Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Dec 1.
Published in final edited form as: J Thromb Haemost. 2015 Nov 25;13(12):2210–2219. doi: 10.1111/jth.13169

Comparison of platelet-derived and plasma FVIII efficacy using a novel native whole blood thrombin generation assay

C K Baumgartner *, G Zhang *, E L Kuether †,*, H Weiler *, Q Shi *,†,‡, RR Montgomery *,†,‡
PMCID: PMC4715732  NIHMSID: NIHMS730513  PMID: 26453193

Abstract

Background

We have recently developed a successful gene therapy approach for hemophilia A in which FVIII expression is targeted to platelets by the αIIb promoter. Levels of platelet expressed FVIII (2bF8) achieved by gene therapy may vary between individuals due to differences in ex vivo transduction and gene expression efficiency. Accurate assays to evaluate 2bF8 efficacy are desirable.

Objective

To compare the hemostatic efficacy of 2bF8 with replacement therapy over a wide therapeutic dose range.

Methods

Efficacy of 2bF8 was assessed using a new transgenic mouse model expressing high 2bF8 levels (LV18tg). Blood from LV18tg mice or FVIIInull mice infused with recombinant FVIII was mixed with FVIIInull blood at different ratios ex vivo to achieve several concentrations of 2bF8 or plasma FVIII. Samples were evaluated with a novel native whole blood thrombin generation assay (nWB-TGA) that uses recalcified whole blood without the addition of tissue factor to initiate coagulation.

Results

FVIII dose dependency was observed in all five thrombin generation (TG) parameters. While the total amount of thrombin generated was similar, 2bF8 significantly accelerated TG compared with plasma FVIII. Remarkably, a 10-fold lower dose of 2bF8 than plasma FVIII (0.2% vs 2%) significantly shortened the onset and peak of TG compared with FVIIInull blood.

Conclusion

Using a new transgenic mouse model we showed that the novel nWB-TGA established here can be used to monitor platelet targeted FVIII gene therapy. The higher therapeutic efficacy of 2bF8 compared with factor replacement therapy seemed to be due to acceleration of TG.

Keywords: factor VIII, gene therapy, hemophilia A, platelets, thrombin

Introduction

Intravenous administration of factor VIII (FVIII) products is the current standard of care for hemophilia A (HA) patients. Due to the short half-life of FVIII [1] frequent treatment is required, which might lead to risks like infections and is cumbersome for patients. A main challenge of FVIII replacement therapy is the development of inhibitory antibodies (inhibitors) against FVIII in up to 30% of severe HA patients rendering the therapy ineffective [2]. FVIII gene therapy is a promising approach to potentially permanently correct the bleeding phenotype of HA patients and improve patients’ quality of life. Our group has previously developed a successful gene therapy approach in which human B-domain deleted FVIII (hBDDFVIII) expression is driven by the megakaryocyte and platelet specific αIIb promoter, resulting in FVIII expression in platelets. Platelet expressed FVIII (2bF8) protects hemophilic mice from lethal blood loss after vessel injury. This therapeutic approach does not induce FVIII inhibitors and is even successful in the presence of pre-existing high titer inhibitors [3–6]. Thus, this approach is among the first to hold promise for inhibitor patients. Levels of 2bF8 achieved by gene therapy may vary between individuals due to differences in ex vivo transduction and gene expression efficiency [3–7]. Therefore, evaluation of therapeutic efficacy over a wide dose range of platelet FVIII is required but cannot be done with standard plasma assays for FVIII.

Tracking the correction of abnormal bleeding phenotypes during the treatment of patients with hemostatic disorders is crucial to evaluate therapeutic success. Global coagulation assays in contrast to single clotting factor assays are desirable to better understand the overall hemostatic condition of patients, but such assays often have reduced sensitivity and specificity. Determining the capacity of blood samples to generate thrombin using thrombin generation (TG) assays provides valuable information on the interplay of pro- and anti-coagulant factors [8, 9]. Historically, TG has been determined in platelet poor or platelet rich plasma. Recently the importance of erythrocytes [10, 11] and other blood cell types like monocytes and macrophages [12–14] in TG and clot formation has become more apparent and prompted the need for a whole blood assay. In 2012 a whole blood TG assay has been developed [15].

TG for monitoring general hemophilia therapy has been suggested [16, 17]. In most studies the endogenous thrombin potential (ETP) and peak thrombin, parameters derived from a TG curve, have been correlated with FVIII activity levels in plasma [17–19] and whole blood samples [15]. Until now, platelet targeted FVIII gene therapy has not been evaluated with TG assays. Common to most previous TG assays (plasma and whole blood based) was the exogenous addition of tissue factor (TF) to initiate coagulation. Although addition of TF is physiologically relevant, we established a native whole blood assay in which coagulation is initiated by recalcification only to better evaluate the intrinsic clotting pathway and therefore the impact of FVIII.

Here we developed a new transgenic mouse model expressing high levels of 2bF8 to evaluate therapeutic efficacy of 2bF8 compared with plasma FVIII over a wide dose range. We evaluated FVIII efficacy by measuring TG with a novel native whole blood thrombin generation assay (nWB-TGA).

