Lipopolysaccharide and Coagulation Factor XII: Biophysics of Contact Activation in Infection

Abstract

Lipopolysaccharide (LPS), a key component of the outer membrane of Gram-negative bacteria, is well-known for its role in triggering inflammation via innate immune receptors. However, evidence suggests that LPS can influence coagulation, in part through activation of the contact pathway. Recent studies from our group and others demonstrate that the supramolecular organization and physicochemical properties of LPS—such as aggregate size, surface charge, and chemotype—critically determine the ability of LPS to activate coagulation factor XII (FXII). While monomeric LPS can modulate FXII activity, only aggregated forms of LPS (e.g., micelles) function as a procoagulant surface, initiating contact activation. This review synthesizes current knowledge on LPS structural heterogeneity and explores how the biophysical properties of LPS govern supramolecular assembly in aqueous environments, ultimately dictating interactions with the contact activation pathway. We further discuss the possible mechanisms by which LPS-driven FXII activation contributes to thromboinflammatory disorders, including disseminated intravascular coagulation and sepsis-associated vascular leakage. Finally, we highlight novel therapeutic strategies—from FXIIa inhibitors to molecules that disrupt LPS supramolecular structures—as potential interventions to mitigate coagulation-driven pathology during bacterial infections. These insights not only reflect our growing understanding of infection-associated thrombosis but may also pave the way for targeted therapies in sepsis and other thromboinflammatory conditions.

Infection and thrombosis are closely linked processes that share common molecular pathways and mutually reinforce one another.¹ During bacterial infections, components of the microbial surface activate host immune and hemostatic responses, leading to a pathological state known as thromboinflammation.² This interplay becomes particularly evident in sepsis and severe Gram-negative bacterial infections, as dysregulated coagulation leads to disseminated intravascular coagulation (DIC), microvascular thrombosis, and multiorgan dysfunction.^3,4

In Gram-negative sepsis patients, growing evidence indicates that procoagulant pathways are activated by bacterial-derived molecules. Among these, lipopolysaccharide (LPS)—a key component of the outer membrane of Gram-negative bacteria—is a potent modulator of host responses.^5,6 The immunostimulatory roles of LPS have been mainly established, with the role of LPS in coagulation having historically been solely assigned to the induction of tissue factor (TF) expression.^7,8 This mechanism is highly relevant in sepsis, as the extrinsic, TF-dependent pathway plays a central role in sepsis-associated coagulopathy.^7,8 In particular, LPS is known to induce TF expression on circulating monocytes, whereas the presence of endothelial TF in vivo remains controversial and may, in part, reflect the binding of TF-positive extracellular vesicles (EVs) derived from monocytes. Collectively, these findings have established TF as the pathological mechanism of action by which LPS contributes to coagulation in bacterial sepsis.⁹

Activation of the contact pathway of coagulation has been recently appreciated as an important contributor to immunothrombosis, alongside established TF-mediated mechanisms.^10–12 The contact pathway is triggered by negatively charged surfaces foreign to the circulation, in contrast to the TF pathway. LPS, by forming supramolecular aggregates, structurally mimics negatively charged surfaces to activate coagulation factor XII (FXII), providing a mechanistic link between bacterial structure and the contact pathway.¹³ Together, the TF and contact pathways work in parallel, with TF driving thrombin generation and the contact pathway adding an inflammatory component, both contributing to the dysregulated coagulation during systemic infection.

Even so, the relative importance of these pathways depends on the type of infection.¹⁴ Gram-positive bacteria often provoke stronger contact activation than Gram-negative species, largely because they release long-chain polyphosphates that potently activate FXII.^15,16 In Gram-negative sepsis, TF-driven thrombin generation is generally considered the dominant mechanism.⁸ Taken together, the data suggest that LPS contributes to DIC as part of a broader picture, with TF playing the dominant role in coagulation during Gram-negative sepsis, while polyphosphates from Gram-positive bacteria can strongly trigger the contact pathway and amplify thrombin generation.

