycles between the TGN and the plasma membrane, was not sialylated by a sialyltransferase at the TGN and that this enzyme and its substrate TGN46 could not physically interact with each other.Our results suggest that SM organizes transmembrane proteins into functional enzymatic domains at the TGN.
Affiliation: Cell and Developmental Biology Programme, Centre for Genomic Regulation, 08003 Barcelona, Spain Universitat Pompeu Fabra, 08002 Barcelona, Spain.Show MeSH
Related in: MedlinePlus
License 1 - License 2
Mentions: Our results thus far show that treatment of HeLa cells with d-cer-C6 and its conversion to C6-SM caused in TGN46 a sialylation defect, a late glycosylation step occurring at the trans-Golgi complex and the TGN. However, a fraction of this immature TGN46 was transported to the cell surface (Fig. 4). Interestingly, a GFP chimera of ST (ST-GFP), an enzyme responsible for protein sialylation, was physically separated from TGN46 after d-cer-C6 treatment (Fig. 1 C). These results raise the question whether the observed glycosylation defects are caused by a physical segregation of the glycosylating enzymes from their substrates. To test whether TGN46 and ST can be mutually accessible in d-cer-C6–treated cells, we used a rapamycin-mediated trapping assay (Pecot and Malhotra, 2004). Upon binding of the small molecule rapamycin to the FK506-binding protein (FKBP; Wiederrecht et al., 1991), the FKBP–rapamycin complex gains a high affinity for the FKBP–rapamycin-binding (FRB) domain of the FKBP–rapamycin-associated protein (Brown et al., 1994; Sabatini et al., 1994). We generated a construct by inserting an FRB domain in the lumenal side of C-terminally GFP-tagged TGN46 (TGN46-FRB-GFP) and the parallel construct by inserting FKBP in the lumenal side of the Golgi localization domain of ST with a C-terminal RFP tag (ST-FKBP-RFP; Fig. 5 A). When expressed in HeLa cells, both constructs localize to the trans-Golgi cisternae/TGN as observed by fluorescence microscopy, confirming that the added domains do not apparently alter the localization of these proteins (Fig. S2 A). Moreover, TGN46-FRB-GFP localization overlaps with endogenous TGN46 (Fig. S2 B). In addition, TGN46-FRB-GFP localized to numerous punctae, which are in fact transport carriers of the TGN to the cell surface (CARTS) because they colocalize with the CARTS-specific cargo mRFP–pancreatic adenocarcinoma up-regulated factor (PAUF; Fig. S2 C; Wakana et al., 2012) but not with proteins specific for the ER exit sites, endosomes, or lysosomes (Fig. S2, D–H). HeLa cells expressing TGN46-FRB-GFP and ST-FKBP-RFP were treated with 20 µM l-cer-C6 or 20 µM d-cer-C6 for 4 h, after which 500 nM rapamycin or DMSO was added and incubation was continued for 2 h in presence of 100 µM cycloheximide to block protein synthesis. In l-cer-C6–treated cells, treatment with rapamycin increased the level of colocalization of the two proteins compared with DMSO-treated cells, as measured by the Pearson’s correlation coefficient between the two channels (Fig. 5, B and C). However, in d-cer-C6–treated cells, rapamycin treatment did not induce a coalescence of the two proteins in the same domains and the Pearson’s correlation coefficient remained unaltered compared with DMSO-treated cells (Fig. 5, B and C). To corroborate that treatment with d-cer-C6 by itself does not alter the trapping capacity of rapamycin, HeLa cells expressing TGN46-FRB-GFP and ST-FKBP-RFP were incubated with 500 nM rapamycin or DMSO for 2 h, after which the incubation was continued for 4 h in the presence of 20 µM l-cer-C6 or 20 µM d-cer-C6 and 100 µM cycloheximide, before fixation for fluorescence microscopy. Pretreatment with rapamycin, but not with DMSO, inhibited the lateral segregation of TGN46-FRB-GFP from ST-FKBP-RFP because these two proteins were present in the same membrane compartments (Fig. 5, D and E). Altogether, these results indicate that TGN46-FRB-GFP and ST-FKBP-RFP are not mutually accessible after d-cer-C6 treatment, suggesting that a physical separation is the cause for the observed sialylation defect in endogenous TGN46.
Affiliation: Cell and Developmental Biology Programme, Centre for Genomic Regulation, 08003 Barcelona, Spain Universitat Pompeu Fabra, 08002 Barcelona, Spain.