Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 1.
Published in final edited form as: J Thromb Haemost. 2022 Dec 22;21(3):467–479. doi: 10.1016/j.jtha.2022.10.005

Tissue Factor Pathway Inhibitor Is a Potential Modifier of Bleeding Risk in Factor XI Deficiency

Stéphanie E Reitsma 1, Lori A Holle 1, Emma G Bouck 1, Dougald M Monroe 2, Alan E Mast 3, John Burthem 4,5, Paula H B Bolton-Maggs 6, Gillian N Gidley 5,7, Alisa S Wolberg 1,*
PMCID: PMC10111213  NIHMSID: NIHMS1888498  PMID: 36696199

Abstract

Background.

Factor XI (FXI) deficiency is associated with increased bleeding risk in some individuals. Neither FXI levels nor clinical clotting assays predict bleeding risk. Compared to controls, FXI-deficient bleeders have reduced clot formation, decreased fibrin network density, and increased susceptibility to fibrinolysis. Tissue factor pathway inhibitor (TFPI) was recently implicated as a modifying factor in individuals with bleeding of unknown cause.

Objective.

Determine the potential of TFPI to modify bleeding risk in FXI-deficient individuals.

Methods.

TFPI effects on thrombin generation and clot formation, structure, and fibrinolysis in FXI-deficient plasma were measured in vitro in the absence or presence of inhibitory anti-TFPI antibody or exogenous recombinant TFPI. Total plasma TFPI was measured in 2 independent cohorts of controls and FXI-deficient individuals classified as bleeders or non-bleeders (cohort 1: N=10 controls, N=16 FXI-deficient; cohort 2: N=48 controls, N=56 FXI-deficient), and correlated with ex vivo plasma clot formation and fibrinolysis parameters associated with bleeding risk.

Results.

In an in vitro model of FXI deficiency, TFPI inhibition enhanced thrombin generation and clot formation, increased network density, and decreased fibrinolysis, whereas increased TFPI had the opposite effects. Compared to controls, plasma from FXI-deficient bleeders had higher TFPI. Total plasma TFPI concentrations correlated with parameters from ex vivo clotting and fibrinolysis assays that differentiate FXI-deficient bleeders and non-bleeders.

Conclusions.

Coagulation and fibrinolysis parameters that differentiate FXI-deficient non-bleeders and bleeders are altered by plasma TFPIα. Total plasma TFPI is increased in FXI-deficient bleeders. TFPI may modify bleeding risk in FXI-deficient individuals.

Keywords: Factor XI, fibrinolysis, tissue factor pathway inhibitor, TFPI, fibrin

INTRODUCTION

Factor (F)XI deficiency is an autosomal disorder. Individuals with heterozygous mutations have moderately-reduced FXI (20–60 IU/dL), whereas people with homozygous or compound heterozygous mutations have low FXI (<15–20 IU/dL).[1, 2] Severe deficiency is present in 1:1,000,000 individuals. Spontaneous bleeding is rare; however, some individuals with FXI deficiency present with tissue-specific bleeding following surgery or injury, predominantly at sites with high fibrinolytic activity (mouth, nose, genitourinary tract), or menorrhagia.[1–3] Interestingly, individuals with similarly reduced FXI antigen or activity or the same F11 mutation can exhibit variable bleeding tendencies, even within family members.[4–8] Some individuals are asymptomatic even after trauma, while others experience bleeding with trauma, or bleeding that begins several hours or even days following trauma. Neither FXI antigen nor activity correlate with clinical risk of bleeding, and activated partial thromboplastin time (APTT) assays do not predict bleeding risk.[8] Consequently, it is difficult to anticipate which FXI-deficient individuals have increased bleeding risk and therefore, challenging to effectively manage their clinical care.[3] Assays that uncover hemostatic mechanisms in FXI-deficient individuals and predict bleeding risk may improve treatment strategies.

Previous studies have explored the ability of global coagulation assays to identify FXI-deficient individuals with bleeding risk. Whole blood thromboelastography does not differentiate FXI bleeders from non-bleeders.[9, 10] However, assays that use corn trypsin inhibitor (CTI)-treated platelet-rich or platelet-poor plasma clotted with a low concentration of tissue factor (TF) have shown promise in differentiating these phenotypes. Several groups have identified parameters from thrombin generation assays that associate with bleeding risk, although the sensitivity differs between studies.[11–13] We showed that in vitro assays that assess clot formation and susceptibility to fibrinolysis can also distinguish FXI-deficient phenotypes.[14, 15] Compared with controls or non-bleeders, CTI-treated plasmas from FXI-deficient bleeders show significantly reduced clot formation, decreased fibrin network density, and increased susceptibility to fibrinolysis. A model incorporating a prolonged APTT with the clot formation rate has good ability to differentiate bleeders from non-bleeders (0.839 [0.715–0.963], area under the receiver operating characteristic curve [confidence interval]).[15] Importantly, the ability of all of these in vitro assays to differentiate FXI-deficient bleeders and non-bleeders suggests a major determinant of bleeding risk in FXI deficiency is present in plasma.

TF pathway inhibitor (TFPI) is a serine protease inhibitor. The TFPIα isoform circulates in plasma and platelets and consists of 3 Kunitz domains and a C-terminal tail containing a series of basic amino acids. The TFPIβ isoform expressed on endothelial cells consists of 2 Kunitz domains and a C-terminal glycosylphosphatidyl inositol anchor. Both isoforms bind the TF/FVIIa/FXa complex and inhibit TF/FVIIa-dependent generation of FXa.[16] TFPIα also binds partially cleaved FV(a) and inhibits early prothrombinase activity.[16–18] Both of these activities downregulate thrombin generation and fibrin formation. In addition, previous studies have described bidirectional links between TFPI and activated FXI(a). By suppressing TF activity, TFPI enhances sensitivity to FXI-dependent feedback activation of the coagulation pathway.[19–21] In turn, FXIa cleavage of endothelial cell-bound TFPIß, as well as TFPIα released from platelets, inhibits TFPI and increases local TF activity.[22] Thus, interplay between FXI and TFPI may tune procoagulant activity across different settings.

TFPI has been implicated as a potential modifying factor in individuals with bleeding of unknown cause. In a cohort of 13 patients with an unclassified bleeding disorder and abnormal thrombin generation, MacDonald et al detected slightly, but non-significantly, increased free TFPI (P=0.06), and significantly increased TFPI activity.[23] In a cohort of 620 patients with mild-to-moderate bleeding tendency including patients with bleeding of unknown cause, Mehic et al detected significantly increased plasma TFPIα, which was associated with delayed thrombin generation.[24] These data are consistent with speculation that low plasma TFPI is protective in patients with factor V deficiency.[25] In a pilot study of 26 individuals (10 controls, 16 FXI-deficient individuals including 8 bleeders and 8 non-bleeders), we previously detected a small, but non-significant difference in TFPI among controls and FXI-deficient individuals.[14] However, the contribution of TFPI to fibrin formation, structure, and stability of FXI-deficient plasma clots, or its potential role as a modifier of bleeding risk in FXI-deficient individuals, was not explored further. Here we aimed to characterize the potential of TFPI to modify bleeding risk in FXI-deficient individuals.