This review focuses on the biophysical properties of LPS that underlie the ability to engage and activate the contact pathway of coagulation. We present a summary of recent studies that demonstrate how the surface characteristics, aggregation state, and supramolecular structure of LPS regulate the activation of FXII. We combine insights from structural biology, colloidal chemistry, and hemostasis that have elucidated the role of LPS as a contact pathway activating surface and discuss the implications of these interactions in the context of infection-driven thrombosis and potential therapeutic intervention.

Overview of the Contact Activation Pathway of Coagulation

The contact activation pathway of coagulation initiates when plasma proteins bind to typically negatively charged biological or synthetic surfaces.^17,18 The contact activation pathway consists of the zymogens FXII and factor XI (FXI), plasma prekallikrein (PK), and the non-enzymatic cofactor high-molecular-weight kininogen (HK).¹⁹ Upon contact with anionic surfaces, FXII autoactivates into an enzymatically active form, FXIIa, which then cleaves PK to generate plasma kallikrein (PKa).^18,20 The ability of PKa to then activate FXII creates a potent positive feedback loop.²¹ As a non-enzymatic cofactor, HK binds both FXII and PK, and catalyzes both binding to anionic surfaces and their binding partners to accelerate FXII autoactivation and PK conversion to kallikrein.²²

Although thought to be dispensable for normal hemostasis—evidenced by the lack of spontaneous bleeding in FXII-deficient individuals—the contact activation pathway plays roles in thrombosis, inflammation, and host defense.^23,24 FXIIa triggers the intrinsic coagulation cascade via FXI activation, leading to thrombin generation and fibrin formation, primarily under pathological conditions.²⁵ Additionally, kallikrein cleaves HK to generate bradykinin (BK), a potent inflammatory mediator that enhances vascular permeability and contributes to hypotension in septic shock.^10,26

Activation of the contact pathway is highly surface-dependent, typically occurring following exposure to negatively charged materials.^27,28 Classical activators of FXII include inorganic surfaces (e.g., glass),²⁹ but a range of biologically relevant activators have been identified, including polyanionic macromolecules (e.g., polyphosphates, nucleic acids),^30–32 EVs,^33,34 misfolded proteins,³⁵ and synthetic nanoparticles.^36,37 Moreover, bacterial surfaces—including cell membrane components such as lipoteichoic acid from Gram-positive bacteria and LPS from Gram-negative bacteria—have been identified as physiological contact pathway activators.^38,39 These surfaces share common physicochemical properties such as high negative charge density and supramolecular organization, both of which are essential for FXII recognition. Understanding how surface properties drive contact pathway activation is crucial for elucidating the molecular mechanisms that connect infection to thromboinflammatory complications.

Structure and Biophysical Properties of Lipopolysaccharide

LPS is an essential component of the outer membrane of Gram-negative bacteria and is critical for bacterial viability, membrane integrity, and host—pathogen interactions.^40–42 In addition to being an endotoxin that is recognized by the innate immune system, LPS has unique properties that dictate its biophysical behavior, which influences its supramolecular organization and ability to interact with host proteins, including the coagulation factors of the contact activation pathway.^40,43

General Structure

In Gram-negative bacteria, the outer leaflet of the outer membrane is formed by the structurally complex amphiphilic molecule known as LPS. It is formed of three main components: The O-antigen, the core oligosaccharide, and lipid A.⁴⁰ The unique structural and chemical characteristics of each component have their own set of structural and chemical properties that influence the immunostimulatory activity, organization of the molecule, and interactions with host systems.⁵

