SEARCH

SEARCH BY CITATION

Keywords:

  • coagulation;
  • estrogen;
  • liver;
  • mice;
  • mRNA

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References
  10. Supporting Information

Summary.  Background: Oral estrogen use is associated with changes in plasma levels of many coagulation proteins. Objective: To gain more insight into the underlying mechanism of estrogen-induced changes in coagulation. Methods: Ovariectomized female mice were used to study the impact of oral 17α-ethinylestradiol (EE) on plasma coagulation, hepatic coagulation gene transcript levels, and dependence on estrogen receptor (ER) α and ERβ. Results: Ten days of oral EE treatment resulted in significantly reduced plasma activity levels of factor (F)VIII, FXII, combined FII/FVII/FX and antithrombin, whereas FIX activity significantly increased. Regarding hepatic transcript levels, oral EE caused significant decreases in fibrinogen-γ, FII, FV, FVII, FX, FXII, antithrombin, protein C, protein Z, protein Z inhibitor and heparin cofactor II mRNA levels, whereas FXI levels significantly increased and transcript levels of FVIII, FIX, protein S and α2-antiplasmin remained unaffected. All EE-induced coagulation-related changes were neutralized by coadministration of the non-specific ER antagonist ICI182780. In addition, ERα-deficient mice lacked the EE-induced changes in plasma coagulation and hepatic transcript profile, whereas ERβ-deficient mice responded similarly to non-deficient littermate controls. A crucial role for the ER was further demonstrated by its rapid effects on transcription, within 2.5–5 h after EE administration, suggesting a short chain of events leading to its final effects. Conclusions: Oral EE administration has a broad impact on the mouse coagulation profile at the level of both plasma and hepatic mRNA levels. The effects on transcription are rapidly induced, mostly downregulatory, and principally mediated by ERα.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References
  10. Supporting Information

Estrogens in contraceptives and hormone replacement therapy cause changes in the plasma coagulation profile. These changes include increases in the levels of the procoagulant factors II, VII, IX, X, XII, XIII, and fibrinogen, and reductions in the natural anticoagulant protein S and antithrombin, resulting in an increased risk for venous thrombosis (reviewed in [1,2]). Changes in coagulation were first identified in women taking oral contraceptives, but they were not seen in women using transdermal patches [3,4], suggesting that the first-pass effect by the liver is important for inducing these changes, presumably by affecting the hepatic transcription of coagulation factor genes. However, several studies have demonstrated that 17α-ethinylestradiol (EE)-containing patches may also confer an increased venous thrombotic risk as compared to estradiol-containing patches [5,6], indicating that the type of estrogen is also of importance with respect to modulating coagulation and thereby the risk for venous thrombosis.

Estrogens largely mediate their effects by binding to the estrogen receptor (ERα or ERβ), which subsequently functions as a transcription factor by binding to estrogen response elements (EREs) in the promoter or enhancer regions of estrogen-responsive genes. Functional EREs have been identified with certainty for a limited number of coagulation factors, including FVII and FXII [7,8]. Recent human and mouse genome-wide searches for high-affinity EREs demonstrated that near-consensus EREs occur in many of the genes belonging to the procoagulant and anticoagulant pathways [9]. For the human genome, these include the liver-specific coagulation factors II, V, VII, IX, X, XI, XII, protein S, protein Z and heparin cofactor II, whereas this list is less extensive for the mouse genome, comprising FVII, FX, FXIII, and α2-antiplasmin. For the mouse, however, functional genome-wide screening of hepatic ERα-binding regions identified them in the fibrinogen, FII, FXI, antithrombin, protein C, protein S, protein Z, plasminogen and heparin cofactor II genes [10]. These genome-wide results predict that estrogens potentially modulate transcription of a large number of coagulation genes expressed in the liver through the direct action of ERs. In this respect, it is surprising that coagulation-dedicated cDNA microarray analyses of livers of mice treated with estrogen revealed that none of the liver-specific coagulation genes was regulated by 17β-estradiol [11].