METHODS

Materials.

Commercial FXI-deficient plasma was from HRF, Inc™ (Raleigh, NC, USA). FXI was from Haematologic Technologies, Inc (Essex Junction, VT, USA). Recombinant full-length TFPIα was produced in human embryonic kidney cells, purified on a heparin column, and characterized as described (Supplemental Figure 1).[26, 27] Recombinant TF (Innovin®) was from Siemens Healthcare Diagnostics (Newark, DE, USA). Reagents to measure thrombin generation (PPP-low, calibrator, and fluorescent substrate/CaCl2 [FluCa]) were from Diagnostica Stago (France). CaCl2 for clotting assays was from Sigma-Aldrich (St Louis, MO, USA). Phosphatidylcholine, phosphatidylethanolamine, and phosphatidylserine were from Avanti Polar Lipids (Alabaster, AL, USA). Large unilamellar phospholipid vesicles (41% phosphatidylcholine/44% phosphatidylethanolamine/15% phosphatidylserine) were made as described.[14] Recombinant tissue plasminogen activator (tPA) was from American Diagnostica, Inc (Stamford, CT, USA). AlexaFluor488-conjugated fibrinogen was from Molecular Probes (Eugene, OR, USA). Control isotype IgG antibodies were from Affinity Biologicals (Ontario, Canada).

Thrombin generation assays.

FXI-deficient plasma was pretreated with vehicle (20 mM HEPES pH 7.4, 150 mM NaCl) or FXI to achieve 0%, 10% (3 nM), or 100% (30 nM) FXI (final). Plasmas were considered to have normal (100%, 1.5 nM) endogenous total TFPI and then pretreated with recombinant TFPIα to achieve 150% (2.25 nM) or 200% (3 nM) final total TFPI, or an anti-TFPI-K1 domain antibody (2H8; 50 μg/mL) to model 0% total TFPI. Thrombin generation was measured by calibrated automated thrombography. Briefly, prewarmed plasma (40 μL) was incubated with 10 μL PPP-low reagent or thrombin calibrator for 10 min at 37°C in a 96-well round-bottom microtiter plate. Reactions were initiated by automatically dispensing 10 μL fluorogenic substrate and CaCl2 to each well. Final TF, phospholipid, fluorogenic substrate, and CaCl2 concentrations were 1 pM, 4 μM, 416 μM, and 16 mM, respectively. Thrombin parameters (lag time, time to peak, peak, and endogenous thrombin potential [ETP]) were calculated using Thrombinoscope software version 5.0.0.742.

Clotting and fibrinolysis assays.

Clot formation and fibrinolysis in the in vitro model of FXI deficiency were assessed using a turbidity assay. Briefly, FXI-deficient plasma was pretreated with FXI or TFPIα as described above. For the clotting reactions, TF (1:30,000 dilution of Innovin final, corresponding to ~0.25 pM final) was added to a well with phospholipids (4 μM final), followed by addition of prewarmed plasma (85% final). Reactions were initiated with addition of CaCl2 (10 mM final). Fibrinolysis reactions included tPA (0.5 μg/mL final). Changes in turbidity indicating fibrin formation and lysis, were measured at 405 nm at room temperature every 12 seconds for 120 minutes using a SpectraMax 384Plus plate reader (Molecular Devices, Sunnyvale, CA, USA). Clot formation parameters were defined as: lag time, time to the inflection point before turbidity increase; rate, slope of a line fitted to the maximum rate of turbidity increase (“Vmax”) using 5–10 points to determine the line (Softmax Pro 5.4, Molecular Devices); time to plateau/peak (TTP) was the time to the turbidity plateau (absence of tPA) or peak (presence of tPA); peak turbidity change was the maximum clot turbidity minus the starting turbidity. In fibrinolysis assays, the area under the curve was calculated using Graphpad Prism version 9.3.1.

Fibrin clot structure analysis.

Fibrin structure in the in vitro model of FXI deficiency was assessed using laser scanning confocal microscopy. FXI-deficient plasma pretreated with FXI or TFPIα as described above was spiked with AlexaFluor488-conjugated fibrinogen (80 μg/mL final, 2.6% of total fibrinogen). Clotting was triggered by recalcification (10 mM CaCl2 final) and addition of TF (1:30,000 dilution of Innovin final; ~0.25 pM) in the presence of phospholipids (4 μM final) in Laboratory-Tek II chamber cover glass slides and imaged as described.[28] Fibrin density was quantified by summing individual sections to create Z-projections and thresholding to visualize fibers and minimize background noise using ImageJ software (version 1.41o). The area covered by pixels above the threshold cutoff was determined using the ImageJ “Measure” function.

Subject characteristics.

Samples and data from two independent cohorts of FXI-deficient individuals and controls were studied.[14, 15] Cohort 1 was from a previous study of 26 individuals in Israel: 10 healthy controls with no personal or family history of thrombosis or bleeding disorders and no relevant medications, and 16 individuals with severe FXI deficiency (FXI:C≤9 IU/dL).[14] FXI-deficient individuals were divided into bleeders (N=8) and non-bleeders (N=8) based on history of bleeding following at least two separate tooth extractions performed without prophylaxis. Bleeders were defined as having oozing of blood for ≥1 hour after extraction, bleeding 24 hours after extraction, or return to the clinic or hospital to achieve hemostasis. Non-bleeders did not experience excessive bleeding.[14] Demographic characteristics for subjects in Cohort 1 were previously published.[14] Cohort 2 was from a previous study of 105 individuals in the United Kingdom: 48 healthy controls with no personal or family history of thrombosis or bleeding disorders and no relevant medications, and 57 individuals with severe (N=9; FXI:C≤15 IU/dL) or partial (N=48; FXI:C 16–60 IU/dL) FXI deficiency.[15] FXI-deficient individuals were divided into non-bleeders (N=39) and bleeders (N=18) based on experience following tonsillectomy and/or dental extraction. Bleeders were defined as those requiring blood product transfusion or return to the operating room or dentist for re-suturing or packing. Non-bleeders did not experience excessive bleeding.[15] Demographic characteristics for subjects in Cohort 2 included in the present study are shown in Supplemental Table I.

TFPI measurements.

