abstract
-
The purpose of this study was to investigate the assembly of membrane proteins by examining the topological organization and interactions of the membrane proteins G and M of vesicular stomatitis virus (VSV).
Exhaustive proteolytic digestion of VSV resulted in the complete degradation of G protein, however, a small peptide fragment having an apparent molecular weight of approximately 9,000 D remained associated with the viral membrane. Characterization of this membrane embedded frargment by tryptic peptide analysis and amino acid sequence determination demonstrated that it was derived from the COOH-terminal end of G protein. Furthermore, comparison of the partial amino acid sequence of this peptide fragment with the predicted amino acid sequence of G protein indicated that this peptide contains an uninterrupted hydrophobic domain of sufficient length to span the viral envelope. Thus, the G protein of VSV is anchored in the viral membrane by a hydrophobic domain located near the COOH-terminus.
In addition to the COOH-terminus, the NHZ-terminus of G protein was also shown to be protected from proteolytic attack by the integrity of the viral envelope. This may be related to the tertiary structure of G protein that is imposed by the viral membrane or the fact that the NHz-terminus may be in close proximity to the surface of the membrane and thus protected from proteolytic attack.
It has recently been reported that the VSV G protein contains tightly bound fatty acid residues. Proteolytic digestion of VSV labeled with [³H]-palmitic acid demonstrated that all the fatty acid residues present in G protein are localized exclusively in the membrane interacting domain. Thus, the lipophilic fatty acids in conjunction with the hydrophobic domain may play an important role in the interaction of G protein with the viral envelope.
VSV grown in the presence of w-[9-³H] diazirinophenoxy nonanoate resulted in the biosynthetic incorporation of this photoreactive fatty acid into the viral phospholipids as well as into the membrane anchoring domain of G protein. Photolysis of the virus resulted in extensive phospholipid crosslinking to the G protein but not to the M protein. This confirms that the COOH-terminal region of G protein is in intimate contact with the hydrophobic core of the lipid bilayer and directly demonstrates that M protein does not penetrate the viral membrane to a significant extent. In addition, a new product was obtained following photolysis and identified as a G-G dimer on the basis of its molecular weight and immunoreactivity. This product arose presumably from protein crosslinking mediated by the photoreactive fatty acid attached to G protein. Thus, the biosynthetic incorporation of this photoreactive fatty acid makes it possible to not only identify integral membrane membrane proteins but also to make photoaffinity probes of membrane proteins which are normally fatty acid acylated.
The nature of the association of M protein with membranes was examined by reconstituting the purified protein into artificial phospholipid vesicles. The M protein was shown to have a strong affinity for such vesicles and the association, once made, could not be disrupted by salt treatment. This suggests that the interaction was at least partly hydrophobic in nature. The nucleocapsid protein N was shown to have no affinity for artificial lipid vesicles, however, it could associate with vesicles in the presence of M protein. The same results were obtained when in vitro synthesized proteins were used for reconstitution. These results support the concept that the M protein is involved in virus maturation through its ability to interact with both the plasma membrane and the ribonucleoprotein core containing N protein.
The maturation of membrane proteins is thought to be directed by discrete polypeptide domains present in the protein which are recognized and decoded by specialized mechanisms of the cell. As an approach to study the function of these various polypeptide domains, a hybrid gene was constructed from plasmids containing cDNA copies of the complete coding sequences of the G and M mRNAs. A chimeric gene, which contained the signal sequence coding region of G protein fused to the bulk of the coding sequence specific for M, was constructed and placed into an expression vector. Introduction of this hybrid gene into mammalian cells by DNA mediated gene transfer resulted in the synthesis of a 36,000 D polypeptide immunoreactive with anti-M antibody. This plasmid should thus be useful for the functional examination of the signal sequence of G protein. As well, it should be useful for examining polypeptide domains which are involved in glycosylation since M protein contains a cryptic glycosylation target site.
The studies reponed in this thesis have been presented, in part, in the following publications.
1. Ghosh, H.P., Capone, J., Irving, R., Kotwal, C., Hofmann, T., Levine, G., Rachubinski, R., Shore, G., and Bergeron, J. (1981) Viral Membrane Proteins: Assembly and Structure. In Replication of Negative Strand Viruses. D.H. Sishop and R.W. Campans ed. p. 655 Elsivier North Holland.
2. Capone, J., Toneguzzo, F., and Ghosh, H.P. (1982) Synthesis and Assembly of Membrane Glycoproteins: Membrane Anchoring COOH-Terminal Domain of Vesicular Stomatitis Virus Envelope Glycoprotein G contains Fatty Acids. J. Biol. Chem. 257, 16
3. Leblanc, P., Capone, J., and Gerber, G.E. (1982) Synthesis and Biosynthetic Utilization of Radioactive Photoreactive Fatty Acids. J. BioI. Chem. 257. 14586
4. Capone, J., Leblanc, P., Gerber, G.E., and Ghosh, H.P. (1983) Localization of Membrane Proteins by the use of a Photoreactive Fatty Acid Incorporated in vivo into Vesicular Stomatitis Virus. J. Biol. Chem. 258, 1395.
5. Kotwal, G., Capone, J., Irving, R., Toneguzzo, F., Bilan, P., Rhee, S., Hofmann, T., and Ghosh, H.P. (1983) Viral Membrane Glycoproteins: Comparison of the Amino Terminal Amino Acid Sequences of the Precursor and Mature Glycoproteins from different Serotypes of Vesicular Stomatitis Virus. Virology (in press).