Materials and methods

Mice

FVIIInull mice (FVIII exon 17 disrupted) [20] were a kind gift from H. Kazazian (University of Pennsylvania School of Medicine). C57BL6/129S mice served as WT controls. Transgenic mice expressing high levels of hBDDFVIII in platelets (LV18tg mice) were generated using 2bF8 lentivirus-mediated transgenesis similar to our approach reported for the generation of platelet expressing FIX (2bF9) transgenic mice [21]. Briefly, fertilized oocytes from C57BL6/129S mice were in vitro transduced with 2bF8 lentivirus (2bF8-LV) [4]. At a 2-cell-stage embryos were transferred into pseudo-pregnant females. Transgene-positive pups were identified by PCR as described in our previous report [3]. LV18tg mice were bred with FVIIInull mice to produce single copy 2bF8 mice on a FVIIInull background. Copy number of 2bF8 was assessed by quantitative real-time PCR as previously described [5]. The 2bF8 integration site was determined by inverse PCR as previously described for 2bF9 [21]. Animal studies complied with a protocol approved by the Institutional Animal Care and Use Committee of the Medical College of Wisconsin.

PCR analysis

White blood cell DNA was isolated using the QIAamp Blood Mini Kit (Qiagen, Valencia, CA, USA). PCR GoTaq Master Mix (Promega, Madison, WI, USA) was used to amplify the desired fragments. A 150bp fragment was amplified using primers 5’-TGTGTCCCGCCCCTTCCTTT-3’ and 5’-GAGCAAATTCCTGTACTGAC-3’ to determine FVIIInull background as described [20]. The undisrupted murine WT FVIII gene was identified by a 233bp product using primers 5’-GCAAGGGAAGTGATATCACT-3’ and 5’-TCCTGTACTGACACTTGTCTC-3’. To distinguish between heterozygous and homozygous transgene carrying mice a PCR strategy using 3 primers was established. Primer 1 was located in the endogenous gene 5’ of the 2bF8 transgene (5’-TTCCACAGGAAAATGGTCACTG-3’), primer 2 was located inside the transgene (5’-TTCGCTTTCAAGTCCCTGTTC-3’) and primer 3 was located in the endogenous gene 3’ of the transgene (5’-AGGCAGTATGAGGCCCAAGAAT-3’). Primer 1 and 2 amplified a 400bp product identifying the transgene and primer 1 and 3 amplified a 270bp product identifying the transgene negative endogenous gene. All primers were obtained from Integrated DNA Technologies (Coralville, IA, USA).

Whole Blood, plasma and platelet lysate collection

Blood of anesthetized mice was drawn from the inferior vena cava (IVC) into 3.8% sodium-citrate or, where indicated, into 3.8% sodium-citrate containing 500µg mL−1 corn trypsin inhibitor (CTI; Haematologic Technologies Inc, Essex Junction, VT, USA) at a 9:1 ratio. Platelet poor plasma and platelet lysates were obtained by processing of blood collected from the IVC or retro-orbitally as previously described [5].

FVIII product

Recombinant hBDDFVIII (rhFVIII, Xyntha, Pfizer Inc, New York, NY, USA) was used at 50U kg−1 for intravenous infusion of mice and as indicated for in vitro experiments.

FVIII activity assays

Functional FVIII activity in platelet lysates or plasma was determined using a modified chromogenic assay (Coatest FVIII/4 kit, DiaPharma Group, West Chester, OH, USA) as previously described [3, 5] and is hereafter referred to as FVIII:C. Dilutions of rhFVIII were used to create a standard curve for measuring FVIII:C levels.

In vitro preparation of whole blood samples with different concentration of FVIII

For in vitro FVIII spiking experiments, 190 µL whole blood drawn from the IVC was spiked with 10 µL of 12U mL−1 rhFVIII to reach a 100% FVIII level (100% = 1.2U FVIII mL−1 plasma = 0.6U FVIII mL−1 whole blood). Blood was incubated on a nutator for 15 minutes (min) to allow binding of FVIII to VWF. To obtain further FVIII dilutions, blood spiked to a 100% FVIII level was serially diluted with FVIIInull blood. For different dilutions of 2bF8 or plasma FVIII, blood from LV18tg mice or FVIIInull mice infused with rhFVIII was analyzed either undiluted or serially diluted with FVIIInull blood. To ensure optimal mixing of blood, serial dilutions were done with a short, low speed vortex pulse.

Preparation of washed red blood cells (RBC)

Sodium-citrated whole blood was centrifuged at 1800g for 5 min at room temperature. The pellet containing RBC was washed with wash-buffer (5.5mM Glucose, 0.5% BSA, 10mM Hepes in 0.9% saline), centrifuged at 1500g for 5 min and washed again 5 times with wash-buffer. RBC were resuspended in wash-buffer and counted (scil Vet ABC hematologic analyzer, scil animal care company, Gurnee, IL, USA). Hematocrit was set to approximately 37% with wash-buffer.

Native whole blood thrombin generation assay (nWB-TGA)

We altered the previously published whole blood TGA protocol [15]. Whole blood drawn from the IVC (15 µL) was combined with a reagent premix consisting of 5 µL 100mM CaCl2 (16.7mM final), 5 µL 10mM Hepes (1.7mM final) and 5 µL 1.8mM (p-tosyl-Gly-Pro-Arg-amide)2-Rhodamine 110 (GPR-R110; 300 mM final; Invitrogen, Carlsbad, CA, USA). A small volume (5 µL) of the reaction mix was applied in duplicates on filter paper disks (Whatman filter paper #1, GE Healthcare, Pittsburgh, PA, USA) pre-inserted into black 96 strip-well plates (Nunc, Roskilde, Denmark) and covered with 40 µL of mineral oil (Sigma-Aldrich, St Louis, MO, USA). Plates were placed in a 37°C preheated Enspire multimode plate reader (PerkinElmer, Waltham, MA, USA), samples were excited at 498 nm and emission read at 521 nm in one minute intervals for a total of 90 min. In a calibration experiment 15 µL of a thrombin calibrator (TGA calibrator, Technoclone, Vienna, Austria) at 4 different dilutions (1:2, 1:4, 1:20, 1:200) were combined with 10 µL of washed RBC and 5 µL of 1.8mM GPR-R110. Based on the calibration experiment, relative fluorescence units acquired from samples were converted into TG curves using online available software (Technoclone, Vienna, Austria). Lag times were obtained from sample evaluation sheets in the software and thrombin generation rates manually calculated as peak thrombin/(peak time – lag time). Samples with different FVIII dose levels were analyzed in two separate runs. The first run included samples with the four lowest FVIII dose levels. The second run was started roughly 90 min thereafter and included the remaining samples with higher FVIII dose levels.

