Hemophilic Mice Research Articles

Pharmacological concentrations of recombinant factor VIIa restore hemostasis independent of tissue factor in antibody‐induced hemophilia mice

ESSENTIALS: The role of tissue factor (TF) in recombinant factor VIIa (rFVIIa) therapy in hemophilia is unclear. An acquired mouse hemophilia model with very low or normal levels of human TF was used in the study. rFVIIa is equally effective in correcting the bleeding in mice expressing low or normal levels of TF. Pharmacological doses of rFVIIa restore hemostasis in hemophilia independent of TF. Recombinant factor VIIa (rFVIIa) has been used widely for treating hemophilia patients with inhibitory autoantibodies against factor VIII or IX. Its mechanism of action is not entirely known. A majority of in vitro studies suggested that pharmacological concentrations of rFVIIa restore hemostasis in hemophilia in a phospholipid-dependent manner, independent of tissue factor (TF). However, a few studies suggested that a TF-dependent mechanism has a primary role in correction of bleeding by rFVIIa in hemophilia patients. Here, we investigated the potential contribution of TF in rFVIIa-induced hemostasis in hemophilia employing a model system of FVIII antibody-induced hemophilia in TF transgenic mice. Mice expressing low levels of human TF (LTF mice), mice expressing relatively high levels of human TF (HTF mice) and wild-type mice (WT mice) had neutralizing anti-FVIII antibodies administered in order to induce hemophilia in these mice. The mice were then treated with varying concentrations of rFVIIa. rFVIIa-induced hemostasis was evaluated with the saphenous vein bleeding model. Administration of FVIII inhibitory antibodies induced the hemophilic bleeding phenotype in all three genotypes. rFVIIa administration rescued the bleeding phenotype in all three genotypes. No significant differences were observed in rFVIIa-induced correction of bleeding between LTF and HTF mice that had FVIII antibodies administered. Our results provide strong evidence supporting the suggestion that the hemostatic effect of pharmacological doses of rFVIIa stems from a TF-independent mechanism.

Effects of FVIII immunity on hepatocyte and hematopoietic stem cell–directed gene therapy of murine hemophilia A

Immune responses to coagulation factors VIII (FVIII) and IX (FIX) represent primary obstacles to hemophilia treatment. Previously, we showed that hematopoietic stem cell (HSC) retroviral gene therapy induces immune nonresponsiveness to FVIII in both naive and preimmunized murine hemophilia A settings. Liver-directed adeno-associated viral (AAV)-FIX vector gene transfer achieved similar results in preclinical hemophilia B models. However, as clinical immune responses to FVIII and FIX differ, we investigated the ability of liver-directed AAV-FVIII gene therapy to affect FVIII immunity in hemophilia A mice. Both FVIII naive and preimmunized mice were administered recombinant AAV8 encoding a liver-directed bioengineered FVIII expression cassette. Naive animals receiving high or mid-doses subsequently achieved near normal FVIII activity levels. However, challenge with adjuvant-free recombinant FVIII induced loss of FVIII activity and anti-FVIII antibodies in mid-dose, but not high-dose AAV or HSC lentiviral (LV) vector gene therapy cohorts. Furthermore, unlike what was shown previously for FIX gene transfer, AAV-FVIII administration to hemophilia A inhibitor mice conferred no effect on anti-FVIII antibody or inhibitory titers. These data suggest that functional differences exist in the immune modulation achieved to FVIII or FIX in hemophilia mice by gene therapy approaches incorporating liver-directed AAV vectors or HSC-directed LV.

The Effect of Prophylactic Recombinant Factor IX and Glycopegylated Factor IX to Normalize Articular Wound Healing Following Hemarthrosis Examined in Hemophilia B Mice

BackgroundRecurrent bleeding into joints results in a crippling arthritis, the consequence of intra-articular (IA) pathologic lysed red blood cell iron and inflammatory signals from macrophages/monocytes. Hemophilic mice that experience joint bleeding develop synovial and osteochondral changes that mimic the changes seen in human hemophilia. We previously examined in hemophilia B mice the efficacy of conventional recombinant factor IX (rFIX) compared to a half-life extended polyethyleneglycol-conjugated factor IX (N9-GP) to support wound healing when administered "on demand" after the initiation of induced joint hemorrhage. In the "on-demand" scenario, N9-GP improved wound healing when compared to rFIX, although neither treatment group demonstrated completely normal joint healing (as observed in hemostatically normal mice that experienced the same injury). We test the hypothesis that the prophylactic use of N9-GP may support wound healing in synovium, cartilage, and bone and may do so more effectively than rFIX.Method: FIX deficient mice were treated prophylactically with either rFIX or N9-GP at the dose of 250 IUnits human factor IX/kg 30 minutes, 24hr or 72hr prior to the induction of unilateral hemarthosis. Injured hemostatically normal (iWT) mice were treated with I.V. normal saline as the controls. Joint diameter was measured daily to quantify joint swelling. Plasma and tissues were harvested from cohorts of mice (n = 8-10/group) at two weeks after hemarthrosis. Pathologic synovial hyperplasia, neovascularization, iron deposition and macrophage infiltration of the synovium were quantified by histologic examination. Integrity of trabecular bone neighboring the hemarthrosis was quantified by microComputed Tomography (mCT).Result: WT mice developed no detectable joint swelling at days 1-14 after the joint bleeding challenge, whereas mice treated prophylactically with rFIX were protected from joint swelling only in the cohort treated 24 hours prior to hemarthrosis. Joint swelling did not develop in groups that received prophylactic N9-GP using any of the 3 schedules. No mouse treated with rFIX (n = 28 total) or N9-GP (n=27 total) developed FIX inhibitors or FIX-binding IgG. Synovitis was graded 0 to 10 by Valentino mouse hemophilic synovitis score, with a mean score in the iWT mice of 1.0±0.5 and in the injured untreated FIX-/- mice of 4.9±1.4. Synovitis was greatly diminished although not eliminated by prophylactic N9-GP given 30 minutes prior to injury (1.7±1.2), and this result contrasted with the relatively less good healing provided by rFIX given 30 minutes prior to injury (3.9±1.1). Synovitis, iron staining, and persistence of macrophages within the joint were each diminished by N9-GP and rFIX prophylaxis. The degree of protection diminished as the interval between dosing and injury increased. Better protection was seen using N9-GP when compared to rFIX at each timepoint. Strikingly, rFIX at this dose provided minimal or no prophylactic protection from synovial neoangiogenesis (vessels/hpf) at two weeks after hemarthrosis even when given at the shortest interval preceding the injury. Injured, untreated FIX-/- mice developed 17.6±5.1 vessels/hpf compared to rFIX given 30 minutes prior to injury (16.1±4.8), 24 hours prior (17.8±5.4) or 72 hours prior (17.1±6.3). In contrast, iWT mice displayed 7.1±3.1 at two weeks after injury, which was comparable to N9-GP given 30 minutes prior to injury (9.4±2.4) and better than N9-GP given 24 hours prior (12.6±3.3) or 72 hours prior (13.1±5.6). Consistent with previous reports, hemophilic mice when compared to iWT mice demonstrated acute losses of bone mineral density in the bone adjacent to the joint experiencing bleeding, as well as structural defects in the number, thickness, and spacing of the bone trabeculae. Prophylactic rFIX or N9-GP given at 30 minutes or 24 hours prior to injury resulted in protection that did not differ statistically from findings in iWT mice. When prophylaxis was given at 72 hours prior to injury, significant volumetric bone mineral density loss of at least 20% was evident in the rFIX group (vBMD 98 mgHA) compared to iWT (131 mgHA) and N9-GP given 72 hours prior (125 mgHA).Conclusion: The support of normal wound healing in joint and bone following hemarthrosis was significantly improved by N9-GP prophylaxis when compared to rFIX in hemophilia B mice, with neoangiogenesis particularly improved. DisclosuresKnudsen:Novo Nordisk: Employment. Monroe:Novo Nordisk: Honoraria, Research Funding. Ezban:Novo Nordisk: Employment. Bateman:Novo Nordisk: Honoraria, Research Funding; Baxter: Honoraria, Research Funding. Monahan:CSL Behring: Consultancy, Honoraria; Pfizer: Honoraria; Chatham LLC: Consultancy; Baxter/Baxalta: Consultancy, Honoraria, Research Funding; Prolor: Research Funding; Bayer: Consultancy; Novo Nordisk: Consultancy, Honoraria, Research Funding; Asklepios BioPharmaceutical: Consultancy, Patents & Royalties: Author I.P. licensed by UNC to AskBio, Research Funding.