Total TFPI antigen in Cohort 1 was measured by ELISA in CTI-treated plasma using a mouse monoclonal IgG2B antibody (for capture) and goat polyclonal antibody (for detection) directed against TFPI-K2 residues D29-K282 (Human TFPI ELISA, RayBiotech Inc, Norcross, USA, accession number P10646) according to the manufacturer’s instructions; these measurements were previously published.[14] Total TFPI antigen for Cohort 2 was measured using the same method. TFPIα antigen was measured in a subset of CTI-treated plasmas from Cohort 2 using an in-house ELISA as described.[29] For all experiments with Cohort 2, values measured in multiple plates were normalized against normal pooled plasma assayed at the same time on each ELISA plate to compensate for inter-assay variability; these values are depicted as percent of normal pooled plasma (%NPP). As expected[29], TFPIα correlated significantly and positively with total TFPI (Supplemental Figure 2); values for total TFPI were used for correlation analyses.

Statistical analysis.

Descriptive statistics were summarized using means, standard deviations, medians, and interquartile ranges. Data normality was assessed using Shapiro-Wilk tests. Statistical significance and correlations were tested using Graphpad Prism (version 9.3.1). Data from the in vitro model of FXI deficiency were assessed by Spearman correlation ranking for FXI concentration (left side of graphs) and TFPI concentration (right side of graphs). Relationships between total TFPI and parameters from clotting and lysis assays of the clinical samples were analyzed using Spearman’s correlations.

RESULTS

TFPI reduces thrombin generation in FXI-deficient plasma.

Previous studies have differentiated FXI-deficient bleeders and non-bleeders using thrombin generation assays.[11–13] Therefore, we first measured the effects of TFPI on thrombin generation potential in vitro, using commercially-obtained FXI-deficient plasma. We performed these assays in the absence of FXI (0%), as well as 10% FXI to recapitulate low FXI seen in many individuals with severe FXI-deficiency.[3] We spiked plasma with a TFPI-blocking antibody or exogenous TFPIα to achieve 0, normal (100%), 150%, or 200% of total plasma TFPI levels, respectively. Representative curves are shown in Figure 1A. Compared to 100% FXI, loss of FXI slightly, but significantly prolonged the lag time and TTP, and significantly decreased the thrombin peak and ETP (Figure 1B–E). In the presence of low (10%) FXI, increasing TFPIα prolonged the lag time and TTP, and decreased the thrombin peak and ETP in a concentration-dependent manner (Figure 1B–E).

Figure 1. TFPI decreases thrombin generation in FXI-deficient plasma.

Figure 1.

Open in a new tab

FXI-deficient plasma was reconstituted with 100% (30 nM), 10% (3 nM) or 0% (0 nM) FXI in the presence or absence of 2H8 anti-TFPI antibody (“0% TFPI”), buffer (100% total TFPI), or recombinant TFPI (to 150% and 200% total plasma concentration), as described in the methods. Thrombin generation was measured by calibrated automated thrombography. (A) Representative thrombin generation curves, (B) Lag time, (C) time to peak, (D) peak, and (E) endogenous thrombin potential (ETP). The data show median ± interquartile range; each point represents a separate measurement from N=4–8 experiments. Statistical trend analyses were performed on FXI concentration (left side of the panels) and TFPI concentrations (right side of the panels) and indicated with an r and P-value. The condition 10% FXI, 100% TFPI is reprised on the right side of each panel to appreciate the concentration-response to TFPI.

TFPI suppresses clot formation in FXI-deficient plasma.

Since clot formation assays can also differentiate bleeders from non-bleeders,[14, 15] we then characterized the effects of TFPI on clot formation in vitro. For these experiments, we triggered clot formation with recalcification and low TF, and measured fibrin formation by turbidity. Representative curves are shown in Figure 2A. Compared to 100% FXI, loss of FXI prolonged the lag time and TTP and decreased the rate of fibrin formation (Figure 2B–D). In the presence of low (10%) FXI, increasing TFPIα prolonged the lag time and TTP and reduced the rate of fibrin formation in a concentration-dependent manner (Figure 2B–D). Increasing TFPIα also slightly decreased the peak turbidity change (Figure 2E). Thus, as in the thrombin generation assays, in the setting of low FXI, increasing TFPI inhibited clot formation.

Figure 2. TFPI suppresses clot formation in FXI-deficient plasma.

Figure 2.

Open in a new tab

FXI-deficient plasma was reconstituted as in Figure 1 and described in the methods. Clot formation was initiated with TF and measured as an increase in turbidity. (A) Representative clot formation curves. (B) Lag time, (C) time to plateau, (D) fibrin formation rate, and (E) peak turbidity change. The data show median ± interquartile range; each point represents a separate measurement from N=4–8 experiments. Statistical trend analyses were performed on FXI concentration (left side of the panels) and TFPI concentrations (right side of the panels) and indicated with an r and P-value. The condition 10% FXI, 100% TFPI is reprised on the right side of each panel to appreciate the concentration-response to TFPI.

TFPI decreases fibrin network density in FXI-deficient plasma.

Given the ability of TFPIα to reduce clot formation, we then characterized the effect of TFPI on fibrin network structure. Here we produced clots in FXI-deficient plasma spiked with FXI, anti-TFPI antibody, or exogenous TFPIα as described in the methods, and imaged the formed clots by confocal microscopy. In these experiments, FXI deficiency was modeled with 0% FXI to limit the high rate of contact activation from the glass chamber slides used for confocal microscopy and increase sensitivity to TFPI. In the presence of 100% TFPI (normal level), clots produced in the absence of FXI had significantly reduced fibrin network density compared to FXI-sufficient clots (Figure 3A–B). In the absence of FXI, increasing TFPI reduced fibrin network density in a concentration-dependent manner.

Figure 3. TFPI decreases fibrin network density in FXI-deficient plasma.

Figure 3.

Open in a new tab

FXI-deficient plasma was reconstituted as in Figure 1, in the presence of AlexFluor488-conjugated fibrinogen, as described in the methods. Clot formation was initiated with TF, and formed clots were imaged by laser scanning confocal microscopy. (A) Representative confocal images. (B) Quantification of fibrin network density. The data show median ± interquartile range; each point represents a separate measurement from N=4 experiments. Statistical trend analysis was performed on TFPI concentrations and indicated with an r and P-value.

TFPI increases susceptibility of FXI-deficient clots to fibrinolysis.

Bleeding in FXI-deficient individuals occurs most frequently in tissues with high endogenous fibrinolytic activity, and plasma clots from FXI-deficient bleeders are particularly sensitive to fibrinolysis in vitro.[14, 15, 30] To determine the effect of TFPI on the susceptibility of FXI-deficient clots to fibrinolysis, we triggered clotting in the presence of tPA, and monitored fibrin formation and lysis as an increase and subsequent decrease in turbidity, respectively (Figure 4A). Similar to that seen in the clot formation assays, reduced FXI delayed the lag time and TTP, and reduced the fibrin formation rate, peak turbidity change, and area under the curve (Figure 4B–F). In FXI-deficient plasma, increasing TFPIα prolonged the lag time and TTP and reduced the rate of fibrin formation and peak turbidity change in a concentration-dependent manner (Figure 4B–E). TFPIα also decreased the area under the curve at the highest concentration tested (Figure 4F).

