Platelet factor XIII-A regulates platelet function and promotes clot retraction and stability

Abstract

Background: Factor XIII (FXIII) is an important proenzyme in the hemostatic system. The plasma-derived enzyme activated FXIII cross-links fibrin fibers within thrombi to increase their mechanical strength and cross-links fibrin to fibrinolytic inhibitors, specifically α₂-antiplasmin, to increase resistance to fibrinolysis. We have previously shown that cellular FXIII (factor XIII-A [FXIII-A]), which is abundant in the platelet cytoplasm, is externalized onto the activated membrane and cross-links extracellular substrates. The contribution of cellular FXIII-A to platelet activation and platelet function has not been extensively studied.

Objectives: This study aims to identify the role of platelet FXIII-A in platelet function.

Methods: We used normal healthy platelets with a cell permeable FXIII inhibitor and platelets from FXIII-deficient patients as a FXIII-free platelet model in a range of platelet function and clotting tests.

Results: Our data demonstrate that platelet FXIII-A enhances fibrinogen binding to the platelet surface upon agonist stimulation and improves the binding of platelets to fibrinogen and aggregation under flow in a whole-blood thrombus formation assay. In the absence of FXIII-A, platelets show reduced sensitivity to agonist stimulation, including decreased P-selectin exposure and fibrinogen binding. We show that FXIII-A is involved in platelet spreading where a lack of FXIII-A reduces the ability of platelets to fully spread on fibrinogen and collagen. Our data demonstrate that platelet FXIII-A is important for clot retraction where clots formed in its absence retracted to a lesser extent.

Conclusion: Overall, this study shows that platelet FXIII-A functions during thrombus formation by aiding platelet activation and thrombus retraction in addition to its antifibrinolytic roles.

Keywords: clot retraction; factor XIII; fibrinogen; fibrinolysis; platelet; platelet activation.

Figure 1

Platelet factor FXIII-A (FXIII-A) is abundant in healthy platelets and partially colocalizes with actin and actin nodules. Platelets were isolated from whole blood from normal healthy donors and factor XIII (FXIII)–deficient patients. (A) Immunocytochemistry was performed using fluorescein isothiocyanate (FITC) anti–FXIII-A antibody to detect platelet FXIII-A content in fixed, permeabilized platelets that were resting or stimulated with 10-μM thrombin receptor activator peptide 6 (TRAP-6) for 3 minutes. Normal healthy platelets (top panel), and FXIII-A–deficient platelets (bottom panel). Images representative of normal controls (n = 3) and FXIII-deficient patients (n = 2). Scale bar represents 20 μm. (B) Normal healthy platelets were spread on collagen (100 μg/mL) and thrombin (1 U/mL)–coated slides in the presence of a fluorescent probe, N-TAMRA cadaverine (ex 547), to measure FXIII-A activity for 45 minutes. Cells were then fixed, permeabilized, and stained using AF647-phalloidin to detect actin and FXIII-A using FITC anti–FXIII-A antibody. Scale bar represents 5 μm. (C–E) Normal platelets were washed and spread at 37 °C for 60 minutes on either collagen (100 μg/mL) or fibrinogen (100 μg/mL). Platelets were fixed, permeabilized, and stained with stochastic optical reconstruction microscopy (STORM)–appropriate AF532-labeled phalloidin anti–FXIII-A primary antibody with STORM-appropriate AF647-labeled secondary. Platelets were submerged in ONI Imaging Buffer for direct STORM imaging and imaged in Total internal refractive flourescence (TIRF). (C) Representative images of platelets spread on collagen: merged image (left), AF532-phalloidin (middle; green), and AF647-FXIII-A (right; magenta). Scale bar represents 20 μm. (D) Close-up image of a single platelet spread on collagen (left), with close-up images (right) of actin nodules in merged image (top), AF532-phalloidin (middle; green), and AF-FXIII-A (bottom; magenta). Scale bar on left represents 2 μm, and scale bars on right represents 0.5 μm (E) Close-up image of a single platelet spread on fibrinogen (left), with close-up images (right) of actin nodules in merged image (top), AF532-phalloidin (middle; green), and AF-FXIII-A (right, bottom; magenta). Scale bar on left represents 2 μm, and scale bars on right represent 0.5 μm.