Given the inconsistency between the results of these studies, we performed a series of in vivo experiments to directly study the effects of oral EE, the estrogenic and most thrombogenic component of oral contraceptives, on the plasma coagulation profile on the one hand, and the impact of a single and multiple EE dose on hepatic coagulation gene transcription and its possible dependency on ERs on the other.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References
  10. Supporting Information

Animals

Heterozygous mice carrying an ERα knockout allele (Esr1+/−) or ERβ knockout allele (Esr2+/−) on a C57Bl/6J background were obtained from Jackson Laboratories (Bar Harbor, ME, USA) and intercrossed to generate female mice with complete receptor deficiency (Esr1−/− or Esr2−/−). To avoid possible baseline differences in plasma and mRNA expression levels resulting from the different origins of the mice, experiments with Esr1 and Esr2 mice included littermate wild-type controls as a reference (Esr1+/+ or Esr2+/+, respectively). Genotypes were confirmed by PCR analysis according to the protocol provided by the Jackson Laboratories. For experiments performed with wild-type female mice only, C57Bl/6J mice were purchased from Charles River (Maastricht, the Netherlands). All mice were housed under a 12-h light/dark cycle, and standard chow diet and drinking water were provided ad libitum. At 8 weeks of age, mice were bilaterally ovariectomized under isoflurane anesthesia, and after a 2-week recovery period, they were randomly assigned to either the experimental group or vehicle treatment group. For the Esr1+/+ and Esr1−/− mice, ovariectomy was performed between an average of 10 and 12 weeks of age.

Hormone treatment

EE and ICI182780 (both from Sigma Aldrich, Steinheim, Germany) stocks were prepared in ethanol, and subsequently diluted in arachid oil with a final concentration of 1% ethanol. For estrogen treatment designated as ‘multiple doses’, mice received a daily gavage of 1 μg of EE in 100 μL of arachid oil for 10 consecutive days. For estrogen treatment designated as ‘single dose’, EE was given only once. The non-specific ER antagonist ICI182780 was injected subcutaneously at a daily dose of 100 μg per mouse, starting 1 day prior to the EE administration in the case of the combined treatment. For the vehicle treatment, mice received 100 μL of arachid oil with an ethanol concentration of 1%, either orally or subcutaneously, as appropriate. In addition, separate experiments were performed to determine whether alternative estrogen treatment protocols yielded essentially different results, and included subcutaneous treatment of ovariectomized mice with 1 μg 17β-estradiol d−1 for 10 days, or oral treatment with 1 μg EE per mouse d−1 for 50 days. Furthermore, non-ovariectomized female mice were treated with 1 μg EE per mouse d−1 for 10 days to determine the additional effect of exogenous estrogen administration.

After the last administration, mice were anesthetized by an intraperitoneal injection of a mixture of ketamine (100 mg kg−1), xylazine (12.5 mg kg−1), and atropine (125 μg kg−1), after which the abdomen was opened by a midline incision and a blood sample on sodium citrate (final concentration of 0.32%) was drawn from the inferior caval vein. Plasma was obtained as previously described [12], and stored at − 80 °C until use. Part of the left liver lobule was isolated and snap-frozen for mRNA analyses, and the uterus was collected and weighed to determine the biological activity of both EE and ICI182780. All experimental procedures were approved by the animal welfare committee of Leiden University.

Plasma analyses

Alkaline phosphatase (ALP), aspartate transaminase (AST) and alanine aminotransferase (ALT) levels were determined with routine clinical chemistry assays.

Plasma activity levels of FVIII, FIX, FXI and FXII were determined in an activated partial thromboplastin time (APTT)-based assay [13] by mixing individual mouse plasmas with human plasma deficient for the respective factor (STA FVIII; Diagnostica Stago, Asnieres, France; FIX-deficient and FXI-deficient plasma from Biopool, Bray, Ireland; and HemosIL FXII from Instrumentation Laboratories, Milan, Italy) and automated APTT reagent (Biomerieux, Durham, NC, USA). Combined FII/FVII/FX activity was assessed by using the Thrombotest (Axis-Shield, Oslo, Norway). For all of these activity assays, clotting times were measured with a semi-automated coagulometer (ACL-300; Instrumentation Laboratories).

In addition to the combined activity, FII, FVII and FX activities were also measured separately. FII activity was determined after complete activation with ecarin (Sigma Aldrich), and this was followed by determination of FII amidolytic activity with the chromogenic substrate S2238 (Chromogenix, Milan, Italy). For FX activity, plasma was activated with Russell’s viper venom (Haematologic Technologies Inc., Essex Junction, VT, USA), and activity was measured by S2765 substrate (Chromogenix) conversion. FVII and antithrombin activity were determined using commercially available kits (Biophen FVII, Hyphen Biomed, Neuville-sur-Oise, France; and Coamatic Antithrombin kit, Chromogenix), according to the manufacturer’s protocol. Plasma antithrombin antigen and fibrinogen antigen levels were assessed by using commercial murine enzyme-linked immunosorbent assay kits from Affinity Biologicals (Ancaster, Canada).

