Surviving a sticky situation: Therapeutic administration of fibrinogen variant γ’ improves outcomes of Staphylococcus aureus septicemia

A defining feature of virulent strains of Staphylococcus, such as S. aureus, is the ability to coagulate blood through the noncanonical conversion of fibrinogen to fibrin. Such “coagulase-positive” staphylococci produce two different coagulases, staphylocoagulase (Coa) and von Willebrand factor-binding protein (vWbp), that each promote coagulation through binding and activation of the zymogen prothrombin (Factor II) in a molecular complex known as staphylothrombin (1, 2). Staphylothrombin, in turn, converts fibrinogen into fibrin. S. aureus possesses multiple surface adhesins known as agglutinins that can bind fibrin and fibrinogen, effectively creating a “shield” around the bacteria in vivo and hindering antibacterial immune defenses and antibiotic treatment (2, 3). The interaction between S. aureus and fibrin(ogen) is a critical facet of the pathogenesis of this important human pathogen in multiple models of infection, including intra-abdominal infection, catheter-associated bloodstream infection, endocarditis, and septicemia (1, 4). S. aureus coagulases and agglutinins also contribute to biofilm and abscess formation in host tissues (5, 6). Based on these findings, there is considerable interest in disrupting the interactions between S. aureus and fibrin(ogen) as an adjunctive therapeutic strategy for severe staphylococcal infections. S. aureus strains lacking both coagulases are less virulent in animal models of sepsis, endocarditis, skin infection, and catheter-related infection (1, 7). Accordingly, attempts have been made to inhibit the function of these virulence factors through immunization or administration of direct thrombin inhibitors (e.g. dabigatran and argatroban), and both approaches have shown promise in preclinical studies (8–10). However, despite decades of preclinical research on numerous S. aureus vaccine candidates, human efficacy has been elusive. Additionally, there are clinical contexts in which the anticoagulant effects of direct thrombin inhibitors would be detrimental if provided as an adjunctive anti-infective therapy. Taken together, these observations highlight the role of coagulase and agglutinin activity in driving S. aureus virulence, and provide the rationale for development of new therapies that target these processes to improve infection outcomes.

In this issue of Journal of Thrombosis and Haemostasis, Negrón et al (11) investigate the therapeutic potential of a human fibrinogen splice variant, fibrinogen γ’, as a novel treatment for S. aureus septicemia. The rationale for this approach is based on prior studies focused on defining the molecular mechanism by which key S. aureus agglutinins interact with fibrinogen. Fibrinogen is a large glycoprotein produced by the liver that plays a key role in blood coagulation and in a number of other physiological processes including angiogenesis, wound healing and the response to infection (12) The protein is composed of 2 sets of 3 polypeptide chains (α₂β₂γ₂) which are arranged in a trinodular structure in which the N-termini of all chains are located in a central E-region and the C-termini are located (or extend from) in the D-region (13). The C-terminus of the γ-chain contains an AGDV sequence within the D-region which plays a role in the interaction of fibrinogen with platelets through integrin αIIbβ3 (14). The same AGDV sequence is exploited by S. aureus since its fibrinogen-binding protein Clumping Factor A (ClfA) binds to this sequence to promote clumping of bacteria and adhesion to host tissues (15, 16). Indeed, mice carrying a mutant form of fibrinogen that lacks the final 5 amino acids of the C-terminal γ chain (Fgg^Δ5), which includes the critical binding motif for ClfA, have increased survival following S. aureus intravenous challenge (17). Moreover, purified fibrinogen γ^Δ5 fails to support S. aureus clumping and adhesion. These data led the authors to hypothesize that provision of variant fibrinogen lacking key residues associated with ClfA binding could improve outcomes from S. aureus septicemia. This hypothesis is first tested by intravenously infecting WT, homozygous Fgg^Δ5/Δ5, and heterozygous Fgg^WT/Δ5 mice with S. aureus and ascertaining whether the presence of variant fibrinogen, even at the reduced levels produced by heterozygous mice, could improve survival. Consistent with the hypothesis, both Fgg^Δ5/Δ5 and Fgg^WT/Δ5 mice displayed increased survival relative to WT mice following S. aureus inoculation. This in vivo observation was correlated to reduced ClfA-mediated clumping of S. aureus in fibrinogen from Fgg^WT/Δ5 mice in vitro.

