Pregnancy loss, tissue factor pathway inhibitor deficiency and resistance to activated protein C

2724 Letters to the Editor

Pregnancy loss, tissue factor pathway inhibitor deficiency and resistance to activated protein C C . G A R D I N E R , * H . C O H E N ,   S . K . A U S T I N , * S . J . M A C H I N *   and I . J . M A C K I E * *Department of Haematology, University College London, London; and  Department of Haematology, University College London Hospitals, London, UK

To cite this article: Gardiner C, Cohen H, Austin SK, Machin SJ, Mackie IJ. Pregnancy loss, tissue factor pathway inhibitor deficiency and resistance to activated protein C. J Thromb Haemost 2006; 4: 2724–6.

Recurrent pregnancy loss is a common problem, with two or more losses in up to 5% of women, and recurrent (three or more consecutive losses) affecting 1–2% of women. Markers of coagulation activation (D-dimer and fibrinopeptide A) are known to increase preceding spontaneous abortion [1], and there is a strong association with antiphospholipid antibodies (aPA) [2] and some hereditary thrombophilia [3]. However, the majority of cases of unexplained pregnancy loss occur in the absence of aPA or hereditary thrombophilic defects. Activated protein C (APC) resistance during normal pregnancy is well documented [4] and acquired APC resistance (measured by a clotting test), independent of the factor V Leiden mutation (FVL), is a risk factor for pregnancy loss [5,6] and pre-eclampsia [7]. Tissue factor pathway inhibitor (TFPI) is known to be a major determinant of thrombin generation-based APC sensitivity [8], i.e. low TFPI levels are associated with APC resistance. We therefore studied the role of endogenous thrombin potential (ETP) and TFPI antigen levels in women with recurrent pregnancy loss. The study population comprised 22 women with a history of recurrent early pregnancy loss (three or more consecutive losses at < 24 weeks, n ¼ 17) or intrauterine fetal death (IUFD; n ¼ 5). Six of the women had the antiphospholipid syndrome (APS), while the other 16 had no detectable aPA. All samples were taken in the non-pregnant state, at least 6 weeks after the end of pregnancy. The following women were excluded from the study: those with protein C or S deficiency, FVL, or who were receiving oral anticoagulation, heparin or oral contraceptives at the time of testing. Thrombin generation was measured with and without exogenous human recombinant APC, using a method based on that of Rosing et al. [9]. Briefly, coagulation was triggered by the addition of 7 pM tissue factor (Innovin, Marburg, Germany), 20 lM phospholipid (Rossix, Mo¨lndal, Sweden) and 16 mM CaCl2 to defibrinated plasma. The ETP, calculated from the area under the curve, was measured continuously by cleavage of a chromogenic substrate (Pefachrome TG, Pentapharm, Basel, Switzerland) using the ACL9000 (InstrumenCorrespondence: Chris Gardiner, Department of Haematology, University College London, London, UK. Tel.: +44 845 1555000 ext. 8527; fax: +44 207 3809886; e-mail: [email protected] Received 8 August 2006, accepted 30 August 2006

tation Laboratory, Milan, Italy). This was performed with and without the addition of 5 nM APC (Eli Lilly, Indianapolis, IN, USA). Results were expressed as ratios relative to pooled normal plasma (PNP): ETP ¼ thrombin formed in patient plasma/thrombin formed in PNP, with no APC. ETP+APC ¼ thrombin formed in patient plasma/thrombin formed in PNP, with 5 nM APC. Total TFPI antigen was assayed using the IMUBIND Total TFPI ELISA kit (American Diagnostica Inc, Stamford, CT, USA) a quantitative sandwich ELISA, which recognizes full-length, truncated and conjugated forms of TFPI [10]. The Mann–Whitney U-test was used to test the differences between the medians and statistical significance was defined as P < 0.05. Normal reference ranges were established in citrated plasma from 20 normal non-pregnant women: ETP median 0.91 (95% reference range 0.70–1.07), ETP+APC 0.96 (0.76–1.07) TFPI antigen 89 ng mL)1 (75–120 ng mL)1). Both ETP and ETP+APC were significantly higher in the women with previous pregnancy morbidity than in normal subjects (median ETP 1.07, P < 0.0001; median ETP+APC 1.32, P < 0.0001). The median TFPI was 75.4 ng mL)1 (range 31.5–120 ng mL)1, P ¼ 0.007), with low TFPI antigen levels in 10 out of 22 (45%) of the women with previous pregnancy morbidity, all of whom had raised ETP and/or ETP+APC (Table 1). Although there was a tendency towards higher ETP and lower TFPI antigen levels in the women with APS, antiphospholipid antibody status had no statistically significant effect on ETP, ETP+APC or TFPI antigen levels. Both ETP+APC and ETP showed a negative correlation with TFPI antigen level (r ¼ )0.48 and )0.24, respectively). Furthermore, the addition of increasing amounts of a polyclonal antibody (rabbit antihuman TFPI IgG; American Diagnostica Inc.), that blocked TFPI function, caused a dose-dependent increase in ETP+APC of up to 50% in normal plasma and a modest increase in ETP (approximately 10%). The mechanisms responsible for the association between thrombophilia and pregnancy morbidity are unclear [11]. Our preliminary data suggest that low levels of plasma TFPI, increased thrombin generation and resistance to APC may be a common finding in women with pregnancy loss/morbidity. We have demonstrated an association between TFPI levels and APC resistance, and have shown that blocking TFPI activity in vitro causes APC resistance. It is clear that the causes of  2006 International Society on Thrombosis and Haemostasis