For all plasma assays, mouse calibration curves paralleled human calibration curves, indicating specific reactions between mouse and human proteins. Pooled normal mouse plasma was used to generate standard curves, which were used to calculate the activity or antigen levels, and subsequently the wild-type vehicle-treated group was set as a reference (100%).

RNA isolation and real-time reverse transcription polymerase chain reaction (RT-PCR)

Individual liver samples (20–30 mg) of 10 animals per group were homogenized in RNAzol (Tel-Test, Friendswood, TX, USA), and RNA isolation and cDNA synthesis was performed as previously described [14], with minor adjustments. Gene-specific quantitative PCR primers were designed with primer express software (Applied Biosystems, Foster City, CA, USA), and are presented in Table S1. Quantitative real-time RT-PCR was performed on the ABI Prism 7900 HT Fast Real-Time PCR System from Applied Biosystems, and data were analyzed with the accompanying Sequence Detection System software. The comparative threshold cycle method with β-actin as internal control was used for quantification and normalization. The vehicle-treated wild-type group was set as a reference, and the ΔCt values of the individual samples were related to the mean ΔCt of the reference group.

Statistical analyses

Data are expressed as mean ± standard error of the mean or as the difference between the experimental and the vehicle-treated group with the standard error of the difference, as appropriate. Data were analyzed with GraphPad Instat software, and statistical differences between groups were evaluated by one-way analysis of variance (anova) with a Dunnett post hoc test to evaluate the dose-finding study and a Bonferroni post hoc test in the case of experiments performed with EE and ICI182780. For the ER-deficient mice, Student’s t-test was used to compare the estrogen-treated with the vehicle-treated mice of the same genotype. P-values < 0.05 were considered to be significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References
  10. Supporting Information

Dose-finding

A dose-finding study was performed as previously described [15], and an in-depth analysis is presented here, showing a hormone-dependent increase in uterus wet weight (Fig. 1A), with a dose of 0.1 μg EE per mouse d−1 restoring the uterus wet weight in ovariectomized mice to that observed in non-ovariectomized female mice, thus mimicking the endogenous estrogen levels. Analyses of circulating liver enzyme levels revealed a slight, but significant, increase in ALP and ALT levels in mice treated with 10 μg EE d−1, whereas these levels, and the levels of AST, were normal for all other EE doses tested (data not shown).

image

Figure 1.  Effects of 10 days of oral EE treatment on the uterus wet weight (A), plasma coagulation factor activity levels (B–C) and hepatic transcript levels (D–F) in ovariectomized female mice (N = 6 per group). Hepatic transcript levels are relative to β-actin as an internal control, and data are presented as mean ± standard error of the mean. Original data are presented in Tables S2 and S3. *P < 0.05, P < 0.01 and P < 0.001 vs. vehicle-treated animals. GpX3, glutathione peroxidase 3.

Download figure to PowerPoint

Oral EE dose-dependently increased FIX activity (Fig. 1B), whereas the levels of FVIII, FXII, combined FII/FVII/FX (Fig. 1C) and antithrombin antigen were decreased. For hepatic mRNA analyses, the glutathione peroxidase 3 (GpX3) and the mannose receptor (Mrc1) genes were used as positive controls, as it has been shown that these genes are estrogen-responsive [16,17], and, as expected, a dose-dependent increase in hepatic transcript levels was observed (Fig. 1D). The changes in levels of combined FII/VII/X and antithrombin coincided with dose-dependent changes in hepatic mRNA levels of these factors (Fig. 1E), whereas the transcript levels of FIX remained constant. In addition, although FXI activity in plasma was not significantly altered by EE treatment, mRNA levels were dose-dependently increased after EE administration (Fig. 1F).

A dose of 1 μg EE per mouse d−1 was selected for further evaluation, as this resulted in significant changes in the coagulation profile without affecting liver enzymes. This evaluation included comparison of the dose of EE with a similar dose of the naturally occurring 17β-estradiol (subcutaneous injection) in ovariectomized mice, which resulted in comparable effects on uterus weight and plasma coagulation (data not shown). Furthermore, oral treatment of non-ovariectomized female mice with 1 μg EE per mouse d−1 caused comparable effects on plasma coagulation as observed under ovariectomized conditions, although, overall, these were less pronounced, and a 50-day treatment regimen (1 μg EE per mouse d−1) also did not yield essentially different results from those obtained with the 10-day treatment period (data not shown).

