Arch Virol (2004) 149: 1611–1617 DOI 10.1007/s00705-004-0300-4
Comparative analysis of adenoviral transgene delivery via tail or portal vein into rat liver J. Herrmann, B. Abriss, E. van de Leur, S. Weiskirchen, A. M. Gressner, and R. Weiskirchen Institute of Clinical Chemistry and Pathobiochemistry, RWTH University Hospital, Aachen, Germany Received September 24, 2003; accepted January 15, 2004 c Springer-Verlag 2004 Published online April 5, 2004
Summary. The potential indications for gene therapy are expanding continuously. Currently, hepatotropic adenoviruses are useful vector systems for targeting liver in experimental animal models. Although this gene delivery technique is widely distributed, there is no common sense about how these viruses should be applied. In general, the local delivery into portal vein and the systemic application via tail vein induces above all substantial transgene expression. We here comparatively analysed both methods and found that the systemic administration of an adenovirus expressing the green fluorescent protein resulted in a stronger infiltration, a more homogenous distribution, and a higher inter-individual reproducibility of reporter gene expression in rat liver than organ-specific administration via the portal vein. Introduction For most applications, an optimal recombinant viral vector should produce a high level of transgene expression. In recent years, vectors based on adenovirus2 (Ad2) or adenovirus-5 (Ad5) have emerged as the most efficient vehicles of transferring genes into liver cells, irrespectively whether these cells are quiescent or proliferative [4]. Nonetheless, although adenoviral vectors are widely applied, the experimental design in regard to viral load as well as the site used for application is discussed controversially. Additionally, it was recently demonstrated that the early levels of transgene expression varies within a species from one animal strain to another [5]. For these reasons, the comparison of results obtained by different investigators is difficult or even impossible. Therefore, we comparatively evaluated hepatic gene transfer mediated by systemic and local application of an adenoviral reporter vector expressing the green fluorescent protein (GFP) in rats of the Sprague-Dawley strain.
1612
J. Herrmann et al.
Material and methods Construction and purification of recombinant adenovirus The construction of the replication deficient recombinant adenoviruses Ad5-CMV-GFP, harboring GFP downstream of a cytomegalovirus promoter (CMV), has been described previously [8]. The purification of the adenoviral reporter was optimized by a two-step-procedure to allow
Fig. 1. Experimental design. A Schematic diagram of adenoviral application pattern, time course, and analytical methods. B Topology of drawn liver samples. The regions of specimensampling (central liver, peripheral liver), sternum and small intestine are indicated by arrows
Adenoviral delivery to rat liver
1613
effective transgene transfer in vivo. Firstly, the virus was separated from HEK-293 lysates and concentrated through CsCl density gradient centrifugation. Therefore, viral solutions were adjusted with 0.5 g of CsCl/ml lysate and centrifuged for 22 h at 220000 × g (4 ◦ C). Secondly, the resulting opaque viral band was further purified through the BD Adeno-XTM Purification Filter system (BD Biosciences, Clontech, Palo Alto, CA, U.S.A.), according to the manufacturer’s instructions. After elution of the trapped adenovirus, viral particles were concentrated by centrifugation (28000 × g, 30 min, 4 ◦ C) and resolved in formulation buffer.
Fig. 2. Immunohistochemistry of liver sections. Representative liver sections of rats that were injected via portal vein A or tail vein B with different concentrations (3 × 109 , 1.5 × 1010 , 3 ×1010 , 3 × 1011 virions/animal) of Ad5-CMV-GFP or saline (control). GFP-positive cells are dark coloured (space bar = 200 µm)