Figure 2

Inhibition of factor XIII-A reduces platelet spreading and limits procoagulant platelet formation, but does not affect actin polymerization. Normal platelets and factor XIII (FXIII)–deficient (FXIII-def) platelets were washed and incubated for 30 minutes in the presence or absence of transglutaminase inhibitor (TGI) before performing spreading studies at 37 °C for 60 minutes on either collagen (100 μg/mL) or fibrinogen (100 μg/mL). Platelets were stained using AF488-labeled phalloidin and imaged using fluorescence confocal microscopy. (A) Representative images of spread platelets on fibrinogen after 60 minutes from normal controls (left), normal controls + TGI (middle), and FXIII-deficient platelets (right). Scale bar represents 10 μm. (B) Representative images of spread platelets on collagen after 60 minutes from normal controls (left), normal controls + TGI (middle), and FXIII-deficient platelets (right). Scale bar represents 10 μm. (C) Platelet spreading on fibrinogen was measured by calculating the percentage area covered by platelets divided by the number of platelets in the field of view; 3 fields of view were taken per condition in each experiment and analyzed by a blinded investigator. Blue: normal platelets—TGI; gray: normal platelets + TGI; green: FXIII-deficient platelets; and purple: FXIII-deficient platelets + TGI. Statistical analysis was performed on normal platelet spreading where paired t-tests were used to determine significance between samples with and without TGI. ∗∗P < .01 and ∗P < .05. Normal donors (n = 10) and FXIII-deficient donors (n = 2; 2 separate experiment repeats per patient). (D) Platelet spreading on collagen was measured by calculating the percentage area covered by platelets divided by the number of platelets in the field of view; 3 fields of view were taken per experiment and analyzed by a blinded investigator. Statistical analysis was performed on normal platelet spreading where paired t-tests were used to determine significance between the sample with and without TGI. ∗∗P < .01. Normal donors (n = 10) and FXIII-deficient donors (n = 2). (E, F) Washed normal platelets were left resting or incubated for 30 minutes in the presence and absence of TGI before stimulating for 20 minutes with collagen-related peptide (CRP) (10 μg/mL; panel E) or thrombin (1 U/mL; panel F). Platelets were then fixed, permeabilized, and stained using AF488-labeled phalloidin to stain F-actin. F-actin was measured using flow cytometry where the median fluorescence intensity of AF488-phalloidin was used as a direct measurement of F-actin levels. Data were normalized as the percentage median fluorescence intensity of the unstimulated control and plotted as percent change from unstimulated control. Data show individual percent change values ± SEM. Statistical analysis was performed using Students t-test on percentage change data comparing samples −TGI to +TGI for each agonist. n = 3. (G) Normal platelets were washed and incubated for 30 minutes in the presence (gray) or absence (blue) of TGI before stimulating at 37 °C with a combination of cross-linked CRP (1 μg/mL) and thrombin (1 U/mL) for 15, 30, 45, and 60 minutes in the presence of 2 mM CaCl₂. Phosphatidylserine (PS) was detected with Annexin V using flow cytometry and expressed using percentage positive ± SEM; statistical analysis was performed using paired t-tests on a separate area under the curve values. ∗P < .05. n = 4. ns, nonsignificant.

Figure 3

Platelet factor XIII-A promotes platelet activation and factor XIII-A inhibition or absence reduces platelet-derived fibrinogen binding to platelets. (A) Platelets were isolated and washed from normal healthy (blue) and factor XIII (FXIII)–deficient (FXIII-def) (green) individuals and incubated in the presence and absence of transglutaminase inhibitor (TGI) (normal: gray; (FXIII)-deficient: purple) for 30 minutes at 30 °C. Platelets were stimulated for 20 minutes at 37 °C with either thrombin receptor activator peptide 6 (TRAP-6) (5 μM; left), U46619 (1 μM; middle left), cross-linked collagen-related peptide (CRP-XL) (1 μg/mL; middle right), or adenosine diphosphate (ADP) (5 μM; right) in the presence of fluorescein isothiocyanate–labeled antifibrinogen antibody to detect fibrinogen binding. Data show individual median fluorescence intensity (MFI) values ± SEM of normal healthy controls (n = 14) and FXIII-deficient patients (n = 2). Statistical analysis in normal controls was performed and compared in samples with and without TGI using paired t-test, ∗∗P < .01. (B) Fluorescence intensity (FU) (MFI) for fibrinogen binding and (C) P-selectin exposure. Data represent mean FU ± SEM of responses to a range of concentrations of CRP-XL (0-3 μg/mL; top panels), TRAP-6 (0-15 μM) (middle panels), and ADP (0-30 μM) (bottom panels) in healthy controls (blue; n = 7) and FXIII-deficient patients (green; n = 2). CRP, collagen-related peptide; ns, nonsignificant; PRP, platelet-rich plasma.