ER antagonist

We subsequently determined whether the effects on the plasma coagulation profile and transcript levels observed with multiple doses of oral EE were mediated by ERα and/or ERβ, by additionally treating mice daily with the non-specific ER antagonist ICI182780 (n = 15 mice per group). This compound was also active in the mouse, because the uterus wet weight of mice treated with both ICI182780 and EE reverted to that of vehicle-treated animals (10.6 ± 0.5 mg vs. 9.5 ± 0.6 mg for the vehicle treatment), whereas EE alone induced a significant increase in the uterus wet weight (107.8 ± 3.9 mg; P < 0.0001). ICI182780 alone had no effect (9.7 ± 0.6 mg). Plasma activity levels of FVIII, FIX, FXII, combined FII/FVII/FX, and antithrombin, as well as antithrombin antigen levels, were significantly altered by estrogen administration (Fig. 2A), which was consistent with the results of the dose-finding. Coadministration with ICI182780 counteracted the EE-induced changes in plasma, whereas ICI182780 treatment alone had no effect on the plasma coagulation profile.

image

Figure 2.  Differences between ovariectomized mice treated with 1 μg of 17α-ethinylestradiol (EE) alone as compared with vehicle-treated mice (□), or mice treated with 1 μg of EE and 100 μg of ICI182780 as compared with vehicle-treated mice (⋄), in the plasma coagulation profile (A; N = 15 per group) and hepatic transcript levels (B; N = 10 per group). Original data are presented in Tables S4 and S5. Data are presented as difference ± standard error of the difference; black symbols, P < 0.05 as compared with vehicle-treated controls (0 μg of EE and 0 μg of ICI182780). ag, antigen level; act, activity level.

Download figure to PowerPoint

Figure 2B shows that the strong estrogen-induced upregulation of GpX3 and Mrc1 transcript levels was completely antagonized by ICI182780 coadministration. Estrogen alone caused significant downregulation of mRNA levels of fibrinogen-γ, FII, FV, FVII, FX, FXII, and antithrombin, whereas FXI transcript levels were significantly increased, which was again comparable to the previously observed changes in the dose-finding study. The panel of coagulation factors was extended, and revealed significant reductions, following EE administration, in mRNA levels of plasminogen, protein C, protein Z, protein Z inhibitor, and heparin cofactor II. Again, ICI182780 coadministration counteracted the EE-induced changes in hepatic transcript levels, resulting in mRNA levels comparable to those in vehicle-treated animals (Fig. 2B), and ICI182780 treatment alone had no effect on hepatic transcript levels. Protein C inhibitor and α2-macroglobulin mRNA levels were also measured, but they were present in too low amounts, which hampered accurate evaluation.

Recombinant mice

As the effects on the coagulation profile were sensitive to oral estrogen administration, and an ER antagonist was able to counteract these effects, experiments were repeated in ovariectomized mice lacking either ERα (Esr1−/−) or ERβ (Esr2−/−) to determine the contribution of the individual ER subtypes to the EE-mediated changes in the coagulation profile. As expected, 10-day oral EE administration did not cause a significant increase in the uterus wet weight of Esr1−/− mice (uterus: 16.7 ± 1.2 mg for the vehicle treatment vs. 16.3 ± 1.4 mg for EE treatment), whereas it did in the Esr1+/+ littermate controls (15.9 ± 1.8 mg vs. 116.5 ± 1.5 mg, < 0.0001). Upon EE administration, Esr1−/− mice showed no effects on plasma coagulation factor levels, whereas the littermate Esr1+/+ controls responded to EE with significantly reduced FXI and increased FIX activity levels (Fig. 3A). Although the plasma coagulation profile of Esr1+/+ mice was only modestly affected by EE treatment, evaluation of hepatic mRNA levels showed significant upregulation of the estrogen-responsive GpX3 and Mrc1 transcripts (Fig. 3B). In addition, hepatic mRNA levels of FXI and protein S were also significantly higher in EE-treated Esr1+/+ mice than in vehicle-treated animals of the same genotype. Fibrinogen-γ, FV, FVII, FX, plasminogen, protein C, protein Z and α2-antiplasmin transcript levels were significantly downregulated following EE treatment, thereby mimicking the observations made in EE-treated wild-type C57Bl/6J mice. In contrast, for the Esr1−/− mice, no significant alterations were observed when vehicle-treated and EE-treated mice were compared (Fig. 3B).

image

Figure 3.  Differences between ovariectomized Esr1+/+ mice treated with 1 μg of EE as compared with vehicle-treated mice (□), or ovariectomized Esr1−/− mice treated with 1 μg of EE as compared with vehicle-treated mice (⋄), in the plasma coagulation profile (A; N = 12–15 per group) and hepatic transcript levels (B; N = 10 per group). Original data are presented in Tables S6 and S7. Data are presented as difference ± standard error of the difference; black symbols, P < 0.05 as compared with vehicle-treated controls of the same genotype. ag, antigen level; act, activity level.