At the base of the molecule, lipid A anchors LPS into the bacterial outer membrane. Lipid A itself is made up of a disaccharide of D-glucosamine linked via β(1→6), which is usually phosphorylated at the 1 and 4′ positions and is acylated with four to seven saturated fatty acid chains.^40,44 The number, length, and positioning of the acyl chains vary between bacterial species and affect both the immunostimulating properties of the molecule and the likelihood that the molecule will associate in supramolecular aggregates in aqueous environments.⁴⁴ Attached to lipid A is the core oligosaccharide, which connects to the O-antigen and is called the core oligosaccharide, as it is made up of inner and outer core. The inner core is quite conserved and includes rare sugars, including 3-deoxy-D-manno-oct-2-ulosonic acid (Kdo) and L-glycero-D-manno-heptose, both of which are essential for bacterial existence.^45,46 The variability of the outer core includes common monosaccharides (glucose, galactose, glucosamine, and N-acetylglucosamine), which differ in order between species, influencing both charge and degree of hydrophilicity of the molecule.^40,47 The O-antigen is a polysaccharide, with repeating oligosaccharide units, which can be highly variable in length and composition (even among strains of the same species). This variability is the foundation for serologically classifying bacterial strains and is one of the primary mechanisms for avoiding an immune response.⁴⁰ Depending on the O-antigen, LPS can be classified as smooth (with the full-length O-antigen), rough (lacking the O-antigen), or semirough.^{40,43,47–49} Semirough LPS contains the core region and part of the outer core but lacks the O-antigen. We have found that these truncated formats are particularly useful for studying how the core portion leads to supramolecular assembly and functioning of LPS.¹³ Figure 1A provides an overview of the structural organization of LPS. The schematic also highlights the different LPS chemotypes—smooth, rough, and semirough—which are important to consider when interpreting the biophysical and biological variations associated with LPS supramolecular structure.

Fig. 1.

Structural organization and heterogeneity of LPS from Gram-negative bacteria. (A) Schematic representation of LPS architecture embedded in the outer membrane of Gram-negative bacteria. LPS is composed of three main domains: Lipid A, which anchors the molecule to the membrane; a core oligosaccharide containing conserved sugars such as Kdo and heptoses; and the highly variable O-antigen polysaccharide. LPS molecules are classified into rough (R-LPS), semirough (SR-LPS), and smooth (S-LPS) chemotypes based on the presence or absence of outer core and O-antigen domains. (B) LPS displays extensive heterogeneity at multiple levels. Chemical variability arises from differences in saccharide composition, acylation patterns, and phosphorylation states. Supramolecular structures formed in aqueous environments include monomers, micelles, and bilayers, depending on concentration, ionic strength, and LPS chemotype. (C) Intermolecular interactions involve electrostatic bridges (e.g., Ca²⁺-mediated coordination between negatively charged phosphate and carboxylate groups) and protein-binding interfaces that depend on both the aggregate state and exposed functional groups. LPS, lipopolysaccharide. Source: [rerif]: Created in BioRender. Silva, A. (2025) https://BioRender.com/v5oaa3x.

Thermodynamics of Lipopolysaccharide Self-Organization

LPSs have a natural tendency to self-assemble into complex supramolecular structures in aqueous environments, owing to their unique amphiphilic character. The ability of LPSs to spontaneously form such structures arises from the balance between hydrophobic forces driving aggregation of the lipid A acyl chains and the hydrophilic, negatively charged oligosaccharide core and O-antigen regions, which favor solvation.^50,51 Depending on environmental conditions—including concentration, ionic strength, pH, and the presence of stabilizing divalent cations, such as Ca²⁺ or Mg²⁺—LPS assemblies can range from small micelles to larger, bilayer-like structures, multilamellar aggregates, or vesicles that resemble biological membranes (Fig. 1B).^52–54

A key parameter defining the onset of LPS aggregation is its critical micelle concentration (CMC). The CMC represents the threshold concentration at which monomeric LPS molecules in solution begin to associate into oligomeric preaggregates, eventually forming larger supramolecular structures such as micelles or other higher-order assemblies.^{50,51,55–57} Reported CMC values for LPS vary depending on bacterial source and environmental conditions, but typically fall within the low micromolar range. For example, LPS from Escherichia coli O111:B4 has a CMC of approximately 1.3 to 1.6 μM in saline solutions.⁵⁷ The presence of divalent cations such as Ca²⁺ or Mg²⁺ can significantly lower the CMC, as these cations neutralize charge repulsion between LPS molecules and promote aggregation at lower concentrations.^54,58 Understanding the CMC is essential for interpreting the biological activities of LPS, since the interaction of LPS with host proteins and cells differs markedly between monomeric and aggregated states.^56,59–62 However, determining a precise CMC for LPS is inherently difficult. This is because LPS does not form discrete, uniform micelles like classical surfactants; instead, LPS assembles into heterogeneous and dynamic aggregates with broad size distributions, whose structures and properties continuously evolve with concentration and ionic environment.⁶³ Moreover, the transition from monomers to aggregates is gradual rather than sharply defined, making it hard to assign a single concentration threshold. The polydisperse and flexible nature of LPS assemblies, combined with sensitivity to small changes in experimental conditions (e.g., buffer composition, pH, cation type), means that reported CMC values for LPS in particular are approximate and highly context-dependent rather than universal constants.^57,63