Frequency and Epitope Specificity of Anti-Factor VIII C1 Domain Antibodies in Acquired and Hereditary Hemophilia Inhibitor Patient Plasma

Acquired hemophilia A (AHA) is a rare autoimmune disease caused by development of inhibitory anti-factor VIII (fVIII) antibodies (also called inhibitors) resulting in severe hemorrhages. In addition, inhibitor development is the most serious complication of today's replacement therapy in patients with hereditary X-linked hemophilia A (HA) disorder. Earlier studies showed that antibodies in AHA and HA inhibitor plasmas are both primarily directed to the A2 and C2 domains suggesting that these two domains are the predominant immunogenic fVIII regions (Fulcher et al, 1987; Prescott et al, 1997; Lollar, 2004). However, the C1 domain also makes a major contribution to the humoral anti-fVIII immune response in hemophilic mice (Healey et al, 2007), which motivated us to analyze the frequency and epitope specificity of anti-C1 antibodies in AHA and HA inhibitor patient plasma.The frequency of domain-specific antibodies were studied by antibody binding to human A2, C1 and C2 domains presented as (i) single human domain (SHD) human/porcine hybrid fVIII and (ii) HSA-fusion proteins. While similar frequencies of A2- and C2-specific antibodies were observed for both applied mapping strategies the use of isolated C1 domain resulted in much higher detection level of anti-C1 antibodies compared to the use of the human C1 domain human/porcine hybrid fVIII protein. As homologue-scanning mutagenesis relies on differences among human and porcine sequences these results suggest the presence of a large number of cross-reactive anti-C1 antibodies binding to species-conserved epitopes. Overall, anti-C1 antibodies were detected in 90 of 115 (78%) AHA and 36 of 63 (57%) HA inhibitor patients.Two well-characterized monoclonal C1 inhibitors, human LE2E9 (Jacquemin et al, 2000) and murine MAb 2A9 (ASH 2014 poster, Batsuli et al) were used for indirect epitope mapping of anti-C1 antibodies in AHA patients by competition binding studies. Our results for AHA patients with non-crossreactive anti-C1 antibodies only (n=11) show that antibody binding to human C1 domain human/porcine hybrid fVIII (HP53) protein was completely blocked in the presence of MAb 2A9. In contrast, antibody binding to the isolated C1 domain was only partially reduced in the presence of MAb 2A9 for a selected number of (high responding) AHA patients (n=10) suggesting the presence of a second population of crossreactive anti-C1 antibodies that exclusively bind to conserved amino acid residues.Competition binding to native and denatured fVIII and HP53 proteins revealed that MAb 2A9 and LE2E9 bind mutually exclusive to a conformational C1 epitope involving amino acid residues that are not conserved between humans and pigs. Consequently, essential binding residues were identified for both C1 inhibitors via the use of HP53 variants, in which surface exposed non-conserved amino acid residues on the human C1 domain were substituted for porcine residues. The results of this mutational analysis showed that despite their competitive binding different amino acid residues are essential for binding of MAb 2A9 and LE2E9. These findings are in agreement with the different specific inhibitory activities of the two C1 inhibitors (97 BU/mg vs 10000 BU/mg).Finally, HSA-C1 point mutants were used to directly map essential epitope residues of anti-C1 antibodies in AHA and HA inhibitor patient plasma.Our study demonstrates that a large number of AHA and HA inhibitor patients (126 of 178; 71%) have anti-C1 antibodies that comprise at least two different populations, crossreactive and non-crossreactive to porcine fVIII. Therefore, in addition to the A2 and C2 domains, the C1 domain seems to significantly contribute to the immune response to fVIII in these patients. As recent data point toward a functional role of the fVIII C1 domain for membrane-, fX-, and von Willebrand factor-binding (Lü et al, 2011) the clinical relevance of anti-C1 antibodies should be analyzed in further studies. DisclosuresTiede:Leo Pharma: Consultancy, Honoraria; Novo Nordisk: Consultancy, Honoraria, Research Funding; Coachrom: Research Funding; SOBI: Consultancy, Honoraria; Biogen Idec: Consultancy, Honoraria; CSL Behring: Consultancy, Honoraria, Research Funding; Pfizer: Consultancy, Honoraria; Boehringer Ingelheim: Consultancy, Honoraria; Octapharma: Other: Investigator, Speakers Bureau; Biotest: Consultancy, Honoraria, Research Funding; Bayer: Consultancy, Honoraria, Investigator, Research Funding; Baxter: Consultancy, Honoraria, Research Funding. Königs:Bayer: Research Funding, Speakers Bureau; Biotest: Research Funding, Speakers Bureau; Pfizer: Research Funding, Speakers Bureau; Sobi: Consultancy; CSL Behring: Research Funding, Speakers Bureau; Intersero: Research Funding; NovoNordisk: Speakers Bureau.

Targeting FVIII Expression to Platelets for Hemophilia A Gene Therapy Does Not Bare an Apparent Thrombosis Risk