Letters to the Editor 2725 Table 1 Clinical and laboratory data for the studied patients ID Diagnosis 1 IUFD 2 IUFD + pre-eclampsia 3 RM 4 IUFD + placental ischemia 5 REPL 6 REPL 7 REPL 8 REPL 9 REPL 10 REPL 11 REPL 12 REPL 13 REPL 14 REPL 15 REPL 16 REPL 17 REPL 18 REPL 19 IUFD 20 REPL 21 REPL 22 IUFD Median Minimum Maximum

TFPI Antiphospholipid antibodies ETP+APC ETP ng mL)1 + +

1.27 1.95

1.22 1.29

75.4 59.0

+ )

0.83 1.26

1.25 1.04

72.0 31.5

) ) ) ) ) ) ) ) + + ) + ) ) ) + ) )

1.00 1.56 2.00 1.58 1.56 1.42 1.66 0.94 1.64 1.36 0.80 0.98 0.92 0.88 1.25 1.88 1.20 1.39

0.92 0.97 1.10 1.24 1.07 1.09 1.19 0.70 1.32 0.85 0.84 1.06 1.13 0.86 0.84 1.24 0.88 1.11

115.0 48.0 92.8 34.2 103.0 78.8 88.0 104.0 74.0 40.2 84.4 93.0 120.0 112.0 55.8 71.0 80.3 69.3

1.32 0.80 2.00

1.08 77.1 0.70 31.5 1.32 120.0

IUFD, interuterine fetal death; REPL, recurrent early pregnancy loss; Anti-b2GP-I, anti-b2 glycoprotein-I antibody; TFPI, total tissue factor pathway inhibitor antigen.

acquired APC resistance are complex. It has been reported that the APC-resistant phenotype observed during hormone replacement therapy is associated with a decrease in both TFPI and protein S levels [12]. Furthermore, Hackeng et al. [13] have recently described an interaction between TFPI and protein S in the regulation of thrombin formation. Inhibitory antibodies to TFPI have been found in women with APS, and these are associated with pregnancy loss [14]. While an autoimmune mechanism for reduced TFPI antigen cannot be excluded, the majority of the women in our study were aPA negative, so an alternative cause is more likely. For several years, low-molecular-weight heparin (LMWH) treatment has been used in the treatment of recurrent pregnancy loss associated with APS [2,15,16], although the precise mechanisms for the protective effect in this setting are unclear. LMWH is frequently used for the treatment of highrisk pregnancies associated with hereditary thrombophilia and it has been reported that this improves the live birth rate [17,18] although no placebo-controlled trials have been performed to date. It is known that placental TFPI may be decreased in gestational vascular complications and this may be restored by maternal LMWH treatment [19]. However, sub-therapeutic anti-Xa levels during treatment of high-risk pregnancies appear to be associated with low levels of plasma TFPI and an increased risk of pregnancy loss [11].  2006 International Society on Thrombosis and Haemostasis

