Transferrin plays a central role in coagulation balance by interacting with clotting factors

Abstract

Coagulation balance is maintained through fine-tuned interactions among clotting factors, whose physiological concentrations vary substantially. In particular, the concentrations of coagulation proteases (pM to nM) are much lower than their natural inactivator antithrombin (AT, ~ 3 μM), suggesting the existence of other coordinators. In the current study, we found that transferrin (normal plasma concentration ~40 μM) interacts with fibrinogen, thrombin, factor XIIa (FXIIa), and AT with different affinity to maintain coagulation balance. Normally, transferrin is sequestered by binding with fibrinogen (normal plasma concentration ~10 μM) at a molar ratio of 4:1. In atherosclerosis, abnormally up-regulated transferrin interacts with and potentiates thrombin/FXIIa and blocks AT's inactivation effect on coagulation proteases by binding to AT, thus inducing hypercoagulability. In the mouse model, transferrin overexpression aggravated atherosclerosis, whereas transferrin inhibition via shRNA knockdown or treatment with anti-transferrin antibody or designed peptides interfering with transferrin-thrombin/FXIIa interactions alleviated atherosclerosis. Collectively, these findings identify that transferrin is an important clotting regulator and an adjuster in the maintenance of coagulation balance and modifies the coagulation cascade.

Fig. 1

Enhanced enzymatic activity of thrombin and FXIIa is associated with elevated transferrin in atherosclerotic plasma. a, b An anti-transferrin antibody (Tf AB) alleviated the potentiating ability of CHD plasma on enzymatic activity of thrombin (a) and FXIIa (b). Data represent mean ± SD (n = 6), **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test. c Amounts of transferrin in plasma from CHD patients and healthy volunteers were determined by ELISA. Data represent mean ± SD (n = 120), **P < 0.01 by unpaired t-test. d Total iron-binding capacity (TIBC) in plasma from CHD patients and healthy volunteers were determined using TIBC kit. Data represent mean ± SD (n = 120), **P < 0.01 by unpaired t-test. e Western blot (top) and quantification (bottom) analysis of transferrin in plasma samples from CHD patients and healthy volunteers. Red Ponceau (RP)-stained blots were used as a loading control. f Western blot (top) and quantification (bottom) analysis of protein extracts from normal arteries (Normal) and atherosclerotic lesions (Plaque). β-actin was used as a control. Data represent mean ± SD (n = 12), **P < 0.01 by unpaired t-test (e, f). g Amounts of transferrin in the plasma from the Apoe-/- mice fed with high fat diet (HFD) and normal diet (ND) were determined by ELISA. Data represent mean ± SD (n = 10), **P < 0.01 by unpaired t-test. h Immunofluorescence staining of transferrin (green) in mice atherosclerotic plaque. Cell nuclei were labeled by DAPI. Scale bar represents 10 μm. Images are representative of at least three independent experiments. i Western blot (top) and quantification (bottom) analysis of transferrin in aortic roots of the Apoe^−/− mice. Data represent mean ± SD (n = 10), **P < 0.01 by unpaired t-test. N.S.: no significance; Tf: transferrin

Fig. 2

Effects of both apo- and holo-transferrin on thrombin, FXIIa and antithrombin. a Potentiating effects of both apo- and holo-transferrin on thrombin. b, c Representative RP-HPLC analysis (b) and quantification (c) of fibrinopeptide A (FbpA) and fibrinopeptide B (FbpB) released from 5 mg of fibrinogen hydrolyzed by 0.1 NIH unit thrombin mixed with 0, 0.2, 1, or 5 μM apo-transferrin, respectively. d Potentiating effects of both apo- and holo-transferrin on FXIIa. e, f Representative western blot (e) and quantification analysis of kallikrein heavy chain (HC ∼52 kDa) (f) released from 10 μg of prekallikrein (PK) hydrolyzed by 0.01 NIH unit FXIIa mixed with 0, 0.2, 1, or 5 μM apo-transferrin (lane 2–5), respectively. Blots of PK, FXIIa heavy chain (FXIIa HC), transferrin, and kallikrein light chain (LC ∼36 and 33 kDa) are also shown. g–j Apo- and holo-transferrin block antithrombin (AT)’s inactivation effect on thrombin (g, h) and FXa (i, j). TAT: thrombin–AT complex. HSA: human serum albumin. Data represent mean ± SD of five independent experiments, *P < 0.05, **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test. N.S.: no significance; Tf: transferrin

Fig. 3

Interactions between transferrin and clotting factors. a–d SPR analysis of the interaction between transferrin and thrombin (a), FXIIa (b), fibrinogen (c) or antithrombin (AT) (d). e, f Native gel shift analysis of interaction between transferrin (8 μg) and thrombin (2, 4, and 8 μg) (e) or FXIIa (2, 4, and 8 μg) (f). g, h Native gel shift analysis of interaction between transferrin (2, 4, and 8 μg) and fibrinogen (8 μg) (g) or AT (8 μg) (h). Arrows indicate the complexes of transferrin–thrombin, transferrin–FXIIa, transferrin–fibrinogen or transferrin–AT. i, k SPR analysis of the interaction between wild-type transferrin (WT-Tf) or transferrin mutant (E333,338R) and thrombin (i) or FXIIa (k). j, l Effects of wild-type transferrin and transferrin mutant on enzymatic activity of thrombin (j) and FXIIa (l). m SPR analysis of the interaction between transferrin and wild-type thrombin (WT-Th) or thrombin mutant (Th-mutant, R117,122A). n Effects of transferrin on enzymatic activity of wild-type thrombin and thrombin mutant. o, q SPR analysis of interaction between transferrin and TH16 or TH16-scr (scrambled control of TH16) (o), and FX18 or FX18-scr (scrambled control of FX18) (q). p Effects of TH16 and TH16-scr on potentiating activity of transferrin on thrombin. r Effects of FX18 and FX18-scr on the potentiating activity of transferrin on FXIIa. Data represent mean ± SD of six independent experiments, **P < 0.01 by unpaired t-test (j, l, n). **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test (p, r). Tf: transferrin

