Staphylococcus epidermidis ST2 strains associated with bloodstream infections contain a unique mobile genetic element encoding a plasmin inhibitor

Abstract

Staphylococcus epidermidis, a common commensal bacterium, is a leading cause of nosocomial catheter-associated bloodstream infections. S. epidermidis sequence type 2 (ST2) is specifically recognized globally for causing invasive disease. In this study, we identified a novel putative integrated conjugative element, pICE-Sepi-ST2, unique to the genomes of S. epidermidis ST2. Our investigation identified pICE-Sepi-ST2 in all ST2 isolates from bloodstream infections. Meanwhile, ST2 isolates from other infection sources, such as catheters, prosthetic joints, and fracture fixations, showed variable pICE-Sepi-ST2 prevalence. pICE-Sepi-ST2 encodes two putative cell wall anchored proteins that we have designated SesX and SesY. Biochemical characterization of SesY revealed that it binds both plasminogen (Plg) and plasmin (Pln) and inhibits Pln's ability to cleave a chromogenic substrate and degrade fibrin clots. Furthermore, all ST2 isolates containing a pICE-Sepi-ST2 also have a mutated sdrG gene. Thus, all ST2 isolates have two genetic modifications that target distinct steps in the hemostatic pathway. SdrG, which inhibits coagulation, is inactivated, and SesY, which inhibits fibrin, is introduced. These findings suggest that the hemostasis pathway is a strategic target for ST2 S. epidermidis bloodstream pathogenesis.

Importance: This study uncovers a new virulence mechanism in Staphylococcus epidermidis ST2 bloodstream isolates. We identify a mobile genetic element (MGE) characteristic of an integrated conjugated element (ICE). pICE-Sepi-ST2 carries the genetic information needed to produce a cell wall-anchored (CWA) protein called SesY. The results indicate that SesY binds to plasminogen (Plg) and plasmin (Pln) and inhibits Pln's degradation of fibrin clots. Genetic analysis showed that all ST2 bloodstream isolates can express the plasmin inhibitor SesY and carry a mutation in the SdrG gene, resulting in the expression of inactive SdrG. Thus, we describe a molecular pathway targeting the coagulation pathway that may be required for S. epidermidis ST2 to cause bloodstream infections.

Keywords: Staphylococcus epidermidis; bloodstream infections; coagulation pathway; integrated conjugated element; mobile genetic element; molecular pathogenesis; plasmin.

Fig 1

Genome comparison of global S. epidermidis isolates obtained from human blood. The circular genome of ST2 S. epidermidis BB403117S was compared using BLAST to 18 other S. epidermidis isolates from 14 different sequence types (STs). The legend of the comparison map includes the name of each S. epidermidis isolate followed by its sequence type (format: Strain_ST). Annotations around the outer ring highlight genomic regions present in ST2 bloodstream isolates but absent in other STs, except for ST54. The genomic data for these isolates were sourced from BV-BRC, and the circular genome comparative map was generated using BRIG (30). For more detailed information on the isolates, see Table S1.

Fig 2

Schematic of pICE-ST2 found in S. epidermidis ST2 bloodstream isolates. (A) Genetic organization of the open reading frame in pICE-ST2. Genes are drawn to scale. Genes are colored based on their function: DNA-binding proteins are shown in orange; CWA proteins are shown in pink; genes for conjugation machinery are shown in red; sortase gene is colored blue; genes with unknown functions are colored gray. (B) Domain organization, location, and topologies of proteins with known predicted functions are shown. Proteins are colored as in panel A based on their predicted gene functions.

Fig 3

Predicted domain organizations and structures of SesY and SesX proteins. (A and C) The predicted domain organization of SesY and SesX is illustrated in panels (A) and (C), respectively. The signal sequence (S) is shown in gray. Following the signal sequences, SesY and SesX have multiple domains labeled numerically as D₁ to D_n. Inverted black arrows indicate the "full-length" protein (SesY_32–341 and SesX_28–982). The cell wall, membrane-spanning region, and short cytoplasmic tail are collectively labeled as W, as shown in gray. The numbers below the schematic indicate the start of each domain, except for the last number, which represents the total number of amino acids in the protein. D-domains are color-coded based on the Alphafold 2 LDDT (per residue confidence metric) score: blue indicates a score over 70, while orange indicates a disordered region with a score below 50. (B and D) Predicted structure of SesY and SesX through Alphafold 2. N- and C-terminus are indicated. Amino acids in the domains are colored based on the LDDT scores.