Our group has previously developed a gene therapy approach for hemophilia A in which FVIII expression is targeted to platelets. Platelet expressed FVIII successfully restores hemostasis in hemophilic mice even in the presence of high titer inhibitory anti-FVIII antibodies and induces immune tolerance to FVIII. Therapeutic efficacy was achieved in our original transgenic mouse line at a level of only 0.75mU FVIII per 108 platelets, which corresponded to an equivalent of 1.25% FVIII in plasma of wild-type (WT) mice. FVIII is not normally expressed in platelets but with platelet-FVIII gene therapy FVIII levels at the site of injury might dramatically increase due to platelet aggregation and activation, and consequent local FVIII release from platelets together with VWF. Furthermore, a higher embolism rate has been suggested in transgenic mice expressing FVIII in platelets on an otherwise FVIII deficient background. Thus, although this approach is very successful in restoring hemostasis, evaluating potential pathological consequences in conjunction with platelet-FVIII is of great importance. Here we explored whether a pro-thrombotic state was induced by platelet expressed FVIII in an attempt to define the breadth of the therapeutic window. To examine this, we analyzed high platelet-FVIII expressing mice using five techniques including 1) a native whole blood thrombin generation assay, 2) ex vivo clot formation using thomboelastometry (ROTEM), 3) assessment of plasma parameters linked with an increased thrombosis risk (D-Dimer, thrombin anti-thrombin complexes (TAT), fibrinogen), 4) in vivo clot formation using a ferric chloride carotid artery injury model, and 5) tissue fibrin deposition. In addition to steady state, mice were subjected to an inflammatory challenge to induce pro-thrombotic conditions.We generated transgenic mice, LV17/18tg, that expressed 18mU FVIII per 108 platelets on a FVIII deficient background. This platelet-FVIII level was 24-fold higher than in our originally described mouse line. While FVIII deficient mice had negligible thrombin generation, comparing WT control to LV17/18tg mice, neither thrombin generation nor ex vivo clot formation was increased above WT levels in the transgenic mice. In WT and LV17/18tg, respectively, peak thrombin was 216 ± 16 and 195 ± 13 nM and endogenous thrombin potential was 1718 ± 41 and 1642 ± 56 nM. Clotting time determined by ROTEM in WT and LV17/18tg, respectively, was 424 ± 20 and 705 ± 37 seconds and maximum clot firmness was 51.3 ± 1.4 and 54.0 ± 1.7 mm. Fibrinogen and D-Dimer levels were similar in WT and LV17/18tg mice. While TAT levels were significantly higher in LV17/18tg (11.6 ± 1.1 ng/ml) than in WT mice (6.0 ± 0.5 ng/ml), interestingly this increase was also observed in FVIIInull mice (11.0 ± 1.2 ng/ml). Therefore, we attributed the elevation of TAT levels to the FVIII deficient background. Why the FVIII deficient mice have elevated TAT levels is not clear and may reflect an abnormality in the FVIII deficient line that we are further investigating. When we monitored time to occlusion in an in vivo thrombus formation model using ferric chloride induced carotid artery injury, vessels in LV17/18tg mice did not occlude faster than in WT mice (time to occlusion: 14.5 ± 0.5 versus 5.9 ± 0.5 min, respectively). Thus, under steady state, platelet-FVIII did not seem to be pro-thrombotic. This led us to evaluate the thrombosis risk under pro-thrombotic conditions. Because inflammation induces a pro-thrombotic state, we challenged mice with 40mg/kg LPS and analyzed the animals 16 hours later. WT and LV17/18tg mice treated with LPS showed a similar significant increase in fibrinogen, D-Dimer and TAT levels over PBS treated control mice. While no fibrin deposition was observed in the liver of WT or LV17/18tg mice, LPS challenge did induce fibrin deposition, but there was no difference between WT and LV17/18tg mice.In conclusion, 24-fold higher platelet-FVIII levels than in our originally described mouse line in which hemostasis was fully restored, did not show any signs of thrombosis in steady state or under inflammatory conditions. Therefore, in this regard, platelet targeted FVIII gene therapy can be considered a safe therapy with a relatively wide therapeutic window that is in excess of 24-fold. DisclosuresBaumgartner:Novo Nordisk: Research Funding. Shi:BloodCenter of Wisconsin: Patents & Royalties: METHOD OF INDUCING IMMUNE TOLERANCE THROUGH TARGETED GENE EXPRESSION. . Montgomery:Biogen: Consultancy; Bayer: Consultancy; CSL Behring: Consultancy; Baxter: Consultancy; Octapharma: Consultancy; Grifols: Consultancy.

A Single Amino Acid in the Gla Domain Mouse FVIIa Allows Its Binding to the Endothelial Protein C Receptor and Enhances Its Coagulant Activity In Vivo

The human protein C (PC) interaction with the endothelial PC receptor (EPCR) is mediated through the PC Gla domain, via key amino acids Phe4 and Leu8. Specifically, substitution of Leu8 with Val from human prothrombin abolishes the PC-EPCR interaction. The Gla domain of human Factor VII (FVII) shares these positions with PC and, consequently, the EPCR binding capacity. In the mouse, a commonly used in vivo model, the sequence determinants of the Gla domain of mouse PC (mPC) interaction with mouse EPCR (mEPCR) are not known. Remarkably, mouse FVII (mFVII) and its activated form (mFVIIa) have poor affinity for mouse EPCR. We previously described a variant of mFVIIa (mFVIIa-FMR) that contained the Leu4->Phe, Leu8->Met and Trp9->Arg from the mPC Gla domain. We found that this molecule was functionally similar to mFVIIa and could bind mEPCR. Using mFVIIa-FMR as surrogate to study the mPC-mEPCR interaction, we highlighted the importance of the Phe4/Met8/Arg9 in the mPC-mEPCR interaction. We also found that mFVIIa-FMR had enhanced hemostatic properties when infused at 3 mg/kg after FeCl3 carotid artery injury in hemophilic mice (vessel occlusion was 2.5 times faster than mFVIIa). In order to further refine whether the mEPCR binding capacity of mPC is coordinated by any/all of Phe4/Met8/Arg9 positions, we previously generated single variants of mPC at these positions using the corresponding amino acids of mFVIIa (that has poor interaction with mEPCR). We found that Phe4 is the sole determinant of specificity of the mPC-mEPCR interaction. Moreover, when Phe4 was placed in mFVIIa, we found that mFVIIa-Phe4 had activity similar to mFVIIa and bound mEPCR on cells (or in solution to soluble mEPCR) with a Kd of ~350nM. This was of similar magnitude to the mEPCR affinity of a mFVIIa variant with the entire mPC Gla domain (~200 nM), indirectly suggesting that Phe4 determines both the specificity and affinity of mPC to mEPCR. Since mFVIIa-FMR showed improved hemostatic properties in vivo as a result of mEPCR binding, enhancing the EPCR-FVIIa binding may generate improved human FVIIa molecules for the treatment of bleeding. Here we wanted to provide proof-of-concept using limited Gla domain modifications. For this, we utilized mFVIIa-Phe4, a minimally modified mFVIIa molecule, described above. Specifically, hemophilia B animals were subjected to a 7.5% FeCl3 injury of their carotid artery for 2 minutes; after 10 minutes mice were infused with 3 mg/kg of mFVIIa or mFVIIa-Phe4. Time to vessel occlusion was determined by monitoring blood flow. Hemostatically normal mice occluded in 13.3 ± 3.0 min. We found that infusion of mFVIIa resulted in vessel occlusion at 8.9 ± 1.7 min. However, mice that received mFVIIa-Phe4 reached vessel occlusion within 4.5 ± 2.4 min, ~2.5 times faster than mFVIIa-infused mice (P<0.01). This was similar to that we previously observed with mFVIIa-FMR infusion after injury in hemophilia B mice. Our results suggest the following: (1) Phe4 in the mPC Gla domain confers the specificity and affinity to mEPCR; (2) a single Phe4 substitution in mFVIIa is the only requirement for enhancing its clotting function in vivo. These data reveal another difference between human and mouse systems that may affect EPCR-dependent functions of other vitamin K-dependent proteins. Moreover, our results suggest the possibility that minimally modified variants of FVIIa with respect to EPCR binding may have more favorable hemostatic properties for clinical use. DisclosuresIvanciu:Bayer Hemophilia Awards Program: Research Funding. Margaritis:Novo Nordisk A/S: Research Funding; Bayer Hemophilia Awards Program: Research Funding.

