Transposon mutagenesis identifies sites upstream of the Neisseria gonorrhoeae pilE gene that modulate pilin antigenic variation

Abstract

Gene conversion mediates the variation of virulence-associated surface structures on pathogenic microorganisms, which prevents host humoral immune responses from being effective. One of the best-studied gene conversion systems is antigenic variation (Av) of the pilin subunit of the Neisseria gonorrhoeae type IV pilus. To identify cis-acting DNA sequences that facilitate Av, the 700-bp region upstream of the pilin gene pilE was targeted for transposon mutagenesis. Four classes of transposon-associated mutations were isolated, distinguishable by their pilus-associated phenotypes: (i) insertions that did not alter Av or piliation, (ii) insertions that blocked Av, (iii) insertions that interfered with Av, and (iv) insertions that interfered with pilus expression and Av. Mutagenesis of the pilE promoter did not affect the frequency of Av, directly demonstrating that pilin Av is independent of pilE transcription. Two stretches of sequence upstream of pilE were devoid of transposon insertions, and some deletions in these regions were not recoverable, suggesting that they are essential for gonococcal viability. Insertions that blocked pilin Av were located downstream of the RS1 repeat sequence, and deletion of the region surrounding these insertions completely abrogated pilin Av, confirming that specific sequences 5' to pilE are essential for the recombination events underlying pilin Av.

FIG. 1.

Transposon and site-directed mutants upstream of pilE. (A) Map of transposon insertions (gray inverted triangles) recovered upstream of pilE. The figure is drawn to scale. Sites of single transposon insertions are dark gray inverted triangles, and sites where multiple insertions were isolated are light gray inverted triangles. The uss and pilin-associated repeats RS1 and RS2 are indicated. Dark gray boxes indicate sequences identical to the pilE constant region, and checkered boxes indicate sequences identical to the cys2 coding sequence in pilE. Diagonally striped boxes indicate PilA binding sites (2), the white triangle indicates the partial Sma/Cla repeat (17), and the open circle denotes the IHF binding site (19). The line from the arrow indicates the direction of pilE transcription. Lines across the top indicate regions in which transposon insertions have different phenotypes. Region 1 contains transposon insertions that have no effect on pilin Av. Regions 2 and 3 contains transposon insertions that retain a piliated phenotype; region 2 insertions completely abrogate pilin Av, while region 3 insertions interfere with Av. Region 4 contains transposon insertions that disrupt pilin Av and also display a piliation defect. Solid lines across the bottom, marked “A” and “B,” indicate spaces in which no transposon insertions were detected. Dashed lines across the bottom indicate the overlapping cloned fragments targeted for transposition. Arrows indicate the borders of the uss deletion. Transposon insertions with numbers above them indicate mutants that have been further characterized and are referred to in the text and subsequent figures. (B) DNA sequence of the region between the cys2 sequence at the 3′ end of RS1 and the pilE start codon. The locations of the upstream cys2 element, RS1, the two PilA binding sites, the IHF binding site, and the pilE start codon are labeled and indicated by the underlined and boldface sequence. The location of each Tn5 insertion is denoted by the dotted boxes containing the 9 bp of sequence that is duplicated as a result of mTn insertion and flanks the mTn insertion site (capitalized sequence). The −10 promoter sequence targeted for mutagenesis is indicated, and the nucleotide changes introduced in the promoter mutant are underlined. Arrows indicate the 5′ and 3′ borders of the deletions.

FIG. 2.

Transposon insertions upstream of pilE display a range of pilin Av defects. Kinetic variation assay measuring the average number of pilus-dependent colony morphology changes occurring over time. A representative piliated transposon mutant from each phenotypic region of transposon insertions is shown. Mutant pilE::mTn#9 represents region 1, pilE::mTn#15 represents region 2, and pilE::mTn#4 represents region 3. The parental strain FA1090 1-81-S2 recA6, grown in the presence of IPTG to activate RecA expression, is also depicted. Error bars represent the standard error of the mean of two experiments done in triplicate or quadruplicate (n = 7 or 8). Mutant pilE::mTn#15 does not antigenically vary, similar to a recA null strain (data not shown). Statistically significant differences (P < 0.05), as determined by Student's t test, are indicated by asterisks for comparison to the isogenic parental strain.

FIG. 3.

Deletion upstream of pilE disrupts pilin Av. (A) Schematic diagram of deletions upstream of pilE. The locations of RS2, RS1, and the pilE promoter and transcriptional start site (arrow) are indicated. The dashed line denotes the region of deletion. Deletion constructs were linked to the Km^r marker to provide a selectable marker linked to the deletion for transformation. (B) The kinetic variation assay was used to measure the average number of pilus-dependent colony morphology changes that occurred over time in the ΔRS1 and ΔRS1-PilA strains. The deletion strains are compared to the parental strain FA1090 1-81-S2 recA6 grown in the presence (Parental) and absence (recA) of IPTG to turn on and off RecA expression, respectively. Error bars represent the standard error of the mean of one representative experiment of three performed, each in triplicate. Statistically significant differences (P < 0.001), as determined by Student's t test, are indicated by asterisks for comparison to the isogenic parental strain.