The pI and pXI assembly proteins serve separate and essential roles in filamentous phage assembly.

Three non-capsid, phage-encoded proteins, pI, pIV and pXI, are required for assembly of the filamentous bacteriophage at the envelope of Escherichia coli. pIV forms the outer membrane component of the assembly site, and pI and pXI are predicted to form the cytoplasmic membrane component. pXI is the result of an in-frame internal translational initiation event in gene I and is identical with the carboxyl-terminal third of pI in amino acid sequence, membrane localization and topology. The two proteins share a cytoplasmic domain predicted to be an amphipathic helix, a transmembrane domain, and a periplasmic domain. By mutating the initiation site for pXI, a phage was made that produced only pI and was shown to absolutely require functional plasmid-encoded pXI for growth. Further mutational analysis was done to examine the functional determinants of the amphipathic helix and periplasmic domains of the pI and pXI proteins. The results show that the amphipathic helix region is very important for pI function but not for pXI function. Mutational analysis of the periplasmic domains of pI and pXI implies that these domains also perform separate functions, and suggests that the interaction between pI and pIV in the periplasm is critical for assembly. The results are discussed with regard to the separate roles that the pI and pXI proteins play in the overall process of phage assembly.

The major coat protein of filamentous bacteriophage f1 specifically pairs in the bacterial cytoplasmic membrane.

Filamentous bacteriophage are long, thin single-stranded DNA viruses that infect male strains of Escherichia coli without killing the host. Each phage contains approximately 2700 copies of the major coat protein, pVIII, which exists as a 5.2 kDa cytoplasmic membrane protein prior to incorporation into phage. Studies from a number of groups analyzing the behavior of wild-type and mutant pVIII in detergents suggested that pVIII might pair under these conditions. In order to test whether pVIII molecules pair in vivo in the cytoplasmic membrane, four plasmidencoded pVIII variants were constructed in which specific residues in the transmembrane region were mutated to cysteine in an attempt to stabilize any pair via disulfide bridges. Variants A35C and I39C were unable to complement phage with an amber mutation in gene VIII. The I39C variant was unable to be packaged into phage particles even though it was inserted into the membrane. In the case of A35C, the inability to complement was not due to a packaging defect because the variant protein could be packaged into phage in the presence of wild-type pVIII. Western blot analysis of cytoplasmic membrane samples revealed that the A35C variant formed stable disulfide dimers in vivo. Expression of A35C interfered with wild-type phage infection, indicating that the assembly machinery may recognize the disulfide dimers in some non-productive way. The results indicate that pVIII may specifically pair along a particular face in the cytoplasmic membrane prior to assembly; however, these pairs must be able to be separated in order for normal assembly to occur.