In purified systems, TFPI prolongs the lag period and acts synergistically with the protein C pathway to inhibit thrombin generation [20]. This is consistent with our finding of TFPI deficiency as a common contributory factor for an APCresistant phenotype associated with recurrent pregnancy loss. As APC has both anticoagulant and anti-inflammatory properties, it seems probable that reduced sensitivity to protein C could contribute to adverse pregnancy outcome as a result of thrombotic and/or inflammatory processes. Our data suggest that TFPI deficiency associated with ETP-dependent APC resistance could be a risk factor for pregnancy loss, and imply a potential role for heparin in the treatment of this condition. Disclosure of Conflict of Interests The authors state that they have no conflict of interest. References 1 Woodhams BJ, Candotti G, Shaw R, Kernoff PB. Changes in coagulation and fibrinolysis during pregnancy: evidence of activation of coagulation preceding spontaneous abortion. Thromb Res 1989; 55: 99–107. 2 Rai RS, Clifford K, Cohen H, Regan L. High prospective fetal loss rate in untreated pregnancies of women with recurrent miscarriage and antiphospholipid antibodies. Hum Reprod 1995; 10: 3301–4. 3 Preston FE, Rosendaal FR, Walker ID, Briet E, Berntorp E, Conard J, Fontcuberta J, Makris M, Mariani G, Noteboom W, Pabinger I, Legnani C, Scharrer I, Schulman S, van der Meer FJ. Increased fetal loss in women with heritable thrombophilia. Lancet 1996; 348: 913–6. 4 Cumming AM, Tait RC, Fildes S, Yoong A, Keeney S, Hay CR. Development of resistance to activated protein C during pregnancy. Br J Haematol 1995; 90: 725–7. 5 Brenner B, Mandel H, Lanir N, Younis J, Rothbart H, Ohel G, Blumenfeld Z. Activated protein C resistance can be associated with recurrent fetal loss. Br J Haematol 1997; 97: 551–4. 6 Rai R, Shlebak A, Cohen H, Backos M, Holmes Z, Marriott K, Regan L. Factor V Leiden and acquired activated protein C resistance among 1000 women with recurrent miscarriage. Hum Reprod 2001; 16: 961–5. 7 Paternoster DM, Stella A, Simioni P, Girolami A, Snijders D. Activated protein C resistance in normal and pre-eclamptic pregnancies. Gynecol Obstet Invest 2002; 54: 145–9. 8 de Visser MC, van Hykkama Vlieg A, Tans G, Rosing J, Dahm AE, Sandset PM, Rosendaal FR, Bertina RM. Determinants of the APTTand ETP-based APC sensitivity tests. J Thromb Haemost 2005; 3: 1488–94. 9 Rosing J, Tans G, Nicolaes GA, Thomassen MC, Van Oerle R, van der Ploeg PM, Heijnen P, Hamulyak K, Hemker HC. Oral contraceptives and venous thrombosis: different sensitivities to activated protein C in women using second- and third-generation oral contraceptives. Br J Haematol 1997; 97: 233–8. 10 Bognacki J, Hammelburger J. Functional and immunologic methods for the measurement of human tissue factor pathway inhibitor. Blood Coagul Fibrinolysis 1995; 6: S65–72. 11 Sarig G, Blumenfeld Z, Leiba R, Lanir N, Brenner B. Modulation of systemic hemostatic parameters by enoxaparin during gestation in women with thrombophilia and pregnancy loss. Thromb Haemost 2005; 94: 980–5. 12 Hoibraaten E, Mowinckel MC, de Ronde H, Bertina RM, Sandset PM. Hormone replacement therapy and acquired resistance to activated protein C: results of a randomized, double-blind, placebo-controlled trial. Br J Haematol 2001; 115: 415–20.

2726 Letters to the Editor 13 Hackeng TM, Sere KM, Tans G, Rosing J. Protein S stimulates inhibition of the tissue factor pathway by tissue factor pathway inhibitor. Proc Natl Acad Sci USA 2006; 103: 3106–11. 14 Martinuzzo M, Iglesias Varela ML, Adamczuk Y, Broze GJ, Forastiero R. Antiphospholipid antibodies and antibodies to tissue factor pathway inhibitor in women with implantation failures or early and late pregnancy losses. J Thromb Haemost 2005; 3: 2587–9. 15 Backos M, Rai R, Baxter N, Chilcott IT, Cohen H, Regan L. Pregnancy complications in women with recurrent miscarriage associated with antiphospholipid antibodies treated with low dose aspirin and heparin. Br J Obstet Gynaecol 1999; 106: 102–7. 16 Rai R, Cohen H, Dave M, Regan L. Randomised controlled trial of aspirin and aspirin plus heparin in pregnant women with recurrent miscarriage associated with phospholipid antibodies (or antiphospholipid antibodies). BMJ 1997; 314: 253–7.