From a thermodynamic perspective, this self-assembly process minimizes the system’s free energy.⁶³ The hydrophobic effect provides a significant entropic gain by releasing ordered water molecules surrounding the lipid A tails, while favorable enthalpic contributions arise from ionic and hydrogen bonding interactions involving the oligosaccharide regions and cation bridges.^64,65 These divalent cations function as electrostatic bridges between LPS molecules by binding to the anionic phosphate groups of lipid A and the carboxyl groups of inner core sugars (e.g., Kdo), neutralizing charge repulsion and promoting tighter packing and stabilization of aggregates.^50,63,64,66

The precise supramolecular architecture of LPS aggregates is dictated by a combination of molecular geometry and electrostatic interactions.^54,67 For example, rough LPS chemotypes—characterized by exposed phosphate groups and truncated core oligosaccharides—form densely packed, condensed aggregates in the presence of divalent cations. In contrast, smooth LPS with extended O-antigen chains form more hydrated, loosely organized aggregates due to steric hindrance from the bulky polysaccharide moieties. Interestingly, despite their more compact molecular organization, rough LPS aggregates exhibit larger hydrodynamic sizes compared with their smooth counterparts, likely due to differences in solvation and supramolecular architecture.^53,58,68 Aggregate size, curvature, surface charge, and hydration shell all contribute to the biological activity of LPS structures, influencing their ability to interact with host molecules, including coagulation factors, complement proteins, antimicrobial peptides, and other immune mediators.^62,66,68,69 The negatively charged components of LPS (phosphates in lipid A and acidic sugars in the core) enable electrostatic interactions with host proteins. Crucially, when LPS aggregates form organized structures, they may provide ideal surfaces to activate FXII via the contact pathway of coagulation.

Interfaces for Protein Binding

The supramolecular state of LPS plays a critical role in determining its interactions with plasma proteins, but both monomeric and aggregated forms engage with proteins through distinct mechanisms. Monomeric LPS molecules, exposing hydrophobic lipid A regions alongside hydrophilic sugar moieties, can transiently interact with proteins by inserting their lipid tails into hydrophobic pockets or engaging sugar-recognizing domains.^40,70 These interactions may modulate protein function through allosteric or electrostatic effects and are often observed with lipid-binding proteins such as lipopolysaccharide-binding protein (LBP), CD14, or lipoproteins,^71–74 which act as molecular carriers in vivo. However, monomers are thermodynamically unstable in aqueous solutions, such as plasma, and their biological activity is generally limited compared with aggregate-bound forms.^56,59,61,62

In contrast, aggregated LPS—such as micelles, vesicles, or multilamellar structures—provides extended, multivalent surfaces with complex physical characteristics, including defined hydrodynamic size, curvature, and spatial arrangement of charged and hydrophobic domains.⁵⁹ These supramolecular interfaces promote stronger protein adsorption, facilitate multivalent interactions, and can induce conformational changes essential for biological activity. Aggregates may function as a contact pathway activator by presenting collective displays of charged and hydrophobic motifs. Figure 1C illustrates how differences in LPS aggregate morphology and surface properties influence protein binding.

Importantly, the morphology of LPS aggregates—and thus their functional interactions—is influenced by factors such as chemotype, concentration, and ionic environment.^54,69 The presence of divalent cations (e.g., Ca²⁺ or Mg²⁺) modulates aggregate properties by altering curvature, size, and net surface charge,⁶⁸ further regulating how LPS engages FXII(a) and other contact factors, including FXI(a) and PK(a).

