Table 1.
M. genitalium clinical isolates used in this study.
Fig 1.
Strain typing and variable regions of mgpB and mgpC targeted for PCR amplification and sequencing.
Small arrows indicate the primers used to PCR amplify each region indicated by grey lines. Each variable region was amplified with two different primer pairs. PCR products were cloned and sequenced from individual plasmids to assess sequence changes over time in infected men. Strain types were determined by sequencing the region indicated by “ST”. Numbers indicate base pairs relative to the start codon of mgpB and mgpC.
Table 2.
Primers used in this studya.
Fig 2.
Immunoblot IgG reactivity of M. genitalium positive patient sera with whole lysates of strain G37.
Sera obtained from four PCR-positive men at first and third clinic visits (V1 and V3, respectively), were diluted 1:1,000 and reacted with M. genitalium whole cell lysates separated on a 7.5% SDS-PAGE gel. “Rb”, specific rabbit sera [40] was used to identify MgpB and MgpC protein bands (arrows). Molecular weight markers (in kDa) are shown at left. The magnitude of antibody reactivity between patients cannot be compared as different film exposure times were used.
Fig 3.
Evolution of MgpB regions B, EF, and G, and MgpC region KLM sequences during infection in Patient 10366.
Amino acid alignments (Multalin [41]) were submitted to Highlighter [42] to generate the output shown. Each horizontal line represents a single cloned sequence. Amino acids that differ from the predominant Visit 1 sequence (top line) are marked with vertical colored bars with different colors corresponding to particular amino acids. Visit 1 and Visit 3 sequences are indicated by grey and yellow block shading at right, respectively. Dashed boxes indicate sequences detailed in Figs 5 and 6. Variable regions not shown for Patients 10378, 10467, and 10477 did not vary substantially between time points and therefore were not analyzed further.
Fig 4.
Evolution of MgpB regions B, EF, and G, and MgpC region KLM sequences during infection in Patients 10378, 10467, and 10477.
Analysis and data presentation are described in legend for Fig 3.
Fig 5.
Detailed alignment of variable segments indicated in Fig 3 for Patient 10366.
The predominant sequence present at Visit 1 is shown in black at the top of each alignment. Each unique sequence is represented on a single line, variant amino acids are marked in color, dots indicate unchanged amino acids, and dashes indicate gaps.
Fig 6.
Detailed alignment of variable segments indicated in Fig 4 for Patients 10378, 10467, and 10477.
The predominant sequence present at Visit 1 is shown in black at the top of each alignment. Each unique sequence is represented on a single line, variant amino acids are marked in color, dots indicate unchanged amino acids, and dashes indicate gaps.
Fig 7.
Assessment of antigenic variation during in vitro passage in Vero cell cocultures.
MgpB region B sequences were compared between Visit 1 (MEGA 1166) and Visit 3 (MEGA 1206) cultured from patient specimens after seven weeks of in vitro passage in Vero cell cocultures. Identical sequences are indicated with identical colors; 34–38 plasmids were sequenced from each culture. Results show that little variation occurred during seven weeks of in vitro growth for each culture.
Fig 8.
Prediction of conformational B cell epitopes in MgpC.
(A). G37 MgpC high-scoring B cell epitopes localize to variable region KLM. DiscoTope 2.0 with a stringent cutoff of -1 (corresponding to 30% sensitivity and 85% specificity), indicated by the horizontal dashed line, was used to predict conformational B cell epitopes using the published MgpC structure (PDB 5mzb, amino acids 25–936 [29]). Variable region KLM is indicated by the purple line, this region is expanded to show detail in panel B (dashed grey lines). (B). Variant amino acids correlate with predicted B cell epitopes within variable region KLM. Red and green lines indicate sequences obtained for Patient 10366 and 10477 isolates, respectively, amino acids that varied between time points indicated by “peaks” (arbitrary units).
Fig 9.
Location of predicted conformational epitopes on the G37 MgpC protein.
The MgpC transmembrane domain (not shown) is oriented downward. Colored residues indicate epitopes that group together on the surface of MgpC; epitopes in the conserved region of KLM are indicated in red. The eight amino acids required for sialic acid binding [29] are shown in magenta, five of which lie within predicted epitopes. Figure produced using Pymol.
Fig 10.
Models of MgpC for Patient 10477 Visit 1 isolate (MEGA 1491, left) and Visit 3 isolate (MEGA 1534, right).
Colored residues indicate predicted epitopes that group together on the surface of MgpC, added to improve data visualization. Amino acids that changed between time points are indicated in black on the Visit 3 model. Amino acids implicated in sialic acid binding are shown in magenta. Residues indicated in red and brown are epitopes predicted in the conserved region of MgpC.