Download figure to PowerPoint

In Esr2−/− mice, EE-induced effects on the uterus wet weight were comparable to those observed in Esr2+/+ mice (20.1 ± 1.1 mg vs. 116.0 ± 5.0 mg for Esr2−/− mice, and 16.3 ± 1.0 mg vs. 114.2 ± 5.1 mg for Esr2+/+ mice, both P-values < 0.0001.). Figure 4A shows that oral EE administration resulted in comparable effects on the plasma coagulation profile for both genotypes, and also, with respect to hepatic transcript levels, Esr2−/− mice responded similarly to EE administration as their wild-type littermate controls (Fig. 4B).

image

Figure 4.  Differences between ovariectomized Esr2+/+ mice treated with 1 μg of EE as compared with vehicle-treated mice (□), or ovariectomized Esr2−/− mice treated with 1 μg of EE as compared with vehicle-treated mice (⋄), in the plasma coagulation profile (A; N = 13–15 per group) and hepatic transcript levels (B; N = 10 per group). Original data are presented in Tables S8 and S9. Data are presented as difference ± standard error of the difference; black symbols, P < 0.05 as compared with vehicle-treated controls of the same genotype. ag, antigen level; act, activity level.

Download figure to PowerPoint

Single EE dose

To extend our understanding of EE modulation of transcription of coagulation genes in the liver, we determined the immediate effects of EE on transcription in mice orally treated with a single dose of EE, alone or in combination with a subcutaneous ICI182780 injection. Hepatic transcript levels were determined 2.5 and 5 h after administration, and 2.5 h after the estrogen administration significant EE-induced changes were observed for fibrinogen-γ, FII, FIX, FX, antithrombin, plasminogen, protein C, and protein S (Fig. 5A). Evaluation of the hepatic transcript levels 5 h after the EE treatment showed more pronounced effects of EE and ICI182780 for fibrinogen-γ, FII, plasminogen and protein S than at the 2.5-h time point, and additional downregulation of FV, FVII, FXII, protein Z, protein Z inhibitor and α2-antiplasmin was observed, whereas FXI mRNA levels were upregulated (Fig. 5B). All EE-induced effects were counteracted in the EE/ICI182780-treated animals. Comparison of these results with the results after a 10-day treatment period demonstrates that only protein S acts differently, as transcript levels are reduced after a single EE dose, but increase after multiple EE doses.

image

Figure 5. Differences between ovariectomized mice treated with 1 μg of 17α-ethinylestradiol (EE) alone as compared with vehicle-treated mice (□), or mice treated with 1 μg of EE and 100 μg of ICI182780 as compared with vehicle-treated mice (⋄), in hepatic transcript levels 2.5 h after administration (A; N = 6 per group) or 5 h after administration (B; N = 6 per group). Original data are presented in Table S10. Data are presented as difference ± standard error of the difference; black symbols, P < 0.05 as compared with vehicle-treated controls (0 μg of EE and 0 μg of ICI182780).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References
  10. Supporting Information

In the present study, we demonstrate that oral administration of the synthetic estrogen EE to ovariectomized mice has a broad impact on the hepatic transcript levels of both procoagulant and anticoagulant genes. The EE effect on transcript levels of the coagulation genes was mostly downregulatory and, as analyzed for a limited set of factors, largely coincided with lowered plasma activity levels. Furthermore, the effects of EE on both the plasma and hepatic transcript levels of coagulation factors were dose-dependent, counteracted by the ER antagonist ICI182780, absent in ERα-deficient mice, and already present shortly after oral administration. We conclude that, in mice, EE has widespread and rapid effects on coagulation that require the involvement of ERα.

The results of the present study are in conflict with those of the study performed by Movérare et al. [11], in which microarray analyses of liver samples from mice treated with 17β-estradiol did not yield altered hepatic coagulation gene transcript levels as compared with vehicle-treated mice. We have no explanation for this difference in findings, other than a difference in treatment protocol, that is, subcutaneous injection of 2.3 μg of 17β-estradiol benzoate per mouse d−1 for 5 days per week for 3 weeks vs. our daily oral administration of 1 μg of EE for 10 days, or the fact that the cut-off value of at least 1.6-fold of the microarray data analyses might be too high to allow detection of the subtle changes in coagulation factor transcript levels. In contrast, two recent genome-wide screens identified high-affinity ER-binding sites in the mouse liver for many genes belonging to both the procoagulant and anticoagulant pathway, including those for fibrinogen, FII, FV, FVII, FX, FXI, antithrombin, plasminogen, protein C, protein S, protein Z, α2-antiplasmin, and heparin cofactor II [9,10]. Thus, these two genome-wide studies predict that estrogens potentially modulate transcription of a large number of coagulation genes expressed in the liver through the direct action of ERs. Our study demonstrates that the transcription of these genes is indeed estrogen-responsive and controlled by ERα. However, our study does not allow us to draw the conclusion that the observed responses upon EE treatment result from a direct interaction between ERα and the gene of interest, as predicted by the genome-wide studies.

