The type-2 Streptococcus canis M protein SCM-2 binds fibrinogen and facilitates antiphagocytic properties

Abstract

Streptococcus canis is a zoonotic agent that causes severe invasive diseases in domestic animals and humans, but little is known about its pathogenesis and virulence mechanisms so far. SCM, the M-like protein expressed by S. canis, is considered one of the major virulence determinants. Here, we report on the two distinct groups of SCM. SCM-1 proteins were already described to interact with its ligands IgG and plasminogen as well as with itself and confer antiphagocytic capability of SCM-1 expressing bacterial isolates. In contrast, the function of SCM-2 type remained unclear to date. Using whole-genome sequencing and subsequent bioinformatics, FACS analysis, fluorescence microscopy and surface plasmon resonance spectrometry, we demonstrate that, although different in amino acid sequence, a selection of diverse SCM-2-type S. canis isolates, phylogenetically representing the full breadth of SCM-2 sequences, were able to bind fibrinogen. Using targeted mutagenesis of an SCM-2 isolate, we further demonstrated that this strain was significantly less able to survive in canine blood. With respect to similar studies showing a correlation between fibrinogen binding and survival in whole blood, we hypothesize that SCM-2 has an important contribution to the pathogenesis of S. canis in the host.

Keywords: SCM; Streptococcus canis; anti-phagocytosis; fibrinogen-binding; pathogen-host interaction.

Figure 1

Comparing SCM-1 and SCM-2. (A) Alignment tree of SCM protein sequences of 75 clinical S. canis isolates from different host species, 14 reference strains and 2 outliers. Maximum Likelihood tree generated with IQtree and visualized on iTOL. Strain names in bold were used in further experiments. (B) Protein Sequence alignment of a representative SCM-1 and SCM-2 protein (S.canis G361 and IMT40096, respectively). Green represents regions with 100% sequence identity between SCM-1 and SCM-2. The absence of a colored bar indicates that no consensus was observable between SCM-1 and SCM-2 in that region. Differences between the amino acid sequences were highlighted in orange. The membrane anchoring LPXTG motif was highlighted in purple.

Figure 2

IgG binding of SCM-2 S. canis strains (A) FACS analysis with Alexa 488 conjugated. IgG Positive controls: SCM-1 S. canis G361 and IMT40165. Negative control: S. canis G2 (n = 3). Data represent mean fluorescence intensity (MFI) ± SD of three independent experiments. Statistical significance was calculated with ordinary one-way ANOVA. (B) Confocal microscopic analysis of bacterial aggregation and IgG binding. Strains were grown overnight at 37°C in TSB. Bacteria were allowed to adhere to poly-L-lysin-coated ibidi-slides, fixed and incubated with Alexa 488-conjugated rabbit IgG (cyan). Bacterial DNA/RNA was stained with ethidium homodimer-1 (red). Bacterial aggregation was visualized by confocal microscopy. Positive control: S. canis IMT40165. Bars indicating 10 μm.

Figure 3

Analysis of the fibrinogen binding capacities of different SCM-2 expressing S. canis strains. FACS analysis with FITC-conjugated human fibrinogen. Positive control: S. canis G2. Negative control: S. canis G361. Data represent MFI ± SD of three independent experiments normalized to the fluorescence of SCM-1 S. canis G361. Fibrinogen binding is abrogated in the KO mutant IMT40096Δscm (n = 4). Statistical significance was calculated with unpaired t-test.

Figure 4

Analysis of the fibrinogen binding capacities of SCM-2 expressing S. canis strains IMT40096 and IMT42870. Interactions of soluble SCM with immobilized fibrinogen were analyzed by surface plasmon resonance spectroscopy (SPR). Representative sensorgrams of three independent experiments of two SCM-2 expressing S. canis (IMT40096, IMT42870). Histidine-tagged recombinant proteins were extracted as previously described (Fulde et al., 2013). The association and dissociation was observed, each of 300 s. Values of the control flow cells were subtracted from each sensorgram. A representative KD value was calculated for the interaction between fibrinogen and SCM using the 1:1 Langmuir binding model.

Figure 5

Whole blood assay of S. canis IMT40096 WT and IMT40096Δscm. Canine blood was heparinized and incubated with 1 × 10⁵ CFU of S. canis for up to 2 h. Survival factors were calculated by ratio of CFU counts at timepoint 0 h versus 1 and 2 h post-inoculation (n = 3). Statistical significance was calculated with unpaired t-test.