Antiviral or proviral action of mycophenolic acid in hepatitis B infection?

Mycophenolic acid (MPA) is the activated form of the prodrug mycophenolate mofetil. It is commonly used in kidney and liver transplantation. In addition, MPA has been demonstrated to have a broad spectrum of antiviral activity in vitro against numerous DNA and RNA viruses, including hepatitis B virus (HBV) and hepatitis C virus.1-3 In contrast, a recent study has reported that MPA surprisingly enhanced HBV replication in cell culture models.4 These findings could spark a clinical debate regarding the use of mycophenolate mofetil in organ transplantation. To gain more insight to this controversy, we evaluated the effects of MPA on HBV in HepG2.2.15 cells, a widely used model that was also used by Hoppe-Seyler et al.4 However, we used the COBAS TaqMan 48 real-time polymerase chain reaction (PCR) system (Roche) for HBV titer quantification. Unexpectedly, 1 μg/mL of MPA treatment significantly increased HBV titers by 70 ± 10% (P < 0.01). In contrast, treatment with 5 and 10 μg/mL MPA resulted in trends of decreasing virus titers, though not significantly (Fig. 1A). For such a relative long-term treatment, 5-10 μg/mL MPA could exert clear antiproliferative effects.2, 3 Consistently, no significant effect on total RNA content was observed with 1 μg/mL MPA treatment, whereas significant reduction was observed in the higher dose groups (P < 0.01) (Fig. 1B). To discount the possibility that differences in cell proliferation compounded our interpretation of the results, we further evaluated viral gene (hepatitis B surface antigen [HBsAg]) expression normalized to household host genes. As shown in Fig. 1C, all three doses of MPA significantly enhanced HBsAg expression by 7.7- to 15.7-fold (P < 0.01). The effects of MPA on HBV titer and viral gene expression. (A) Comparable with previous studies,1, 2, 4 HepG2.2.15 cells were treated with 1, 5, and 10 μg/mL (clinically achievable doses) of MPA for 5 days. After the first 3 days of treatment, culture medium was refreshed and maintained for an additional 2 days of treatment. Supernatant collected from the later period of treatment was subjected to HBV DNA quantification using the COBAS TaqMan 48 real-time PCR system (Roche). Data are presented as 106 IU/mL for three experiments. (B) Quantification of total RNA in cells harvested after MPA treatment using the NanoDrop 2000 spectrophotometer (Thermo Scientific). (C) Quantitative real-time PCR quantification of gene expression of HBsAg normalized to household host genes glucose-6-phosphate dehydrogenase and albumin expression. Data are presented as fold changes. *P < 0.01 (heteroscedastic double-sided Student t test) Our findings strongly suggest that MPA may have proviral effects, at least in the context of HBV infection models and thus in agreement with the observations of Hoppe-Seyler et al.4 It was also highlighted that MPA-dependent p38 mitogen-activated protein kinase activation drives HBV replication.4 Conversely, its antiviral activity has been attributed to depletion of nucleotide pools by inhibiting its canonical target IMPDH or the induction of antiviral interferon-stimulated genes.3 Thus a picture emerges in which MPA exerts kaleidoscopic effects through multiple mechanisms, which are either antiviral or proviral, both depending on the type as well as the context of viral infection. Given its huge clinical and fundamental importance, a comprehensive description of MPA effects in viral pathology is urgently required. Qiuwei Pan*, Anneke J. van Vuuren*, Luc J.W. van der Laan*, Maikel P. Peppelenbosch*, Harry L. A. Janssen*, * Department of Gastroenterology and Hepatology, Erasmus MC-University Medical Center, Rotterdam, The Netherlands.