Comparative study of anti-hepatitis B virus RNA interference by double-stranded adeno-associated virus serotypes 7, 8, and 9

Abstract

Using a hepatitis B virus (HBV) transgenic mouse model, we previously showed that a single dose of double-stranded adeno-associated virus (dsAAV) vector serotype 8 carrying a small hairpin RNA (shRNA) effectively reduces HBV replication and gene expression, but the effect gradually decreases with time. In this report, we compared the anti-HBV RNA interference (RNAi) effect of dsAAV8 with those of dsAAV7 and dsAAV9, two other hepatotropic AAV vectors, and examined whether the sequential use of these heterologous AAV vectors could prolong the anti-HBV effect. Our results showed that shRNA delivered by each of the three dsAAV vectors profoundly reduced the serum HBV titer and liver HBV mRNA and DNA levels in the transgenic mice for up to 22 weeks, with dsAAV8 having the greatest inhibitory effect, followed by dsAAV9 and dsAAV7. The potency of dsAAV8 correlated with the presence of higher levels of vector DNA and anti-HBV shRNA in the liver. An in vivo cross-administration experiment showed that preexisting anti-AAV8 antibody completely blocked the anti-HBV RNAi effect of dsAAV8, but had no effect on the potency of dsAAV7 and dsAAV9. Moreover, we demonstrated that a longer anti-HBV effect could be achieved by the sequential use of dsAAV8 and dsAAV9. These results indicate that effective and persistent HBV suppression might be achieved by a combination of the power of RNAi silencing effect and multiple treatments with different AAV serotypes.Molecular Therapy (2009) 17 2, 352-359 doi:10.1038/mt.2008.245.

Figure 1

Comparison of liver transduction efficiency of different double-stranded adeno-associated virus (dsAAV) serotypes. Groups of ICR mice were intravenously injected with 1 × 10¹² vector genomes per mouse of dsAAV7/CB-GFP, dsAAV8/CB-GFP, or dsAAV9/CB-GFP vector. (a) Photographs of liver cryosections (5 µm) were taken 3 weeks after injection and examined for green fluorescent protein (GFP) expression. Original magnification ×40. Bar = 500 µm. (b) Quantification of the average fluorescence intensity and the percentage of the total area containing GFP-positive cells.

Figure 2

Comparison of different double-stranded adeno-associated virus (dsAAV) serotypes in small hairpin RNA (shRNA)-mediated hepatitis B virus (HBV) suppression. Groups of HBV transgenic mice (n = 4–6) were intravenously injected with 1 × 10¹² vector genomes per mouse of dsAAV7, dsAAV8, or dsAAV9 vector expressing control GL2 or HBV-S1 shRNAs. (a) Two weeks after injection, serum HBV titers were measured by real-time PCR and are presented as a percentage of the pretreatment titer for the group (mean ± SD). (b) Time course of HBV suppression after injection of the different dsAAV serotypes encoding HBV-S1 shRNA (mean ± SD).

Figure 3

Reduction in liver hepatitis B virus (HBV) mRNA and DNA levels in transgenic mice by the different double-stranded adeno-associated virus (dsAAV) serotypes. HBV transgenic mice were treated with the different dsAAV serotypes expressing HBV-S1 small hairpin RNA as described in the legend of Figure 2, then, 12 weeks after injection, liver samples were collected for northern and Southern blot analysis. (a) Northern blot analysis of HBV 3.5-, 2.4-, and 2.1-kb mRNAs, with mouse glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA as the internal control. (b) Southern blot analysis of HBV DNA. The bands corresponding to the integrated transgene (Tg) and the HBV replicative DNA intermediates (DS, double-stranded; RC, relaxed circular; SS, single-stranded) are indicated.

Figure 4

Copy number of the different double-stranded adeno-associated virus (dsAAV) serotype vectors and small hairpin RNA (shRNA) expression in the liver. Liver samples were obtained from the dsAAV-treated hepatitis B virus (HBV) transgenic mice as described in the legend of Figure 2. (a) Analysis of dsAAV vector genomes. Total liver DNA (10 µg) from mice treated with the different dsAAV serotypes was analyzed by Southern blotting after BamHI and XbaI digestion, which releases the 1.05-kb fragment from the AAV genome. The vector plasmid was used as the reference standard for estimation of AAV genome copy number. (b) Total liver RNA (30 µg) was analyzed by northern blotting using radiolabeled probes specific for the antisense strand of HBV-S1 shRNA (“as HBV-S1”) and endogenous microRNA-122.

Figure 5

Preimmunization with double-stranded adeno-associated virus 2/8 (dsAAV2/8) does not block the anti-HBV effect of other dsAAV serotypes expressing HBV-S1 small hairpin RNA (shRNA). Hepatitis B virus (HBV) transgenic mice were initially intravenously injected with 1 × 10¹² vector genomes (vg) per mouse of dsAAV2/8/GL2, then, 4 weeks later, were randomly grouped and received a second injection of 1 × 10¹² vg per mouse of dsAAV7, dsAAV8, or dsAAV9 vector expressing HBV-S1 shRNA. Mice injected with saline served as controls. Serum HBV titers at the indicated time points after the second injection were determined and are presented as a percentage of the pretreatment titer for the group (mean ± SD; n = 3).

Figure 6

Effect of multiple administration of double-stranded adeno- associated virus (dsAAV) vectors expressing HBV-S1 small hairpin RNA in transgenic mice. (a,b) The data for two independent experiments on (a) two mice or (b) three mice. Hepatitis B virus (HBV) transgenic mice (6–8-weeks old) were injected with 1 × 10¹² vector genomes (vg) per mouse of dsAAV2/8/HBV-S1, then, (a) 60 or (b) 62 weeks later, were re-injected with dsAAV2/9/HBV-S1 at a dose of 3.3 × 10¹³ vg/kg. Serum HBV titers at the indicated time points were measured and are presented as a percentage of the pretreatment titer for the same mouse.

Figure7

Hepatitis B virus (HBV) suppression by double-stranded adeno-associated virus 9 (dsAAV9)/HBV-S1 in young and aged transgenic mice. HBV transgenic mice at the age of 6–8-week or 14-month old were injected with 3.3 × 10¹³ vg/kg of dsAAV9/HBV-S1. (a) Serum HBV titers at the indicated time points after injection were measured and presented as described in the legend of Figure 2. (b) Total liver DNA and RNA were collected at the indicated time points. Southern and northern blots were used to analyze the amount of AAV DNA (upper panel) and antisense HBV-S1 shRNA, miR-122, and 5S rRNA (lower panel), respectively.