Pharmacological Concentrations of Recombinant Factor VIIa Restores Hemostasis in Acquired Hemophilia Primarily Independent of Tissue Factor

High doses of recombinant FVIIa (rFVIIa) are used successfully to treat hemophilia patients with inhibitory antibodies against FVIII and FIX. Despite its widespread and successful use of treating bleeding disorders for many years, the understanding of the mechanism of rFVIIa action in hemophilia therapy is incomplete. A majority of the studies, mostly in vitro studies, suggested that activation of FX by rFVIIa bound to phospholipids exposed on the activated platelets is responsible for the hemostatic effect of rFVIIa in hemophilia patients. However, other in vitro studies and mathematical modeling indicated that the action of rFVIIa at therapeutic doses is dominated by the TF-dependent pathway, partly because of the need to overcome the inhibitory effect of zymogen FVII, with a minor contribution from phospholipid-dependent mechanism. In the present study, we investigated whether TF plays a role in rFVIIa-mediated hemostasis in hemophilia by employing transgenic mice expressing low levels of human TF (Low TF mice) or normal levels of human TF (HTF mice; breeding pairs for these mice were provided by Nigel Mackman). Mice were administered with FVIII mAb to induce acquired hemophilia. The saphenous vein incision bleeding model was used to evaluate the hemostasis. In this model, saphenous vein was exposed and bleeding was induced by a sharp incision of the vein with a fine scissors. The bleeding from the cut was monitored for 30 min while blood was collected onto a filter paper by capillary action. After each hemostasis incident, the clot was disrupted gently to reinitiate a new bleeding episode, and average time to achieve hemostasis (ATH) was calculated.First, we investigated the role of TF in hemostasis by comparing the bleeding episodes in low TF and HTF mice following the saphenous vein incision. In the saphenous vein bleeding model, control wild-type mice had a median ATH of 60 sec (n=6) and hemophilia A mice failed to achieve hemostasis within 30 min experimental time frame (n=6). HTF mice had essentially very similar ATH compared to that of the wild-type mice, i.e., 64 sec (n=5). In contrast, low TF mice had a substantially longer median ATH, 90 sec (n=6), which is statistically significantly longer than that of ATH of wild-type or HTF mice. Administration of TF antibodies (1 mg/kg) to neutralize remaining traces of TF in Low TF mice further prolonged the bleeding time to a median ATH of 309 sec (n=6). These data clearly illustrates the role of TF in hemostasis.Next, we created acquired hemophilic condition in low TF and HTF mice by injecting FVIII mAb (1 mg/kg). Low TF mice also received TF antibodies. Administration of FVIII mAb markedly prolonged the bleeding time in both groups of mice (between 300 to 1800 sec), with a median ATH of 900 sec. Administration of rFVIIa (4 mg/kg) to HTF hemophilic mice reduced the ATH to a median of 51 sec (mean ± SD, 50 ± 5 sec, n=6). The same concentration of rFVIIa also restored hemostasis in hemophilic low TF mice with a median ATH of 66 sec (mean ±SD, 70 ± 18 sec, n=6). This value is not statistically significantly different from that of ATH of wild-type mice. Although rFVIIa restored the hemostasis in both HTF and low TF mice, it appears that rFVIIa is slightly, but statistically significantly, more effective in HTF mice compared to low TF mice in restoring the hemostasis (p = 0.025). However, we did not observe any significant differences between these two groups in the volume of blood loss following rFVIIa administration. In summary, our data strongly support the hypothesis that high concentrations of rFVIIa restore hemostasis in hemophilia primarily in TF-independent manner. More detailed studies are needed to completely eliminate the role of TF in rFVIIa mediated hemostasis. DisclosuresNo relevant conflicts of interest to declare.

Vascular Remodeling Underlies Re-Bleeding in Hemophilic Arthropathy

Hemophilic arthropathy is a debilitating condition that can develop as a consequence of frequent joint bleeding despite adequate clotting factor replacement. The mechanisms leading to repeated spontaneous bleeding are unknown. To better understand these mechanisms, we investigated the time course of synovial, vascular, stromal and cartilage changes in response to a single induced hemarthrosis in FVIII-deficient mice, compared the observed changes in vascular architecture to the changes in mouse models of rheumatoid arthritis (RA) and osteoarthritis (OA), and explored the extent to which these findings correlated with the development of hemophilic arthropathy in patients. Following a single induced hemarthrosis in the FVIII-deficient mouse, we found soft tissue hyperproliferation with marked induction of neoangiogenesis and evolving abnormal vascular architecture in the injured joint. While soft tissue changes were rapidly reversible, and cartilage changes were mild and mostly reversible, ongoing structural changes of abnormal vascularity persisted for months. These vascular changes involved pronounced remodeling with expression of α-Smooth Muscle Actin (SMA), Endoglin (CD105) and vascular endothelial growth factor (VEGF), as well as altered joint perfusion as determined by in vivo imaging. Surprisingly, these vascular changes were also seen in uninjured joints, suggesting systemic mediation of the observed effects. Changes in vascular architecture and pronounced expression of α-SMA were unique to hemophilia, and were not found in joint tissue obtained from RA or OA mice or from tissue from patients with OA or RA. Since FVIII-deficient mice do not suffer from spontaneous joint bleeding, human studies needed to be designed to reveal if such vascular changes were associated with joint bleeding and joint deterioration. Therefore, to determine if vascular perfusion changes were also present in hemophilia patients and to what extent they were related with joint bleeding and the degree of arthropathy, a cohort of 26 adult patients with hemophilia was studied prospectively. Radiographic Pettersson Scores and clinical Hemophilia Joint Health Scores (HJHS) were assessed and joint bleed status and vascularity changes were investigated by dynamic in vivo imaging using power Doppler (PD) during high resolution musculoskeletal ultrasound examinations of 156 joints (both elbows, knees and ankles of each patient) at baseline and, subsequently of 10 joints during painful episodes. Subclinical joint bleeding was present in 41% of joints at baseline. Positive PD signals were more frequently present in joints with subclinical bleeding than in joints without (83% vs. 55%; p< 0.01), while the mean PD signal score was significantly higher in bleeding joints (3.1 vs. 1.5; p < 0.01). Importantly, joint bleed status was independently associated with PD signal score (odds ratio [OR] = 1.45 [95% CI: 1.13, 1.86, p = 0.0035]) and with Pettersson score (OR = 1.21 [95%CI: 1.03, 1.43, p = 0.0185]) after adjusting for HJHS. Acute bleeding episodes, irrespective of occurring in previously non-bleeding or in previously subclinically bleeding joints only happened in joints where vascularity changes were present at baseline. Moreover, during these episodes, mean PD signal increased significantly from 3.5 at baseline to 6.0 (maximum 9 per joint; p≤0.05). Therefore, in vivo imaging demonstrated that those vascular changes, first observed in the FVIII-deficient mouse, were significantly associated with bleeding and joint deterioration in hemophilia patients.In aggregate, our observations in hemophilic mice and patients provide strong evidence that abnormal blood vessel formation and dynamic vascular remodeling facilitate and perpetuate joint bleeding in response to local and/or systemic stimuli. While changes in vascularity appear to be an important link to repeated bleeding and progression of arthropathy, the interrelations of bleeding and vascular changes are complicated and will require further study. These findings not only provide new insights into the pathobiology of progression of hemophilic arthroapthy, but also open new avenues to study molecular targets for angiogenesis inhibition to prevent aberrant vessel formation, potentially increasing the effectiveness of clotting factor replacement therapy. DisclosuresVon Drygalski:CSL Behring: Honoraria, Speakers Bureau; Novo Nordisk: Honoraria, Research Funding; Biogen: Honoraria, Research Funding; Hematherix Inc: Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Bayer: Honoraria, Research Funding; Pfizer: Honoraria; Baxalta: Honoraria, Research Funding. Mosnier:Hematherix Inc: Equity Ownership, Membership on an entity's Board of Directors or advisory committees.