Statistical analysis

Statistical differences between experimental groups were determined by the nonparametric Mann-Whitney test allowing analysis of TG data with infinite lag or peak times scored at 90 min (maximum observation period). Statistical analysis was conducted with GraphPadPrism 4 software (GraphPad Software, La Jolla, CA, USA). Mean ± standard error of the mean (SEM) are shown in scatter plots and bar graphs, and represented in the results section. A p-value of <0.05 was considered statistically significant.

Results

Negligible thrombin generation in FVIIInull mice determined by nWB-TGA

To evaluate efficacy of FVIII within platelets and account for the impact of other blood cells, we employed a novel modification of the previously published whole blood TGA [15]. In the nWB-TGA described here, coagulation initiated by recalcification only (without the addition of TF) and modification of filter paper usage resulted in high sensitivity of this assay to FVIII. Blood from FVIIInull or WT mice was analyzed with the nWB-TGA. We used the following five TG parameters to evaluate thrombin generation: Lag time, peak time, peak thrombin, endogenous thrombin potential (ETP), and thrombin generation rate (TGR) (Fig. 1). TG in whole blood of FVIIInull mice differed significantly from that in WT mice (Fig. 2A). In some FVIIInull mice no TG was detectable. Lag and peak times for these mice were recorded as 90 min. While lag time and peak time in WT mice were 8.4 ± 0.6 min and 14.6 ± 1 min, respectively, FVIIInull mice showed a lag time of 55.6 ± 6.9 min and a peak time of 61.7 ± 5.9 min. Peak thrombin in WT mice was 122.2 ± 7.8nM compared with 9.9 ± 2.8nM in FVIIInull mice. ETP was 1093 ± 45.9nM in WT and 115.7 ± 40.1nM in FVIIInull mice. TGR was also significantly different in WT compared with FVIIInull mice (22.7 ± 2.9 versus 1.3 ± 0.3nM min−1) (Fig. 2B). WT blood drawn into CTI resulted in approximately 1.5-fold prolonged lag times (12.1 ± 0.4 min) compared with WT blood without CTI suggesting contact activation as the mechanism for initiation of coagulation in the nWB-TGA (Fig. 2C). Taken together, our data demonstrate that TG in whole blood of FVIIInull mice was negligible compared with TG in WT mice. Differences were significant in all five TG parameters.

Fig. 1. Schematic of thrombin generation curve and the five TG parameters evaluated throughout the manuscript.

Fig. 1

Open in a new tab

Fig. 2. Negligible thrombin generation (TG) in FVIIInull mice compared with WT mice determined by native whole blood thrombin generation assay (nWB-TGA).

Fig. 2

Open in a new tab

Coagulation in sodium-citrated whole blood drawn from the IVC was started by recalcification only – no tissue factor was added. (A) Representative TG curves of WT versus FVIIInull mice. (B) Comparison of the five TG parameters in WT and FVIIInull mice. Data points represent the mean of duplicates performed on single animals. The mean of the WT and FVIIInull group is depicted as solid line. Data shown are derived from 15 independent experiments. (C) Lag time determined with the nWB-TGA of WT blood drawn from the IVC directly into sodium-citrate without (−) or with (+) corn trypsin inhibitor (CTI, n=4 per group). ***p<0.001 (Mann-Whitney test)

Dose dependent thrombin generation in FVIIInull blood spiked with FVIII ex vivo

Next we determined the sensitivity of the nWB-TGA to different concentrations of FVIII. We spiked whole blood from FVIIInull mice ex vivo with rhFVIII to obtain FVIII concentrations ranging from 0% to 100%. FVIII dose dependency in all five TG parameters was observed. Reconstitution of hemophilic blood with rhFVIII to a 1%, 20% and 100% level shortened lag time to 23.5 ± 1.7, 10.7 ± 0.6 and 8.2 ± 0.6 min, respectively. Peak time was 34.1 ± 2.4, 17.2 ± 1.0 and 14.4 ± 1.3 min, peak thrombin was 30 ± 4, 129 ± 11 and 195 ± 2nM thrombin, ETP was 459 ± 50, 1249 ± 78 and 1524 ± 93nM thrombin, and TGR was 3.2 ± 0.7, 23.6 ± 4.8 and 43.8 ± 12.3nM thrombin min−1 at 1%, 20% and 100% FVIII, respectively. A significant increase was observed in all TG parameters at the 1% FVIII dose level compared with FVIIInull blood (Fig. 3). For TGR 0.5% FVIII significantly differed from FVIIInull blood. Thus, all five TG parameters responded in a dose dependent manner to FVIII in the nWB-TGA, with a limit of detection of 1% FVIII found with in vitro spiking experiments.

Fig. 3. Sensitivity of nWB-TGA to different concentrations of rhFVIII ex vivo.

