Severe deficiency of coagulation Factor VII results in spontaneous cardiac fibrosis in mice

Abstract

Mice genetically modified to produce low levels (approximately 1% of wild-type) of coagulation FVII presented with echocardiographic evidence of heart abnormalities. Decreases in ventricular size and reductions in systolic and diastolic functions were found, suggestive of a restrictive cardiomyopathy and consistent with an infiltrative myopathic process. Microscopic analysis of mouse hearts showed severe patchy fibrosis in the low-FVII mice. Haemosiderin deposition was discovered in hearts of these mice, along with increases in inflammatory cell number, ultimately resulting in widespread collagen deposition. Significant increases in mRNA levels of TGFbeta, TNFalpha and several matrix metalloproteinases in low-FVII mice, beginning at early ages, supported a state of cardiac remodelling associated with the fibrotic pathology. Mechanistic time-course studies suggested that cardiac fibrosis in low-FVII mice originated from bleeding in heart tissue, resulting in the recruitment of leukocytes, which released inflammatory mediators and induced collagen synthesis and secretion. These events led to necrosis of cardiomyocytes and collagen deposition, characteristics of cardiac fibrosis. The results of this study demonstrated that haemorrhagic and inflammatory responses to a severe FVII deficiency resulted in the development of cardiac fibrosis, observed echocardiographically as a restrictive cardiomyopathy, with compromised ventricular diastolic and systolic functions.

Figure 1

Ultrasound evaluation of cardiac function in FVII^tTA/tTA mice. Representative M-mode images from the short-axis view of the LV of 12 mo (A) WT and (B) FVII^tTA/tTA mice. Anterior wall (a), posterior wall (p) wave motions. Doppler waveforms of the blood velocity obtained at the mitral valve level in (C) WT mice and (D) FVII^tTA/tTA mice. AO refers to aortic outflow, E and A refer to early and late filling of the LV, respectively. (E, F) Doppler waveforms of the blood velocity at the main pulmonary artery in (E) WT and (F) FVII^tTA/tTA mice. A white bar is indicated in each as a marker of the peak blood velocities.

Figure 2

Histochemistry and immunohistochemistry to detect spontaneous bleeding in hearts of low-FVII mice. (A,B) Prussian Blue staining (100×) of (A) WT) and (B) FVII^tTA/tTA mice at 4. The arrows in B point to haemosiderin deposits (blue). (C) Electron micrographic detail (3500×) of a macrophage (arrow) from the LV in an area of a haemosiderin deposit in a 4 mo mouse heart section. Approximately 6 erythrocytes (an example is labeled e) containing electron-dense iron deposits are seen phagocytosed by the macrophage. Also crumpled membranes of an erythrocyte are observed in a nascent siderosome (s). A collagen-rich matrix (c) is also abundant in the area. The macrophage nucleus is labelled (n). (D, E) Prussian Blue staining (100×) of (D) WT and (E) FVII^tTA/tTA mice at 12 mo. Haemosiderin deposits are clearly seen in E (blue). (F, G) H & E staining of heart tissue sections in 4 mo (F) WT and (G) FVII^tTA/tTA mice (200× magnification). The arrows point to RBC surrounding vessels, showing leakage of blood into heart tissue in the FVII^tTA/tTA mouse. (H, I) Bleeding is also evidenced by the positive (brown) immunostaining for P-selectin in heart tissue sections in 4 mo (I) FVII^tTA/tTA mice, as compared to (H) control WT heart tissue (200×). (J, K) Gross photographs of hearts of (J) WT and (K) FVII^tTA/tTA mice at 4 mo of age old after Evans blue injection. Photographs were taken at 6× magnification using a surgical dissecting microscope.

Figure 3

Histological identification of cardiac inflammation and fibrosis in hearts of 4 mo FVII^tTA/tTA mice. (A–D) Immunostaining for infiltrated CD-45+ cells (brown) in cardiac tissue. Panels A and B show anti-CD45 immunostains of 4 mo (A) WT and (B) FVII^tTA/tTA mice (100×). Panels C and D display anti-CD45 immunostains (brown) of 12 mo (C) WT and (D) FVII^tTA/tTA mice (100×). (E–J) Histological identification of collagen in hearts of 4 mo FVII^tTA/tTA mice. Masson’s trichrome staining of collagen (blue; 20×). Brightfield images of picro-Sirius Red staining (red) from 4 mo (F) WT and (I) FVII^tTA/tTA (I) mice (60×). Epifluorescence detection (red) of collagen in hearts of 4 mo WT (G) and FVII^tTA/tTA (J) mice hearts (40×). In E–G, the arrows point to collagen surrounding normal vessels.

