Algorithmic assessment of vaccine-induced selective pressure and its implications on future vaccine candidates

Abstract

Posttrial assessment of a vaccine's selective pressure on infecting strains may be realized through a bioinformatic tool such as parsimony phylogenetic analysis. Following a failed gonococcal pilus vaccine trial of Neisseria gonorrhoeae, we conducted a phylogenetic analysis of pilin DNA and predicted peptide sequences from clinical isolates to assess the extent of the vaccine's effect on the type of field strains that the volunteers contracted. Amplified pilin DNA sequences from infected vaccinees, placebo recipients, and vaccine specimens were phylogenetically analyzed. Cladograms show that the vaccine peptides have diverged substantially from their paternal isolate by clustering distantly from each other. Pilin genes of the field clinical isolates were heterogeneous, and their peptides produced clades comprised of vaccinated and placebo recipients' strains indicating that the pilus vaccine did not exert any significant selective pressure on gonorrhea field strains. Furthermore, sequences of the semivariable and hypervariable regions pointed out heterotachous rates of mutation and substitution.

Figure 1

Multiple sequence alignment of pilus predicted peptides from 12 strains used in the analysis (Table 1). These peptide sequences were produced from translating DNA sequences (see Table 1 for GenBank accession numbers). There are three domains in the pilus peptide: conserved domain (C: 1–53 amino acids), a semivariable domain (SV: 54–114 amino acids), and a hypervariable region (HV: variable number of amino acids starting at amino acid 132). The color shadings (white, gray, and black) indicate the variability of the sequence; we have white: high variability, gray: slightly variable, and black: highly conserved.

Figure 2

Most parsimonious cladogram of full-length predicted peptides. Pgh 3-2 was used as an outgroup since it is the ancestral strain of the vaccine strains. Strains from infected vaccinees are marked by ∗. For a few strains, small sequence segments at the beginning of the gene were not obtained and were treated as missing values in the analysis.

Figure 3

Consensus cladogram of the semivariable (SV) regions peptides (included amino acids 51–127). Pgh 3-2 was used as an outgroup. Strains from vaccinated individuals are marked by *.

Figure 4

Consensus cladogram of the hypervariable (HV) regions peptides (included amino acids 109–166). Pgh 3-2 was used as an outgroup. Strains from vaccinated individuals are marked by *.