OBJECTIVE To optimize the viral persistence rate in a hydrodynamic injection (HI) based hepatitis B virus (HBV) transfection mouse model. METHODS (1) 5-6-wk-old male C3H/HeN and C57BL/6 mice were hydrodynamically injected with 10 μg endotoxin-free pAAV/HBV1.2 plasmid DNA via the tail vein. Hepatitis B surface antigen (HBsAg), hepatitis B e antigen (HBeAg) and HBV DNA, both in the serum and liver, were detected at different time points post HI by ELISA, immunohistochemical staining or quantitative polymerase chain reaction (PCR); (2) male C3H/HeN and C57BL/6 mice, either hydrodynamically injected mice at 10 wk post HI or naïve mice, were all immunized subcutaneously with 5 μg HBsAg formulated in complete Freund's adjuvant three times at a 2-wk interval. Two weeks after the final immunization, splenocytes were isolated for T cell function analysis by ELISPOT assay; and (3) five weeks post HI, C3H/HeN mice were intragastrically administered 0.1 mg/kg entecavir once a day for 14 d, or were intraperitoneally injected with 1 mg/kg interferon (IFN)-α twice a week for 2 wk, or were treated with PBS as controls. The sera were collected and assayed for HBV DNA on days 0, 7 and 14 after drug treatment. RESULTS (1) Approximately 90% (22/25) of the injected C3H/HeN mice were still HBsAg-positive at 46 wk post HI, whereas HBsAg in C57BL/6 mice were completely cleared at 24 wk. Serum levels of HBeAg in C3H/HeN mice were higher than those in C57BL/6 mice from 4 wk to 46 wk. HBV DNA levels in the hydrodynamically injected C3H/HeN mice were higher than those in the C57BL/6 mice, both in the serum (from 4 wk to 46 wk) and in the liver (detected at 8 wk and 46 wk post HI). Histology showed that hepatitis B core antigen and HBsAg were expressed longer in the liver of C3H/HeN mice than in C57BL/6; (2) HBsAg specific T cell responses after HBsAg vaccination in hydrodynamically injected C3H/HeN and C57BL/6 mice, or naive control mice were detected by ELISPOT assay. After stimulation with HBsAg, the frequencies of IFN-γ producing splenocytes in the hydrodynamically injected C3H/HeN mice were significantly lower than those in hydrodynamically injected C57BL/6 mice, control C3H/HeN and control C57BL/6 mice, which were 0, 17 ± 7, 18 ± 10, and 41 ± 10 SFCs/10(6) splenocytes, respectively, and the mean spot sizes showed the same pattern. Even just stimulated with PMA and ionomysin, T-cell responses elicited in the vaccinated control C3H/HeN were much higher than those in hydrodynamically injected C3H/HeN mice; and (3) For drug treatment experiments on the hydrodynamically injected C3H/HeN mice, serum HBV DNA levels in the entecavir treatment group declined (131.2 folds, P < 0.01) on day 7 after treatment and kept going down. In the group of IFN-α treatment, serum HBV DNA levels declined to a lowest point (6.42 folds, P < 0.05) on 7 d after treatment and then rebounded. CONCLUSIONS We have developed a novel HI-based HBV transfection model using C3H/HeN mice, which had a higher HBV persistence rate than the classic C57BL/6 mouse model.