1614
J. Herrmann et al.
Adenoviral delivery to rat liver
1615
The content of virions per ml was determined by measurement of the optical density (OD) at 260 nm (1OD260 unit = 1 × 1012 virions/ml). Viral stocks were aliquoted and stored in liquid nitrogen until use. Administration of recombinant adenoviruses to animals 200 µl of saline and four different amounts of Ad5-CMV-GFP (Fig. 1A) were administered via tail or portal vein into male Sprague-Dawley rats (3 animals per group), weighing about 200 g. Application via the portal vein was performed after laparotomy with an 1 ml disposable r , Braun, fine-dose syringe with integrated hypodermic needle (0.33 × 1.2 mm, Omnican-F Melsungen, Germany) allowing slow injection of adenoviral solutions. The prick area was gently pressed with a sterile cloth to prevent bleeding. Five days after surgery the animals were sacrificed and samples (∼0.2 g) from peripheral and central regions of the livers (Fig. 1B) were either fixed for immunohistochemical analysis in 4% (w/v) buffered paraformaldehyde or homogenized for SDS-polyacrylamide gel electrophoresis (PAGE) with an Ultra Turrax in 600 µl lysis buffer [50 mM Tris-HCl (pH 8.0), 250 mM NaCl, 2.5 mM EDTA, 2% NP40 (v/v), 0.1% SDS (w/v), 0.5% deoxycholic acid (w/v), 1:500 protease inhibitor cocktail (Sigma, Taufkirchen, Germany), 1:100 phosphatase inhibitor cocktail 2 (Sigma)]. Immunohistochemistry Fixed liver pieces of Ad5-CMV-GFP- or mock-infected rats were embedded in paraffin. Sections (1.5-µm) were blocked against endogenous peroxidases in 3% H2 O2 . GFP was detected with an 1:100 diluted polyclonal rabbit anti-GFP antibody (Santa Cruz Biotech., Santa Cruz, CA, U.S.A.) and a 1:300 diluted secondary biotinylated anti-rabbit IgG (Vector Laboratories, Burlingame, CA, U.S.A.) followed by treatment with an avidin-conjugated peroxidase (Vectastain ABC-Elite Kit, Vector Laboratories). The peroxidase activity was detected with diaminobenzidine (DAKO, Hamburg, Germany) and tissue sections were briefly counterstained with methyl green (Sigma). SDS-PAGE and immunoblotting 15 µg whole liver lysates each were resolved on a 12% (w/v) Tris-Glycine gel (Novex, Groningen, The Netherlands) by SDS-PAGE under reducing conditions. For immunoblotting, proteins were electroblotted onto Protran membranes (Schleicher & Schuell, Dassel, Germany) according to standard procedures. The membranes were blocked for 1.5 h in a solution containing 5% non-fat milk powder, 20 mM Tris-HCl (pH 7.5) and 150 mM NaCl. Thereafter, membranes were incubated for 1 h in an 1:1000-diluted mixture of two mouse monoclonal antibodies (clones 7.1 and 13.1) raised against GFP (Roche, Mannheim, Germany). The primary antibodies were visualised using horseradish peroxidase-conjugated goat anti-mouse IgG (Santa Cruz Biotech.), diluted 1:5000, and the super signal chemiluminescent substrate (Pierce, Rockford, IL, U.S.A.). As internal loading control, β-actin content
Fig. 3. Western blot analysis of GFP expression in liver lysates. Protein lysates (15 µg) of peripheral and central liver areas were taken from animals receiving Ad5-CMV-GFP via portal vein A or tail vein B and analyzed for GFP expression. The amounts of injected virions/animal are given on the top. As a control for equal gel loading, blots were stripped and tested for β-actin expression. C Relative quantification of GFP contents in liver samples. Expression of GFP (measured in BLU) was analysed by Western blot analysis and normalised to β-actin. In the diagram, the mean values ± SD of each experimental group (n = 3) are plotted
1616
J. Herrmann et al.
of each sample was assessed by a monoclonal mouse antibody (clone AC-15) obtained from Sigma. Detection of chemiluminescence was performed with the Lumi-ImagerTM (Roche) and expression values were quantified by measurement of BLU (Boehringer Light Units) using the software LumiAnalyst 3.0 (Roche). The GFP signals were normalised against β-actin and the average of each experimental group was calculated.
