Fig 1.
Organization of the bacteriophage T4 long tail fiber.
(A) A structural model of bacteriophage T4 virion showing the head, the tail, and the long tail fibers. (B) The proximal half of the LTF is formed by gp34 trimer (red), the knee cap is formed by gp35 monomer (green), and the distal half is formed by gp36 trimer (blue) and gp37 trimer (yellow) [10]. The part of the gp37 trimer for which the X-ray structure was determined is outlined by blue rectangle. (C) The crystal structure of the gp37 C-terminal fragment [18]. The three polypeptide chains in the gp37 trimer are shown in red, blue, and green. The ferrous ions are shown as yellow spheres.
Fig 2.
The LTF needle binds to both LPS and OmpC receptors.
(A) SDS-PAGE profiles of purified LTF needle and OmpC trimers. “+” and “-” represent boiling in the presence of SDS. The boiled gp37 sample (lane 2) showed the major 27 kDa monomer band and a faint band of residual trimer present in the boiled sample. The unboiled gp37 sample (lane 3) showed two trimer bands, which probably represent two trimer conformations, intact trimer and partially unfolded trimer. (B) Diazirine-labelled gp37 was incubated with purified LPS from E. coli B for 1 h at 37ºC followed by UV irradiation for 15 min and Superose 6 10/300 GL size-exclusion chromatography (red). Same amounts of diazirine-labelled gp37 (blue) and LPS (green) were loaded on the same column as controls. (C) SDS-PAGE profile of the peak fractions from panel B. The gel was developed by silver staining. Lanes 2 to 13 (elution volumes corresponding to 8 to 16 ml) show fractions of gp37-LPS complex peak. Lane 1 shows LPS control peak (elution volume, 9 ml) and lane 14 shows the gp37 control peak (elution volume, 16 ml). Note the presence of a small amount of cross-linked dimer of gp37 of trimers in lane 14. (D) ELISA assay to capture gp37-OmpC complex. The wells were coated with gp37 and OmpC was added at various molar ratios. The unbound OmpC was removed by repeated washing and the amount of bound OmpC was determined by adding mouse anti-OmpC antibody followed by HRP-conjugated anti- mouse IgG antibody. Various controls (coating buffer + OmpC; BSA + OmpC; or gp37 + OmpC + BSA) were included to validate the specific interaction between gp37 and OmpC.
Fig 3.
Mutational analysis of the LTF needle.
(A) gp37 trimer with one of the polypetide chains highlighted in green. The positions of amino acids on the backbone where an amber stop codon was introduced are shown in red and blue. (B) Suppressor patterns of fourteen gp37 amber mutant phages, each tested with 13 different suppressor E. coli strains. Amino acid substitutions above the native sequence represent functional phenotypes and those below represent lethal phenotypes. Red numbers and letters in both A and B represent the amino acids that are essential for function and blue numbers and letters represent the amino acids that are not essential for function.
Fig 4.
Amino acids lining the LTF tip surface interact with LPS and OmpC receptors.
(A) Combinatorial mutagenesis was used to determine the functional importance of each of the amino acids. See S3 Fig for a schematic of combinatorial mutagenesis and genetic rescue strategies. Amino acid substitutions shown above the native sequence represent functional phenotype (blue) and the ones below (red) represent null phenotype. (B & C) Functional importance of the alanine substitutions. Residues occupying the tip of the LTF were tested for their functional importance by constructing single (B) and double (C) alanine substitution mutants followed by genetic rescue. Phenotypes were scored as functional (blue) or null (red). (D) Critical single residues identified in marker rescue assays are highlighted in red. “*” represents the amino acids that were found to be null in the background of double alanine substitutions. (E) A single polypeptide chain showing functionally important (red) or nonessential (blue) amino acid residues.
Fig 5.
Modulation of LTF-receptor interactions.
(A) SDS-PAGE of purified mutant LTF needle proteins. The first lane on the left show molecular mass standards. A vertical line was added between images that were sliced and pasted. (B) Quantification of LTF needle-OmpC interaction by direct binding ELISA assay. Binding results that correlated with the in vivo genetic assays (Fig 4) are shown as black bars and those that did not correlate are shown either as blue bars (greater binding than WT needle) or red bars (poorer binding than WT needle). (C) Western blot with anti-OmpC antibody showing the presence of OmpC band in the E. coli K12 strain (lane 2), but not in the B strain (lane 1). The purified OmpC protein was included in lane 3 as a positive control. (D) Binding of WT and mutant LTF needle proteins to E. coli. Bound gp37 was detected by Western blotting using monoclonal anti-histag antibodies. Lane 1 is negative control in which the gp37 protein was omitted. Lanes 3, 5, and 7 represent negative controls where E. coli was omitted. Lane 8 is the positive control where purified WT gp37 was loaded into the well. (E) Dependence of protein concentration on the binding of WT and N937A gp37 proteins to E. coli K12 bacteria. Lane 1 is the negative control where the gp37 protein was omitted and lanes 6 and 11 are negative controls where E. coli was omitted. Arrows show the position of the gp37 band. (F) Binding of WT gp37, N937A, and G938A proteins to E. coli bacteria in the presence of increasing concentrations of purified E. coli B LPS.
Fig 6.
LPS and OmpC receptor binding patches on the LTF tip.
Surface representation of the LTF tip showing top (A) and side (B) views with the functionally critical amino acid residues highlighted. Red patch corresponds to amino acids that are important for both LPS and OmpC binding, blue patch for LPS binding, and cyan patch for OmpC binding.