In humans, oral contraceptive use causes changes in the plasma coagulation profile, resulting in a prothrombotic shift of the hemostatic balance and hence an increased venous thrombotic risk. In mice, however, the effects of oral EE on plasma and/or hepatic mRNA levels of procoagulant factors were unexpectedly mostly downregulatory, with the exception of FIX and FXI. Therefore, we considered the possibility that our EE dose was outside the therapeutically relevant window, and the results were, in part, artefacts or side effects of a relatively high oral estrogen dose. However, several observations argue against such artefacts: although the liver weight of mice treated with 1 μg of EE was increased as compared with the vehicle-treated mice, no effects on liver enzymes were observed, and histologic analyses of hematoxylin/eosin-stained liver sections did not reveal morphologic abnormalities such as hepatocyte degeneration or infiltration of inflammatory cells, which are typical for EE-induced hepatotoxicity [18]. As the downregulatory effects are already present with low doses of EE (from 0.1 μg EE per mouse d−1 onwards), this implies a specific effect of the estrogen treatment. The observed effects may be related to the synthetic nature of the estrogen; however, subcutaneous injections of 17β-estradiol resulted in comparable effects on both plasma and transcript levels, indicating that there is no difference in response to either EE or 17β-estradiol. Furthermore, non-ovariectomized female mice also showed comparable (downregulatory) effects on coagulation after EE treatment, indicating that there is no additional effect of EE on endogenous estrogen levels. Besides this, with respect to plasma analyses, our observations are in line with data from Wong et al. [19], who showed that procoagulant factor activity levels are lower in female mice than in male mice. This may, at least in part, be a result of higher estrogen levels in females. Taking these findings together, we conclude that the observed EE-induced effects on both mouse plasma and hepatic transcript levels are within a biologically and pharmacologically relevant range.

The question remains how EE induces a fast ER-mediated decrease in transcript levels for a large number of coagulation genes. Estrogen generally regulates gene expression by transcriptional activation. However, an increasing number of studies have demonstrated that ligand-bound ERs may also mediate transcriptional repression. A recent genome-wide gene expression profiling study of 47 000 murine transcripts showed that, of the 78 genes regulated by EE in the liver for which the change was at least 1.5-fold, 17 were downregulated [20]. As this cut-off of > 1.5-fold change is higher than the fold changes observed in coagulation factors, not only might this explain why coagulation genes were not found in this genome-wide study, but it might also imply that many more genes are affected by EE administration, either in a positive or a negative fashion.

Another possible explanation for the reduced transcript levels is that estrogen induces or represses expression of a transcription factor that is important in the control of gene expression. For example, Gao et al. [21] has shown that STAT3 expression levels in estrogen-treated obese mice are upregulated, resulting in reduced transcript levels of genes involved in hepatic lipid biosynthesis. With respect to coagulation genes, hepatocyte nuclear factor 4α (HNF4α) could be such a transcription factor, as it is known that it can regulate the transcription of several coagulation factors and can interfere with estrogen signaling [22,23]. However, in our study, the rapid EE-induced effects on coagulation factor mRNA levels were independent of significant changes in hepatic HNF4α transcript levels (data not shown). It has become clear that activating protein-1 (AP-1) proteins play a role in gene repression [24], and in this respect it is striking that AP-1 elements are often found together with EREs [10], which also might provide an explanation for the reduced transcript levels observed in the present study. Besides altering transcriptional activity, and thereby lowering mRNA levels, EE could also affect the mRNA stability itself. However, for now, this is a subject for speculation, and further research is needed to clarify this overall downregulatory mechanism.

