31. Creation of a Hemophilia B Mouse Expressing Defective Human Factor IX

Abstract

The majority of cases of human hemophilia B are the result of missense mutations in the coagulation factor IX gene and defective circulating factor IX is detectable in most patients. The available mouse factor IX knockout models of hemophilia B (FIXKO mouse) reproduce the bleeding phenotype of human hemophilia B, but because they produce no factor IX they fail to reproduce the dominant human phenotype. In addition, following exposure to factor IX via a number of routes (e.g. intravenous factor IX protein treatment; intramuscular factor IX gene therapy), the FIXKO mice generate anti-factor IX antibodies that inhibit coagulation. This is in contrast to the observation that inhibitor antibody formation in factor IX-deficient humans treated with purified factor IX is unusual (~2–5% of patients). The generation of inhibitor antibodies complicates studies of safety and long-term expression following gene therapy. We have created a human factor IX mouse model of hemophilia B (R333Q-hFIX mouse) by homologous recombination in embryonic stem cells. The mouse expresses no mouse factor IX, but instead expresses a missense mutant human factor IX from the mouse FIX promoter. Mutant human factor IX mRNA transcript and circulating human factor IX are detectable throughout development, but factor IX activity is 3% functional factor IX (as measured by aPTT clotting times) was observed without the development of inhibitor antibodies. In contrast, given the same treatment, FIXKO mice developed inhibitor antibodies (range 2.5–34.4 Bethesda Inhibitor Units) and clotting function was unimproved. ELISAs measuring specific IgG subclass antibodies against human factor IX confirmed the absence of antibody response in the R333Q-hFIX mice. Mice with the missense mutation R333Q had no anti-factor IX IgG either before or after factor IX gene thereapy (four months serial testing), whereas the knockout mice developed antibodies predominantly of the IgG1 subclass. An additional cohort of mice were treated with an eight times higher dose of AAV2 human factor IX gene therapy and the R333Q-hFIX mice again tolerated the human factor IX, while FIXKO mice developed inhibitor antibodies. Finally, the mice originally treated with the lower dose of AAV2 vector were re-challenged with an AAV serotype 1 vector containing the identical human factor IX expression cassette. Despite expression of normal human factor IX levels (>5 mcg/ml) R333Q-hFIX mice did not break tolerance. Finally, we describe the use of the monoclonal antibody A-1 to preferentially detect human factor IX expressed after gene therapy, but not the background missense human factor IX R333Q. These studies suggest that the presence of a functionally inactive human factor IX present throughout development results in relative immunologic non-responsiveness during adulthood to a factor IX gene therapy challenge in the R333Q mouse. Our R333Q-hFIX mice strain will complement the FIXKO mice for studying factor IX circulating kinetics, tolerance and gene therapy.