Non‐invasive viral gene transfer of factor IX to colonic epithelial cells in hemophilia B mice

Gene therapy is potentially an ideal therapeutic strategy for the treatment of hemophilia B. In recent studies, hepatocytes, skeletal muscle, endothelial cells, bone marrow stroma cells, as well as other cell types, have been targeted for gene transfer [1– 3]. However, the routine routes of administration of viral vectors include intravenous infusion, intramuscular injection and hypodermic embedment [4–6], which are invasive and carry the risk of both bleeding and infection. For this reason, a potentially non-invasive strategy may be to target colonepithelial cells via an enema to deliver a recombinant adenoassociated viral vector serotype 2 (rAAV2) that encodes human factor (F) IX (hFIX). To this endwe selected the gut epithelium as a target cell hFIX gene transfer, and analyzed the effects of this both in vitro and in vivo. Human and murine gut epithelial cell lines (human colon carcinoma SW480 cell line/mouse colon cancer C26 cell line) were transduced in vitro with rAAV2/hFIX. The expression and biological activity of hFIX were studied by ELISA, and one-stage clotting assay, respectively. Elevated levels of hFIX protein and activity were detected, which persisted for 21 days post-infection. The maximum hFIX (±SD) from infected SW480 cells was 98 (±4.34) ng 10 cell 24 h, whereas for C26 cells these levels were 78 (±5.12) ng 10 cell 24 h. The procoagulant hFIX activity was in accordance with the levels of hFIX protein. Using a one-stage assay, the maximum FIX values (±SD) were 21.23 (±3.21)% and 17.23 (±2.29)%, respectively, in comparison to FIX activity in normal human plasma. After transduction, conditioned media from both the SW480 and C26 cells showed markedly higher procoagulant activity than that from non-infected control cells. These results demonstrated that both the SW480 and C26 cells were able to produce active hFIX protein after transduction with the rAAV2/hFIX viral vector. We next investigated whether biologically active hFIX could be expressed in vivo following administration of rAAV2/hFIX by enema. Normal Kunmingmice and hemophilia Bmice were given a 1 · 10 vg kg dose of rAAV2/hFIX by enema, and thereafter the expression levels of hFIX protein were analyzed. On the 7th day post-infection, hFIX protein was detected both in the plasma and in the intestinal juice of normal Kunming mice (n = 16). Maximum plasma levels (23.06 ± 1.68 ng mL) were detected 21 days after infection, and decreased to baseline levels by the 35th day. Similarly, hFIX protein was detected in the plasma of hemophilia B mice treated with rAAV2/hFIX on the 14th [9.19 (±0.70) ng mL] and 21st [12.98 (±1.16) ng mL] day after viral administration (n = 4) (Fig. 1). On the 21st day after the enema, the activity of hFIX protein was 2.79 (±1.03)% in hemophilia B mice treated with rAAV2/hFIX, and 0.10 (±0.09)% in control mice treated with sodium chloride. The activated partial thromboplastin time (±SD) for these mice was 85.1 (±1.7) s and 92.6 (±2.3) s (n = 3), respectively. In addition, the bleeding volumes in 5 minutes of rAAV2/hFIX transduced hemophilia B mice [84 (±15)lL] (n = 3) were decreased after tail-cut, in comparison to non-transduced control mice [106 (±17) lL] (n = 3). The tail bleed volume for normal Kunmingmicewas 53(±12) lL (n = 3). In conclusion, hemophilia B mice express biologically active hFIX after administration of rAAV2/hFIX by enema. The expression of antibodies against rAAV2 and hFIX was analyzed by ELISA and a modified Bethesda assay. These results showed that in Kunming mice, antibodies against rAAV2 could be detected on the 14th day post-enema, and reached a maximum value (10.05 ± 0.80 BU mL) (n = 16) on the 21st day, then began to decrease. In separate experiments, circulating antibodies against hFIX were detected in the plasma of transduced Kunming mice on the 7th day, which increased rapidly and reached a maximum value (19.3 ± 1.8 BU mL) (n = 16) on the 14th day, and thereafter decreased. DNA analysis, as well as histochemical and histopathological analyses, was performed to evaluate the safety of rAAV2/ Correspondence: Fangping Chen, Department of Haematology, Xiangya Hospital, Central South University, Changsha 410008, China. Tel.: + 86 731 4327330; fax: + 86 731 4327332. E-mail: xychenfp@public.cs.hn.cn