Fig. 4

Elevated levels of transferrin–thrombin/FXIIa complexes in CHD patient plasma and atherosclerotic plaque. a Western blot analysis of transferrin–prothrombin (Tf–PTh) and transferrin–FXII complexes in healthy (Normal) and CHD plasma. Red Ponceau (RP)-stained blots were uesd as the loading control. b, c Quantification of the transferrin–PTh (b) and transferrin–FXII (c) complexes. d Co-immunoprecipitation of transferrin and prothrombin or FXII in human normal plasma. e Human atherosclerotic plaque was labeled with either anti- transferrin antibody (green) or anti-thrombin antibody (red) to detect presence of the transferrin–thrombin complex (top), or labeled with either anti-transferrin antibody (red) or anti-FXIIa antibody (green) to detect presence of the transferrin–FXIIa complex (bottom). Cell nuclei were labeled with DAPI. Arrows indicate transferrin–thrombin- or transferrin–FXIIa-positive structures. Scale bar represents 30 μm. Images are representative of at least three independent experiments. f–h Western blot analysis (f) and quantification of transferrin–prothrombin complex (g) and the transferrin–FXII complex (h) in the supernatants of the homogenized thoracic aorta tissue from normal controls and atherosclerotic patients. Data represent mean ± SD (n = 12), **P < 0.01 by unpaired t-test. Tf: transferrin

Fig. 5

Effects of transferrin overexpression and knockdown on atherosclerotic development and hypercoagulability. a–f Plasma concentrations of transferrin in five groups of Apoe^−/− mice fed a HFD for 6 weeks (transferrin overexpression (PLP-Tf) and its blank PLP, knockdown (RNR-Tf) and its blank RNR, and normal Apoe^−/− mice (NC)) (a). Relative activity of thrombin (b) and FXIIa (c), APTT (d), PT (e) in their plasma and tail bleeding time (f) are also shown. g Representative images of carotid artery blood flow (top) in FeCl₃-treated mice by laser speckle perfusion imaging, and the region of interest (green rectangle) was placed in the carotid artery to quantify blood flow change. Relative blood flow in the region of interest is shown (bottom) by using perfusion unit. Red: blood flow; Blue and black area: background; The color bar on the right side indicates the perfusion unit scale (0–302). h Representative images of oil-red O-stained atherosclerotic plaques (top) and quantitative analysis of stained area (bottom) are shown. Data represent mean ± SD (n = 6), **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test. i Atherosclerotic plaques from the mice fed a HFD for 4 weeks were labeled with either anti-transferrin, anti-thrombin, or anti-FXIIa antibodies. Cell nuclei were labeled by DAPI. Arrows indicate transferrin–thrombin- or transferrin–FXIIa-positive structures. Scale bar represents 30 μm. Images are representative of at least three independent experiments. N.S.: no significance; Tf: transferrin

Fig. 6

Transferrin interferences exert anti-AS effects in vivo. The HFD-fed Apoe^−/− mice were subjected to anti-transferrin antibody (Tf AB) or control IgG treatment twice/week for 6 weeks. a Representative images (top) of oil-red O-stained plaques and quantitative analysis (bottom) of the stained area are shown. b Effects of TH16 and FX18 on FeCl₃-induced carotid artery thrombus formation in C57BL/6J mice. Representative images of carotid artery blood flow (top) and quantitation (bottom) are shown. Red: blood flow; Blue and black area: background; The color bar on the right side indicates the perfusion unit scale (0–302). c Effects of TH16 and FX18 on mouse AS development. Representative images (top) of oil-red O-stained plaques and quantitative analysis (bottom) of the stained area are shown. d Graphical representation of transferrin’s central role and its interactions with clotting factors to maintain coagulation balance. Transferrin participates in three types of interactions for coagulation balance including: 1) most of transferrin (TRF, ~40 μM) is sequestered by binding with fibrinogen (~10 μM) at a molar rate of 4:1; 2) transferrin blocks inactivation effect of AT towards thrombin and FXa by binding with AT at a molar rate of 2:1; 3) transferrin interacts and potentiates thrombin and FXIIa at a molar rate of 1:1. Data represent mean ± SD (n = 6–8), **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test. N.S.: no significance

Fig. 7

Effects of transferrin overexpression, knockdown, anti-transferrin antibody treatment, and interference peptides on coagulation. a Plasma concentrations of transferrin in four groups of C57BL/6J mice (transferrin overexpression (PLP-Tf), knockdown (RNR-Tf), anti-transferrin antibody-treated (Tf AB), and normal control mice (NC)). b–f Relative activity of thrombin (b) and FXIIa (c), APTT (d), PT (e) in their plasma and tail bleeding time (f) are also shown. g–i Effects of TH16, FX18, TH16-scr, and FX18-scr on plasma recalcification time (g), clotting time (h), and tail bleeding time (i) in C57BL/6J mice. Data represent mean ± SD (n = 6–8), **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test. N.S.: no significance; Tf: transferrin