Fig. 3

Open in a new tab

Blood from FVIIInull mice was spiked with rhFVIII ex vivo to obtain a 100% FVIII level. Further FVIII levels were achieved by serially diluting the 100% FVIII level blood with FVIIInull blood. The five TG parameters, lag time, peak time, peak thrombin, ETP, and TGR, are shown. Data shown are derived from 9 independent experiments. *p<0.05 (Mann-Whitney test).

Generation of a novel transgenic mouse model expressing high levels of 2bF8

In order to compare efficacy of 2bF8 with plasma FVIII we developed a new mouse model, LV18tg, that expressed high levels of 2bF8 on an otherwise FVIIInull background. Mice with a single copy of the 2bF8 gene cassette were selected and further propagated. We determined by inverse PCR that the 2bF8 insert integrated in a reverse transcriptional orientation into the deoxyribonuclease 1-like 2 protein gene on chromosome 17 (Fig. 4A). A 3 primer PCR approach allowed distinguishing between transgene heterozygous (LV18tg+/−) and homozygous mice (LV18tg+/+). In two separate PCRs - one for the endogenous WT FVIII gene and one for the FVIII knock out - disruption of exon 17 of the endogenous FVIII gene was confirmed (Fig. 4B). LV18tg+/− mice expressed FVIII:C of 6 ± 0.3mU per 108 platelets and LV18tg+/+ mice expressed 12 ± 0.4mU per 108 platelets (Fig. 4C) compared with 1.4 ± 0.3mU FVIII per 108 platelets in homozygous mice of our previously described transgenic model [3]. Thus, the new transgenic mouse model expressed high levels of 2bF8 and only homozygous mice were used for this study.

Fig. 4. Characterization of novel transgenic mouse model expressing high levels of 2bF8 on FVIIInull background (LV18tg mice).

Fig. 4

Open in a new tab

(A) Schematic diagram of 2bF8 integration site in the deoxyribonuclease 1 like protein 2 gene (DNase1l2) on chromosome 17 of transgenic mice referred to as LV18tg mice in the manuscript. (B) PCR of genomic DNA from white blood cells of WT, LV18tg heterozygous (LV18tg+/−) and LV18tg homozygous (LV18tg+/+) mice identifying 2bF8 transgene insertion (first row), murine FVIII (mFVIII, second row), and the exon 17-disrupted murine FVIII (Ex17KO, third row). H2O was used as negative control for PCRs. For the transgene insertion PCR 3 primers were used with primer 1 located in the endogenous DNase1l2 gene 5’ of the 2bF8 transgene, primer 2 within the transgene and primer 3 in the endogenous DNase1l2 gene 3’ of the transgene. This strategy allows distinguishing heterozygous from homozygous animals (transgene negative: 270bp band; transgene positive: 400bp band). (C) FVIII:C in platelets of WT (n = 10), LV18tg+/− (n = 7) and LV18tg+/+ (n = 10) mice determined by chromogenic FVIII:C assay of platelet lysates.

Accelerated thrombin generation of platelet-derived FVIII (2bF8) compared with plasma FVIII

Since in LV18tg+/+ transgenic mice every platelet (100% of platelets) expressed FVIII, assuming 109 platelets per mL mouse blood and that plasma is about 50% of blood volume, we calculated that 2bF8 levels in LV18tg+/+ mice corresponded to ~20% plasma FVIII in WT mice. This was based on the fact that WT mice (no FVIII in platelets) had 1.2U FVIII mL−1 plasma when rhFVIII was used as standard [3]. WT mice could not serve as source for plasma FVIII in this study because the 2bF8 gene cassette introduced into LV18tg mice encoded for human B-domain deleted FVIII. Therefore, we infused FVIIInull mice with recombinant human B-domain deleted FVIII (rhFVIII). A dose of 50U kg−1 rhFVIII was administered intravenously to reach plasma FVIII levels observed in WT mice (100% FVIII level). We combined blood from LV18tg mice (2bF8) or from rhFVIII infused FVIIInull mice (plasma FVIII) with FVIIInull blood at different ratios to obtain FVIII concentrations shown in Fig. 5. Similar to ex vivo spiking with rhFVIII, dose dependent TG was observed with 2bF8 and plasma FVIII when samples were analyzed with the nWB-TGA (Fig. 5).

Fig. 5. Comparison of efficacy of 2bF8 versus plasma FVIII.

Fig. 5

Open in a new tab

Blood from LV18tg+/+ mice (2bF8) or from FVIIInull mice that were injected intravenously with 50U kg−1 rhFVIII (plasma FVIII) 5 minutes prior to blood draw was ex vivo mixed with FVIIInull blood to target indicated FVIII levels. Further FVIII levels were achieved by serially diluting the 100% FVIII level blood or undiluted LV18tg+/+ blood with FVIIInull blood. The five thrombin generation parameters (A) lag time, (B) peak time, (C) peak thrombin, (D) ETP, and (E) TGR at different FVIII dose levels are shown. Black bars represent 2bF8, white bars represent plasma FVIII, and hatched bars represent FVIIInull blood. Data shown are derived from 5 (plasma FVIII) and 6 (2bF8) independent experiments. *p<0.05, **p<0.01 (Mann-Whitney test). (F) Representative TG curves of blood at 1% 2bF8 or plasma FVIII levels.