Figure 4

Collagen deposition in hearts of 12 mo FVII^tTA/tTA mice. Low-resolution brightfield images of picro-Sirius Red staining (red) in (A) 12 mo WT and (B) 12 mo FVII^tTA/tTA mice. In B, the arrows indicate locations of some of the collagen deposits in the LV, RV and the interventricular septal wall (20×). Masson’s trichrome staining of collagen (blue) in heart sections from (C) 12 mo WT and (F) 12 mo FVII^tTA/tTA mice (20×). Higher resolution brightfield images of picro-Sirius Red staining (red) in (D) 12 mo WT and (G) 12 mo FVII^tTA/tTA mice (60×). Epifluorescence detection (red) of collagen in hearts of (E) 12 mo WT and (H) 12 mo FVII^tTA/tTA hearts (40×). In C, D, and E, the arrows point to collagen surrounding normal vessels.

Figure 5

Quantitative assessment of markers of cardiac inflammation, bleeding, and fibrosis in FVII^tTA/tTA mice at 4 mo and 12 mo of age. (A) CD45, (B) haemosiderin, (C) collagen, (D) macrophage, (E) SMC+ vessel counts in the LV, with (F) an example of a 4 mo FVII^tTA/tTA mouse heart section image (400×) used for counting. In this case, the arrowheads point to some of the vessels present. The quantitative data were obtained from morphometric analysis of histology sections (n = 6, 5, 4, and 4 hearts). At least 3 sections/mouse were counted for 4 mo WT, 4 mo FVII^tTA/tTA, 12 mo WT, and 12 mo FVII^tTA/tTA counts, respectively.

Figure 6

Q-RT-PCR analysis of transcripts associated with tissue remodelling and fibrosis in FVII^tTA/tTA mice. Results for mRNA levels of (A) TNF-α (B) MMP-2, (C) MMP-3, (D) MMP-9, (E) MMP-12, and (F) TIMP-1. (G and H) Immunohistochemical staining of MMP-9 in 4 mo of age (G) WT and (H) FVII^tTA/tTA mouse hearts (100×). (I) Transcript levels of TGF-β in heart tissue. The black and grey bars represent values of WT and FVII^tTA/tTA mice, respectively. Data are shown as the mean ± S.E.M, n = 5, for WT mice, and n = 6 for FVII^tTA/tTA mice at 4 mo of age; n = 4 for WT and n = 4 for FVII^tTA/tTA mice at 12 mo of age. *P<0.05.

Figure 7

Time course of development of cardiac fibrosis in FVII^tTA/tTA mice. (A–E) Histology of FVII^tTA/tTA hearts at 5 wk of age. (A) H&E staining (100×). (B) Anti-P-Selectin immunostaining (100×). The arrows point to platelet accumulation (brown). (C) Anti-CD45 staining (100×). Abundant brown-stained cells indicate CD45⁺-leukocyte infiltration. (D) Picro-Sirius Red staining (100×). (E) Prussian Blue staining (100×). (F–I) Histology of FVII^tTA/tTA hearts at 6 wk of age (100×). (F) H&E staining. (G) Anti-CD45 immunostaining (H) Picro-Sirius Red. (I) Prussian Blue. (J–L) Histology of FVII^tTA/tTA hearts in 8 wk mice (100×). (L) Prussian Blue. (K) Picro-Sirius Red. (L) Anti-CD45 immunostaining for leukocytes. (M–P) TEM images of the LV of 6 wk of age WT (M and N) and FVII^tTA/tTA (O and P) mice. (M) Intact blood vessels (bv) containing red blood cells (rbc) are seen in the ventricular parenchema within the collagenous matrix (*) in the WT specimen (2000×). (N) Active blood vessels (bv) are located in the interstitial space (I) between cardiomyocytes (cm) and do not modify the morphology of the cardiac cells (3500×). (O and P) The LV of FVII^tTA/tTA mice. (O) Cardiomyocyte (cm)-containing RBC from a forming vessel. Extravascular RBC can also be seen in the area. An endothelial cell can be seen in recently formed vessels as well as endothelial cell processes (arrow) (3000×). (P) A fibrotic area of the LV. Macrophages (m) with included phagocytic vesicles after cell debris clearance and abundant fibroblasts (f) embedded in a rich collagenous matrix (*) are present. Cm surround the fibrotic areas (2500×).