Heparanase Procoagulant Activity in Women Using Oral Contraceptives
Heparanase Procoagulant Activity in Women Using Oral Contraceptives
In estrogen receptor-positive cells, estrogen increased heparanase procoagulant activity while in the absence of the estrogen receptor this effect was not observed, supporting an estrogen receptor-dependent activity (Figs 1–3). These results potentially indicate a new mechanism of hypercoagulability in women using estrogen. In order to gain clinical relevance to this notion we studied the plasma of 34 women using OC and compared them with 41 women without hormonal therapy. The results indicate a significant increase in TF/heparanase activity, mostly attributed to heparanase procoagulant activity, although TF activity also increased (Fig. 4). Heparanase level by ELISA showed no difference between the groups. The ELISA of heparanase dominantly detects the enzymatic active heparanase (50 + 8 kDa) and less sensitive to the pro-heparanase (65 kDa) (Shafat et al., 2006) that is active in the hemostatic system (Nadir et al., 2011a). Therefore, it is possible that the ELISA results do not correlate to the procoagulant activity of heparanase. Previously, we studied the plasma of 35 women at the end of pregnancy compared with the plasma of 20 non-pregnant women (Nadir et al., 2011b). There is a considerable similarity in the results of pregnant women and the OC user group regarding the pattern of changes and level of activities. In pregnant versus non-pregnant women, the TF/heparanase, heparanase, TF and Xa activities were 0.53 versus. 0.17, 0.33 versus 0.05, 0.2 versus 0.12 and 0.34 versus 0.22 (absorbance 405–490 nm), respectively. Accordingly, in the OC users compared with non-users the TF/heparanase, heparanase, TF and Xa activities were 0.6 versus. 0.26, 0.4 versus 0.12, 0.2 versus 0.14 and 0.09 versus 0.05 (absorbance 405–490 nm), respectively. In heparanase ELISA, as in our present data, no significant difference was observed. In the present study results of Xa levels are about four times lower compared with our previous study as the test was done using different methodology (25 µl of plasma with 20 min incubation, and in the previous article, 50 µl of plasma with 30 min incubation). Hence, with the limitation of two separate studies, in the pregnant women group, as in the OC group, TF/heparanase activity was increased and the dominant parameter of the complex was the increase in heparanase procoagulant activity. Downstream the extrinsic coagulation cascade, levels of factor Xa and D-dimer were significantly increased in the OC users, although TAT was comparable between the groups. The fact that no difference was found in the TAT may result from sensitivity limitations of the assay and is in accordance with previous published data (Pinto et al., 1997; Wiegratz et al., 2008). Figure 5 demonstrates the wide range of heparanase procoagulant activity in the study group and Table II shows that OC users are at increased risk to have high heparanase procoagulant activity. The clinical significance of the assay as a predictor of thrombotic risk and as a routine test before and/or immediately after starting OC should be addressed in a large clinical trial. Additionally, large-scale study is needed to answer questions regarding the effect of different estrogen dose, progesterone dose and formulation in pills, age and smoking on heparanase procoagulant activity.
Tamoxifen, a partial estrogen receptor antagonist most widely used drug in endocrine therapy of breast cancer, is known to increase the risk of thrombosis by 2–3-fold (Hernandez et al., 2009). Cohen et al. (2007) demonstrated that tamoxifen induced heparanase expression in MCF-7 cells. According to our results, tamoxifen partially inhibited the procoagulant effect of heparanase in the presence of estrogen but increased heparanase procoagulant activity in the absence of estrogen (Fig. 2A). Following chemotherapy treatment the menstrual cycle may stop intermittently or permanently, causing a drop in estrogen level in the plasma. Tamoxifen is usually started after chemotherapy for 2–5 years. It is possible that in women with a drop in normal estrogen level, tamoxifen may induce increased heparanase procoagulant activity and a risk of thrombosis. This notion is in accordance with the data of Hernandez et al. (2009) showing that in women at the age of 45–69, older women were at increased risk of thrombosis during tamoxifen treatment.
According to the results, there is a discrepancy in the contribution of heparanase activity to the TF/heparanase complex in the women's plasma compared with the medium of the cancer cell lines. The contribution is calculated as heparanase activity/TF + heparanase activity. While in the plasma the contribution in women without pills was 45%, and in women using pills was 65% (Fig. 4), in medium of MCF-7 cells without estrogen the heparanase procoagulant activity was 1% and in the presence of estrogen it reached 20% (Fig. 2). The reason for the discrepancy could be secretion of TF bearing apoptotic bodies to the medium over the 16 h of incubation in serum-free medium or different setups of hormonal effect in health versus malignancy.
The main limitation of the current study is that the two estrogens given to the women and cell cultures: EE and E2, respectively, may have different effects on the coagulation system, although an increase in heparanase procoagulant activity was demonstrated in both of them. Different ways of estrogen administration, i.e. orally or transdermally may also influence the procoagulant effect. These important issues should further be investigated.
In summary, the present work suggests a new mechanism for activation of the coagulation system during estrogen treatment. Although the sample size of the study group was limited, significant differences in the activation of the extrinsic coagulation pathway were demonstrated. The clinical relevance of the heparanase procoagulant activity assay as a screening tool in thrombophilia work-up ought to be elucidated in the future.