At almost all FVIII dose levels, 2bF8 had significantly shorter lag and peak times than plasma FVIII indicating faster onset and peak of TG (Fig 5A and 5B). Remarkably, at as little as the 0.2% plasma FVIII equivalent, 2bF8 showed a significantly shorter lag time and peak time than FVIIInull blood, while the 0.2% plasma FVIII level did not differ from FVIIInull blood. Rather, 2% plasma FVIII was required to significantly reduce lag and peak time compared with FVIIInull blood (Fig. 5A and 5B). Although at the 1% FVIII level peak thrombin for both 2bF8 and plasma FVIII significantly differed from FVIIInull blood, 2bF8 showed higher peak thrombin than plasma FVIII (Fig. 5C). For the ETP a similar dose response for 2bF8 and plasma FVIII was found, suggesting similar overall potential for TG. (Fig. 5D). For TGR significantly higher values than FVIIInull blood were observed at 1% of 2bF8 and at 2% of plasma FVIII. For most FVIII dose levels, TGR was significantly higher with 2bF8 than with plasma FVIII (Fig. 5E). Thus, a 10-fold lower dose of 2bF8 than plasma-FVIII was required to similarly shorten lag and peak time, indicating that 2bF8 seemed to accelerate TG compared to plasma FVIII as illustrated for the 1% FVIII levels in Fig. 5F.

Discussion

Platelet targeted FVIII gene therapy has been proven successful in mice [3–7, 22, 23] and dogs [24]. We previously showed that FVIII expressed in platelets under the control of the αIIb promoter (2bF8) restored hemostasis in a murine model of HA even in the presence of high titer inhibitors [3–6]. Because levels of 2bF8 may vary between individuals due to variable ex vivo transduction and gene expression efficiency, we here evaluated efficacy of 2bF8 over a wide dose range.

Measuring thrombin generation (TG) with TG assays (TGA) has been suggested for monitoring factor replacement [16, 17] and inhibitor bypassing therapy [25, 26] of HA patients. TGA have also been recommended as an ex vivo tool to determine responsiveness of patient samples to FVIII or bypassing agents and thereby tailor individual therapy [26–28]. While sensitivity of TGA to differences in plasma FVIII levels has been reported, sensitivity to FVIII expressed in platelets has not yet been shown. In this study we used TG as read out for efficacy of 2bF8 compared with plasma FVIII. To specifically investigate FVIII expressed in platelets and because other blood cells have been shown to contribute to coagulation [10–14], we employed a whole blood TGA. We modified the protocol established by Ninivaggi et al [15] and chose a strategy that emphasized the intrinsic pathway allowing a better sensitivity to FVIII. In our nWB-TGA, re-calcification of whole blood was used to initiate TG without the addition of TF. While most reports using TF to initiate TG correlated FVIII:C with ETP or peak thrombin, we were able to show FVIII dose dependency in all five TG parameters (Figs 2 and 3). The work that pioneered the whole blood TGA, which together with a study from the same research group [29] is the only report on a whole blood TGA so far, used TF and correlated FVIII:C to ETP and peak thrombin. However, no correlation of FVIII:C with lag time, peak time or TGR was observed [15]. Lack of TF in the nWB-TGA, thereby avoiding initiation of coagulation through the extrinsic pathway, was likely the reason for this striking FVIII dose dependency - particularly in the lag time. In our assay, contact activation at the surface of the filter paper disks seemed to be the mechanism triggering TG [30] as addition of CTI significantly prolonged lag times in WT blood. Our results from the nWB-TGA are in agreement with studies from plasma-based TGA that triggered clotting through the intrinsic pathway to increase sensitivity to FVIII. Initiation of clotting without TF but using FIXa instead, allowed accurate measurement of FVIII levels below 1U dL−1 [31]. Using a subsampling method, induction of TG with FIXa resulted in a FVIII dose dependent lag time [32]. Interestingly, very weak activation through the extrinsic pathway also seemed to increase FVIII sensitivity. A roughly 6-fold reduction of the TF concentration commonly used in a plasma-based assay (0.179pM TF versus 1pM TF) revealed a FVIII dose dependent lag time [33]. The aforementioned studies are two rare reports that describe a FVIII dose dependent lag time. Further studies are required to clearly identify the lack of TF as reason for high FVIII sensitivity of the nWB-TGA. The pioneer work for the whole blood TGA elegantly demonstrated that red cells rather than platelets are the major source of procoagulant membranes [15]. Although, accordingly, our preliminary data indicated that platelet activation by ADP did not increase TG (data not shown), further investigations are necessary to determine the impact of platelets in the nWB-TGA. While we observed variations in some TG parameters within individual mouse strains, there was significant difference between WT and FVIIInull mice without overlap. The relatively wide range could be due to natural variations in some coagulation factors, e.g. FVIII levels [34], or aggressive behavior of some animals. Evidence suggests that coagulation factors, including FVIII and VWF, significantly change after exercise [35–37]. To demonstrate potential clinical usefulness, analyzes of blood from a healthy human control showed that, with the exception of the inter-assay variation of the TGR, both intra-assay and inter-assay coefficient of variations (CV) of all TG parameters were <15% (data not shown).

The development of a new mouse model expressing high levels of 2bF8 (LV18tg mice) allowed evaluation of 2bF8 efficacy over a wide dose range. Homozygous LV18tg mice expressed 12mU FVIII per 108 platelets (Fig 4), which corresponded to a plasma FVIII level of about 20% in WT mice. LV18tg mice were generated by lentivirus transduction of oocytes and transplantation into pseudo-pregnant females. In contrast, previously published 2bF8 transgenic mice (2bF8trans) generated by electroporation of ES cells with the 2bF8 construct expressed 8-fold lower 2bF8 levels (1.5mU FVIII per 108 platelets) [3]. In mammalian systems the presence of multiple homologous copies of a transgene within a concatameric array was shown to have repressive effects on transgene expression [38, 39]. This could explain the high expression levels in our new transgenic model (LV18tg mice), which had a single transgene insertion. In contrast, 2bF8trans mice had 11 consecutive copies of the 2bF8 transgene insertion which might have caused lower transgene expression [3]. Increased DNA methylation and formation of a suppressive chromatin environment might cause repression of repetitive transgene arrays [38]. Additionally, the integration site impacts transgene expression levels [40]. Accordingly, a second mouse line with a single 2bF8 transgene inserted at a different genomic location had somewhat reduced levels of 2bF8 (data not shown). Whether one of the suggested mechanisms is involved in regulation of 2bF8 expression remains to be determined.

