Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Cirrhotic rat livers with extensive fibrosis can be safely transduced with clinical-grade adenoviral vectors. Evidence of cirrhosis reversion


Adenoviral vectors efficiently target normal liver cells; however, a clear-cut description of the safety boundaries for using adenovectors in hepatic cirrhosis has not been settled. With this in mind, we used a first-generation, replication-deficient adenoviral vector carrying the E. coli lacZ gene (Ad5βGal) to monitor therapeutic range, biodistribution, toxicity and transduction efficiency in Wistar rats made cirrhotic by two different experimental approaches resembling alcoholic cirrhosis and biliary cirrhosis in humans. Further, we show proof of concept on fibrosis reversion by a ‘therapeutic’ Ad-vector (AdMMP8) carrying a gene coding for a collagen-degrading enzyme. Dose–response experiments with Ad5βGal ranging from 1 × 108–3 × 1012 viral particles (vp) per rat (250 g), demonstrated that adenovirus-mediated gene transfer via iliac vein at 3 × 1011 vp/rat, resulted in an approximately 40% transduction in livers of rats made cirrhotic by chronic intoxication with carbon tetrachloride, compared with approximately 80% in control non-cirrhotic livers. In rats made cirrhotic by bile-duct obstruction only, 10% efficiency of transduction was observed. Biodistribution analyses showed that vector expression was detected primarily in liver and at a low level in spleen and kidney. Although there was an important increase in liver enzymes between the first 48 h after adenovirus injection in cirrhotic animals compared to non-transduced cirrhotic rats, this hepatic damage was resolved after 72–96 h. Then, the cDNA for neutrophil collagenase, also known as Matrix Metalloproteinase 8 (MMP8), was cloned in an Ad-vector and delivered to cirrhotic rat livers being able to reverse fibrosis in 44%. This study demonstrates the potential use of adenoviral vectors in safe transient gene therapy strategies for human liver cirrhosis.


Liver cirrhosis is a worldwide health problem. Cirrhosis is a major liver disease for which there are no completely satisfactory therapies. Cirrhotic livers are characterized by extensive fibrosis throughout the entire hepatic parenchyma, especially around central and portal veins. The deposition of excessive fibrous or collagenous proteins in the subendothelial space or Space of Disse results in decreased free exchange flow between hepatocytes and sinusoidal blood. The cellular effects of these collagenous materials and other non-collagenous components, especially on hepatocytes, cause synthetic and metabolic dysfunction characteristic of advanced liver disease.12 Removing the fibrous septa might result in benefit for subjects undergoing liver fibrosis due to the functional re-establishment of the hepatocyte–sinusoid flow exchange. Thus, delivery of genes coding for collagen-degrading enzymes may represent a novel therapy strategy for liver cirrhosis.

Various vector types have been shown to be efficient at delivering genes to normal livers.34567 Similarly, the use of viral and nonviral vectors for gene delivery to functionally compromised livers has been instrumental to establish ‘proof of concept’ in several experimental models. Various strategies involving adenoviral vectors have been used.89 However, a key issue concerning toxicology, biodistribution and safety when using adenoviral vectors in cirrhotic animals, remains unresolved. Part of this study was designed to evaluate biodistribution, liver toxicity and efficiency of transduction of a reporter gene in hepatic cells of cirrhotic Wistar rats. We also wanted to delimit the ‘therapeutic window’ for potential and safe use of these vectors in a given clinical setting. Adenoviral vectors were chosen because they have been shown to be efficient vehicles for delivering genes to the liver and their distribution has been extensively analyzed in healthy animals. However, the potential toxicity of these vectors, especially to the liver, was a major concern. Animals with decreased liver function may have increased sensitivity to hepatotoxic effects of adenoviral vector administration. Therefore, we analyzed liver enzyme profiles of cirrhotic rats after intravenous adenoviral delivery with gene expression in the liver. Another part of this study was designed to determine proof of concept when using a neutrophil collagenase cDNA, cloned in an adenoviral vector (AdMMP8). Here, it is clearly shown that a ‘safe dose’ (3 × 1011 vp/kg) of adenoviral particles, induces a vigorous and fast degradation of excessive collagenous material deposited in livers of experimental animals resembling secondary biliary cirrhosis in humans. This effect resulted in marked improvement of functional hepatic tests (alanine transaminase or ALT, aspartate transaminase or AST and bilirubins) and ascitis. Importantly, our results are supported by recent publications showing that systemic administration of adenoviral vectors can effectively target diseased livers.1011