Figure 4

Platelet factor XIII-A inhibition or absence reduces platelet thrombus formation on fibrinogen but not collagen. (A, B) Whole-blood platelet thrombus formation under flow was performed with healthy donor blood at 500/s in capillary chambers coated with 100 μg/mL of fibrinogen with added DiOC6 (4 μg/mL) to label platelets and recorded over 600 seconds (s) using fluorescence confocal microscopy. (A) Data represents fluorescence intensity (FU) ± SEM of platelet deposition in the absence (blue) and presence (gray) of transglutaminase inhibitor (TGI) for over 600 seconds. Statistical analysis was performed on endpoint thrombus formation at the 600-second time point and compared in samples with and without TGI using paired t-test, ∗P < .05. n = 3. (B) Representative images of endpoint thrombus formation at the 600-second time point in healthy donor blood in the absence (left) and presence (right) of TGI. Scale bar represents 20 μm. (C) Platelet thrombus formation was performed at 1000/s on 100 μg/mL of collagen with whole blood from normal healthy donors (blue; left panel) and factor XIII (FXIII)–deficient (FXIII-def) donors (green; left panel); whole blood from normal healthy donors was also assayed in the absence (blue; right panel) and presence (black; right panel) of TGI for over 600 seconds. Data represent mean FU ± SEM. Statistical analysis was performed on endpoint thrombus formation at the 600-second time point and compared in samples with and without TGI using the paired t-test. Normal donors: n = 3; FXIII-deficient donors: n = 2. (D) Representative images of endpoint thrombus formation at the 600-second time point in FXIII-deficient patient blood (left) and healthy donor blood in the absence (middle) and presence (right) of TGI. Scale bar represents 20 μm. ns, nonsignificant.

Figure 5

Inhibition or absence of platelet (plt) factor XIII-A reduces the extent of clot retraction and enhances the rate of fibrinolysis under flow. (A) Clot retraction was performed in nonsiliconized glass tubes using either normal or factor XIII (FXIII)–deficient (FXIII-def) washed platelets in FXIII-depleted plasma; clotting was initiated for 30 minutes with thrombin (1 U/mL) and CaCl₂ (2 mM), and after 30 minutes, clots were imaged and weighed. Clot weight (g) was measured and compared in normal donors (n = 9) in the presence (gray) and absence (blue) of transglutaminase inhibior (TGI) and FXIII-deficient patients (n = 2 patients; 1 patient was recalled for a second time) in the presence (purple) and absence (green) of TGI. Data represents mean ± SEM of clot weight. Statistical analysis was performed on normal donors and samples with and without TGI compared using paired t-test; ∗∗P < .01. (B) Chandler model thrombi were formed from FXIII-deficient plasma (black dashed line) and the incorporation of either normal healthy washed platelets (blue) in the absence and presence (gray) of TGI or FXIII-deficient washed platelets (green) in the absence and presence (purple) of TGI. Data presented represents the median fluorescence intensity (FU) values normalized to the internal FXIII-deficient plasma control. FXIII-deficient patients (n = 2) and normal healthy controls (n = 2). (C) Thrombus formation under flow was performed in microcapillary chambers and visualized using fluorescence confocal microscopy. Whole reconstituted blood made up from FXIII-depleted plasma, either normal healthy platelets (blue) or FXIII-deficient platelets (green) with the red cells from the same donor in the presence of DiOC6 (4 μg/mL) to label platelets, AF647-labeled fibrinogen (75 μg/mL), and tPA (20 nM). Thrombi were fully formed at a shear rate of 500/second, which was achieved when fibrin and platelets covered the entire surface of the well before flow was switched to a lysis buffer containing 125 nM tPA in Tyrode’s buffer and thrombi were lysed to completion. Data represent the normalized FU values of AF647-fibrin(ogen) from the point at which full coverage was achieved and lysis was initiated to completion, where lysis times are represented in (D). Data represent normal healthy donors (n = 3) and FXIII-deficient patient donors (n = 2; mean ± SEM). (E) Representative images from various time points over the course of fibrinolysis: top images are from samples containing normal platelets in FXIII-depleted plasma and bottom images represent FXIII-deficient platelets in FXIII-depleted plasma. FXIII-deficient patient samples lysed faster, hence the shorter times on representative images. Fibrin(ogen): red; platelets: green. Scale bar represents 100 μm.