Results and discussion Presently, there is no exhausting data available for the applicability to experimental target the liver by adenoviral devices. The reported methodologies are discrepant or even contradictory, e.g. the amounts of adenoviruses used for the targeting of rat liver range from 1 × 109 to 8 × 1010 pfu [2, 7]. Indeed, the optimal viral concentration might be strongly dependent on the aim of each study, but it is not intelligible that viral loads differ by several orders of magnitudes under different experimental conditions. Another critical issue is the manner of viral application, e.g. some studies prefer the injection via tail vein while other investigators use administration via portal vein [6, 7]. In most cases there are no rational reasons why preferring one of these methods, but there is no doubt that the systematic application via tail vein, which is used in our laboratory for directing transgene expression in rat liver [1, 3], is more simple in handling than the local delivery through the portal vein. To address this question more comprehensive, we initiated a study and analysed the two different routes of administration using an adenoviral reporter expressing GFP (Ad5-CMV-GFP). We infected male Sprague-Dawley rats with different amounts of Ad5-CMV-GFP, ranging from 3 × 109 to 3 × 1011 virions/animal, and evaluated the transgene expression after 5 days in liver specimen taken from central and peripheral regions of liver lobules (Fig. 1). As expected, the infusion of higher viral particle numbers resulted in more cells positive for GFP, which was directly visible by inspection of liver samples under the UV light (not shown) and in sections of fixed samples analysed by immunohistochemistry (Fig. 2). Surprisingly, the injection via the tail vein resulted in an approximately tenfold more effective expression of transgene at both, low (3 × 109 virions/animal) and high (3 × 1011 virions/animal) viral concentrations. Additionally, both applications resulted in a slightly higher GFP expression in peripheral regions of the liver. The same relative expression in the different liver areas was confirmed by Western blot analysis (Fig. 3). Furthermore, animals receiving their adenoviral injections systemically showed an overall stronger expression of the transgene compared to those injected via the portal vein. Moreover, the inter-individual reproducibility within each experimental group was higher when rats were treated by the tail vein injection procedure. One possible explanation of the reduced bioavailability of adenoviruses locally entering the liver might be a high first-pass hepatic turnover resulting in lower transgene expression. Theoretically, it might also be possible that the decreased reproducibility after portal injection is due to a transduction variance resulting from surgery. However, this hypothesis is in contrast to our findings showing that livers taken from animals receiving sham operation plus infection via tail vein
Adenoviral delivery to rat liver
1617
reveal no relevant differences in transgene expression compared to animals without surgery. In summary, our study reveal that in contrast to local administration the systemic application of adenoviruses via the tail vein decreases the differences between each animal and results in a more homogenous distribution and higher efficiency of transgene expression in rat liver. References 1. Arias M, Sauer-Lehnen S, Treptau J, Janoschek N, Theuerkauf I, Buettner R, Gressner AM, Weiskirchen R (2003) Adenoviral expression of a transforming growth factor-β1 antisense mRNA is effective in preventing liver fibrosis in bile-duct ligated rats. BMC Gastroenterology 3: 29 2. Chen LM, Chao L, Chao J (1997) Adenovirus-mediated delivery of human kallstatin gene reduces blood pressure of spontaneously hypertensive rats. Hum Gene Ther 8: 341–347 3. Dooley S, Hamzavi J, Breitkopf K, Wiercinska E, Said HM, Lorenzen J, ten Dijke P, Gressner AM (2003) Smad7 prevents activation of hepatic stellate cells and liver fibrosis in rats. Gastroenterology 125: 178–191 4. Ghosh SS, Takahashi M, Thummala NR, Parashar B, Chowdhury NR, Chowdhury JR (2000) Liver-directed gene therapy: promises, problems and prospects at the turn of the century. J Hepatol 32 (Suppl. 1): 238–252 5. Lefesvre P, Attema J, Lemckert A, Havenga M, van Bekkum D (2003) Genetic heterogeneity in response to adenovirus gene therapy. BMC Mol Biol 4: 4 6. Nakamura T, Sakata R, Ueno T, Sata M, Ueno H (2000) Inhibition of transforming growth factor β prevents progression of liver fibrosis and enhances hepatocyte regeneration in dimethylnitrosamine-treated rats. Hepatology 32: 247–255 7. Qi Z, Atsuchi N, Ooshima A, Takeshita A, Ueno H (1999) Blockade of type β transforming growth factor signaling prevents liver fibrosis and dysfunction in the rat. Proc Natl Acad Sci USA 96: 2345–2349 8. Weiskirchen R, Kneifel J, Weiskirchen S, van de Leur E, Kunz D, Gressner AM (2000) Comparative evaluation of gene delivery devices in primary cultures of rat hepatic stellate cells and rat myofibroblasts. BMC Cell Biol 1: 4 Author’s address: Dr. Ralf Weiskirchen, Institute of Clinical Chemistry and Pathobiochemistry, RWTH-University Hospital, Pauwelsstr. 30, D-52074 Aachen, Germany; e-mail:
[email protected]