Figure 4. TFPI increases susceptibility of FXI-deficient clots to fibrinolysis.

Figure 4.

Open in a new tab

FXI-deficient plasma was reconstituted as in Figure 1 and described in the methods. Clot formation was initiated with TF in the presence of tissue plasminogen activator, and clot formation and lysis were measured by an increase and decrease in turbidity, respectively. (A) Representative fibrinolysis curves. (B) Lag time, (C) time to peak, (D) fibrin formation rate, (E) peak turbidity change, and (F) area under the curve. The data show median ± interquartile range; each data point represents a separate measurement from N=4–8 experiments. Statistical trend analyses were performed on FXI concentration (left side of the panels) and TFPI concentrations (right side of the panels) and indicated with an r and P-value. The condition 10% FXI, 100% TFPI is reprised on the right side of each panel to appreciate the concentration-response to TFPI.

Total plasma TFPI is elevated in FXI-deficient bleeders and correlates with ex vivo clot formation and fibrinolysis parameters.

Our data from the in vitro model of FXI deficiency demonstrated effects of TFPI on clotting and fibrinolysis parameters previously associated with bleeding in FXI-deficient individuals. To test the hypothesis that plasma TFPI contributes to bleeding risk in a population of FXI-deficient individuals, we quantified total TFPI in plasma samples from two independent cohorts and correlated these measurements with parameters derived from clot formation and fibrinolysis assays from these groups.[14, 15]

Cohort 1 (pilot cohort) consisted of 26 subjects (10 controls and 16 FXI-deficient individuals [8 non-bleeders, 8 bleeders]).[14] As previously observed[14], there was considerable overlap in total plasma TFPI in non-bleeders and bleeders, but slightly though significantly increased total TFPI in bleeders compared to controls (P=0.024, Figure 5A). In this small cohort, Spearman correlations of total TFPI with previously published clotting and fibrinolysis measurements[14] (Figure 5B–K) suggested that across all samples, total TFPI correlated significantly with the fibrin formation rate in clotting assays (Figure 5D, Table 1) and the TTP and fibrin formation rate in fibrinolysis assays (Figure 5G–H, Table 1). These relationships were driven particularly by the bleeders, in which total TFPI correlated significantly with the clotting rate, and the fibrinolysis TTP, rate, peak turbidity change, and area under the curve. In each case, increased total TFPI was associated with reduced clot formation and/or stability.

Figure 5. TFPI is increased in FXI-deficient bleeders and correlates with ex vivo clot formation and fibrinolysis parameters in cohort 1.

Figure 5.

Open in a new tab

(A) TFPI measured by ELISA in plasmas from healthy controls (black) and FXI-deficient individuals characterized as non-bleeders (blue) or bleeders (red).[14] Groups were compared by one-way ANOVA with Fisher’s LSD multiple comparisons test, *P<0.05. (B-E) Spearman correlations for TFPI and ex vivo clot formation parameters from [14]. (F-K) Spearman correlations for TFPI and ex vivo fibrinolysis parameters from [14]. Each point represents a separate individual.

Table 1.

Spearman’s correlations between TFPI and clotting and fibrinolysis parameters in Cohorts 1 and 2.

Spearman’s coefficient, r (P value)
All Samples Controls Non-Bleeders Bleeders
Cohort 1
Clotting
Lag time 0.38 (0.06) 0.19 (0.61) 0.05 (0.93) 0.52 (0.20)
Time to plateau 0.34 (0.09) −0.18 (0.63) 0.52 (0.20) 0.29 (0.50)
Rate −0.51 (0.01) −0.16 (0.66) −0.07 (0.88) −0.81 (0.02)
Peak turbidity change −0.21 (0.31) −0.16 (0.66) 0.24 (0.58) −0.62 (0.12)
Fibrinolysis
Lag time 0.20 (0.32) −0.01 (>0.99) 0.00 (>0.99) 0.57 (0.15)
Time to peak 0.48 (0.01) 0.04 (0.92) 0.36 (0.39) 0.76 (0.04)
Rate −0.46 (0.02) −0.20 (0.58) 0.19 (0.66) −0.83 (0.02)
Peak turbidity change −0.34 (0.09) 0.02 (0.97) 0.02 (0.98) −0.76 (0.04)
Area under the curve −0.25 (0.23) 0.20 (0.58) 0.12 (0.79) −0.83 (0.02)
Lysis time −0.21 (0.30) 0.24 (0.51) 0.14 (0.75) −0.67 (0.08)
Cohort 2
Clotting
Lag time 0.41 (<0.0001) 0.34 (0.02) 0.29 (0.08) 0.73 (0.0006)
Time to plateau 0.50 (<0.0001) 0.44 (0.002) 0.38 (0.02) 0.62 (0.01)
Rate −0.40 (<0.0001) −0.37 (0.01) −0.28 (0.08) −0.65 (0.003)
Peak turbidity change −0.15 (0.14) −0.17 (0.25) −0.02 (0.89) −0.47 (0.05)
Fibrinolysis
Lag time 0.47 (<0.0001) 0.31 (0.03) 0.38 (0.02) 0.78 (0.0001)
Time to peak 0.49 (<0.0001) 0.30 (0.04) 0.47 (0.001) 0.73 (0.0001)
Rate −0.19 (0.05) 0.09 (0.53) −0.16 (0.33) −0.51 (0.03)
Peak turbidity change −0.13 (0.18) 0.20 (0.18) −0.15 (0.38) −0.38 (0.12)
Area under the curve −0.11 (0.26) 0.25 (0.08) −0.11 (0.52) −0.45 (0.06)
Lysis time −0.09 (0.35) 0.32 (0.03) −0.17 (0.31) −0.58 (0.01)
Open in a new tab

Cohort 2 (validation cohort) consisted of plasma samples from an independent group of controls and FXI-deficient individuals: 105 subjects (48 healthy controls and 57 FXI-deficient individuals [39 non-bleeders, 18 bleeders]).[15] As in Cohort 1, although total TFPI did not differ between FXI-deficient non-bleeders and bleeders, bleeders had higher total TFPI compared to controls (P=0.026, Figure 6A). Total TFPI correlated positively with FXI:C in controls (r=0.34, P=0.02), but did not correlate with FXI:C in FXI-deficient individuals (Supplemental Figure 3). Relationships between total TFPI and previously published clot formation and fibrinolysis parameters in Cohort 2[15] were generally similar to that seen in Cohort 1, but were better powered to reveal significant associations (Figure 6B–K, Table 1). Across all samples in both clot formation and fibrinolysis assays, total TFPI correlated positively with the lag time and TTP, and negatively with the fibrin formation rate (Figure 6B–D, F–H, Table 1). Both controls and FXI-deficient individuals contributed to these associations, although correlations were strongest for the bleeders, driven by both the highest TFPI levels and poorest clotting characteristics. Total TFPI was also negatively correlated with the area under the curve (P=0.06) and clot lysis time (P=0.01) in bleeders (Figure 6K, Table 1). Trends were similar in a sub-analysis of individuals with partial or severe deficiency (data not shown). Collectively, these data suggest that in plasma from FXI-deficient individuals, increased total TFPI amplifies the negative impact of FXI deficiency on clot formation and fibrinolysis parameters previously shown to differentiate non-bleeders and bleeders.[14, 15] Collectively, these data support the hypothesis that plasma TFPI contributes to bleeding risk in FXI-deficient individuals.