In this study we showed for the first time that efficacy of FVIII expressed in platelets can be assessed with TGA. Similar to ex vivo spiking of whole blood with rhFVIII using the novel nWB-TGA we found dose dependency of 2bF8 in all five TG parameters (Fig 5). We showed that an equivalent of 0.2% of 2bF8 was sufficient to accelerate TG compared to FVIIInull blood, while 0.2% of plasma FVIII was not effective. In contrast, a 10-fold higher concentration of plasma FVIII (2%) was required to observe a similar effect. These data are in agreement with our previous studies showing that very low levels of 2bF8 are more efficient than equivalently low levels of plasma FVIII and suffice to fully restore hemostasis. All mice survived tail clipping with 0.5% of 2bF8 levels while only about 20% of FVIIInull mice infused with rhFVIII to 0.5% plasma levels survived this test [3]. We previously demonstrated therapeutic efficacy of 2bF8 in mice even in the presence of inhibitors [3–6]. In blood from such mice, however, no TG was detected using the nWB-TGA (data not shown). This is not surprising as the nWB-TGA is a static assay, while in vivo, 2bF8 released directly into the growing clot is protected from circulating inhibitors. Interestingly, we found that TG was induced faster with 2bF8 than with plasma FVIII (shorter lag and peak time, and higher TGR) while ETP was similar (Fig 5). These results suggest that the potential mechanism of superiority of 2bF8 over plasma FVIII might be acceleration of TG rather than increase of the overall capacity to generate thrombin. Mice with even higher 2bF8 levels than LV18tg mice could address the question if further increase of 2bF8 enhances the overall capacity to generate thrombin.

We conclude that therapeutic efficacy of platelet expressed FVIII can be evaluated with TGA. 2bF8 exhibits higher efficacy than plasma FVIII determined by nWB-TGA. Additionally, the novel nWB-TGA introduced here might be a useful tool to monitor HA and potentially hemophilia B patients during factor replacement, bypassing and platelet targeted gene therapy.

Acknowledgement

We thank Mac Monroe and Alisa Wolberg at the University of North Carolina (Chapel Hill) for helpful discussion on TGA. This work was supported by the Novo Nordisk Access-to-Insight initiative (C.K.B.), NIH-5P01HL044612 (R.R.M.), NIH-5P01HL081588 (R.R.M.), NIH-7R01HL112641 (R.R.M.), NIH-R01HL102035 (Q.S.), the MACC fund (Q.S.), the Children’s Hospital Foundation (Q.S.), and funding from the Medical College and the Children’s Hospital of Wisconsin to the Pediatric-Cardiac-Thrombosis-and-Hemostasis-Research-Center (R.R.M.)

Footnotes

Addendum:

C. K. Baumgartner designed and performed experiments, analyzed and interpreted data, and wrote manuscript. G. Zhang and E. L. Kuether designed and performed experiments. H. Weiler facilitated transgenic mouse generation. Q. Shi and R. R. Montgomery helped design experiments, analyzed and interpreted data, and made critical comments on manuscript.

Disclosure of Conflict of Interest:

The authors state that they have no conflict of interest.