Since a 50% reduction in the C-terminal γ-chain ClfA-binding sequence in Fgg^WT/Δ5 mice was sufficient to improve sepsis outcomes, the authors next queried whether prophylactic or therapeutic reconstitution of fibrinogen deficient mice with human variant forms of fibrinogen lacking the key ClfA binding residues could improve outcomes of S. aureus bacteremia. A natural human splice variant fibrinogen γ’ was chosen for these experiments, as this variant has the final AGDV sequence of the γ-chain replaced with a 20 amino acid substitution (18). This variant fibrinogen occurs in around 12% of all circulating human fibrinogen, and shows altered clot structure through changes in fibrin polymerization (19). Fibrinogen-deficient and WT mice reconstituted with purified human fibrinogen γ-γ’ heterodimers (phFibγ’-γ) displayed a significant survival advantage over mice reconstituted with phFibγ-γ homodimers following subsequent challenge with a lethal dose of S. aureus. The increased survival correlated with decreased bacterial burdens in the kidneys, heart, and lung. Interestingly, this improved survival was not directly correlated to quantitative changes in ClfA-dependent clumping or adhesion in vitro when comparing phFibγ’-γ and phFibγ-γ. However, phFibγ’-γ elicited qualitatively smaller clumps when incubated with S. aureus. Finally, the authors test whether administration of phFibγ’-γ to fibrinogen mutant mice 30 minutes after bacterial inoculation could improve survival during S. aureus sepsis. Mice treated with phFibγ’-γ had a significant survival advantage over mock treated mice or mice treated with phFibγ-γ. Collectively, these data establish the preclinical feasibility of therapeutic administration of naturally occurring human fibrinogen variants to improve outcomes of S. aureus sepsis.

The data presented by Negrón and colleagues have translational significance, and thus are worthy of future studies to more closely model the expected clinical applications. Although therapeutic administration of phFibγ’-γ improves sepsis outcomes in fibrinogen-deficient mice, it will be important to determine 1) the temporal window for successful therapeutic intervention using phFibγ’-γ, 2) the efficacy of phFibγ’-γ in the context of standard of care antibiotic therapy, and 3) the efficacy of phFibγ’-γ in mice with various levels of fibrinogen depletion, rather than deficiency. Future studies should also further define the mechanism by which phFibγ’-γ improves survival in the setting of S. aureus infection. Although qualitative differences in clumping were observed when incubating S. aureus with phFibγ’-γ versus phFibγ-γ, there were no significant quantitative differences in clumping or adhesion between these two fibrinogen preparations. As the authors discuss, the C-terminal portion of fibrinogen γ, which is conserved in humans and mice, mediates interactions with platelets, and human fibrinogen γ’ can sequester and inhibit the activity of thrombin. Accordingly, a role for altered coagulation or platelet binding cannot be ruled out at this time. Future studies are therefore necessary to confirm the mechanistic basis for protection from sepsis mortality in mice treated with phFibγ’-γ. One final caveat is that the γ’ splice variation does not occur in mice, and while the lack of the AGDV sequence on injected human fibrinogen γ’ protects mice from S. aureus infection, translation to the protection against infection in humans (or other species) will need to be confirmed in future clinical studies.

S. aureus is one of the leading causes of bloodstream infection, and mortality approaches 30% even in the setting of appropriate antibiotic treatment (20). New treatment strategies, such as those that disrupt the activity of key S. aureus virulence factors, are desperately needed. The study by Negrón and colleagues highlights the clinical value of targeting the interaction between S. aureus and fibrinogen, a crucial component of the pathogenesis of this organism. Because fibrinogen γ’ is a natural human splice variant, and because patients with septicemia often receive treatment with blood products such as fresh frozen plasma to correct coagulopathy, provision of fibrinogen γ’ is worthy of further clinical evaluation.