LETTERS TO THE EDITOR
Long-term estrogen treatment of mice with a prothrombotic phenotype induces sustained increases in thrombin generation without affecting tissue fibrin deposition
Article first published online: 30 OCT 2012
© 2012 International Society on Thrombosis and Haemostasis
Journal of Thrombosis and Haemostasis
Volume 10, Issue 11, pages 2392–2394, November 2012
How to Cite
CLEUREN, A. C. A., VAN OERLE, R., REITSMA, P. H., SPRONK, H. M. and VAN VLIJMEN, B. J. M. (2012), Long-term estrogen treatment of mice with a prothrombotic phenotype induces sustained increases in thrombin generation without affecting tissue fibrin deposition. Journal of Thrombosis and Haemostasis, 10: 2392–2394. doi: 10.1111/j.1538-7836.2012.04916.x
- Issue published online: 30 OCT 2012
- Article first published online: 30 OCT 2012
- Accepted manuscript online: 5 SEP 2012 10:55AM EST
- Received 6 May 2012, accepted 30 August 2012
Venous thrombosis represents a serious complication of oral contraceptive use and hormone replacement therapy. The estrogen component, often 17α-ethinylestradiol, is considered to be the predominant thrombotic constituent and we have previously shown that oral ethinylestradiol (EE) in mice also has profound effects on the plasma coagulation profile, at least at the level of individual pro- and anticoagulant factors . The overall effect of alterations in the hemostatic balance can be determined by assessing thrombin generation and subsequent calculation of the endogenous thrombin potential (ETP). In women using oral contraceptives and hormone replacement therapy, the ETP is increased and it is believed that this has a predictive value regarding the venous thrombotic risk [2,3]. Therefore, we now report the overall effect of estrogen-induced alterations on the mouse hemostatic balance by assessing thrombin generation. In addition, we determined the relation between thrombin generation and the net effect on a spontaneous thrombotic phenotype (i.e. fibrin depositions in lung tissues of the prothrombotic factor V Leiden (FVQ/Q) and thrombomodulin proline mutant (TMpro/pro) mice).
FVQ/Q and TMpro/pro mice [4,5], both on a C57BL/6J background, were bred at the animal facilities of the Leiden University and Maastricht University, respectively. After ovariectomy, mice were orally treated with 1 μg EE per day or a vehicle for either 10 days or 10 weeks, while ovariectomized wild-type C57BL/6J animals (Charles River, Maastricht, the Netherlands) were treated for 10 days only as a reference (n = 12 per group). After the last EE or vehicle administration, mice were exsanguinated and platelet-poor plasma was obtained . In addition, lungs of 10-week-treated mice were isolated for fibrin deposition analyses .
Thrombin generation by means of the calibrated automated thrombogram was assessed after 10 days of treatment to determine the overall plasma coagulability. Hereto, 1:6 diluted plasma was triggered with a low tissue factor concentration (1 pm) to include the contribution of all EE-induced changes in procoagulant and anticoagulant factors of both the extrinsic and intrinsic pathway [7,8]. Thrombin generation curves showed a decreased initiation and propagation of thrombin generation under EE administration (Fig. 1A,B). This is in line with our previous data showing reduced levels of procoagulant factors upon estrogen treatment . ETP values were increased in plasmas of EE-treated FVQ/Q and TMpro/pro mice, effects that were comparable to the effects observed in wild-type animals (448 ± 30 nm*min vs. 930 ± 41 nm*min, P < 0.001). These increased values were mainly due to a prolonged thrombin activity as the result of a diminished inhibition, because the tail of the curve started later in this group (Fig. 1A,B). As antithrombin is the main inhibitor of thrombin activity, we measured the antithrombin activity levels in plasma (Coamatic antithrombin kit; Chromogenix, Milan, Italy) and found that estrogen administration caused a significant 15% reduction in antithrombin activity levels, which negatively correlated with the ETP values (Pearson r = −0.51, P = 0.002).
To determine the relation between thrombin generation and a spontaneous thrombotic phenotype, FVQ/Q and TMpro/pro mice were daily vehicle- or EE-treated for 10 weeks. During these 10 weeks, none of the mice died or showed signs of abnormalities. In addition, careful inspection of the mice upon sacrifice revealed no signs of macrovascular thrombosis. Thrombin generation curves after 10 weeks of treatment showed a comparable pattern to the 10-day treatment, including the effects on ETP (for FVQ/Q mice 446 ± 40 nm*min vs. 746 ± 98 nm*min, P < 0.01, and for TMpro/pro mice 683 ± 75 nm*min vs. 901 ± 70 nm*min, P < 0.05), peak height, velocity index and tail values (data not shown), indicating a sustained increase in thrombin generation during the 10 weeks of oral ethinylestradiol treatment.
In order to establish whether this sustained increase coincided with an enhanced activity of the coagulation system in vivo, plasma thrombin-antithrombin complex levels were measured (Enzygnost TAT micro; Dade Behring, Leusden, the Netherlands), but no difference between treatment groups was detected (Fig. 1C). As FVQ/Q and TMpro/pro mice can display fibrin depositions as a marker of increased thrombin generation and thus a spontaneous thrombotic phenotype [4,5], we determined fibrin deposition levels in lung homogenates of 10-week-treated animals by Western blotting . Although fibrin depositions of FVQ/Q and TMpro/pro mice were in the lower range of detection (2–10 ng mg−1 tissue), they were higher than the fibrin concentrations present in wild-type mice, because these did not exceed the detection threshold of 2 ng mg−1. However, oral estrogen administration for 10 weeks did not result in significant increased fibrin depositions in either the FVQ/Q (4.0 ± 0.5 ng mg−1 vs. 5.5 ± 0.7 ng mg−1) or TMpro/pro mice (5.3 ± 0.6 ng mg−1 vs. 4.8 ± 0.4 ng mg−1; Fig. 1D).
To exclude the possibility that fibrin depositions were not affected because of EE-induced changes in fibrinolysis, plasmin-antiplasmin (PAP) levels were measured (mouse PAP ELISA; Cusabio, Kampenhout, Belgium). For both the vehicle- and estrogen-treated groups, PAP levels were below 3.1 pg mL−1, indicating that the low fibrin depositions in general, and the lack of an effect of EE administration in particular, cannot be explained by increased fibrinolysis.
Previous studies with FVQ/Q and TMpro/pro mice have shown that exposure to several genetic and environmental risk factors, including deletions in tissue factor pathway inhibitor and hypoxia, can aggravate the thrombotic phenotype of these mice [5,9,10]. Although we were not able to detect enhanced fibrin depositions upon prolonged exposure to oral ethinylestradiol, this lack of a thrombogenic response is in line with our previous data, which demonstrated that short-term EE administration did not result in an increased thrombogenicity in a stasis-induced thrombosis model .
Taken together, the current data indicate that long-term oral ethinylestradiol treatment of factor V Leiden and thrombomodulin proline mutant mice can induce a sustained increase in thrombin generation as determined by the calibrated automated thrombogram. However, this does not coincide with increased levels of thrombin-antithrombin complexes in vivo, nor does it result in fibrinolysis-dependent or independent changes in fibrin depositions. Therefore, the increased thrombin generation does not translate into a spontaneous macrovascular or microvascular thrombotic phenotype, which argues against the use of these mice in studying the effects of estrogens on thrombosis.
We thank P. Pluijmen and D. Fens from the Cardiovascular Research Institute Maastricht for their technical assistance. This study was financially supported by the Netherlands Heart Foundation (grant 2006B045).
Disclosure of Conflict of Interests
The authors state that they have no conflict of interest.