Figure 6. TFPI is increased in FXI-deficient bleeders and correlates with ex vivo clot formation and fibrinolysis parameters in cohort 2.

Figure 6.

Open in a new tab

(A) TFPI was measured by ELISA in plasmas from healthy controls (black) and FXI-deficient individuals characterized as non-bleeders (blue) or bleeders (red). Groups were compared by Kruskal-Wallis test with Dunn’s multiple comparisons test, *P<0.05. (B-E) Spearman correlations for TFPI and ex vivo clot formation parameters from [15]. (F-K) Spearman correlations of TFPI and ex vivo fibrinolysis parameters from [15]. Each point represents a separate individual.

DISCUSSION

Bleeding tendencies seen in certain FXI-deficient individuals cannot be predicted from plasma FXI antigen or activity alone. We and others have shown that in vitro assays measuring thrombin generation or clot formation, structure, and stability in plasma can differentiate FXI-deficient bleeders and non-bleeders.[11–15] These findings suggest bleeding in FXI-deficient individuals is mediated by both the FXI level and as-yet unidentified plasma protein(s). Herein, we characterized the effect of plasma TFPI on clot formation, structure, and stability in an experimental model of FXI deficiency in vitro, and in two separate cohorts of well-phenotyped FXI-deficient individuals ex vivo. Our findings demonstrate direct effects of TFPI on clot formation and stability in FXI-deficient plasma, and associate these changes with parameters that differentiate the clinical phenotype of FXI-deficient individuals. These findings suggest plasma TFPI is a potential modifier of bleeding risk in FXI-deficient individuals, with implications for understanding underlying biological mechanisms and developing potential predictive algorithms and therapeutic approaches.

Previous studies have identified several mechanism(s) by which TFPI can influence hemostasis, including in the setting of FXI deficiency. TFPI directly inhibits FXa, as well as the FVIIa/TF/FXa complex, and is a highly effective inhibitor of TF activity.[31] By dampening TF activity, TFPI increases dependence on FXI(a) feedback activity to support hemostasis.[19–21] Individuals lacking this backup feedback activity may be unable to mount an effective hemostatic response following injury. TFPI also tightly binds to partially cleaved forms of FV and inhibits early prothrombinase (FVa/FXa) activity.[17, 18] Whether elevated TFPI increases bleeding risk via its ability to inhibit TF, prothrombinase, or both remains unclear; however, either anticoagulant function would be expected to reduce clot formation and stability. Thus, FXI-deficient individuals may be particularly sensitive to small changes in TFPI levels.

There is not a clear mechanism to explain why TFPI is increased in FXI-deficient bleeders. TFPI varies 2-fold in the general population, although some individuals have up to 10-fold higher or lower concentrations.[29] Use of hormonal contraceptives can reduce free TFPI levels [32, 33]; however, contraceptive use was not different between non-bleeders and bleeders (1/39 non-bleeders and 2/18 bleeders in Cohort 2). Several of the bleeders had total plasma TFPI concentrations at the upper end of the normal range. Thus, one possibility is that bleeders are simply those individuals who fall naturally at the higher end of the normal TFPI range in plasma. Alternately, it is possible that a specific mechanism present in bleeders increases circulating TFPI. Previous studies measured TFPIα in plasma from persons with hemophilia A or B[25][34, 35], although it is unclear if the TFPI level associates with different bleeding risk in those patients. In any of these situations, bleeding may result from increased TFPI or from the underlying mechanism that leads to elevated TFPI. For example, rare alternative splicing variants in F5 encoding the FV B-domain produce FVshort, which tightly binds TFPIα and increases circulating TFPIα levels 5- to 20-fold.[17, 36–39] We did not test for these F5 variants in our subjects, but mean total TFPI levels were only ~1.25-fold higher in FXI-deficient bleeders versus controls, substantially lower than those in people with F5 variants. Interestingly, FVshort comprises ~1% of the circulating FV in healthy individuals, indicating that alternative splicing events occur at low rates even in normal F5.[37] Thus, some individuals with FXI deficiency may have slightly increased FVshort that leads to elevated plasma TFPIα and increases risk for bleeding.

The identification of TFPI as a potential modifier of hemostasis in FXI deficiency in humans may have implications for understanding the use of mice as a model of FXI deficiency. Humans and mice share 80% amino acid sequence identity in FXI.[40] However, unlike humans, FXI-deficient mice do not exhibit prolonged bleeding in standard models of hemostasis.[41, 42] Mechanism(s) for this difference are not understood, but it is interesting to speculate that this relates to differences in the relationship between FXI and TFPI. First, both humans and mice express TFPIα and TFPIβ. However, expression differs somewhat between species.[43, 44] Although both species express TFPIα in platelets, humans also express substantive TFPIα in plasma, whereas mice have only trace amounts in plasma.[45] Mice also express a TFPIγ isoform at high plasma concentrations. TFPIγ possesses significant functional activity and may increase the need for strong local procoagulant activity to support hemostasis following vascular injury.[45, 46] As such, mice may have acquired compensatory mechanisms that enhance TF activity and reduce dependence on FXI. Second, whereas FXI circulates in plasma in both humans and mice, mice also maintain a pool of FXI sequestered on vessel wall glycosaminoglycans that is released to the plasma upon heparin infusion[47]. Heparin infusion also increases plasma TFPI levels in mice[45, 48], suggesting these molecules may interact at the vessel wall. Third, FXI(a) can directly cleave and release TFPIβ from cultured endothelial cells and therefore, increase local TF activity.[22] However, the lack of bleeding in FXI-deficient mice suggests FXIa-mediated cleavage of TFPIβ is not necessary for hemostasis, perhaps indicating that TF activity is present in sufficient excess in mice to drive hemostasis in a fully FXI-independent manner.