Towards Restoring Coagulation in Hemophilia By Antibody-Mediated Inhibition of Antithrombin

Currently hemophilia A is mainly treated with factor VIII (FVIII). However, 20-25% of the hemophiliacs develop inhibiting antibodies (inhibitors) against FVIII. Hemophiliacs with inhibitors are treated with high dose FVIII or with therapies that bypass the need for FVIII such as activated FVII (rFVIIa, NovoSeven) or activated prothrombin concentrate. These have some disadvantages including rapid clearance, price and a risk of thrombo-embolism.In hemophilia, the balance between coagulation and anticoagulation is shifted towards the latter. This balance can be restored by administration of clotting factors but also by reducing anticoagulation mechanisms. Antithrombin (AT) constitutes a main anticoagulant mechanism as it inhibits thrombin, FXa and other activated clotting factors. In silico studies as well as observations that AT deficiency reduces bleeding in hemophilic mice and humans strongly support the concept that lowering functional AT levels bypasses the need for FVIII to generate thrombin. A Phase 1 clinical trial by Alnylam Pharmaceuticals showed that lowering AT activity by RNA-interference can boost thrombin generation in hemophilia. However, RNA-interference is an experimental therapy with unknown long term effects and is not suitable for acute treatment of bleeding as it takes weeks to become effective.We hypothesize that monoclonal antibodies (mAbs) that block AT activity can be used as FVIII bypass therapy. Such a mAb-based therapy is suitable for on demand and for prophylactic therapy, is relatively cheap and does not require administration of activated clotting factors.We have generated a panel of mouse anti-AT mAbs exhibiting varying binding characteristics to human AT. To assess the inhibitory activity of anti-AT mAbs, FXa activity assay was performed after FXa was incubated with normal plasma supplemented with the different anti-AT mAbs. Five anti-AT mAbs were identified that inhibit AT activity at around equimolar ratio. Inhibitory anti-AT mAbs were able to dose-dependently enhance thrombin generation in normal and hemophilic plasma. Heparin caused a marked prolongation of aPTT of normal and FVIII-deficient plasma, which was almost completely normalized by addition of the inhibitory mAb anti-AT 5G1.1, but not by anti-AT 6E7.2, indicating that these mAbs exert their inhibitory activity via different mechanisms.These data provide initial in vitro proof-of-concept that inhibitory anti-AT mAbs can be used to improve coagulation and may be used as FVIII bypass therapy. DisclosuresBoross:Prothix B.V.: Employment. Hack:Prothix B.V.: Equity Ownership, Patents & Royalties.

Characterization of a High-Affinity Fully Human IgG2 Antibody Against Tissue Factor Pathway Inhibitor As a Bypass Agent for the Treatment of Hemophilia

The tissue factor initiated coagulation pathway is intact in patients with hemophilia; however, this pathway is normally inhibited by tissue factor pathway inhibitor (TFPI), resulting in insufficient thrombin generation to stop bleeding and prevent recurrence of bleeding. Antagonizing this natural inhibition through anti-TFPI antibodies is a potential mechanism to effectively restore hemostasis in patients with hemophilia.A human phage-displayed antibody library was used to identify optimal anti-TFPI antibodies. From the panning campaign, unique antibodies were identified that bound to human TFPI and were cross-reactive to murine TFPI. In vitro characterization demonstrated that 6 of the antibodies could block TFPI activity, partially or completely restoring factor Xa activity. These 6 antibodies also shortened clotting time of hemophilic human (Hem A) plasma in a diluted prothrombin assay. When the ROTEM assay (Tem International GmbH, Basel, Switzerland) was used to measure clotting in human Hem A plasma from congenitally deficient individuals (factor VIII [FVIII] levels <0.38%), anti-TFPI antibodies significantly shortened clotting time, alone or in combination with FVIII or activated factor VII. In addition, the antibodies were effective in shortening clotting time in human blood containing anti-FVIII antibodies, indicating that anti-TFPI antibodies can be applied to inhibitor patients. One of the identified anti-TFPI antibodies was tested in a hemophilia A mice tail vein transection model and improved survival rate to 60%, significantly higher than the 10% survival rate of mice treated with an isotype control antibody. Additionally, combined with low-dose FVIII, this antibody could further increase the survival rate in hemophilic mice, and it was selected for further optimization.Optimization covered 2 different aspects: (1) affinity maturation to increase its affinity to human and murine TFPI; and (2) sequence optimization to reduce sequence deviation from germ-line sequences and to remove critical residues prone to undergo unwanted chemical modifications during production and/or storage. Both processes consisted of 2 rounds of sequence optimization and screening. Following affinity maturation, target affinity was increased by >200-fold on human and >500-fold on murine TFPI. Sequence optimization was performed on the backbone of the final affinity-matured variant. BAY 1093884, the final optimized variant, is a fully human IgG2 antibody with <10 pM binding affinity to human and murine TFPI. DisclosuresBauzon:Bayer: Employment. Liu:Bayer: Employment. Grudzinska-Goebel:Bayer: Employment. Tebbe:Bayer: Employment. Pan:Bayer: Employment. Cifrese:Bayer: Employment. Wang:Bayer: Employment. Sim:Bayer HealthCare LLC: Employment.

Immune Tolerance Developed in Platelet-Targeted FVIII Gene Therapy in Hemophilia Mice Is CD4+ T Cell-Mediated

