Figure 1.
Schematic representation of the mutagenesis strategy utilizing rpsL counter selection.
A 100(shown as patterned region in WT sequence) was selected for the kpsD and waaL; StrepR strains harboring the pRedET plasmid were transformed with PCR product consisting of the rpsL-neo cassette flanked by 50 bp regions homologous to sequences on either side of the chosen point of insertion; double cross-over events during homologous recombination make the mutant strains StrepS KanR due to integration of the rpsL-neo cassette; for complementation, mutant strains were transformed with 100 bp oligonucleotides bearing the same sequence as the WT across the region of insertion; complemented strains were again StrepR due to loss of the rpsL-neo cassette and restoration of the parental genotype; streptomycin resistant (StrepR), streptomycin sensitive (StrepS), kanamycin resistant (KanR). The methodology depicted in this figure is a modification of the rpsL counter-selection technique previously used for site-directed mutagenesis in bacterial artificial chromosomes [57]–[59].
Figure 2.
Phenotypic characterization of capsule mutants (kpsD) and complemented (kpsDC) UPEC strains.
A) Agar slab showing K1 phage reactions of 1177 and RS218 strains with CFT073 (expressing K2 antigen) as negative control; dashed lined through the center denotes line of phage suspension. Wild-type and complemented strains expressing K1 antigen are lysed by the phage and exhibit inhibited growth upon crossing the line of phage suspension. Growth of capsule mutants and the negative control strain is unaffected. B) Agarose gel showing precipitin bands arising from cross-linking of K2 antigen and antisera post electrophoresis. CFT073 WT and capsule complement show a positive reaction; no band is observed for the kpsD mutant due to loss of surface expression of the K2 antigen.
Figure 3.
Analysis of lipopolysaccharide profiles by silver staining.
Silver stained LPS profiles from strains listed below the gels. LPS banding patterns vary according to serotype; CFT073:O6 (A), E. coli 1177:O1 (B), RS218:O18 (C). Wild-type, O antigen mutant (waaL) and its complemented strain (waaLC) are ordered from left to right. The thick band at the bottom represents the lipid A core and is intact in all the strains. The bands stacked on top of the core represent different O antigen side chains of different lengths that disappear in the waaL mutant strains. Banding patterns are restored in the complemented strains and appear to be the same as the WT.
Figure 4.
Bacteria (approximately 4×108 CFU/ml) were incubated in fresh human serum for 90 min and colony counts were used to determine serum sensitivity index (SSI) for each strain, shown at the top of the graph. SSI of 0 represents a resistant strain; 1,2,3,4,5 represent equivalent reduction in log values of colony forming units (CFU/ml) post incubation with serum. The WT strains CFT073, 1177 and RS218 were resistant; CFT073waaL, 1177waaL and RS218waaL mutants were all highly sensitive to human serum (SSI; P<0.05, paired t test) and the complemented strains (waaLC) were resistant. CFT073kpsD and RS218kpsD mutants displayed reduced sensitivity (SSI = 2 and 1, respectively); complemented strains (kpsDC) were resistant. The 1177kpsD mutant was completely resistant to serum killing.
Figure 5.
Survival of UPEC CFT073 (A) and 1177 (B) WT, mutant and complemented strains in whole blood killing assays.
Bacteria were incubated in fresh whole blood for 3–5 h and colony counts were used to compare survival phenotypes of mutants and complemented strains to the WT parents. Fitness of the CFT073 and 1177 capsule (kpsD) mutants was significantly impaired compared to the WT, and was restored completely in the respective capsule complemented (kpsDC) strains. The survival data for the RS218 WT derivatives could not be analyzed due to the high sensitivity of WT strain to killing by whole blood. P values, as determined by the independent samples, t test were as follows: # 0.038, * 0.042.
Figure 6.
Fitness of CFT073 capsule (kpsD) and O antigen (waaL) mutants during in vivo competition with CFT073 WT and corresponding complemented (kpsDC, waaLC) strains.
C57B/L6 mice were infected transurethally with a 1∶1 mixture of CFT073kpsD and CFT073 WT or CFT073kpsDC; CFT073waaL and CFT073 WT or CFT073waaLC. (A) Each marker represents LOG10 total CFU recovered from each mouse per 0.1 g of bladder tissue. Lines connect data points for the same mouse, and horizontal bars represent median values. (B) Each marker represents the LOG10 competitive index calculated for each individual mouse; competitive indices are the ratio of the mutant 0.1 g of bladder tissue to that of WT or complemented strain. Dashed lines represent hypothetical competitive index of 1 (LOG101 = 0), which indicates no difference in fitness between the two strains. Horizontal bars represent group medians, and each competition group had 6 to 8 mice. All mutants were significantly outcompeted by CFT073 WT or their respective complemented strains for bladder colonization (P<0.05). The O6 antigen was significantly more important than the K2 capsular antigen for bladder colonization (P<0.05).
Table 1.
Bacterial strains used in this study