Cre/lox Studies Identify Resident Macrophages as the Major Source of Circulating Coagulation Factor XIII-A.

OBJECTIVE: To establish the cellular source of plasma factor (F)XIII-A. APPROACH AND RESULTS: A novel mouse floxed for the F13a1 gene, FXIII-Aflox/flox (Flox), was crossed with myeloid- and platelet-cre-expressing mice, and cellular FXIII-A mRNA expression and plasma and platelet FXIII-A levels were measured. The platelet factor 4-cre.Flox cross abolished platelet FXIII-A and reduced plasma FXIII-A to 23±3% (P<0.001). However, the effect of platelet factor 4-cre on plasma FXIII-A was exerted outside of the megakaryocyte lineage because plasma FXIII-A was not reduced in the Mpl-/- mouse, despite marked thrombocytopenia. In support of this, platelet factor 4-cre depleted FXIII-A mRNA in brain, aorta, and heart of floxed mice, where FXIII-Apos cells were identified as macrophages as they costained with CD163. In the integrin αM-cre.Flox and the double copy lysozyme 2-cre.cre.Flox crosses, plasma FXIII-A was reduced to, respectively, 75±5% (P=0.003) and 30±7% (P<0.001), with no change in FXIII-A content per platelet, further consistent with a macrophage origin of plasma FXIII-A. The change in plasma FXIII-A levels across the various mouse genotypes mirrored the change in FXIII-A mRNA expression in aorta. Bone marrow transplantation of FXIII-A+/+ bone marrow into FXIII-A-/- mice both restored plasma FXIII-A to normal levels and replaced aortic and cardiac FXIII-A mRNA, while its transplantation into FXIII-A+/+ mice did not increase plasma FXIII-A levels, suggesting that a limited population of niches exists that support FXIII-A-releasing cells. CONCLUSIONS: This work suggests that resident macrophages maintain plasma FXIII-A and exclude the platelet lineage as a major contributor.

Transglutaminase 2 moderates the expansion of mouse abdominal aortic aneurysms.

OBJECTIVE: Previously published work has indicated that transcripts encoding transglutaminase 2 (TG2) increase markedly in a rat model of abdominal aortic aneurysm. This study determines whether TG2 and the related TG, factor XIII-A (FXIII-A), protect against aortic aneurysm development in mice. METHODS: C57BL/6J wild-type, Tgm2 -/- knockout, F13a1 -/- knockout, and Tgm2 -/- /F13a1 -/- double knockout mice were subjected to laparotomy and periaortic application of CaCl2. RESULTS: Tgm2 -/- mice showed slightly greater aortic dilatation at 6 weeks after treatment when compared with wild type. However, vessels from Tgm2 -/- mice, but not wild-type mice, continued to dilate up to 6 months after injury and by 24 weeks, a greater number of Tgm2 -/- mice had developed aneurysms (16/17 vs 10/19; P = .008). Laparotomy resulted in a high death rate in F13a1 -/- knockout mice, more frequently from cardiac complications than from hemorrhage, but among F13a1 -/- mice that survived for 6 weeks after CaCl2 treatment, abdominal aortic aneurysm diameter was unaltered relative to wild-type mice. Laparotomy resulted in a higher death rate among Tgm2 -/- /F13a1 -/- double knockout mice, owing to an increased frequency of delayed bleeding. Surprisingly, Tgm2 -/- /F13a1 -/- double knockout mice showed a trend toward decreased dilatation of the aorta 6 weeks after injury, and this finding was replicated in Tgm2 -/- /F13a1 -/- mice subjected to carotid artery injury. Levels of transcripts encoding TG2 were not increased in the aortas of injured wild-type or F13a1 -/- knockout mice relative to uninjured mice, although changes in the levels of other transcripts accorded with previous descriptions of the CaCl2 aneurysm model in mice. CONCLUSIONS: Knockout of Tgm2, but not F13a1 exacerbates aortic dilatation, suggesting that TG2 confers protection. However, levels of TG2 messenger RNA are not acutely elevated after injury. FXIII-A plays a role in preventing postoperative damage after laparotomy, confirming previous reports that it prevents distal organ damage after trauma. TG2 promotes wound healing after surgery and, in its absence, the bleeding diathesis associated with FXIII-A deficiency is further exposed.