References

  • 1.Björkman S, Blanchette VS, Fischer K, Oh M, Spotts G, Schroth P, Fritsch S, Patrone L, Ewenstein BM, Collins PW. Comparative pharmacokinetics of plasma- and albumin-free recombinant factor VIII in children and adults: the influence of blood sampling schedule on observed age-related differences and implications for dose tailoring. J Thromb Haemost. 2010;8:730–736. doi: 10.1111/j.1538-7836.2010.03757.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ehrenforth S, Kreuz W. Incidence of development of factor VIII and factor IX inhibitors. Lancet. 1992;339:594–598. doi: 10.1016/0140-6736(92)90874-3. [DOI] [PubMed] [Google Scholar]
  • 3.Shi Q, Wilcox DA, Fahs SA, Weiler H, Wells CW, Cooley BC, Desai D, Morateck PA, Gorski J, Montgomery RR. Factor VIII ectopically targeted to platelets is therapeutic in hemophilia A with high-titer inhibitory antibodies. J Clin Invest. 2006;116:1974–1982. doi: 10.1172/JCI28416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Shi Q, Wilcox DA, Fahs SA, Fang J, Johnson BD, Du LM, Desai D, Montgomery RR. Lentivirus-mediated platelet-derived factor VIII gene therapy in murine haemophilia A. J Thromb Haemost. 2007;5:352–361. doi: 10.1111/j.1538-7836.2007.02346.x. [DOI] [PubMed] [Google Scholar]
  • 5.Shi Q, Fahs SA, Wilcox DA, Kuether EL, Morateck PA, Mareno N, Weiler H, Montgomery RR. Syngeneic transplantation of hematopoietic stem cells that are genetically modified to express factor VIII in platelets restores hemostasis to hemophilia A mice with preexisting FVIII immunity. Blood. 2008;112:2713–2721. doi: 10.1182/blood-2008-02-138214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Kuether EL, Schroeder JA, Fahs SA, Cooley BC, Chen Y, Montgomery RR, Wilcox Da, Shi Q. Lentivirus-mediated platelet gene therapy of murine hemophilia A with pre-existing anti-factor VIII immunity. J Thromb Haemost. 2012;10:1570–1580. doi: 10.1111/j.1538-7836.2012.04791.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Schroeder JA, Chen Y, Fang J, Wilcox DA, Shi Q. In vivo enrichment of genetically manipulated platelets corrects the murine hemophilic phenotype and induces immune tolerance even using a low multiplicity of infection. J Thromb Haemost. 2014;12:1283–1293. doi: 10.1111/jth.12633. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Brummel-Ziedins KE, Vossen CY, Butenas S, Mann KG, Rosendaal FR. Thrombin generation profiles in deep venous thrombosis. J Thromb Haemost. 2005;3:2497–2505. doi: 10.1111/j.1538-7836.2005.01584.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Kremers RMW, Peters TC, Wagenvoord RJ, Hemker HC. The balance of pro- and anticoagulant processes underlying thrombin generation. J Thromb Haemost. 2015;13:437–447. doi: 10.1111/jth.12798. [DOI] [PubMed] [Google Scholar]
  • 10.Whelihan MF, Mann KG. The role of the red cell membrane in thrombin generation. Thromb Res. 2013;131:377–382. doi: 10.1016/j.thromres.2013.01.023. [DOI] [PubMed] [Google Scholar]
  • 11.Horne MK, Cullinane AM, Merryman PK, Hoddeson EK. The effect of red blood cells on thrombin generation. Br J Haematol. 2006;133:403–408. doi: 10.1111/j.1365-2141.2006.06047.x. [DOI] [PubMed] [Google Scholar]
  • 12.Hiden M, Haidl H, Rosenkranz A, Cimenti C, Leschnik B, Muntean W. Monocytes enhance rVIIa induced thrombin generation in absence of platelets and microparticles. 37th Hemoph Symp Hambg. 2006:95–105. [Google Scholar]
  • 13.Libby P, Simon DI. Inflammation and Thrombosis. Circulation. 2001;103:1718–1721. doi: 10.1161/01.cir.103.13.1718. [DOI] [PubMed] [Google Scholar]
  • 14.Aleman MM, Gardiner C, Harrison P, Wolberg AS. Differential contributions of monocyte- and platelet-derrived microparticles towards thrombin generation and fibrin formation and stability. J Thromb Haemost. 2011;9:2251–2261. doi: 10.1111/j.1538-7836.2011.04488.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ninivaggi M, Apitz-Castro R, Dargaud Y, de Laat B, Hemker HC, Lindhout T. Whole-blood thrombin generation monitored with a calibrated automated thrombogram-based assay. Clin Chem. 2012;58:1252–1259. doi: 10.1373/clinchem.2012.184077. [DOI] [PubMed] [Google Scholar]
  • 16.Young G, Sørensen B, Dargaud Y, Negrier C, Brummel-Ziedins K, Key NS. Thrombin generation and whole blood viscoelastic assays in the management of hemophilia: current state of art and future perspectives. Blood. 2013;121:1944–1950. doi: 10.1182/blood-2012-08-378935. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Lewis SJ, Stephens E, Florou G, Macartney NJ, Hathaway LS, Knipping J, Collins PW. Measurement of global haemostasis in severe haemophilia A following factor VIII infusion. Br J Haematol. 2007;138:775–782. doi: 10.1111/j.1365-2141.2007.06722.x. [DOI] [PubMed] [Google Scholar]
  • 18.Gilmore R, Harmon S, Gannon C, Byrne M, O’Donnell JS, Jenkins PV. Thrombin generation in haemophilia A patients with mutations causing factor VIII assay discrepancy. Haemophilia. 2010;16:671–674. doi: 10.1111/j.1365-2516.2009.02190.x. [DOI] [PubMed] [Google Scholar]
  • 19.Dargaud Y, Béguin S, Lienhart A, Dieri RAl, Trzeciak C, Bordet JC, Hemker HC, Negrier C. Evaluation of thrombin generating capacity in plasma from patients with haemophilia A and B. Thromb Haemost. 2005;93:475–480. doi: 10.1160/TH04-10-0706. [DOI] [PubMed] [Google Scholar]
  • 20.Bi L, Lawler A, Antonarakis S, High K, Gearhart J, Kazazian H. Targeted disruption of the mouse factor VIII gene produces a model of haemophilia A. Nat Genet. 1995;10:119–121. doi: 10.1038/ng0595-119. [DOI] [PubMed] [Google Scholar]
