A novel in vivo siRNA delivery system specifically targeting liver cells for protection of ConA-induced fulminant hepatitis

Abstract

Background: Fulminant hepatitis progresses to acute liver failure (ALF) when the extent of hepatocyte death exceeds the liver's regenerative capacity. Although small interfering RNA (siRNA) appears promising in animal models of hepatitis, the approach is limited by drawbacks associated with systemic administration of siRNA. The aim of this study is to develop a hepatocyte-specific delivery system of siRNA for treatment of fulminant hepatitis.

Methodology/principal findings: Galactose-conjugated liposome nano-particles (Gal-LipoNP) bearing siRNA was prepared, and the particle size and zeta potential of Gal-LipoNP/siRNA complexes were measured. The distribution, cytotoxicity and gene silence efficiency were studied in vivo in a concanavalin A (ConA)-induced hepatitis model. C57BL/6 mice were treated with Gal-LipoNP Fas siRNA by i.v. injection 72 h before ConA challenge, and hepatocyte injury was evaluated using serum alanine transferase (ALT) and aspartate transaminase (AST) levels, as well as liver histopathology and TUNEL-positive hepatocytes. The galactose-ligated liposomes were capable of encapsulating >96% siRNA and exhibited a higher stability than naked siRNA in plasma. Hepatocyte-specific targeting was confirmed by in vivo delivery experiment, in which the majority of Gal-LipoNP-siRNA evaded nuclease digestion and accumulated in the liver as soon as 6 h after administration. In vivo gene silencing was significant in the liver after treatment of Gal-Lipo-siRNA. In the ConA-induced hepatitis model, serum levels of ALT and AST were significantly reduced in mice treated with Gal-lipoNP-siRNA as compared with control mice. Additionally, tissue histopathology and apoptosis showed an overall reduction of injury in the Gal-LipoNP siRNA-treated mice.

Conclusions/significance: This study is the first to our knowledge to demonstrate reduction of hepatic injury by liver-specific induction of RNA interference using Gal-LipoNP Fas siRNA, highlighting a novel RNAi-based therapeutic potential in many liver diseases.

Figure 1. Stability of galactosylated liposomes/siRNA complex in mouse plasma.

Cy3-labeled siRNA was complexed with galactosylated liposomes or naked siRNA and was subsequently incubated with fresh mouse plasma at 37°C for various time periods. After incubation for 0, 1, 3, 6,12, 24, and 48 h, siRNA from galactosylated liposomes was extracted with Triton X-100 and detected by polyacrylamide gel electrophoresis.

Figure 2. Distribution of liposomes/siRNA complex in vivo.

C56BL/6 mice were i.v. injected with naked Cy3-labeled siRNA (Naked siRNA), non-targeted liposomes (lipoNP/siRNA) and targeted liposomes with Cy3-labeled siRNA complex (Gal-LioNP/siRNA). Untreated mice served as naïve control. At 6 h, 24 h and 48 h following administration, mice were sacrificed and heart, liver, spleen, lung, kidney were collected. The remaining siRNAs in organs were detected by fluorescence microscope.

Figure 3. Silencing Fas in vivo.

(A&B) Effect of siRNA dose on gene silencing in mice. Gal-liposomes/Fas-siRNA complex was i.v. injected into C57BL/6 mice (n = 6, per group) at a siRNA dose from 0.125 nmol/g to 0.75 nmol/g. Gl2 siRNA and PBS treatments were used as negative controls. Gene silencing of Fas in liver was determined using quantitative RT-PCR (A) and western blot (B). (C) fluorescence intensity. Mice were treated with Cy3-labeled Gal-liposomes/Fas-siRNA complex as described in (A). The liver tissues were sectioned and the fluorescence was detected under microscope as described in Materials and Methods. (D&E) Dynamic observation of gene silencing. Gal-liposomes/Fas-siRNA complex was i.v. injected into C57BL/6 mice at a siRNA dose of 0.25 nmol/g (n = 6, per group). Gl2 siRNA and PBS treatments were used as negative controls (n = 6, per group). Gene silencing in liver in different time point was determined using quantitative RT-PCR (D) and western blot (E).

Figure 4. Non-specific inflammatory response after injection of siRNA/liposome complex.

(A) Liver IFN-γ, TNF-α and IL-6 mRNA levels after administration of galactosylated liposomes/siRNA complex in mice. Groups of three mice were i.v. injected with Gal-LipoNP complexed with Fas siRNA, Gl2 siRNA and PBS, respectively. At 4 h following administration, liver tissue were collected, IFN-γ, TNF-α and IL-6 levels were measured by quantitative PCR. (B)Serum ALT and AST levels after administration of galactosylated liposomes/siRNA complex. Groups of three mice were i.v. injected with Gal-LipoNP complexed with Fas siRNA, Gl2 siRNA and PBS, respectively. Blood samples were collected at 24 h following administration of experimental agents. AST and ALT levels were measured as described in Materials and Methods.

Figure 5. Protecting ConA-induced hepatic injury using Gal-LipoNP/Fas siRNA.

C57BL/6 mice were treated with Gal-LipoNP containing Fas siRNA or GL2 siRNA, LipoNP that incorporated with Fas siRNA, Other control groups included naked siRNA administration by hydrodynamic injection and PBS group. 72 h after gene silencing, mice were administered (15 mg/kg) ConA intravenously. After 24 h injury, liver tissues were collected and total RNAs were extracted. Fas Gene suppression (A) and ALT and AST levels (B) were measured. The livers were stained with H&E (C). The apoptotic cells in the liver tissues were detected using TUNEL assay (D), and the percentage of apoptotic cells were counted (E). (a) C57BL/6 mice treated with PBS; (b) Gal-LipoNP containing GL2 siRNA; (c) naked Fas siRNA administration by hydrodynamic injection ; (d) LipoNP containing Fas siRNA; (e) Gal-LipoNP containing Fas siRNA. Values are given as mean ± SD (n = 6, per group). (*) Statistical significance when compared with Gal-LipoNP/Fas siRNA and Gal-LipoNP/GL2 siRNA groups was denoted at p<0.05.