Transglutaminase 2 limits the extravasation and the resultant myocardial fibrosis associated with factor XIII-A deficiency.

BACKGROUND AND AIMS: Transglutaminase (TG) 2 and Factor (F) XIII-A have both been implicated in cardiovascular protection and repair. This study was designed to differentiate between two competing hypotheses: that TG2 and FXIII-A mediate these functions in mice by fulfilling separate roles, or that they act redundantly in this respect. METHODS: Atherosclerosis was assessed in brachiocephalic artery plaques of fat-fed mixed strain apolipoprotein (Apo)e deficient mice that lacked either or both transglutaminases. Cardiac fibrosis was assessed both in the mixed strain mice and also in C57BL/6J Apoe expressing mice lacking either or both transglutaminases. RESULTS: No difference was found in the density of buried fibrous caps within brachiocephalic plaques from mice expressing or lacking these transglutaminases. Cardiac fibrosis developed in both Apoe/F13a1 double knockout and F13a1 single knockout mice, but not in Tgm2 knockout mice. However, concomitant Tgm2 knockout markedly increased fibrosis, as apparent in both Apoe/Tgm2/F13a1 knockout and Tgm2/F13a1 knockout mice. Amongst F13a1 knockout and Tgm2/F13a1 knockout mice, the extent of fibrosis correlated with hemosiderin deposition, suggesting that TG2 limits the extravasation of blood in the myocardium, which in turn reduces the pro-fibrotic stimulus. The resulting fibrosis was interstitial in nature and caused only minor changes in cardiac function. CONCLUSIONS: These studies confirm that FXIII-A and TG2 fulfil different roles in the mouse myocardium. FXIII-A protects against vascular leakage while TG2 contributes to the stability or repair of the vasculature. The protective function of TG2 must be considered when designing clinical anti-fibrotic therapies based upon FXIII-A or TG2 inhibition.

Normal Bone Deposition Occurs in Mice Deficient in Factor XIII-A and Transglutaminase 2.

Transglutaminase activity has been widely implicated in bone deposition. A predominant role has been proposed for factor (F)XIII-A and a subsidiary role suggested for the homologous protein, transglutaminase 2. Full-length FXIII-A is an 83kDa protransglutaminase that is present both in plasma and also in haematopoietic and connective tissue lineages. Several studies have reported expression in murine cells, including osteocytes, of a 37 kDa protein that reacts with the monoclonal anti-FXIII-A antibody AC-1A1. This protein was presumed to be a catalytically active fragment of FXIII-A-83 and to play a major role in bone deposition. We detected a 37 kDa AC-1A1 reactive protein in FXIII-A mRNA negative cell lines and in tissues from FXIII-A(-/-) mice. By mass spectrometric sequencing of AC-1A1 immunoprecipitates, we identified this protein as transaldolase-1, and confirmed that recombinant transaldolase-1 is recognised by AC-1A1. We have also shown that bone deposition is normal in FXIII-A(-/-).TG2(-/-) double knockout mice, casting doubt on the role of transglutaminases in bone mineralisation. Various studies have used antibody AC-1A1 for immunohistochemistry or immunofluorescence. We observe strong FXIII-A dependent staining in paraffin embedded mouse heart sections, with relatively low background in non-expressing mouse cells. In contrast, FXIII-A independent staining predominates in cultured human cells using a standard immunofluorescence procedure. Immunofluorescence is present in membrane compartments that are expected to lack transaldolase, indicating that other off-target antigens are recognised by AC-1A1. This has significant implications for studies that have used this approach to define the subcellular trafficking of FXIII-A in osteocytes.