Structure and function of conjugative pili: monoclonal antibodies as probes for structural variants of F pili

Abstract

The lac-tra operon fusion plasmid pTG801 contains the known F plasmid DNA transfer (tra) genes required by Escherichia coli to elaborate functional F pili (T. Grossman and P. M. Silverman, J. Bacteriol. 171:650-656, 1989). Here, we show that these pili are actually structural variants of normal F pili and that the F plasmid must contain additional genes that affect pilus structure and function. We confirmed a previous report that two monoclonal antibodies that recognize epitopes at and near the amino terminus of F pilin do not decorate the sides of normal F pili, as determined by immunogold electron microscopy. However, both antibodies laterally decorated pTG801 pili. The epitope for one of the antibodies has been shown to include the amino-terminal acetyl group of F pilin, which must therefore also be present on pTG801 pilin. Normal antibody staining was restored to pTG801 pili when cells contained, in addition to pTG801, the compatible plasmid pRS31, which must therefore include at least one gene affecting F-pilus structure. One candidate, traD, was excluded as the sole such gene, since traD+ derivatives of a pTG801 strain still elaborated pili that could be laterally decorated with antibody. Moreover, although traD alone restored RNA bacteriophage R17 infectivity to pTG801 cells, as expected, it did not mimic pRS31 in restoring to pTG801 pili other characteristics of normal F pili. We conclude that pRS31 contains as yet uncharacterized genes required for elaboration of structurally normal F pili. Finally, we identified vesicular material, especially abundant in cultures of pTG801 transformants, that stained heavily with the anti-F-pilin monoclonal antibodies. This material may reflect the inner membrane pool of F pilin.

The Escherichia coli K-12 F plasmid gene traX is required for acetylation of F pilin

“…Since the pilin polypeptide expressed by phage EDX134 reacted poorly with polyclonal anti-pilus serum, Laine et al (23) suggested that acetylation of pilin subunits might require traG, the only F locus known to affect F piliation that was not expressed by the phage. Subsequent electron microscope analyses reported by Grossman et al (12) supported this suggestion, since F-pilus filaments expressed by pTG801 hosts appeared to react with both JEL92 and JEL93. As this plasmid expresses a tra segment extending from the PstI site in traY to the EcoRI site in traG (13), its products should differ from EDX134 products only in including the TraG polypeptide required for piliation.…”

Towards a structural biology of bacterial conjugation

“…Additionally, filaments observed in bacterial cultures, without purification, and composed of acetylated subunits tend to remain isolated, with little tendency to aggregate, whereas those composed of unacetylated subunits form massive aggregates (Grossman and Silverman, 1989). Finally, while cells expressing unacetylated F-pilin are sensitive to donorspecific RNA bacteriophages, which adsorb to the pilus sides, bacteriophage binding to pili composed of unacetylated subunits is difficult to visualize by electron microscopy, suggesting a reduction in affinity (Grossman et al, 1990). Addition of traX plasmids to cells elaborating pili composed of unacetylated subunits at least partially complemented all of these defects (T. Grossman and P. M. Silverman, unpublished).…”

“…However, at least in some conditions, domain I is available to antibodies only when the ␣-amino group of the amino-terminal alanine residue is not acetylated. Normally, when acetylation occurs in a reaction requiring the traX gene product (Maneewannakul et al, 1995), domain I appears to be masked, in so far as antibodies to domain I epitopes fail to bind, as judged by immunogold electron microscopy (Frost et al, 1986;Grossman et al, 1990). Additionally, filaments observed in bacterial cultures, without purification, and composed of acetylated subunits tend to remain isolated, with little tendency to aggregate, whereas those composed of unacetylated subunits form massive aggregates (Grossman and Silverman, 1989).…”

“…The model for F-pilin places domain I facing the extracellular environment, for which there is direct evidence (Grossman et al, 1990). However, at least in some conditions, domain I is available to antibodies only when the ␣-amino group of the amino-terminal alanine residue is not acetylated.…”

“…Distribution of cell envelope F-pilin in inner-and outer membrane fractions. Membrane fractions were isolated from an FЈlac strain by isopycnic banding in a sucrose gradient, stained with monoclonal antibody JEL93 and immunogold-protein A (30 nm gold particles), and visualized by electron microscopy, all as described by Grossman et al (1990) and Paiva et al (1992). Inner membrane is shown in the left panel, and outer membrane is shown in the right panel.…”