Immune response to factor VIII (FVIII) is not only a severe complication in protein replacement therapy, but also a major concern in gene therapy of hemophilia A. Our previous studies have demonstrated that platelet-targeted FVIII (2bF8) gene therapy together with in vivo drug-selection of transduced cells can not only rescue the bleeding diathesis but also induce anti-FVIII specific immune tolerance in FVIIInull mice. In the current study, we investigated 1) whether our non-selectable lentiviral vector (LV) for the induction of platelet-FVIII expression is sufficient to induce immune tolerance and 2) which cell compartment is tolerized after platelet gene therapy. Platelet-specific FVIII expression was introduced by 2bF8LV-transduction of hematopoietic stem cells followed by syngeneic transplantation into FVIIInull mice preconditioned with 660 cGy total body irradiation (TBI) or Busulfan (Bu) plus ATG (anti-thymocyte globulin). After bone marrow transplantation and reconstitution, animals were analyzed by PCR, qPCR, FVIII:C assay, and tail clipping test to confirm the success of 2bF8 gene therapy. Sixteen weeks after transplantation, animals were challenged with recombinant human FVIII (rhF8) via retro-orbital venous administration at a dose of 50 U/kg weekly for 4 weeks. The titers of anti-FVIII inhibitory antibodies (inhibitors) were determined by Bethesda assay. The CFSE-labeled CD4 T cell proliferation assay and ELISPOT-based memory B cell maturation assay were used to determine which cell compartment is tolerized to FVIII after 2bF8 gene therapy. The level of platelet-FVIII expression was 1.44 ± 0.39 mU/108 platelets (n = 6) in the 660 cGy group, which is not significantly different from the level obtained from the Bu+ATG group [3.04 ± 1.19 mU/108 platelets (n = 6)]. Even after rhF8 challenge, no antibodies were detected in 2bF8LV-transduced recipients in either group. In contrast, all animals in the control group that did not undergo gene therapy developed various levels of inhibitors (204±97 BU/ml, n=7). The frequency of regulatory T cells in both 660 cGy TBI (11.01±0.52%) and Bu+ATG (11.02±0.68%) groups were significantly higher than the control group (8.05±0.57%). T cell proliferation assay demonstrated that CD4+ T cells from 2bF8 LV-transduced recipients that had been challenged with rhF8 did not proliferate when restimulated with rhF8 in vitro. The daughter CD4+ T cells in the group with 10 U/ml of rhF8 were 5.84 ± 2.49% (n = 6), which was not significantly different from the control group without no rhF8 stimulation (0 U/ml) (5.33 ± 1.72%). CD4+ T cells from primed FVIIInull mice did proliferate after rhF8 restimulation. The proliferated daughter cell was 13.12 ± 6.76% (n = 7) in the group with rhF8 (10 U/ml) re-stimulation, which is significantly higher than the group without rhF8 co-culture (4.99 ± 1.16%). Since FVIII-specific memory B cell maturation is CD4+ T cell dependent, we isolated CD4+ T and memory B cells from 2bF8LV-transduced or FVIIInull mice after rhF8 immunization and co-cultured with rhF8 followed by ELISPOT assay to examine the antibody secreting cells. No spots were detected when memory B cells from rhF8-primed FVIIInull mice were co-cultured with CD4+ T cells from 2bF8LV-transduced recipients. In contrast, when memory B cells from either rhF8 immunized 2bF8LV-transduced or untreated FVIIInull mice were cultured with CD4+ T cells from rhF8-primed FVIIInull mice, there were 142 and 205 anti-FVIII antibody secreting cells, respectively, detected per 106 cells seeded. These results indicate that CD4+ T cells from 2bF8LV-transduced mice are tolerized to rhF8 stimulation. In conclusion, 2bF8 lentiviral gene transfer without in vivo selection of genetically manipulated cells can introduce FVIII-specific immune tolerance in hemophilia A mice and this immune tolerance is CD4+ T cell-mediated. DisclosuresBaumgartner:Novo Nordisk: Research Funding. Shi:BloodCenter of Wisconsin: Patents & Royalties: METHOD OF INDUCING IMMUNE TOLERANCE THROUGH TARGETTED GENE EXPRESSION..

Human Factor VIII Expression and Normalization of Bleeding Following AAV Gene Therapy in a Double Knockout Mouse Model of Hemophilia

Hemophilia A is an X-linked bleeding disorder caused by mutations in the gene encoding the Factor VIII coagulation protein (FVIII). Bleeding episodes in patients are reduced by prophylactic therapy or treated acutely using recombinant or plasma derived FVIII. Recently, Nathwani et al demonstrated in preclinical and clinical studies sustained expression of coagulation factor IX using AAV8 technology to deliver the human gene to the liver, driven by a liver specific promoter. The same group demonstrated FVIII expression in mice and primates using a modified B-domain truncated form of FVIII delivered in an AAV8 capsid.We have made an AAV5 construct containing a B-domain deleted FVIII gene (AAV5-SQ) with a liver specific promoter and evaluated it in a double knockout mouse model of hemophilia.The double knockout mice (DKO) were created by crossing factor VIII deficient mice with RAG2 deficient mice (RAG2 KO). The RAG2 KO mice lacked the ability to mount an adaptive immune response thereby allowing sustained expression of a human protein without the development of an antibody response.Eight week old male DKOs were randomly distributed into three groups, twenty per group, and treated via a single IV injection with either vehicle, or AAV5-SQ at 2 x 1013 or 1 x 1014 vg/kg. C57BL/6J mice comprised a fourth group and were treated with vehicle via a single IV injection to demonstrate wild type bleeding times and blood loss. Bleeding times and blood loss were assessed in these mice 8 weeks post-dose, at 16 weeks of age. In addition, forty 16 week old DKO mice were randomly divided into two groups and treated with a single IV injection of rhFVIII-SQ protein (rhSQ, Xyntha®) at either 50 or 200 IU/kg. Bleeding times were assessed in these mice 30 minutes post-dose, at 16 weeks of age.Eight weeks post dosing with either AAV5-SQ or vehicle, the tail bleeding time and blood loss were measured following transection of the tip of the tail for evaluation of the functional efficacy of AAV5-SQ gene therapy. Wild-type mice receiving vehicle had a mean of 0.040 ± 0.073 g blood loss and 5.11 ± 5.61 min bleeding time. DKO mice treated with vehicle had a mean blood loss and bleeding time of 0.741 ± 0.207 g and 28.96 ± 1.40 min, respectively. Mice receiving AAV5-SQ at 2x1013 vg/kg showed significantly reduced blood loss (0.387 ± 0.384 g, p=0.0008 vs DKO+ vehicle; p=0.0003 vs WT) and bleeding time (17.12 ± 11.58 min, p=0.00005 vs DKO+ vehicle; p=0.0013 vs WT) while 1x1014 vg/kg AAV5-SQ treatment corrected blood loss and bleeding times to wild-type levels (0.104 ± 0.203 g [p=0.192 vs WT, p= 5.49x10-12 vs DKO + vehicle] and 5.58 ± 9.32 mins [p=0.847 vs WT, p= 1.75x-13 vs DKO + vehicle], respectively). The effect of AAV5-SQ treatment on blood loss and bleeding time was comparable to the effects of rhSQ. DKO mice receiving 50 IU/kg of rhSQ had a mean blood loss and bleeding time of 0.492 ± 0.297 g and 18.14 ± 9.39 min, respectively, which was not significantly different from mice receiving AAV5-SQ at 2x1013 (p=0.343 for blood loss, p=0.760 for bleeding time). DKO mice receiving 200 IU/kg of rhSQ had a mean blood loss and bleeding time of 0.134 ± 0.191 g and 4.29 ± 6.16 min, respectively, which was not significantly different from mice receiving AAV5-SQ at 1x1014 (p=0.635 for blood loss, p=0.608 for bleeding time).In a separate experiment, 4 groups of DKO mice, n=10 per group, were injected with either vehicle, AAV5-SQ at 2x1013, AAV5-SQ at 2x1014 vg/kg or rhSQ at 50 IU/kg. Blood was collected 8 weeks after AAV5-SQ treatment or 5 and 30 min after rhSQ for evaluation of plasma hFVIII-SQ protein levels and activity. Expressed hFVIII-SQ levels were measured by electrochemiluminescence assay. Factor VIII-SQ protein levels at 2x1013 vg/kg AAV5-SQwere 46.8±44.0 ng/ml and 355±166ng/ml at 2x1014 vg/kg. At 50 IU/kg of rhSQ the plasma protein levels were 79.1±11.3 ng/ml at 5 min and 44.7±16.6 ng/ml at 30min post dosing. Western blot analysis of the plasma from these mice showed the expressed protein to be similar in size to rhSQ.In summary, AAV5-SQ injected into DKO hemophilic mice resulted in a dose dependent expression of B-domain deleted FVIII protein and a corresponding correction of bleeding time and blood loss. At the highest dose tested complete correction was achieved. Similar corrections in bleeding were observed at approximately the same plasma levels of FVIII protein produced either endogenously by AAV5-SQ or following exogenous administration of B-domain deleted FVIII. DisclosuresBunting:BioMarin Pharmaceutical: Employment. Zhang:BioMarin Pharmaceutical: Employment. Xie:BioMarin Pharmaceutical: Employment. Bullens:BioMarin Pharmaceutical: Employment. Mahimkar:BioMarin Pharmaceutical: Employment. Fong:BioMarin Pharmaceutical: Employment. Sandza:BioMarin Pharmaceutical: Employment. Colosi:BioMarin Pharmaceutical: Employment. Long:BioMarin Pharmaceutical: Employment. Vehar:BioMarin Pharmaceutical: Employment. Carter:BioMarin Pharmaceutical: Employment.