Nanoscale to Networks: How Lipopolysaccharide Organization Drives Coagulation

We have recently shown that LPSs from the Gram-negative bacterial cell envelope are potent activators of the contact phase of coagulation, connecting microbial infection to thromboinflammatory outputs.⁷⁵ Indeed, in extension of initial work performed several decades ago, which demonstrated for the first time that endotoxins are able to initiate coagulation through the activation of FXII,³⁹ our recent data have demonstrated that LPS-induced FXII activation is strongly influenced by the LPS supramolecular organization and by the physicochemistry of the endotoxin structure.⁷⁵

Indeed, only aggregated LPS species (e.g., micelles) have been shown to promote FXII autoactivation. In contrast, monomeric LPS lacks this capacity.⁷⁵ These findings underscore the relevance of LPS supramolecular organization for procoagulant function. In addition, LPS structure—activity relationship studies show that LPS chemotype is a major determinant of activity, mediated by differences in lipid A acylation, oligosaccharide composition, and the resulting aggregate morphology. For instance, semirough LPS derived from E. coli O26:B6 is considerably more potent in activating FXII than smooth (O111:B4) or rough (Rd2) LPS, indicating that the presence and size of polysaccharide chains can influence access to negatively charged regions crucial for FXII binding.⁷⁵

Biophysical properties of LPS aggregates, including surface charge, hydrophobicity, and aggregate structure, are important parameters that influence the ability of LPS aggregates to induce contact pathway activation. The electrostatic interactions are mostly determined by the phosphate and carboxyl groups of the sugar moieties, whereas lipid A participates in the hydrophobic packing, controlling the aggregate curvature and stability. Together, these parameters define how FXII and other contact proteins interact with the LPS surface. Comparisons to both synthetic polyanionic surfaces and known endogenous FXII activators indicate that LPS provides a unique mix of structural complexity and chemical diversity, making it an especially potent and multifaceted procoagulant platform.^36,76,77

From a mechanistic perspective, LPS aggregates appear to function as a dynamic “procoagulant platform” that facilitates the members of the contact activation pathway to arrange themselves spatially.⁷⁸ The resultant spatial proximity may catalyze interactions between FXII and other molecular components, thereby creating an environment that is favorable for FXII autoactivation. Additionally, FXII may experience allosteric changes as a result of binding to LPS, which could enhance the catalytic efficiency of FXIIa. Divalent cations such as Ca²⁺ are expected to stabilize these aggregates by bridging negatively charged residues, thereby fortifying the scaffold structure and potentially altering the accessibility of FXII-binding sites.⁵⁸ This dual mechanism, which combines scaffold assembly and conformational activation, offers a compelling framework for understanding how LPS activates the contact pathway in the settings of infection and inflammation.

In our latest work, we systematically investigated the mechanisms by which LPS chemotypes and supramolecular structures control FXII activation and FXIIa catalytic activity.¹³ For FXIIa, we found that E. coli O26:B6 LPS had a two-phase effect: FXIIa catalytic activity was suppressed at higher concentrations and increased at low concentrations. These effects were associated with structural changes in the organization of LPS, from monomers to micelles and then to bilayers, suggesting that the enzymatic function of FXIIa is dynamically influenced by the structural state of LPS. Of the supramolecular forms, micellar LPS was the only one to directly activate FXII to FXIIa. However, vesicle-like LPS, which is stabilized by Ca²⁺ ions that neutralize phosphate charges and reduce electrostatic repulsion, was unable to support the proteolytic autoactivation of FXII due to changes in surface curvature or interaction site masking.^13,75

Furthermore, rather than triggering FXII activation, monomeric LPS regulated the enzymatic activity of FXIIa, FXIa, and PKa. This modulation appears to be allosteric and could involve brief interactions with exosites or conformational elements that affect substrate recognition or catalysis.⁷⁵ These findings show that even non-activating LPSs can fine-tune protease activity and affect the reaction kinetics of the contact activation pathway. Together, our results provide a mechanistic understanding of how bacterial structural diversity during infection regulates the host coagulation cascade and show that the procoagulant phenotype of bacterial components, such as LPS, is intimately dependent upon their supramolecular plasticity.

