Mating pair formation homologue TraG is a variable membrane protein essential for contact-independent type IV secretion of chromosomal DNA by Neisseria gonorrhoeae

Abstract

Neisseria gonorrhoeae uses a type IV secretion system (T4SS) to secrete chromosomal DNA into the surrounding milieu. The DNA is effective in transforming gonococci in the population, and this mechanism of DNA donation may contribute to the high degree of genetic diversity in this species. Similar to other F-like T4SSs, the gonococcal T4SS requires a putative membrane protein, TraG, for DNA transfer. In F-plasmid and related systems, the homologous protein acts in pilus production, mating pair stabilization, and entry exclusion. We characterized the localization, membrane topology, and variation of TraG in N. gonorrhoeae. TraG was found to be an inner-membrane protein with one large periplasmic region and one large cytoplasmic region. Each gonococcal strain carried one of three different alleles of traG. Strains that carried the smallest allele of traG were found to lack the peptidoglycanase gene atlA but carried a peptidoglycan endopeptidase gene in place of atlA. The purified endopeptidase degraded gonococcal peptidoglycan in vitro, cutting the peptide cross-links. Although the other two traG alleles functioned for DNA secretion in strain MS11, the smallest traG did not support DNA secretion. Despite the requirement for a mating pair stabilization homologue, static coculture transformation experiments demonstrated that DNA transfer was nuclease sensitive and required active uptake by the recipient, thus demonstrating that transfer occurred by transformation and not conjugation. Together, these results demonstrate the TraG acts in a process of DNA export not specific to conjugation and that different forms of TraG affect what substrates can be transported.

Fig 1

Subcellular fractionation and Western blot analysis of gonococcal strains. The wild-type strain MS11 (WT), the GGI deletion strain ND500 (ΔGGI), and strain PK166 expressing CAT (cat⁺) were separated into soluble and membrane fractions, and the fractions were subjected to Western blot analysis with anti-TraG, anti-CAT, and anti-PilQ antibodies.

Fig 2

(A) Quantitative analysis of the enzymatic activities of TraG′-′PhoA fusions expressed in N. gonorrhoeae. Strains were grown in log-phase in liquid medium and induced to express the TraG′-′PhoA fusion proteins. The enzymatic activity of ′PhoA was calculated and plotted in phosphatase units. (B) Quantitative analysis of the enzymatic activities of TraG′-′LacZ fusions expressed in N. gonorrhoeae. Strains were grown in log phase in liquid medium and induced to express the TraG′-′LacZ fusion proteins. The enzymatic activity of LacZ was calculated and plotted in Miller units. Each reported value is the result of three independent experiments and the reported errors are the standard deviations. *, P < 0.05 as determined by using the Student t test.

Fig 3

Topology model for TraG. The results from the fusion protein studies are consistent with a five membrane-span model. Active ′PhoA fusions are shown with triangles. Active LacZ fusions are shown with lollipop shapes. The numbers indicate the location of the TraG junction in amino acids.

Fig 4

(A) Southern blot analysis of EcoRI- and ClaI-digested chromosomal DNA from 10 low-passage gonococcal isolates. A region internal to the conserved portion of traG was used as the probe. traG is not present at all or is found on one of two different sized restriction fragments in these isolates. (B) Three different versions of the GGI. Class I is the version found in strain MS11 and contains traG1 and atlA. Class II contains both traG2 (traG_sac-4) and atlA, while class III is the version found in strain PID2059. Class III contains traG3 and eppA but lacks atlA. The shaded areas indicate regions that are different from the nucleotide sequence of strain MS11.

Fig 5

EppA degrades peptidoglycan. (A) Zymogram of purified EppA electrophoresed through a polyacrylamide gel containing gonococcal peptidoglycan. Zones of clearing indicate areas where EppA bound to or degraded peptidoglycan. (B) Coomassie blue-stained gel showing quantities of EppA used for part A. Lane 1, 2.4 μg; lane 2, 1.8 μg; lane 3, 1.2 μg; lane 4, 0.9 μg; lane 5, 0.6 μg; lane 6, no EppA. C. EppA degraded ³H-labeled gonococcal sacculi producing soluble peptidoglycan fragments that were quantified by scintillation counting. This reaction was inhibited by addition of EDTA or 1,10-phenanthroline.

Fig 6

HPLC analysis shows EppA produces a variety of soluble peptidoglycan fragments from digestion of gonococcal peptidoglycan, but these peptidoglycan fragments are not produced in the presence of EDTA. Peptidoglycan fragments were detected spectrophotometrically by absorbance at 210 nm.

Fig 7

LC/MS analysis of peptidoglycan fragments generated by EppA. Proposed structures for the peptidoglycan fragments numbered in the figure are listed in Table 3.

Fig 8

Fluorometric detection of DNA secreted by gonococcal strains. The concentration of DNA in culture supernatants was measured by treating culture supernatants with a fluorescent, DNA-binding dye and comparing them to DNA standards of known concentrations. Secreted DNA was normalized to total protein in the cell pellet. The average background fluorescence value was subtracted from the values for all strains. Strains measured were MS11 (WT), traG1 deletion mutant PK186 (ΔtraG1), complemented traG1 strain PK191 (ΔtraG1 + traG1⁺), HH534 (traG2⁺), HH528 (traG3⁺ eppA⁺), PID2059, PK161 (traG3⁺ eppA⁺ + atlA⁺), PK162 (PID2059 + atlA⁺), and ND500 (ΔGGI). All strains are MS11 background except those labeled PID2059. Each reported value is the result of at least three separate experiments, and the errors are the standard deviations. *, P < 0.05 compared to wild type (as determined by using the Student t test).

Fig 9

Coculture transformation in static liquid culture. (A) Strains JD1545 (cnp::cat, recA6) and MS11Spc (rpsE) were mixed and grown in shallow liquid culture without shaking for 2.5 h in the absence or presence of 25 μg of DNase I/ml. Transformation was assessed by enumerating the Cm^r Sp^r transformants per Sp^r recipient. (B) Strain JD1545 (cnp::cat, recA6) was cocultured with MS11Spc or MS11Spc pilT (NT04), and transformation frequency was determined as in panel A.