Figure 1.
All constructs were based on the wild-type HBV expression plasmid pCH-9/3091 which contains a complete HBV genome. Open reading frames are depicted as yellow boxes. Transcription of the pregenomic (pg) RNA is under control of the CMV-IE promoter. The start sites for transcripts controlled by the endogenous preS1, preS2/S and HBx promoters are indicated by the rightward pointing arrows. “ε” denotes the RNA packaging signal whose interaction with the polymerase initiates pgRNA encapsidation and reverse transcription. The diamond labeled pA symbolizes the HBV polyadenylation signal. In first generation HBV vectors, represented by pCH-S-TG, the transgene (TG) replaced the viral S gene. Here, eGFP, RFP2 and Luc were used as transgenes. These vectors still produce core and X protein, plus truncated forms of polymerase (region covered with dark slash lines) and PreS1/PreS2 (pS1, pS2) gene products. For vector production, functional polymerase and surface proteins are provided in trans by a helper plasmid, e.g. pCH3142 which is identical to pCH-9/3091 except it lacks the “ε” signal. In the pCH-M5-TG vectors, expression of the endogenous HBV gene products was ablated by premature stop codons (blue crosses), or mutation of the preS1 and preS2 initiation codons (downward pointing triangles) so as to engage all preS/S transcripts as TG mRNAs; here TG included truncated MMP-8 (tMMP8). For the chimeric Ad-HBV vectors, Ad-CH-TG, the entire HBV vector expression cassettes from the pCH-M5-TG plasmids were incorporated into ΔE1/ΔE3 Ad-vector backbones. Ad-C-MMP8 contained full-length MMP-8 under CMV promoter control, but no HBV sequences.
Table 1.
PCR primers for MMP8, tMMP8, HGF, c-Met, and GAPDH mRNA.
Figure 2.
Trans-complementation rescues HBV vectors lacking endogenous gene products into infectious virions with enhanced transgene expression.
(A) Rescue of replication in intracellular nucleocapsids by helper plasmid pCH3142. HepG2 cells were cotransfected with equal amounts of pCH3142 and the indicated pCH-S or pCH-M5 plasmids. Transfection with the wild-type HBV expression plasmid pCH-9/3091 served as reference. DNA from cytoplasmic nucleocapsids was analyzed by Southern blotting using a 32P-labeled HBV-specific probe. The positions of the major replicative intermediates, i.e. relaxed circular (RC) DNA, double-stranded linear DNA (DS) and single-stranded DNA (SS) are indicated. (B) and (C) Enhanced transgene expression by HBV vectors devoid of endogenous gene products. HepG2 cells were transfected with either the first generation pCH-S-GFP plasmid, or the new pCH-M5-GFP vector and GFP expression at 48 h post transfection was monitored by fluorescence microscopy (B). Alternatively, HepG2 cells were cotransfected with the Renilla luciferase encoding plasmids pCH-S-Luc or pCH-M5-Luc plus a plasmid encoding firefly luciferase (pGL3-control). At 48 h post transfection, Renilla luciferase activity from the HBV vector was then normalized to firefly luciferase activity in the same cells (C). (D) Rescue of infectious HBV vector particle formation by wild-type HBV. HepG2 cells were cotransfected with the vector pCH-M5-GFP and, instead of the helper plasmid pCH3142 as in (A), with the wild-type HBV expression plasmid pCH-9/3091. Viral particles from the supernatant were then tested for infectivity on differentiated HepaRG cells, using a nominal moi of 5,000 vge/cell. GFP expression indicating the presence of infectious vector particles was assessed by fluorescence microscopy 6 days post infection (panel “Fluorescence”). An overlay with the brightfield image of the same field is shown in panel “Merged”.
Figure 3.
In vitro and in vivo transgene expression by chimeric Ad-HBV vectors.
(A) tMMP8 expression in HepG2 cells. HepG2 cells were inoculated with the chimeric virus particles Ad-CH-tMMP8 (Ad-tMMP8) and Ad-CH-RFP2 (Ad-RFP2), or the conventional Ad-C-MMP8 particles; untreated HepG2 cells served as control. Four days later, cell lysates were analyzed by immunoblotting using a polyclonal MMP-8 antibody that recognizes an amino acid segment present in both MMP-8 and tMMP-8. Detection of tubulin on the same blot served as loading control. (B) and (C) Ad-HBV vector encoded transgene delivery into the liver. Ad-CH-RFP2 particles were injected into the tail vein of normal rats, with rats injected with saline as control. Two weeks later, liver tissues were immediately freeze-sectioned and analyzed for RFP2 expression by fluorescence microscopy. Strong red fluorescence was detected in rats treated with the Ad-HBV vector (B) whereas only limited autofluorescence was seen in rats treated with saline (C). (D), (E) and (F) Rats with liver fibrosis were treated with Ad-CH-tMMP8 (D), Ad-C-MMP8 (E), or Ad-CH-RFP2 (F), respectively. Two weeks later, rat livers were sectioned and immunohistochemically stained with MMP-8 antibody. MMP-8 and tMMP-8 were visualized as brown precipitates.
Figure 4.
Chimeric Ad-HBV vector delivering truncated MMP-8 achieves similar reduction of hydroxyproline content in cirrhotic liver as a conventional full-length MMP-8 Ad vector.
Hydroxyproline (HYP) contents in livers from normal rats (group E), or rats with TAA-induced cirrhosis (groups A–D) and treated with the chimeric Ad-HBV vectors Ad-CH-tMMP8 (group A) or Ad-CH-RFP2 (group C), or the conventional Ad vector Ad-C-MMP8 (group B) or saline (group D) were determined 2 weeks, 4 weeks, and 8 weeks after treatment, using a standard colorimetric assay. Both MMP-8 vectors achieved a significant reduction in HYP content compared to the Ad-CH-RFP2 vector and saline. Error bars represent standard deviation (n = 3).
Figure 5.
Chimeric Ad-HBV vector delivering truncated MMP-8 achieves similar amelioration of fibrosis and cirrhosis as a conventional full-length MMP-8 Ad vector.
(A) Gross morphology and histology. Livers and liver sections from rats of the same treatment groups as in Fig. 4 were analyzed for gross morphology (Gross view), and by hematoxylin - eosin (HE) staining, Sirius red collogen staining, or immunohistochemical staining specific for type I collagen. (B) Overall collagen contents. Sirius red staining followed by polarizing microscopy was used to determine the collagen positive areas per liver sections. (C) Knodell fibrosis scores. The highly elevated collogen levels and Knodell scores in the cirrhotic animals treated with the RFP2 Ad vector or saline were strongly reduced by treatment with either the full-length MMP-8 Ad vector or the chimeric Ad-HBV vector delivering truncated tMMP8. Error bars represent standard deviations (n = 3).
Figure 6.
Similar enhancement of hepatocyte proliferation promoting factors by chimeric Ad-HBV vector for truncated MMP-8 and conventional full-length MMP-8 Ad vector.
(A) c-Met mRNA. (B) Hepatocyte growth factor mRNA. mRNA levels for c-Met, hepatocyte growth factor (HGF) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) in liver tissue from the same treatment groups as in Fig. 3 were determined by quantitative RT-PCR at the indicated timepoints post treatment. Values for c-Met and HGF were then correlated with those for GAPDH in the same sample. Error bars represent standard deviations (n = 3).