  • 21.Zhang G, Shi Q, Fahs SA, Kuether EL, Walsh CE, Montgomery RR. Factor IX ectopically expressed in platelets can be stored in alpha-granules and corrects the phenotype of hemophilia B mice. Blood. 2010;116:1235–1243. doi: 10.1182/blood-2009-11-255612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Yarovoi HV, Kufrin D, Eslin DE, Thornton MA, Haberichter SL, Shi Q, Zhu H, Camire R, Fakharzadeh SS, Kowalska MA, Wilcox DA, Sachais BS, Montgomery RR, Poncz M. Factor VIII ectopically expressed in platelets: efficacy in hemophilia A treatment. Blood. 2003;102:4006–4013. doi: 10.1182/blood-2003-05-1519. [DOI] [PubMed] [Google Scholar]
  • 23.Yarovoi HV, Nurden AT, Montgomery RR, Nurden P, Poncz M. Intracellular interaction of von Willebrand factor and factor VIII depends on cellular context: lessons from platelet-expressed factor VIII. Blood. 2005;105:4674–4676. doi: 10.1182/blood-2004-12-4701. [DOI] [PubMed] [Google Scholar]
  • 24.Du LM, Nurden P, Nurden AT, Nichols TC, Bellinger DA, Jensen ES, Haberichter SL, Merricks E, Raymer RA, Fang J, Koukouritaki SB, Jacobi PM, Hawkins TB, Cornetta K, Shi Q, Wilcox DA. Platelet-targeted gene therapy with human factor VIII establishes haemostasis in dogs with haemophilia A. Nat Commun. 2013;4:2773. doi: 10.1038/ncomms3773. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Turecek PL, Váradi K, Keil B, Negrier C, Berntorp E, Astermark J, Bordet J-C, Morfini M, Linari S, Schwarz HP. Factor VIII Inhibitor-Bypassing Agents Act by Inducing Thrombin Generation and Can Be Monitored by a Thrombin Generation Assay. Pathophysiol Haemost Thromb. 2003;33:16–22. doi: 10.1159/000071637. [DOI] [PubMed] [Google Scholar]
  • 26.Dargaud Y, Lienhart A, Negrier C. Brief report Prospective assessment of thrombin generation test for dose monitoring of bypassing therapy in hemophilia patients with inhibitors undergoing elective surgery. Blood. 2010;116:5734–5738. doi: 10.1182/blood-2010-06-291906. [DOI] [PubMed] [Google Scholar]
  • 27.Klintman J, Astermark J, Berntorp E. Combination of FVIII and by-passing agent potentiates in vitro thrombin production in haemophilia A inhibitor plasma. Br J Haematol. 2010;151:381–386. doi: 10.1111/j.1365-2141.2010.08378.x. [DOI] [PubMed] [Google Scholar]
  • 28.Doshi BS, Gangadharan B, Doering CB, Meeks SL. Potentiation of thrombin generation in hemophilia A plasma by coagulation factor VIII and characterization of antibody-specific inhibition. PLoS One. 2012;7:e48172. doi: 10.1371/journal.pone.0048172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ninivaggi M, Kelchtermans H, Kuijpers MJ, Hemmeryckx B, Heemskerk JWM, Lindhout T, Hoylaerts MF, Laat B De. Whole blood thrombin generation in Bmal1-deficient mice. Thromb Haemost. 2014;112:271–275. doi: 10.1160/TH13-11-0910. [DOI] [PubMed] [Google Scholar]
  • 30.de Maat S, Tersteeg C, Herczenik E, Maas C. Tracking down contact activation - from coagulation in vitro to inflammation in vivo. Int J Lab Hematol. 2014;36:374–381. doi: 10.1111/ijlh.12222. [DOI] [PubMed] [Google Scholar]
  • 31.Ninivaggi M, Dargaud Y, van Oerle R, de Laat B, Hemker HC, Lindhout T. Thrombin generation assay using factor IXa as a trigger to quantify accurately factor VIII levels in haemophilia A. J Thromb Haemost. 2011;9:1549–1555. doi: 10.1111/j.1538-7836.2011.04358.x. [DOI] [PubMed] [Google Scholar]
  • 32.McIntosh JH, Owens D, Lee CA, Raut S, Barrowcliffe TW. A modified thrombin generation test for the measurement of factor VIII concentrates. J Thromb Haemost. 2003;1:1005–1011. doi: 10.1046/j.1538-7836.2003.00209.x. [DOI] [PubMed] [Google Scholar]
  • 33.Salvagno GL, Astermark J, Lippi G, Ekman M, Franchini M, Guidi GC, Berntorp E. Thrombin generation assay: a useful routine check-up tool in the management of patients with haemophilia? Haemophilia. 2009;15:290–296. doi: 10.1111/j.1365-2516.2008.01877.x. [DOI] [PubMed] [Google Scholar]
  • 34.Fahs SA, Hille MT, Shi Q, Weiler H, Montgomery RR. A conditional knockout mouse model reveals endothelial cells as the principal and possibly exclusive source of plasma factor VIII. Blood. 2014;123:3706–3713. doi: 10.1182/blood-2014-02-555151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Rock G, Tittley P, Pipe A. Coagulation factor changes following endurance exercise. Clin J Sport Med. 1997;7:94–99. doi: 10.1097/00042752-199704000-00004. [DOI] [PubMed] [Google Scholar]
  • 36.Lockard MM, Gopinathannair R, Paton CM, Phares DA, Hagberg JM. Exercise training-induced changes in coagulation factors in older adults. Med Sci Sports Exerc. 2007;39:587–592. doi: 10.1249/mss.0b013e31802eff4b. [DOI] [PubMed] [Google Scholar]
  • 37.Beltrame LG, Abreu L, Almeida J, Boullosa DA. The acute effect of moderate intensity aquatic exercise on coagulation factors in haemophiliacs. Clin Physiol Funct Imaging. 2015;35:191–196. doi: 10.1111/cpf.12145. [DOI] [PubMed] [Google Scholar]
  • 38.Garrick D, Fiering S, Martin DIK, Whitelaw E. Repeat-induced gene silencing in mammals. Nat Genet. 1998;18:56–59. doi: 10.1038/ng0198-56. [DOI] [PubMed] [Google Scholar]
  • 39.Calero-Nieto FJ, Bert AG, Cockerill PN. Transcription-dependent silencing of inducible convergent transgenes in transgenic mice. Epigenetics and Chomatin. 2010;3:1–13. doi: 10.1186/1756-8935-3-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Girton JR, Johansen KM. Chromatin structure and the regulation of gene expression: the lessons of PEV in Drosophila. Adv Genet. 2008;61:1–43. doi: 10.1016/S0065-2660(07)00001-6. [DOI] [PubMed] [Google Scholar]

RESOURCES