Localization of the Lys, Asp, Glu, Leu tetrapeptide receptor to the Golgi complex and the intermediate compartment in mammalian cells

To test whether the observed shift of the KDEL receptor towards the trans side of the stack was a specific response to viral infections, or rather a general response to cellular stress, we cultured uninfected SA:48 cells under two different extremes of temperature, 20°C (2 h) or heat shock, 43°C (4 h). These cells were then prepared for immunocytochemistry and double labeled as before for the KDEL-R and the G protein tag. A quantitative analysis of these experiments gave the following results. After the 20°C incubation, 56% of the Golgi stack-associated labeling for the KDEL-R was detected on the trans side, while after the heat shock treatment, the corresponding value was 46 %. These results suggest that the shift of the KDEL receptor towards the TGN side of the stack may be a general response to conditions of cellular stress. A recent paper from the Pelham group has provided in vitro data showing that binding of KDEL ligands to permeabilized Golgi membranes is highest at acidic pH (optimal pH. '~5) . We therefore asked whether the localization of the KDEL receptor was affected by bafilomycin A1, a specific inhibitor of the vacuolar proton ATPases (Altendorf et al., 1989) . For this, the SA:48 HeLa ceils were filled with 16-nm gold-BSA (to mark late endosomes and Figure 9 . Double labeling of SLO treated L cells infected with the ts 045 VSV and switched to 20°C (plus cycloheximide) using anti-KDEL-R (KR; arrowheads, 10 nm gold) and anti-G protein spike (GP; small arrows, 5 nm gold). Note the extensive TGN elements (T) that label significantly for both the G protein and for the KDEL-R. The close packing of G protein into morphologically distinct quasicrystalline assays in the TGN at 20°C has been previously documented (see Griitiths et al., 1985 Griitiths et al., , 1989 . These assays (parallel bars), as well as the numerous buds that accumulate (/arge arrows), are diagnostic features of the TGN under this condition. The G protein labeling also extends throughout the stack although it is often lower on the cis side (note the stack on the lower left part of A). Some residual G is also present in the rough ER (R), and a small amount evidently leaked to the plasma membrane (P). Bars, 100 nm. lysosomes, after an overnight chase) and subsequently with 4-nm gold-BSA for 5 min (early endosomes) before fixation. For the last 8 h before fixation, the cells were treated with 500 nM bafilomycin A1. This treatment effectively neutralized low pH compartments since by light microscopy we observed the loss of acridine orange staining of endosomes/ lysosomes (results not shown). The endocytic compartmentmarked, bafilomycin-treated, SA:48 cells were then doublelabeled with anti G tail (6 nm gold) and anti-KDEL-R (10 nm gold). Under this condition, there was no significant labeling of any endocytic structure (not shown). When we quantitated the KDEL-R labeling over the Golgi stack, as above, there was no change in the distribution of the receptor (with 18% of the receptor on the trans side of the stack and 82 % on the cis). These data argue that bafilomycin has no effect on the distribution of the KDEL-R.