IDO1 suppresses inhibitor development in hemophilia A treated with factor VIII.

The development of inhibitory antibodies to factor VIII (FVIII) is a major obstacle in using this clotting factor to treat individuals with hemophilia A. Patients with a congenital absence of FVIII do not develop central tolerance to FVIII, and therefore, any control of their FVIII-reactive lymphocytes relies upon peripheral tolerance mechanisms. Indoleamine 2,3-dioxygenase 1 (IDO1) is a key regulatory enzyme that supports Treg function and peripheral tolerance in adult life. Here, we investigated the association between IDO1 competence and inhibitor status by evaluating hemophilia A patients harboring F8-null mutations that were either inhibitor negative (n = 50) or positive (n = 50). We analyzed IDO1 induction, expression, and function for any relationship with inhibitor occurrence by multivariable logistic regression and determined that defective TLR9-mediated activation of IDO1 induction is associated with an inhibitor-positive status. Evaluation of experimental hemophilic mouse models with or without functional IDO1 revealed that tryptophan metabolites, which result from IDO1 activity, prevent generation of anti-FVIII antibodies. Moreover, treatment of hemophilic animals with a TLR9 agonist suppressed FVIII-specific B cells by a mechanism that involves IDO1-dependent induction of Tregs. Together, these findings indicate that strategies aimed at improving IDO1 function should be further explored for preventing or eradicating inhibitors to therapeutically administered FVIII protein.

Silencing of protease-activated receptors attenuates synovitis and cartilage damage following a joint bleed in haemophilic mice.

Joint bleeding results in blood-induced arthropathy. We investigate whether a joint bleed alters protease-activated receptor (PAR) expression, and whether treatment with small interfering RNA (siRNA) targeted against PAR1-4 attenuates synovitis and cartilage damage. Protease-activated receptor expression was evaluated upon a joint bleed in haemophilic mice and in humans. In addition, mice with a joint bleed were randomized between treatment with PAR1-4 siRNA or control and evaluated for the presence of synovitis and cartilage damage. Also, human cartilage was transfected with PAR1-4 siRNA or control, and evaluated for plasmin-induced cartilage damage. Following a joint bleed, we observed an increase in synovial PAR1, -2 and -4 expression, and an increase in chondrocyte PAR2 and -3 expression in mice (all P < 0.05). Also an increase in synovial PAR1 and chondrocyte PAR4 expression in patients was observed (both P < 0.05). Treatment of a joint bleed in haemophilic mice with PAR1-4 siRNA attenuates synovitis and cartilage damage (both P < 0.01). Treatment of human cartilage tissue explants with PAR1-4 siRNA reduced plasmin-induced cartilage damage (P < 0.01). This study demonstrates that synovial and chondrocyte PAR expression is altered upon a joint bleed, and that treatment with PAR1-4 siRNA attenuates synovitis and plasmin-induced cartilage damage.

Hemostatic agents of broad applicability produced by selective tuning of factor Xa zymogenicity

AbstractThere is a clinical need to develop safe therapeutic strategies to mitigate bleeding. Previously, we found that a novel zymogen-like factor Xa variant (FXa-I16L) was effective in correcting the coagulation defect in hemophilic mice. Here we expand the mutational framework to tune the FX(a) zymogen-like state. Alteration of FXa zymogenicity yields variants (V17M, I16L, I16M, V17T, V17S, and I16T) with a wide range (≤1000-fold) of reduced function toward physiologic substrates and inhibitors. The extent of zymogen-like character, including resistance to antithrombin III, correlates well with plasma half-life (<2 minutes to >4 hours). Importantly, biologic function, including that of the most zymogen-like variant (FXa-I16T), was greatly enhanced when bound to FVa membranes. This resulted in improvement of clotting times and thrombin generation in hemophilic plasma. The FXa variants were remarkably effective in mouse injury models. In these systems, the data show that the more active the protease, the more difficult it is to overcome the protective mechanism of circulating inhibitors to achieve a therapeutic benefit. Depending on the treatment situation, the more zymogen-like variants (V17S and I16T) were most useful when given before injury whereas variants exhibiting greater activity but shorter half-lives (I16L and I16M) were most effective when administered after injury. This new class of FXa variants provides a useful and flexible platform for selectively bioengineering biologic function and half-life to target different clinical bleeding scenarios.

479. ZFN-Mediated In Vivo Genome Editing Results in Supraphysiological Levels of Lysosomal Enzymes Deficient in Hunter and Hurler Syndrome and Gaucher Disease

Lysosomal storage diseases (LSDs) represent a group of inherited metabolic disorders associated with mutations in genes encoding lysosomal enzymes leading to systemic accumulation of toxic storage materials. Manifestations include organomegaly, skeletal dysplasias, cardiopulmonary obstruction, and severe neurologic impairment, often leading to death by age 10. Current LSD therapies include enzyme replacement therapy, substrate reduction therapy and hematopoietic stem cell transplant (HSCT). However, all of these therapies are expensive, incompletely effective, and HSCT is associated with significant risk of morbidity and mortality. Due to the unmet need genome editing strategies are proposed to permanently modify patient cells by genetically complementing the LSD defect. This is achieved by utilizing engineered zinc finger nucleases (ZFNs) to introduce a DNA cut at the target locus, mediating integration of a therapeutic LSD donor cDNA. To ensure long-term expression of the transgenes in vivo we target the albumin locus as a “safe harbor” in hepatocytes via co-delivery of the albumin ZFNs and LSD donors by adeno-associated virus (AAV). This system exploits the high transcriptional activity of the native albumin enhancer/promoter; uses stably modified hepatocytes to potentially allow long-term expression of the inserted transgene; and utilizes an endogenous promoter obviating this requirement in the AAV payload. We have previously exploited AAV-mediated in vivo targeting of the murine albumin “safe harbor” locus for the synthesis of therapeutic levels of FVIII and FIX to overcome the clotting defect in hemophilic mice. Using the same approach, we co-delivered mouse albumin ZFNs with donor constructs encoding either human iduronate-2-sulfatase (IDS) deficient in Hunter syndrome, α-L-iduronidase (IDUA) for Hurler syndrome, or Glucocerebrosidase (GBA) for Gaucher disease via AAV in WT mice. We show stable integration of the LSD donors at the albumin locus, resulting in liver-specific expression and secretion of these proteins into plasma. This led to a 4-fold (GBA), 10-fold (IDUA) or 100-fold (IDS) increase in enzymatic activity in the plasma, demonstrating that the secreted proteins are functional. Importantly, increased activity was also detected in secondary tissues (spleen) showing efficient uptake and activity in distal tissues. Moreover, preliminary data in MPSI and MPSII mice suggests these levels of IDS and IDUA expression may lead to correction of the enzyme deficiency. IDS, IDUA and GBA expression remained stable throughout the study (up to 2 months), while expression of FIX has been stable for >1 yr suggesting this process results in long-term protein expression. In summary, our data provide proof of concept for ZFN-mediated targeting of the albumin locus in hepatocytes as an in vivo protein replacement platform to express different proteins associated with lysosomal storage diseases.