The inability to predict bleeding in FXI deficiency presents physicians with a significant clinical problem.[3] Undertreatment leaves patients at risk for bleeding during surgical procedures and childbirth, and overtreatment with plasma products and purified FXI increases risk of volume overload and thrombotic events, respectively.[3] Identifying a potential biomarker and mechanism contributing to bleeding risk is therefore important and may enable more effective screening of FXI-deficient patients for bleeding risk prior to surgical procedures. Previous studies have suggested plasma TFPI is a primary determinant of thrombin generation in persons with hemophilia.[49, 50] Accordingly, TFPI inhibition increases thrombin generation and fibrin formation in vitro and shortens the bleeding time in animal models of hemorrhage.[49, 51–54] These findings have inspired therapeutic strategies to reduce or block TFPI and enhance hemostatic potential in individuals with hemophilia A and B, as well as other hemostatic defects.[51–56] In light of our findings, it may be not only feasible, but also appropriate, to use anti-TFPI therapeutics to promote hemostasis in FXI-deficient bleeders.

Although our in vitro and ex vivo data support the hypothesis that TFPI modifies bleeding risk in FXI-deficient individuals, our findings do not exclude other plasma components that may also contribute to the phenotype independently or in concert with altered TFPI. For example, von Willebrand factor,[57, 58] the thrombin-activatable fibrinolysis inhibitor,[30] polyphosphate,[59] and skeletal muscle myosin[60] have each also been implicated as potential modifiers in FXI-deficient individuals or in FXI(a)-related procoagulant functions. These and other proteins and/or metabolites may also contribute to the observed differences and ultimately, agnostic methods might be needed to survey plasma for potential modifiers.

Strengths of our study include the combination of an experimental model that allowed precise control of FXI and TFPI levels in in vitro assays, with direct characterization of plasma TFPI in two independent cohorts of FXI-deficient individuals with severe or partial FXI deficiency. Our study also has limitations. First, although blood type is a potential modifier of bleeding phenotype, we do not have information on blood type for either of our cohorts.[61] However, elevated TFPIα detected in patients with bleeding of unknown cause was not related to blood type,[24] making it unlikely blood type contributed to the differences in total plasma TFPI detected here. Second, TFPI in circulation can exist in several forms, including lipoprotein-bound TFPI, TFPIα (also called free plasma TFPI), and variably C-terminally truncated forms of TFPIα.[25, 29, 62] The correlation analyses performed here were based on measurement of total plasma TFPI, which does not distinguish between these forms. However, total TFPI correlates significantly with TFPIα ([29] and Supplemental Figure 2), and our in vitro experiments show TFPIα can modify clot formation, structure, and lysis, consistent with the established anticoagulant role for this isoform.

In conclusion, our findings from an experimental model of FXI deficiency, as well as plasma samples from two independent cohorts, suggest plasma TFPI is a potential modifier of bleeding risk in FXI-deficient individuals. Knowledge of TFPI levels may improve algorithms to predict bleeding risk. TFPI may be potential therapeutic target to reduce bleeding complications in FXI-deficient bleeders.

Supplementary Material

Supplementary Material

ESSENTIALS.

  • Bleeding risk in factor XI (FXI)-deficiency is variable and correlates poorly with FXI:C

  • We studied TFPI effects on thrombin generation and clotting in FXI-deficient plasma

  • In an experimental model, TFPI alters thrombin generation, clot formation, structure, and lysis

  • In clinical samples, TFPI correlates with parameters that distinguish bleeders from non-bleeders

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the late Dr. Uri Seligsohn and Drs. Michal Zucker and Ophira Salomon for their expertise in collecting the samples in cohort 1, and Laura Gray for technical assistance.

FUNDING

This study was supported by funding from the National Institutes of Health, National Heart, Lung, and Blood Institute (R01HL126974 to ASW).

Footnotes

CONFLICT OF INTEREST

The authors declare no competing financial interests.

SUPPORTING INFORMATION

Supplemental data can be found at the journal website.

