Mapping the Golgi retention signal in the G1 membrane glycoprotein of Uukuniemi virus, a member of the Bunyaviridae family

Sammanfattning: Newly synthesized secretory and integral membrane proteins destined for the plasma membrane are transported along the exocytic pathway from the endoplasmic reticulum (ER) via the Golgi complex (GC) to the cell surface. Some of the transported proteins contain information, which target them to their correct intracellular locations. Enveloped viruses have been useful tools for almost three decades to study the mechanisms of protein transport and sorting along the exocytic pathway. The simple structure of viruses force them to rely on host cell mechanisms. This includes the synthesis, processing, modification and transport o their membrane proteins. Enveloped viruses acquire their lipoprotein coat by budding through one of the cellular membranes. Some viruses bud at the plasma membrane whereas others mature at intracellular membranes. The site of budding is thought to be determined by the accumulation of viral membrane glycoproteins in the budding compartment. Such viral glycoproteins are likely to contain signals that retain them in the budding compartment. As a model system to study retention of proteins in the GC, we are using the viral heterodimeric spike glycoproteins G1 and G2 of Uukuniemi virus, a phlebovirus within the Bunyaviridae family. G1 and G2 are synthesized as a precursor protein (p110), which we have confirmed is cotranslationally cleaved immediately after the signal sequence of G2. We present evidence that the signal sequence of G2 remains attached to the C-terminus of G1. The glycoproteins are inserted in the membrane as type I membrane proteins with their C-terminal tails exposed in the cytoplasm. Using the T7 RNA polymerase-driven vaccinia virus expression system we found that G1 expressed alone was targeted to and retained in the GC, whereas G2 was dependent on G1 to exit the ER. Thus, we conclude that the signal for targeting the G1-G2 heterodimer to the GC resides in G1. We have constructed various chimeric proteins by replacing different domains of G1 with the corresponding domains of vesicular stomatitis virus (VSV) G protein, chicken lysozyme, or CD4, to locate the domain in G1 harboring the Golgi retention signal. In these experiments, the chimeras were expressed by using the Semliki Forest virus system. The results indicated that neither the ectodomain nor the transmembrane domain (TMD) of G1 is essential for Golgi localization. Instead, the cytoplasmic tail of G1 is both necessary and sufficient for Golgi retention. This was supported by the fact that the green fluorescent protein (GFP) could be targeted to the GC by attaching the G1 tail to either the N- or the C- terminus of GFP, a protein normally dispersed throughout the cytoplasm. To map the retention signal in more detail, we constructed CD4 chimeras with progressive C-terminal deletions of the G1 tail. The results indicated that the Golgi retention signal is located approximately between residues 10 to 50 counting from the TMD border. In this short region, two cysteines were found to be palmitylated. Replacement of both cysteines with alanins did not influence Golgi retention. To further narrow down the region harboring the Golgi retention signal, we expressed the cytoplasmic tail of G1 as a c-myc epitope-tagged 81 -residue peptide. The peptide was targeted to the GC but was also found associated with vacuolar structures, the identity of which remained unclear. Deletion of the 10 most N- terminal residues abolished the vacuolar staining. Progressive C-terminal deletions resulted in the detection of the shortest peptide, spanning from residues 10 to 40, still capable of becoming targeted to the GC. Additional deletions and alanine replacements revealed the importance of residues 10 to 15 and 35 to 40. The mechanism by which the G1 tail is able to retain the G1-G2 heterodimeric complex in the GC is still unclear. Different alternatives include recycling between the Golgi-cisternae, interactions between the spike protein heterodimers forming aggregates too large to be included in transport vesicles, or retention by binding to resident Golgi proteins or lipids.