In summary, our data demonstrate that oral EE treatment can have a major and rapid impact on mouse coagulation. Future studies should identify the sequence of molecular steps through which ERα evokes the overall decreased transcription levels of coagulation factors. Although this question remains, for now, unanswered, we believe that this work describes novel and important insights into sequence of events contributing to sex hormone-induced changes in coagulation.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References
  10. Supporting Information

We thank H. L. Vos for critical reading of the manuscript. This study was supported by grant 2006B045 from the Netherlands Heart Foundation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References
  10. Supporting Information
  • 1
    Rosendaal FR. Venous thrombosis: the role of genes, environment, and behavior. Hematology Am Soc Hematol Educ Program 2005: 112.
  • 2
    Middeldorp S. Oral contraceptives and the risk of venous thromboembolism. Gend Med 2005; 2(Suppl. A): S39.
  • 3
    Vehkavaara S, Silveira A, Hakala-Ala-Pietila T, Virkamaki A, Hovatta O, Hamsten A, Taskinen MR, Yki-Jarvinen H. Effects of oral and transdermal estrogen replacement therapy on markers of coagulation, fibrinolysis, inflammation and serum lipids and lipoproteins in postmenopausal women. Thromb Haemost 2001; 85: 61925.
  • 4
    Canonico M, Oger E, Plu-Bureau G, Conard J, Meyer G, Levesque H, Levesque H, Trillot N, Barrellier MT, Wahl D, Emmerich J, Scarabin PY. Hormone therapy and venous thromboembolism among postmenopausal women: impact of the route of estrogen administration and progestogens: the ESTHER study. Circulation 2007; 115: 8405.
  • 5
    Jick S, Kaye JA, Li L, Jick H. Further results on the risk of nonfatal venous thromboembolism in users of the contraceptive transdermal patch compared to users of oral contraceptives containing norgestimate and 35 microg of ethinyl estradiol. Contraception 2007; 76: 47.
  • 6
    Johnson JV, Lowell J, Badger GJ, Rosing J, Tchaikovski S, Cushman M. Effects of oral and transdermal hormonal contraception on vascular risk markers: a randomized controlled trial. Obstet Gynecol 2008; 111: 27884.
  • 7
    Farsetti A, Misiti S, Citarella F, Felici A, Andreoli M, Fantoni A, Sacchi A, Pontecorvi A. Molecular basis of estrogen regulation of Hageman factor XII gene expression. Endocrinology 1995; 136: 507683.
  • 8
    Di Bitondo R, Hall AJ, Peake IR, Iacoviello L, Winship PR. Oestrogenic repression of human coagulation factor VII expression mediated through an oestrogen response element sequence motif in the promoter region. Hum Mol Genet 2002; 11: 72331.
  • 9
    Bourdeau V, Deschenes J, Metivier R, Nagai Y, Nguyen D, Bretschneider N, Gannon F, White JH, Mader S. Genome-wide identification of high-affinity estrogen response elements in human and mouse. Mol Endocrinol 2004; 18: 141127.
  • 10
    Gao H, Falt S, Sandelin A, Gustafsson JA, Dahlman-Wright K. Genome-wide identification of estrogen receptor alpha-binding sites in mouse liver. Mol Endocrinol 2008; 22: 1022.
  • 11
    Moverare S, Skrtic S, Lindberg MK, Dahlman-Wright K, Ohlsson C. Estrogen increases coagulation factor V mRNA levels via both estrogen receptor-alpha and -beta in murine bone marrow/bone. Eur J Endocrinol 2004; 151: 25963.
  • 12
    Tchaikovski SN, van Vlijmen BJ, Rosing J, Tans G. Development of a calibrated automated thrombography based thrombin generation test in mouse plasma. J Thromb Haemost 2007; 5: 207986.
  • 13
    Nagai N, Lijnen HR, Cleuren AC, Rosendaal FR, van Hoef B, Hoylaerts MF, van Vlijmen BJ. Factor V Leiden mutation is associated with enhanced arterial thrombotic tendency in lean but not in obese mice. Thromb Haemost 2007; 98: 85863.
  • 14
    de Visser YP, Walther FJ, Laghmani EH, Boersma H, van der Laarse A, Wagenaar GT. Sildenafil attenuates pulmonary inflammation and fibrin deposition, mortality and right ventricular hypertrophy in neonatal hyperoxic lung injury. Respir Res 2009; 10: 30.
  • 15
    Cleuren AC, van Hoef B, Hoylaerts MF, van Vlijmen BJ, Lijnen HR. Short-term ethinylestradiol treatment suppresses inferior caval vein thrombosis in obese mice. Thromb Haemost 2009; 102: 9931000.
  • 16
    Lansink M, Jong M, Bijsterbosch M, Bekkers M, Toet K, Havekes L, Emeis J, Kooistra T. Increased clearance explains lower plasma levels of tissue-type plasminogen activator by estradiol: evidence for potently enhanced mannose receptor expression in mice. Blood 1999; 94: 13306.
  • 17
    Lindberg MK, Moverare S, Skrtic S, Gao H, Dahlman-Wright K, Gustafsson JA, Ohlsson C. Estrogen receptor (ER)-beta reduces ERalpha-regulated gene transcription, supporting a ‘ying yang’ relationship between ERalpha and ERbeta in mice. Mol Endocrinol 2003; 17: 2038.
  • 18
    Yamamoto Y, Moore R, Hess HA, Guo GL, Gonzalez FJ, Korach KS, Maronpot RR, Negishi M. Estrogen receptor alpha mediates 17alpha-ethynylestradiol causing hepatotoxicity. J Biol Chem 2006; 281: 1662531.
  • 19
    Wong JH, Dukes J, Levy RE, Sos B, Mason SE, Fong TS, Weiss EJ. Sex differences in thrombosis in mice are mediated by sex-specific growth hormone secretion patterns. J Clin Invest 2008; 118: 296978.
  • 20
    Ahlbory-Dieker DL, Stride BD, Leder G, Schkoldow J, Trolenberg S, Seidel H, Otto C, Sommer A, Parker MG, Schutz G, Wintermantel TM. DNA binding by estrogen receptor-alpha is essential for the transcriptional response to estrogen in the liver and the uterus. Mol Endocrinol 2009; 23: 154455.
  • 21
    Gao H, Bryzgalova G, Hedman E, Khan A, Efendic S, Gustafsson JA, Dahlman-Wright K. Long-term administration of estradiol decreases expression of hepatic lipogenic genes and improves insulin sensitivity in ob/ob mice: a possible mechanism is through direct regulation of signal transducer and activator of transcription 3. Mol Endocrinol 2006; 20: 128799.
  • 22
    Farsetti A, Moretti F, Narducci M, Misiti S, Nanni S, Andreoli M, Sacchi A, Pontecorvi A. Orphan receptor hepatocyte nuclear factor-4 antagonizes estrogen receptor alpha-mediated induction of human coagulation factor XII gene. Endocrinology 1998; 139: 45819.
  • 23
    Stauffer DR, Chukwumezie BN, Wilberding JA, Rosen ED, Castellino FJ. Characterization of transcriptional regulatory elements in the promoter region of the murine blood coagulation factor VII gene. J Biol Chem 1998; 273: 227787.
  • 24
    Carroll JS, Meyer CA, Song J, Li W, Geistlinger TR, Eeckhoute J, Brodsky AS, Keeton EK, Fertuck KC, Hall GF, Wang Q, Bekiranov S, Sementchenko V, Fok EA, Silver PA, Gingeras TR, Liu XS, Brown M. Genome-wide analysis of estrogen receptor binding sites. Nat Genet 2006; 38: 128997.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References
  10. Supporting Information

