Role for RTX-family toxin HlyA of extraintestinal pathogenic Escherichia coli in serum resistance

Abstract

Extraintestinal pathogenic Escherichia coli (ExPEC) is a major cause of urinary tract infections, bacteraemia, and sepsis. CFT073 is a prototypic, urosepsis isolate of sequence type (ST) 73. ST73 isolates are associated with higher virulence scores than other pandemic clonal groups, such as ST131. This laboratory, among others, has previously shown that strain CFT073 is serum-resistant, with virulence factors such as the exopolysaccharide capsule and other extracellular polysaccharides imparting resistance to the complement system. In this study, it was shown that culture supernatants were protective in standardized serum killing assays, when compared to cultures standardized in fresh medium. Diluting cultures in fresh medium in place of conditioned medium significantly increased sensitivity of CFT073 to serum, indicating that a secreted factor may provide resistance to serum. Haemolysin, a pore-forming toxin, is secreted by CFT073 in a calcium-dependent manner. This study found that a CFT073 hlyA mutant is significantly more sensitive to 50% serum than the wild-type, implicating haemolysin in the response of CFT073 to serum. In addition to acting as a toxin upon secretion, it has been shown previously that HlyA forms a complex with lipopolysaccharide (LPS), which permits modulation of host immune responses by HlyA whilst cell-associated. The effect of HlyA on capsule expression and serum resistance was examined and characterized in this study, with results indicating that perhaps the HlyA-LPS complex interacts with surface capsule. This study is the first to identify haemolysin as a virulence factor promoting resistance to serum in CFT073, acting whilst associated with the cell.

Keywords: Escherichia coli; ExPEC; HlyA; LPS; bloodstream infection; capsule; exotoxin; haemolysin; sepsis; serum resistance.

Figure 1.

HlyA mutants are sensitive to serum during exponential growth. Cultures were grown to (A) exponential OD_600nm = 0.6 and (B) stationary phase OD_600nm = 3.0 (standardized to OD_600nm = 0.6) in conditioned LB medium before diluting 1:2 in NHS. For the supernatant swap assay, cultures were grown to (C) exponential and (D) stationary phase as before, prior to standardizing in the conditioned LB of the other strain (e.g. wild-type was standardized in HlyA-deficient supernatant and vice versa) before diluting 1:2 in NHS. Percentage survival was calculated as a fraction of CFU/ml at T90 in NHS over CFU/ml at T90 in HIS x 100. Statistical significance calculated by unpaired one-way ANOVA and shown relative to wild-type for each assay. Y-axis in Log 10. N = 4 biological replicates.

Figure 2.

HlyA mutants display posttranscriptional changes to capsule. Cultures were grown to exponential OD_600nm = 0.6 and whole cell lysates were separated by SDS-PAGE and subject to LPS straining (A; N = 3, N =1 shown). Lane 1; wild-type, 2; Δksl, 3; ΔwaaL, 4; Δksl ΔwaaL, and 5; ΔhlyA. Whole cell lysates were also subject to western blot (B; N = 3). Additionally, exponential cells were subject to ELISA (C; N = 4) and RT-qPCR (D; N = 4) to examine detectable K2 capsule and capsule gene expression, respectively. Statistical significance calculated by one-way ANOVA and shown relative to wild-type. Anti-K2 = anti-K2 antibody used in ELISA.

Figure 3.

Complementation of hlyA restores haemolysis and serum resistance. (A) Wild-type colonies are haemolytic, (B) ΔhlyA colonies are not haemolytic on blood agar, and (C) complemented mutants display a restoration of haemolysis, to levels comparable to wild-type. (D) Survival of wild-type, ΔhlyA, and ΔhlyA pHlyCABD at exponential and stationary phase upon exposure to 50% NHS. Strains were grown to OD_600nm = 0.6 (exponential) or OD6_00nm = 5.0 and incubated in 50% HIS and NHS for 90 min at 37°C. Statistical significance determined by One-way ANOVA and Dunnet’s multiple comparisons test. Significance shown relative to wild-type. n = 4. ^****; P < .0001. N = 4 biological replicates.

Figure 4.

Feedback loop exists between waa genes and the hly operon. (A) CFT073, mutant derivatives and complemented waaG mutant were plated on blood agar plates to observe haemolytic activity (if any). These results were mirrored by quantitative haemolysis assays performed in (B), as before, wherein 1% TritonX100 served as a positive control for haemolysis and uninoculated LB broth served as a negative control. Statistical significance determined by One-way ANOVA and Dunnett’s multiple comparisons test. Significance shown relative to wild-type. n = 4. ***; P < .001, ^****; P < .0001. N = 4 biological replicates

Figure 5.

HlyA does not confer resistance to nonpathogenic strain in vitro. (A) DH5α colonies are nonhaemolytic, whilst DH5α pHlyCABD are haemolytic. (B) Exponential (and stationary phase—not shown) cultures were standardized to 1 × 10⁸, exposed to 50% NHS and HIS for 90 min at 37°C prior to plating, overnight incubation, and CFU/ml counting. Wild-type = CFT073. Statistical significance determined by One-way ANOVA and multiple comparisons. ^**** = P < .0001. N = 4 biological replicates.

Figure 6.

Hypothetical model for the role of HlyA in serum resistance. Graphical representation of the CFT073 outer envelope, including LPS and capsule. This study has shown that the R1 core and O6 antigen contribute to K2 capsule retention, likely through ionic interactions as detailed in other species such as Klebsiella (Fresno et al. , Singh et al. 2022), as well as E. coli K1 (Jiménez et al. 2012). HlyA also associates with LPS (likely the core) through ionic interactions and as a result of findings in this study, is hypothesized to associate with K2 in a similar manner.