REFERENCES

  • 1.Seligsohn U Factor XI deficiency in humans. J Thromb Haemost. 2009;7 Suppl 1:84–7. [DOI] [PubMed] [Google Scholar]
  • 2.Asakai R, Chung DW, Davie EW, Seligsohn U. Factor XI deficiency in Ashkenazi Jews in Israel. N Engl J Med. 1991;325(3):153–8. [DOI] [PubMed] [Google Scholar]
  • 3.Wheeler AP, Gailani D. Why factor XI deficiency is a clinical concern. Expert Rev Hematol. 2016;9(7):629–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Seligsohn U High gene frequency of factor XI (PTA) deficiency in Ashkenazi Jews. Blood. 1978;51(6):1223–8. [PubMed] [Google Scholar]
  • 5.Bolton-Maggs PH, Young Wan-Yin B, McCraw AH, Slack J, Kernoff PB. Inheritance and bleeding in factor XI deficiency. Br J Haematol. 1988;69(4):521–8. [DOI] [PubMed] [Google Scholar]
  • 6.Ragni MV, Sinha D, Seaman F, Lewis JH, Spero JA, Walsh PN. Comparison of bleeding tendency, factor XI coagulant activity, and factor XI antigen in 25 factor XI-deficient kindreds. Blood. 1985;65(3):719–24. [PubMed] [Google Scholar]
  • 7.Sharman Moser S, Chodick G, Ni YG, Chalothorn D, Wang MD, Shuldiner AR, et al. The Association between Factor XI Deficiency and the Risk of Bleeding, Cardiovascular, and Venous Thromboembolic Events. Thromb Haemost. 2022;122(5):808–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Santoro C, Di Mauro R, Baldacci E, De Angelis F, Abbruzzese R, Barone F, et al. Bleeding phenotype and correlation with factor XI (FXI) activity in congenital FXI deficiency: results of a retrospective study from a single centre. Haemophilia. 2015;21(4):496–501. [DOI] [PubMed] [Google Scholar]
  • 9.Pike GN, Cumming AM, Thachil J, Hay CR, Burthem J, Bolton-Maggs PH. Evaluation of the use of global haemostasis assays to monitor treatment in factor XI deficiency. Haemophilia. 2017;23(2):273–83. [DOI] [PubMed] [Google Scholar]
  • 10.Livnat T, Shenkman B, Martinowitz U, Zivelin A, Dardik R, Tamarin I, et al. The impact of thrombin generation and rotation thromboelastometry on assessment of severity of factor XI deficiency. Thromb Res. 2015;136(2):465–73. [DOI] [PubMed] [Google Scholar]
  • 11.Desage S, Dargaud Y, Meunier S, Le Quellec S, Lienhart A, Negrier C, et al. Report of surgeries, their outcome and the thrombin generation assay in patients with Factor XI deficiency: A retrospective single-centre study. Haemophilia. 2022;28(2):301–7. [DOI] [PubMed] [Google Scholar]
  • 12.Pike GN, Cumming AM, Hay CR, Bolton-Maggs PH, Burthem J. Sample conditions determine the ability of thrombin generation parameters to identify bleeding phenotype in FXI deficiency. Blood. 2015;126(3):397–405. [DOI] [PubMed] [Google Scholar]
  • 13.Rugeri L, Quelin F, Chatard B, De Mazancourt P, Negrier C, Dargaud Y. Thrombin generation in patients with factor XI deficiency and clinical bleeding risk. Haemophilia. 2010;16(5):771–7. [DOI] [PubMed] [Google Scholar]
  • 14.Zucker M, Seligsohn U, Salomon O, Wolberg AS. Abnormal plasma clot structure and stability distinguish bleeding risk in patients with severe factor XI deficiency. J Thromb Haemost. 2014;12(7):1121–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Gidley GN, Holle LA, Burthem J, Bolton-Maggs PHB, Lin FC, Wolberg AS. Abnormal plasma clot formation and fibrinolysis reveal bleeding tendency in patients with partial factor XI deficiency. Blood Adv. 2018;2(10):1076–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Wood JP, Ellery PE, Maroney SA, Mast AE. Biology of tissue factor pathway inhibitor. Blood. 2014;123(19):2934–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Wood JP, Bunce MW, Maroney SA, Tracy PB, Camire RM, Mast AE. Tissue factor pathway inhibitor-alpha inhibits prothrombinase during the initiation of blood coagulation. Proceedings of the National Academy of Sciences of the United States of America. 2013;110(44):17838–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Wood JP, Petersen HH, Yu B, Wu X, Hilden I, Mast AE. TFPIalpha interacts with FVa and FXa to inhibit prothrombinase during the initiation of coagulation. Blood Adv. 2017;1(27):2692–702. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Keularts IM, Zivelin A, Seligsohn U, Hemker HC, Beguin S. The role of factor XI in thrombin generation induced by low concentrations of tissue factor. Thromb Haemost. 2001;85(6):1060–5. [PubMed] [Google Scholar]
  • 20.Lakshmanan HHS, Estonilo A, Reitsma SE, Melrose AR, Subramanian J, Zheng TJ, et al. Revised model of the tissue factor pathway of thrombin generation: Role of the feedback activation of FXI. J Thromb Haemost. 2022;20(6):1350–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.von dem Borne PA, Meijers JC, Bouma BN. Feedback activation of factor XI by thrombin in plasma results in additional formation of thrombin that protects fibrin clots from fibrinolysis. Blood. 1995;86(8):3035–42. [PubMed] [Google Scholar]
  • 22.Puy C, Tucker EI, Matafonov A, Cheng Q, Zientek KD, Gailani D, et al. Activated factor XI increases the procoagulant activity of the extrinsic pathway by inactivating tissue factor pathway inhibitor. Blood. 2015;125(9):1488–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.MacDonald S, White D, Langdown J, Downes K, Thomas W. Investigation of patients with unclassified bleeding disorder and abnormal thrombin generation for physiological coagulation inhibitors reveals multiple abnormalities and a subset of patients with increased tissue factor pathway inhibitor activity. Int J Lab Hematol. 2020;42(3):246–55. [DOI] [PubMed] [Google Scholar]
  • 24.Mehic D, Tolios A, Hofer S, Ay C, Haslacher H, Rejto J, et al. Elevated levels of tissue factor pathway inhibitor in patients with mild to moderate bleeding tendency. Blood Adv. 2021;5(2):391–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Duckers C, Simioni P, Spiezia L, Radu C, Gavasso S, Rosing J, et al. Low plasma levels of tissue factor pathway inhibitor in patients with congenital factor V deficiency. Blood. 2008;112(9):3615–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Chang JY, Chantrathammachart P, Monroe DM, Key NS. Studies on the mechanism of action of the aptamer BAX499, an inhibitor of tissue factor pathway inhibitor. Thromb Res. 2012;130(3):e151–7. [DOI] [PubMed] [Google Scholar]
  • 27.Pedersen AH, Nordfang O, Norris F, Wiberg FC, Christensen PM, Moeller KB, et al. Recombinant human extrinsic pathway inhibitor. Production, isolation, and characterization of its inhibitory activity on tissue factor-initiated coagulation reactions. The Journal of biological chemistry. 1990;265(28):16786–93. [PubMed] [Google Scholar]
  • 28.Gray LD, Hussey MA, Larson BM, Machlus KR, Campbell RA, Koch G, et al. Recombinant factor VIIa analog NN1731 (V158D/E296V/M298Q-FVIIa) enhances fibrin formation, structure and stability in lipidated hemophilic plasma. Thromb Res. 2011;128(6):570–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ellery PER, Hilden I, Sejling K, Loftager M, Martinez ND, Maroney SA, et al. Correlates of plasma and platelet tissue factor pathway inhibitor, factor V, and Protein S. Res Pract Thromb Haemost. 2018;2(1):93–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Colucci M, Incampo F, Cannavo A, Menegatti M, Siboni SM, Zaccaria F, et al. Reduced fibrinolytic resistance in patients with factor XI deficiency. Evidence of a thrombin-independent impairment of the thrombin-activatable fibrinolysis inhibitor pathway. J Thromb Haemost. 2016;14(8):1603–14. [DOI] [PubMed] [Google Scholar]