Table S1. Quantitative polymerase chain reaction (QPCR) primer sequences.

Table S2. Plasma coagulation factor activity (act) and antigen (ag) levels of ovariectomized mice treated with increasing doses of oral 17α-ethinylestradiol (EE) for 10 consecutive days.

Table S3. Hepatic transcript levels of ovariectomized mice treated with increasing doses of oral 17α-ethinylestradiol (EE) for 10 consecutive days.

Table S4. Plasma coagulation factor activity (act) and antigen (ag) levels of ovariectomized mice treated with 1 μg of oral 17α-ethinylestradiol (EE) and/or 100 μg subcutaneous ICI182780 for 10 consecutive days.

Table S5. Hepatic transcript levels of ovariectomized mice treated with 1 μg of oral 17α-ethinylestradiol (EE) and/or 100 μg of subcutaneous ICI182780.

Table S6. Plasma coagulation factor activity (act) and antigen (ag) levels of ovariectomized Esr1+/+ and Esr1−/− mice treated with 1 μg of oral 17α-ethinylestradiol (EE).

Table S7. Hepatic transcript levels of Esr1+/+ and Esr1−/− mice treated with oral 17α-ethinylestradiol (EE).

Table S8. Plasma activity (act) and antigen (ag) levels of Esr2+/+ and Esr2−/− mice treated with oral 17α-ethinylestradiol (EE).

Table S9. Hepatic transcript levels of Esr2+/+ and Esr2−/− mice treated with oral 17α-ethinylestradiol (EE).

Table S10. Hepatic transcript levels 2.5 and 5 hours after mice were treated with oral 17α-ethinylestradiol (EE) and/or subcutaneous ICI182780.

FilenameFormatSizeDescription
JTH_3930_sm_tables.pdf113KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.