Fig 4

Binding of SesY to Plg and Pln. (A) Host ligands were immobilized on 4HBX microtiter plates to evaluate for binding of His-tagged recombinant proteins rSesY and FnbPB_N2N3. Detection was performed using an anti-His antibody. Bovine serum albumin (BSA) served as a negative control. (B) Different concentrations of recombinant proteins rSesY and FnbPB_N2N3 (positive control) were evaluated for binding to immobilized human Plg and Pln on 4HBX microtiter plates. Binding was detected in a dose-dependent manner using α-His Tag HRP conjugated antibody (1:3,000). The binding curve and dissociation constant were approximated via non-linear regression using a one-site, specific binding model. (C) Inhibition of rSesY binding to immobilized Pln was assessed in varying concentrations of a lysine derivative, tranexamic acid (circle), and arginine (square). rSesY bound to Pln was detected with α-His Tag HRP conjugated antibody at a 1:3,000 dilution. The presence of tranexamic acid decreased rSesY binds to Pln by 85%. Results from three replicates (n = 3) are represented. Error bars indicate the standard deviation (mean ± SD).

Fig 5

SesY inhibits Pln activity. (A) Graph representing color development from chromogenic Pln substrate S-2251 digestion at different time points during 1.5 h in the presence of 5 µM of rFnbpB_N2N3, rSesY, and BSA. Control represents wells with no additional protein added to the reaction. (B) The initial velocities of Pln are plotted against various concentrations of the Pln S-2251 substrate to generate Michaelis-Menten curves. This was done in the presence of increasing concentrations of the recombinant protein rSesY (0 µM, 1 µM, 2.5 µM, 5 µM, 10 µM) to evaluate its effect on Pln activity. (C) Lineweaver-Burk plots are generated from Michaelis-Menten curves to analyze the enzyme (Pln)-substrate (S-2251)-inhibitor (rSesY) relationship. Noncompetitive inhibition is shown in the presence of rSesY since K_m is unaffected and V_max is reduced. Results from three replicates (n = 3) are represented. Error bars indicate the standard deviation (mean ± SD).

Fig 6

SesY inhibits the Pln-mediated degradation of fibrin. (A) Percentage of fibrin gel remaining after digestion of pre-formed fibrin gel at 12 h in the presence of Pln and 5 µM of different proteins: rFnbpB_N2N3, rSesY, and BSA. Buffer control represents wells with 0 µM of additional protein. No Pln represents wells with pre-formed fibrin gel without the addition of Pln. (B) Percentage of fibrin gel remaining after digestion of pre-formed fibrin gel at 12 h time point in the presence of Pln and different concentrations of rSesY (0 µM, 1 µM, 2.5 µM, 5 µM, 10 µM). Buffer control represents wells with 0 µM of additional protein. No Pln represents wells with pre-formed fibrin gel without the addition of Pln. Results from three replicates n = 3 is represented. Error bars indicate the standard deviation. One-way ANOVA followed by Dunnett’s multiple-comparison tests (mean ± SD, ****P < 0.000001 [A] and P < 0.00001 [B]).

Fig 7

Schematic of truncations in the SdrG protein found in S. epidermidis ST2 isolates. The figure illustrates the different domains of the SdrG protein: the signal sequence is labeled S; the N-terminal domains are labeled N1, N2, and N3; B1 and B2 are repeats of unknown function; R stands for the serine-aspartate repeat region; W stands for the cell wall region; and M for the membrane-spanning region. The dotted line indicates the position of the stop codon. The figure shows the amino acid sequence followed by an asterisk (*) to represent the truncation of the protein at that position. Stop codons found in ST2 bloodstream isolates are highlighted in red.