Extrahepatic sources of factor VIII potentially contribute to the coagulation cascade correcting the bleeding phenotype of mice with hemophilia A.

A large fraction of factor VIII in blood originates from liver sinusoidal endothelial cells although extrahepatic sources also contribute to plasma factor VIII levels. Identification of cell-types other than endothelial cells with the capacity to synthesize and release factor VIII will be helpful for therapeutic approaches in hemophilia A. Recent cell therapy and bone marrow transplantation studies indicated that Küpffer cells, monocytes and mesenchymal stromal cells could synthesize factor VIII in sufficient amount to ameliorate the bleeding phenotype in hemophilic mice. To further establish the role of blood cells in expressing factor VIII, we studied various types of mouse and human hematopoietic cells. We identified factor VIII in cells isolated from peripheral and cord blood, as well as bone marrow. Co-staining for cell type-specific markers verified that factor VIII was expressed in monocytes, macrophages and megakaryocytes. We additionally verified that factor VIII was expressed in liver sinusoidal endothelial cells and endothelial cells elsewhere, e.g., in the spleen, lungs and kidneys. Factor VIII was well expressed in sinusoidal endothelial cells and Küpffer cells isolated from human liver, whereas by comparison isolated human hepatocytes expressed factor VIII at very low levels. After transplantation of CD34(+) human cord blood cells into NOD/SCIDγNull-hemophilia A mice, fluorescence activated cell sorting of peripheral blood showed >40% donor cells engrafted in the majority of mice. In these animals, plasma factor VIII activity 12 weeks after cell transplantation was up to 5% and nine of 12 mice survived after a tail clip-assay. In conclusion, hematopoietic cells, in addition to endothelial cells, express and secrete factor VIII: this information should offer further opportunities for understanding mechanisms of factor VIII synthesis and replenishment.

Plant-based oral tolerance to hemophilia therapy employs a complex immune regulatory response including LAP+CD4+ T cells

AbstractCoagulation factor replacement therapy for the X-linked bleeding disorder hemophilia is severely complicated by antibody (“inhibitor”) formation. We previously found that oral delivery to hemophilic mice of cholera toxin B subunit-coagulation factor fusion proteins expressed in chloroplasts of transgenic plants suppressed inhibitor formation directed against factors VIII and IX and anaphylaxis against factor IX (FIX). This observation and the relatively high concentration of antigen in the chloroplasts prompted us to evaluate the underlying tolerance mechanisms. The combination of oral delivery of bioencapsulated FIX and intravenous replacement therapy induced a complex, interleukin-10 (IL-10)–dependent, antigen-specific systemic immune suppression of pathogenic antibody formation (immunoglobulin [Ig] 1/inhibitors, IgE) in hemophilia B mice. Tolerance induction was also successful in preimmune mice but required prolonged oral delivery once replacement therapy was resumed. Orally delivered antigen, initially targeted to epithelial cells, was taken up by dendritic cells throughout the small intestine and additionally by F4/80+ cells in the duodenum. Consistent with the immunomodulatory responses, frequencies of tolerogenic CD103+ and plasmacytoid dendritic cells were increased. Ultimately, latency-associated peptide expressing CD4+ regulatory T cells (CD4+CD25−LAP+ cells with upregulated IL-10 and transforming growth factor-β (TGF-β) expression) as well as conventional CD4+CD25+ regulatory T cells systemically suppressed anti-FIX responses.

Binding of Factor VIII to Lipid Nanodiscs Increases its Clotting Function in a Mouse Model of Hemophilia A.

BackgroundHemophilia A is a congenital bleeding disorder caused by defective or deficient factor VIII (FVIII). The active form of FVIII is the co-factor for the serine protease factor IXa (FIXa) in the membrane-bound intrinsic tenase (FVIIIa-FIXa) complex. The assembly of the FVIIIa-FIXa complex on the activated platelet surface is critical for successful blood clotting.ObjectivesTo characterize the role of lipid nanodiscs (ND) for on FVIII function in vivo and test the lipid ND as a delivery system for FVIII. To evaluate the potential of binding recombinant FVIII to ND as improved treatment for Hemophilia A.MethodsRecombinant porcine FVIII (rpFVIII) was expressed and characterized in solution, and when bound to ND. The rpFVIII, ND and rpFVIII-ND complexes were characterized via transmission electron microscopy. Functional studies were carried out using aPTT tests and time resolved tail snip studies of hemophilic mice.ResultsFunctional rpFVIII was successfully assembled on lipid ND. When injected in hemophilic mice, the rpFVIII-ND complexes showed a pronounced pro-coagulant effect, which was stronger than that of rpFVIII alone. While injection of the ND alone showed a pro-coagulant effect this effect was not additive, implying that the rpFVIII-ND complexes have a synergistic effect on the clotting process in hemophilic mice.ConclusionsBinding of rpFVIII to ND prior to its injection in hemophilic mice significantly improves the therapeutic function of the protein. This represents a meaningful step towards a new approach to modulate blood coagulation at the membrane-bound FVIII level and the assembly of the intrinsic tenase complex.

Hitting the Target Without Pulling the Trigger

Targeted genomic integration typically requires designer nucleases that augment gene targeting efficiencies by creating a double-strand DNA (dsDNA) break at the target locus. This break provides the “molecular trigger” necessary to boost the efficiency of homologous recombination by several orders of magnitude. Eliminating the nuclease once it has done its job is highly desirable but remains a challenge. In a recent issue of Nature, Barzel et al. described a promising nuclease-free method for gene targeting by homologous recombination with unexpected high efficiency.1 Promoterless adeno-associated virus (AAV) vectors were designed to specifically target the albumin locus, which is highly expressed in the liver. Consequently, targeted genomic integration of a coagulation factor IX (FIX) complementary DNA into the albumin locus directed sustained FIX expression at therapeutic levels and corrected the bleeding diathesis in hemophilic mice. This approach may ultimately pave the way toward a safer and efficacious liver-directed gene therapy that minimizes the risk of insertional oncogenesis and off-target effects. This is particularly relevant for diseases that manifest themselves in infants and require early treatment by gene therapy.