Discussion
In estrogen receptor-positive cells, estrogen increased heparanase procoagulant activity while in the absence of the estrogen receptor this effect was not observed, supporting an estrogen receptor-dependent activity (Figs 1–3). These results potentially indicate a new mechanism of hypercoagulability in women using estrogen. In order to gain clinical relevance to this notion we studied the plasma of 34 women using OC and compared them with 41 women without hormonal therapy. The results indicate a significant increase in TF/heparanase activity, mostly attributed to heparanase procoagulant activity, although TF activity also increased (Fig. 4). Heparanase level by ELISA showed no difference between the groups. The ELISA of heparanase dominantly detects the enzymatic active heparanase (50 + 8 kDa) and less sensitive to the pro-heparanase (65 kDa) (Shafat et al., 2006) that is active in the hemostatic system (Nadir et al., 2011a). Therefore, it is possible that the ELISA results do not correlate to the procoagulant activity of heparanase. Previously, we studied the plasma of 35 women at the end of pregnancy compared with the plasma of 20 non-pregnant women (Nadir et al., 2011b). There is a considerable similarity in the results of pregnant women and the OC user group regarding the pattern of changes and level of activities. In pregnant versus non-pregnant women, the TF/heparanase, heparanase, TF and Xa activities were 0.53 versus. 0.17, 0.33 versus 0.05, 0.2 versus 0.12 and 0.34 versus 0.22 (absorbance 405–490 nm), respectively. Accordingly, in the OC users compared with non-users the TF/heparanase, heparanase, TF and Xa activities were 0.6 versus. 0.26, 0.4 versus 0.12, 0.2 versus 0.14 and 0.09 versus 0.05 (absorbance 405–490 nm), respectively. In heparanase ELISA, as in our present data, no significant difference was observed. In the present study results of Xa levels are about four times lower compared with our previous study as the test was done using different methodology (25 µl of plasma with 20 min incubation, and in the previous article, 50 µl of plasma with 30 min incubation). Hence, with the limitation of two separate studies, in the pregnant women group, as in the OC group, TF/heparanase activity was increased and the dominant parameter of the complex was the increase in heparanase procoagulant activity. Downstream the extrinsic coagulation cascade, levels of factor Xa and D-dimer were significantly increased in the OC users, although TAT was comparable between the groups. The fact that no difference was found in the TAT may result from sensitivity limitations of the assay and is in accordance with previous published data (Pinto et al., 1997; Wiegratz et al., 2008). Figure 5 demonstrates the wide range of heparanase procoagulant activity in the study group and Table II shows that OC users are at increased risk to have high heparanase procoagulant activity. The clinical significance of the assay as a predictor of thrombotic risk and as a routine test before and/or immediately after starting OC should be addressed in a large clinical trial. Additionally, large-scale study is needed to answer questions regarding the effect of different estrogen dose, progesterone dose and formulation in pills, age and smoking on heparanase procoagulant activity.
Tamoxifen, a partial estrogen receptor antagonist most widely used drug in endocrine therapy of breast cancer, is known to increase the risk of thrombosis by 2–3-fold (Hernandez et al., 2009). Cohen et al. (2007) demonstrated that tamoxifen induced heparanase expression in MCF-7 cells. According to our results, tamoxifen partially inhibited the procoagulant effect of heparanase in the presence of estrogen but increased heparanase procoagulant activity in the absence of estrogen (Fig. 2A). Following chemotherapy treatment the menstrual cycle may stop intermittently or permanently, causing a drop in estrogen level in the plasma. Tamoxifen is usually started after chemotherapy for 2–5 years. It is possible that in women with a drop in normal estrogen level, tamoxifen may induce increased heparanase procoagulant activity and a risk of thrombosis. This notion is in accordance with the data of Hernandez et al. (2009) showing that in women at the age of 45–69, older women were at increased risk of thrombosis during tamoxifen treatment.
According to the results, there is a discrepancy in the contribution of heparanase activity to the TF/heparanase complex in the women's plasma compared with the medium of the cancer cell lines. The contribution is calculated as heparanase activity/TF + heparanase activity. While in the plasma the contribution in women without pills was 45%, and in women using pills was 65% (Fig. 4), in medium of MCF-7 cells without estrogen the heparanase procoagulant activity was 1% and in the presence of estrogen it reached 20% (Fig. 2). The reason for the discrepancy could be secretion of TF bearing apoptotic bodies to the medium over the 16 h of incubation in serum-free medium or different setups of hormonal effect in health versus malignancy.
The main limitation of the current study is that the two estrogens given to the women and cell cultures: EE and E2, respectively, may have different effects on the coagulation system, although an increase in heparanase procoagulant activity was demonstrated in both of them. Different ways of estrogen administration, i.e. orally or transdermally may also influence the procoagulant effect. These important issues should further be investigated.
In summary, the present work suggests a new mechanism for activation of the coagulation system during estrogen treatment. Although the sample size of the study group was limited, significant differences in the activation of the extrinsic coagulation pathway were demonstrated. The clinical relevance of the heparanase procoagulant activity assay as a screening tool in thrombophilia work-up ought to be elucidated in the future.