17 Dolitzky M, Inbal A, Segal Y, Weiss A, Brenner B, Carp H. A randomized study of thromboprophylaxis in women with unexplained consecutive recurrent miscarriages. Fertil Steril 2006; 86: 362–6. 18 Brenner B, Hoffman R, Carp H, Dulitsky M, Younis J. Efficacy and safety of two doses of enoxaparin in women with thrombophilia and recurrent pregnancy loss: the LIVE-ENOX study. J Thromb Haemost 2005; 3: 227–9. 19 Aharon A, Lanir N, Drugan A, Brenner B. Placental TFPI is decreased in gestational vascular complications and can be restored by maternal enoxaparin treatment. J Thromb Haemost 2005; 3: 2355–7. 20 van Ôt Veer C, Golden NJ, Kalafatis M, Mann KG. Inhibitory mechanism of the protein C pathway on tissue factor-induced thrombin generation. Synergistic effect in combination with tissue factor pathway inhibitor. J Biol Chem 1997; 272: 7983–94.

What causes the enhancement of activity of factor VIIa by tissue factor? C . M . C O L I N A , * D . V E N K A T E S W A R L U ,   R . D U K E , * à L . P E R E R A * à and L . G . P E D E R S E N * à *Department of Chemistry, UNC-CH, Chapel Hill, NC;  Department of Chemistry, NCAT, Greensboro, NC; and àLaboratory of Structural Biology, NIEHS, RTP, NC, USA

To cite this article: Colina CM, Venkateswarlu D, Duke R, Perera L, Pedersen LG. What causes the enhancement of activity of factor VIIa by tissue factor? J Thromb Haemost 2006; 4: 2726–9.

Full-length tissue factor enhances the activity of activated factor VII (FVIIa) for the zymogen FX by a factor of 105 [1,2]. On the other hand, soluble tissue factor (sTF) enhances binding by a factor of approximately 76 [3]. The molecular cause of these enhancements has been of considerable interest. As we earlier participated in studies of the dynamics of the light chain of FVIIa [4], the zymogen FVII [5], and the complex sTF– FVIIa [6], we here employ current theoretical methodology [7,8] to assess the dynamic differences in sTF–FVIIa and FVIIa in a consistent manner using molecular dynamics. The goal is to compare the equilibrated solution structure of the complexed and free FVIIa and thereby find differences that might provide clues as to the enhancement of sTF. The basis of our simulations is the X-ray crystal structure of Banner et al. [9] of human sTF–FVIIa. Although this work provided an enormous advance in our knowledge of sTF– FVIIa, several fragments were missing from the structure: Correspondence: Lee G. Pedersen, Department of Chemistry, Campus Box 3290, UNC-CH, Chapel Hill, NC 27599, USA. Tel.: +1 919 962 1578; fax: +1 919 962 2388; e-mail: [email protected] Received 9 August 2006, accepted 8 September 2006

FVIIa residues 143–152 (chymotrypsin numbering); and sTF residues 1–4, 85–89, and 211–219. Also, the structure contained an active site inhibitor. We employed homology modeling to complete the FVIIa structure, and we employed the 2HFT [10] XRCS to complete the sTF structure (residues 1–210). The active site inhibitor was removed. All crystallographic calcium ions were included. The AMBER 8/PMEMD8 molecular dynamics program was employed [8]. Overall, the protocol was similar to that employed in a previous study [5]. Two simulations were performed: (i) the solvated sTF–FVIIa complex; and (ii) FVIIaf. The latter was derived from extraction of FVIIa from the sTF–FVIIa complex simulation at 7 ns, followed by a separate simulation. Results Global behavior

Plots of the backbone root mean square deviation (rmsd) (simulation vs. starting structure) are essentially horizontal (Fig. 1A) by 20 ns for FVIIa in the sTF–FVIIa complex, with some larger fluctuations for the FVIIaf simulations. These data support the conclusion that the systems are solution-equilibrated by 20 ns. There is, however, a large difference in the rmsd and in the fluctuations for free FVIIa vs. FVIIa in the  2006 International Society on Thrombosis and Haemostasis