The current findings establish a mechanistic basis for dysregulation of the coagulation cascade during endotoxemia and septic shock. Many Gram-negative pathogens, including Pseudomonas aeruginosa, Klebsiella pneumoniae, and E. coli, exhibit phase variation between smooth and rough LPS phenotypes.⁴⁰ Our results suggest that infections with rough or semirough LPS strains may be more likely to result in dysregulated coagulation due to increased FXII activation. The observed variability in LPS procoagulant activity could account for differential thrombotic risk among bacteremic patients with similar infection severity.^79–81 These discoveries could enable clinicians to tailor treatments based on LPS chemotype. For infections with semirough LPS—which strongly activate FXII—targeted anticoagulants (e.g., FXII inhibitors) or polyanion-neutralizing therapies (e.g., cationic peptides) may be optimal. In contrast, infections expressing smooth LPS, which weakly activate FXII, may not require as strong or concentrated doses of an inhibitor. By defining how LPS structure influences FXII activation, these findings bridge coagulation and host defense, opening new avenues for treating infection-driven thrombosis. Figure 2 illustrates the key steps of this process, including LPS release from the bacterial envelope, its supramolecular organization into micelles or vesicle-like structures, and the subsequent interactions with components of the contact pathway that lead to FXII activation and thromboinflammatory responses. Figure 2 highlights how LPS structural diversity and aggregate morphology modulate procoagulant activity, providing a visual framework for understanding these mechanisms in the context of infection.

Fig. 2.

LPS release, supramolecular organization, and interaction with the contact pathway of coagulation. (A) LPS is released into the bloodstream during Gram-negative infections either through bacterial lysis or after phagocytosis and degradation by immune cells. (B) Once in circulation, LPS self-assembles into different supramolecular structures—monomers, micelles, or bilayers—depending on its concentration, chemotype, and local physicochemical conditions. (C) These distinct LPS forms differentially interact with contact pathway proteins. Monomeric LPS can bind FXII and other factors but does not initiate coagulation; instead, it may modulate enzymatic activity. Micellar LPS promotes FXII autoactivation and supports propagation of the intrinsic coagulation and kallikrein—kinin pathways. Vesicle-like LPS structures reduce FXII activation potential and may represent an endogenous regulatory mechanism that limits procoagulant activity. FXII, coagulation factor XII; LPS, lipopolysaccharide; PKa, plasma kallikrein. Source: [rerif]: Created in BioRender. Silva, A. (2025) https://BioRender.com/bx0wj65.

Therapeutic Implications of Lipopolysaccharide—Contact Activation Pathway Interactions

The relationship between LPS and the coagulation system is one of the primary causes of Gram-negative infection complications.^14,26 By triggering FXII and the contact activation pathway, LPS contributes to DIC, which causes widespread clotting and organ damage in extreme situations like sepsis.³ LPS aggregates can contribute to thrombosis by serving as surfaces that catalyze clot formation.^82–84 Using a translational sepsis model, Obi et al showed that endotoxemia promotes early large-vein stasis venous thrombosis in a manner dependent on TLR4 (toll-like receptor 4) and ICAM-1 (intercellular adhesion molecule 1), and interestingly, this effect was enhanced under neutropenic conditions. These results highlight TLR4 and ICAM-1 as potential targets for preventing or treating sepsis-associated thrombosis.⁸⁴ Notably, Wang highlighted a bell-shaped dose—response in venous thrombosis to LPS, which may be related to enhanced endogenous thrombolysis at higher LPS doses.⁸² Importantly, higher LPS concentrations in solution increase the likelihood of forming molecular aggregates, which, in vitro, can directly influence FXII activation and show a bell-shaped profile. Studies show that the supramolecular form of LPS—whether as monomers, micelles, or bilayers—affects FXII activation in distinct ways. Micelles are capable of directly activating FXII, monomers appear to modulate the activity of FXIIa, and bilayers show minimal effect. The presence of calcium further stabilizes LPS aggregates and can modify their surface properties, which in turn influences FXII activation.^13,39,75 Importantly, we demonstrate that FXII activation by LPS follows a bell-shaped curve, with maximal activation observed when LPS is in the micellar form.¹³ At higher LPS concentrations, FXII activation decreases, likely due to the excessive surface area of the aggregates. This phenomenon has also been supported by mathematical modeling from Galochkina et al.⁷⁸ These results indicate that the structure and concentration of LPS aggregates are critical determinants of FXII activation. However, in complex biological systems, the stability of these aggregates and their in vivo effects remain unclear and are still under investigation.