  • 31.Baugh RJ, Broze GJ Jr., Krishnaswamy S. Regulation of extrinsic pathway factor Xa formation by tissue factor pathway inhibitor. The Journal of biological chemistry. 1998;273(8):4378–86. [DOI] [PubMed] [Google Scholar]
  • 32.Dahm AEA, Sandset PM. Tissue factor pathway inhibitor and bleeding tendency: can hormonal state explain the differences? Blood Adv. 2021;5(11):2516–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Harris GM, Stendt CL, Vollenhoven BJ, Gan TE, Tipping PG. Decreased plasma tissue factor pathway inhibitor in women taking combined oral contraceptives. Am J Hematol. 1999;60(3):175–80. [DOI] [PubMed] [Google Scholar]
  • 34.Ellery PER, Hilden I, Thyregod P, Martinez ND, Maroney SA, Gill JC, et al. Measurement of plasma and platelet tissue factor pathway inhibitor, factor V and Protein S in people with haemophilia. Haemophilia. 2019;25(6):1083–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Tardy-Poncet B, Piot M, Chapelle C, Berger C, Tardy B. Difference in TFPI levels between haemophilia A and B patients. Haemophilia. 2011;17(2):312–3. [DOI] [PubMed] [Google Scholar]
  • 36.Peterson JA, Gupta S, Martinez ND, Hardesty B, Maroney SA, Mast AE. Factor V east Texas variant causes bleeding in a three-generation family. J Thromb Haemost. 2022;20(3):565–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Vincent LM, Tran S, Livaja R, Bensend TA, Milewicz DM, Dahlback B. Coagulation factor V(A2440G) causes east Texas bleeding disorder via TFPIalpha. The Journal of clinical investigation. 2013;123(9):3777–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zimowski KL, Petrillo T, Ho MD, Wechsler J, Shields JE, Denning G, et al. F5-Atlanta: A novel mutation in F5 associated with enhanced East Texas splicing and FV-short production. J Thromb Haemost. 2021;19(7):1653–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Cunha ML, Bakhtiari K, Peter J, Marquart JA, Meijers JC, Middeldorp S. A novel mutation in the F5 gene (factor V Amsterdam) associated with bleeding independent of factor V procoagulant function. Blood. 2015;125(11):1822–5. [DOI] [PubMed] [Google Scholar]
  • 40.Gailani D, Sun MF, Sun Y. A comparison of murine and human factor XI. Blood. 1997;90(3):1055–64. [PubMed] [Google Scholar]
  • 41.Mohammed BM, Cheng Q, Matafonov A, Monroe DM, Meijers JCM, Gailani D. Factor XI promotes hemostasis in factor IX-deficient mice. J Thromb Haemost. 2018;16(10):2044–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Gailani D, Lasky NM, Broze GJ, Jr., A murine model of factor XI deficiency. Blood Coagul Fibrin. 1997;8(2):134–44. [DOI] [PubMed] [Google Scholar]
  • 43.Maroney SA, Ellery PE, Wood JP, Ferrel JP, Martinez ND, Mast AE. Comparison of the inhibitory activities of human tissue factor pathway inhibitor (TFPI)alpha and TFPIbeta. J Thromb Haemost. 2013;11(5):911–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Maroney SA, Ellery PE, Mast AE. Alternatively spliced isoforms of tissue factor pathway inhibitor. Thromb Res. 2010;125 Suppl 1:S52–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Girard TJ, Grunz K, Lasky NM, Malone JP, Broze GJ, Jr., Re-evaluation of mouse tissue factor pathway inhibitor and comparison of mouse and human tissue factor pathway inhibitor physiology. J Thromb Haemost. 2018;16(11):2246–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Broze GJ, Anbukumar DS, Lasky NM, Girard TJ. The Tissue Factor Pathway Inhibitor Gamma Knockout Mouse Is a Better Model Than the Wild Type Mouse for Studying Tissue Factor-Related Human Diseases. Blood. 2019;134(Supplement_1):484-. [Google Scholar]
  • 47.Mohammed BM, Cheng Q, Matafonov A, Verhamme IM, Emsley J, McCrae KR, et al. A non-circulating pool of factor XI associated with glycosaminoglycans in mice. J Thromb Haemost. 2019;17(9):1449–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Maroney SA, Ferrel JP, Pan S, White TA, Simari RD, McVey JH, et al. Temporal expression of alternatively spliced forms of tissue factor pathway inhibitor in mice. J Thromb Haemost. 2009;7(7):1106–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Erhardtsen E, Ezban M, Madsen MT, Diness V, Glazer S, Hedner U, et al. Blocking of tissue factor pathway inhibitor (TFPI) shortens the bleeding time in rabbits with antibody induced haemophilia A. Blood Coagul Fibrin. 1995;6(5):388–94. [DOI] [PubMed] [Google Scholar]
  • 50.Chelle P, Montmartin A, Damien P, Piot M, Cournil M, Lienhart A, et al. Tissue factor pathway inhibitor is the main determinant of thrombin generation in haemophilic patients. Haemophilia. 2019;25(2):343–8. [DOI] [PubMed] [Google Scholar]
  • 51.Hilden I, Lauritzen B, Sorensen BB, Clausen JT, Jespersgaard C, Krogh BO, et al. Hemostatic effect of a monoclonal antibody mAb 2021 blocking the interaction between FXa and TFPI in a rabbit hemophilia model. Blood. 2012;119(24):5871–8. [DOI] [PubMed] [Google Scholar]
  • 52.Gorczyca ME, Nair SC, Jilma B, Priya S, Male C, Reitter S, et al. Inhibition of tissue factor pathway inhibitor by the aptamer BAX499 improves clotting of hemophilic blood and plasma. J Thromb Haemost. 2012;10(8):1581–90. [DOI] [PubMed] [Google Scholar]
  • 53.Dockal M, Hartmann R, Fries M, Thomassen MC, Heinzmann A, Ehrlich H, et al. Small peptides blocking inhibition of factor Xa and tissue factor-factor VIIa by tissue factor pathway inhibitor (TFPI). The Journal of biological chemistry. 2014;289(3):1732–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Barg AA, Brutman-Barazani T, Avishai E, Budnik I, Cohen O, Dardik R, et al. Anti-TFPI for hemostasis induction in patients with rare bleeding disorders, an ex vivo thrombin generation (TG) guided pilot study. Blood Cells Mol Dis. 2022;95:102663. [DOI] [PubMed] [Google Scholar]
  • 55.Berntorp E, Fischer K, Hart DP, Mancuso ME, Stephensen D, Shapiro AD, et al. Haemophilia. Nat Rev Dis Primers. 2021;7(1):45. [DOI] [PubMed] [Google Scholar]
  • 56.Mast AE, Ruf W. Regulation of coagulation by tissue factor pathway inhibitor: Implications for hemophilia therapy. J Thromb Haemost. 2022;20(6):1290–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Gueguen P, Galinat H, Blouch MT, Bridey F, Duchemin J, Le Gal G, et al. Biological determinants of bleeding in patients with heterozygous factor XI deficiency. Br J Haematol. 2012;156(2):245–51. [DOI] [PubMed] [Google Scholar]
  • 58.Bolton-Maggs PH, Patterson DA, Wensley RT, Tuddenham EG. Definition of the bleeding tendency in factor XI-deficient kindreds--a clinical and laboratory study. Thromb Haemost. 1995;73(2):194–202. [PubMed] [Google Scholar]
  • 59.Choi SH, Smith SA, Morrissey JH. Polyphosphate is a cofactor for the activation of factor XI by thrombin. Blood. 2011;118(26):6963–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Morla S, Deguchi H, Zilberman-Rudenko J, Gruber A, McCarty OJT, Srivastava P, et al. Skeletal muscle myosin promotes coagulation by binding factor XI via its A3 domain and enhancing thrombin-induced factor XI activation. The Journal of biological chemistry. 2022;298(2):101567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Mehic D, Hofer S, Jungbauer C, Kaider A, Haslacher H, Eigenbauer E, et al. Association of ABO blood group with bleeding severity in patients with bleeding of unknown cause. Blood Adv. 2020;4(20):5157–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Castoldi E, Simioni P, Tormene D, Rosing J, Hackeng TM. Hereditary and acquired protein S deficiencies are associated with low TFPI levels in plasma. J Thromb Haemost. 2010;8(2):294–300. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material
NIHMS1888498-supplement-Supplementary_Material.pdf (3MB, pdf)

RESOURCES