TP228 is a low-copy number, multidrug resistant plasmid that replicates in Escherichia coli and which is stabilized by an active partitioning system. The TP228 partition complex consists of three main components: the ParF and ParG proteins that assemble on the centromere site parH to form the segrosome complex. ParF (206 amino acids) is a polymerizing ATPase that plays an essential role in directing the faithful movement of plasmids to daughter cells at cell division. In this study, ParF was subjected to pentapeptide scanning mutagenesis which generated nine unique random pentapeptide insertions in the protein. Insertions at different positions in ParF identified regions in the protein that are crucial for segregation, for ParF self-association, and for the ParF-ParG interaction in vivo. Two of the mutant ParF proteins were entirely impaired in plasmid partitioning, ParF self-association, and interaction with ParG. Three mutants conferred severe reductions in, but did not entirely abolish, plasmid segregation, whereas the interactions with wild-type ParF and ParG were abrogated in these ParF mutants. A pentapeptide insertion between residues Lys151 and Glu152 did not affect ParF self-association but did impair both the ParF-ParG interaction and plasmid segregation which is a novel observation. Longer oligopeptide insertions between these positions exerted similar effects to those observed with the pentapeptide insertion: insertions up to 33 amino acids did not perturb the ParF-ParF interaction indicating that this region of the protein is tolerant of insertions with respect to ParF self-association. Substitution and deletion analysis of residues Lys151 and Glu152 were examined to assess further the importance in ParF-ParG interaction and plasmid partitioning of the alpha6 helix structural element that contains these residues. The impacts of K151A substitution and Delta151 deletion were similar to the effects produced by the pentapeptide insertion between Lys151 and Glu152. In contrast, the ParF-E152A protein conferred wild-type activity in ParF self-association and ParG interaction assays and as well as in partition assays. The ParFDelta152 mutant was impaired in plasmid segregation. In addition, pentapeptide insertions at positions 160 in the β7 β-strand element, 185 in the turn before the alpha8 helix, and 205 at the C-terminus of ParF did not produce any detectable change in protein function. However, insertions longer than five amino acids between positions Lys160-Ala161 and Ser185-Ser186 dramatically reduced plasmid partition activity. The pentapeptide mutant ParF205GVPLF was unaffected in partition activity and also self-associated and interacted with ParG. The role of the protein C-terminus in ParF-mediated plasmid segregation was assessed further by truncation mutations. ParFDelta206 that lacks the final residue in ParF and ParFDelta205-206 in which two amino acids were deleted did not perturb plasmid segregation, whereas deletion of three amino acids from the ParF C-terminus abolished accurate plasmid partitioning. Nevertheless, the ParFDelta204-206 protein was not perturbed in ParF self-association or in the ParF-ParG interaction indicating the involvement of the C-terminus of ParF in an unknown aspect of plasmid partitioning. Overall, pentapeptide scanning mutagenesis was a powerful mutagenesis strategy to dissect the organization of the TP228 segregation complex by identifying regions in ParF that are tolerant to insertions, defective in known functions of the protein, and disrupted for the ParF-ParG interaction. The data provide new insights into the structure and function of ParF which is a potential target for new antibacterial compounds that disrupt the segregation of multi-resistance plasmids.