In addition to coagulation, the contact activation pathway is intricately linked to inflammation through the kallikrein—kinin system. When PKa is activated, BK is released, leading to blood vessel permeability and inflammation.²¹ Therefore, LPS-driven contact activation worsens tissue damage by promoting swelling and leakage. Furthermore, there is growing evidence that the contact activation pathway and the complement system, another immune pathway activated by LPS, work together to increase inflammation and clotting, which leads to more severe disease states.^4,85–87

These findings point to promising new therapeutic possibilities. FXII inhibitors are being explored as potential treatments for septic coagulopathy and thrombotic complications.² In particular, therapies targeting FXIIa, such as Garadacimab (CSL312) and other specific monoclonal antibodies, are currently under investigation for their ability to reduce both inflammation and thrombosis, while preserving normal hemostatic function.^88,89 Promising alternatives for traditional anticoagulants include monoclonal antibodies that target FXI or PK, or inhibitors of kallikrein activity, which may reduce pathological clotting while preventing bleeding risks.^24,90–95

Under conditions of weak FXII activation, alternative therapeutic strategies may be needed, particularly if the inhibition of FXII alone does not yield sufficient anti-inflammatory or antithrombotic effects. In such cases, targeting downstream components, such as PK or bradykinin B2 receptors, may be effective, as they can still be activated even by low concentrations of FXIIa, especially in inflammatory settings.⁹⁶ Complement inhibitors may also provide benefit, since low levels of LPS can trigger the complement cascade, contributing to inflammation and vascular injury. Blocking components like C3 or C5 can attenuate the synergistic inflammatory response between the contact and complement systems,⁹⁷ although to date, none are approved for treatment of infection complications outside of atypical hemolytic uremic syndrome.⁹⁸ Modulating the innate immune response to LPS directly—using TLR4 antagonists or agents that enhance LPS detoxification, such as recombinant LBP⁹⁹ or alkaline phosphatase¹⁰⁰—may further reduce systemic effects even when coagulation activation is subtle.

Another strategy involves targeting the biophysical properties of LPS itself—modulating its aggregation state or blocking its interaction surfaces could reduce its ability to activate the contact pathway.¹⁰¹ Novel strategies include the use of surfactant-like substances that interfere with LPS supramolecular assembly or nanoparticle decoys that imitate LPS surfaces to competitively inhibit protein binding.¹⁰² Limiting coagulation activation has also been demonstrated to be possible with synthetic peptides or small molecules that block LPS—phosphate group interactions.^103–105

Conclusion

This review clarifies a significant change in our understanding of LPS. Although LPS has historically been thought of only as an endotoxin, this molecule functions as a complex, structured surface with wide-ranging effects, including catalyzing activation of the contact pathway of coagulation. A key element in this process is the supramolecular organization of LPS, which influences inflammation and thrombosis during infections. Despite progress, some significant concerns remain unresolved: Which particular properties of LPS, such as surface charge, aggregate size, or acylation patterns, are the best indicators of its capacity to activate FXII? Are the monomeric forms of LPS involved in controlling coagulation factor activity? Can we take these molecular insights and turn them into tools to evaluate the thrombogenic potential of different bacterial strains or to create targeted treatment strategies? A collaborative approach involving hemostasis, microbiology, and biophysics will be necessary to answer these questions. By taking an interdisciplinary approach, we can better understand the processes involved and discover new ways to combat thrombosis linked to infections.