Results and discussion

Two rat cirrhosis models were used: carbon tetrachloride (CCl4-treatment) induced experimental cirrhosis that resembles alcoholic human disease, as well as cirrhosis post-chronic infection with hepatitis C virus and the biliary duct ligation model (BDL) mimics biliary cirrhosis in humans.1213 Doses of Ad5βGal ranging from 1 × 108–3 × 1012 viral particles (vp) per rat (weighing approximately 250 g) were delivered systemically through the iliac vein and a total of five rats per dose were tested. The goal of these experiments was to establish the toxicity window of adenovector dosage. The dose level of 1 × 1012 (4 × 1012 vp/kg) resulted in high mortality (100% in cirrhotic animals and approximately 60% in normal rats). A 3 × 1012 vp (1.2 × 1013 vp/kg) dose killed all cirrhotic animals and 80% of healthy animals, whereas 3 × 1011 total vp (1.2 × 1012 vp/kg) was reasonably tolerated in normal and cirrhotic rats with an optimal degree of efficiency of liver transduction (Figure 1). The threshold between dosages is quite narrow. However, recent data obtained in our laboratory using 6 × 1011 vp/kg of a therapeutic Ad vector which efficiently reverses experimental liver fibrosis,8 provides confidence and warrants further investigation on the use of Ad vectors on liver fibrosis. In a previous paper,8 we used a first generation, clinical-grade adenovector bearing the cDNA for urokinase-plasminogen activator capable of reversing liver cirrhosis in an efficient manner with no further complications.8 The dose used was 6.6-fold and 20-fold lower than 4 × 1012 and 1.2 × 1013 vp/kg toxic doses respectively, reported in this paper. Transduction was assessed by X-gal staining of tissue sections 72 h after vector administration.

Figure 1

Efficiency of transduction of cirrhotic rat livers compared with normal livers. Doses of Ad5βGal vector ranging from 1 × 108–3 × 1012 virus particles (vp) were injected into the iliac vein of non-cirrhotic (a), bile duct ligated (BDL)- (b) and CCl4-treated animals (c). β-Gal staining was determined 72 h after injection. Five animals were tested per dose. The percentage of transduced area was quantitatively monitored by computer-assisted morphometric analysis of tissue sections. The doses of 1 × 1012 and 3 × 1012 were fatal and therefore could not be analyzed.

Morphometric analyses of multiple tissue sections revealed a transduction efficiency of approximately 40% in animals transduced with 3 × 1011 Ad5βGal virus particles after CCl4-treatment for 5 weeks and decreasing only slightly to 38% after 8 weeks of intoxication (Figure 2). Ad5βGal was internalized mostly by hepatocytes, though sinusoidal cells were also shown slightly permissive to the Ad vector. Decreased β-gal transduction efficiency in CCl4 animals as measured by internalization of Ad5βGal, might not be mediated through αvβ5 integrin, since we did not find significant differences in specific staining between CCl4-injured animals and normal rats (data not shown). We speculate that such a transduction efficiency difference could be due to decreased expression of coxsackie/adenovirus receptor (CAR) or to an impediment in hepatocyte exposure to adenoviral vectors due to hemodynamic alterations, ie shunting of blood circulation in the severely damaged liver.

Figure 2

Histologic analysis of transduction and fibrosis in livers from normal rats (a–c) and rats treated with CCl4 for 5 weeks (d–f) or 8 weeks (g–i). (a, d and g) β-Galactosidase staining 72 h after infusion of 3 × 1011 vp of Ad5βGal via the iliac vein (magnification, ×200). (b, e and h) Staining for collagenous protein using Sirius Red on paraffin sections (magnification, ×100). (c, f and I) Same as b, e and h, respectively, but stained with Masson's trichromic staining (magnification, ×100). Note the increase in collagenous material and distortion of liver architecture which is apparent by 5 weeks (e and f) and complete cirrhosis by 8 weeks (h and i) with characteristic bridging septa of collagen between central veins and portal tracts.

In the BDL model, extrahepatic cholestasis due to prolonged obstruction of bile flow resulted in even more extensive morphological and biochemical changes. This included an extensive proliferation of bile ducts in enlarged portal tracts, with inflammation and necrosis, and the formation of periportal fibrosis, as well as pericentral collagen deposition (Figure 3). The degree of transduction with 3 × 1011 vp in animals after 2 or 4 weeks of biliary obstruction was about 10% Figure 3. In spite of the evolution of the fibrotic process from 2 to 4 weeks of bile duct ligation, this percentage did not change considerably. The relatively low transduced area correlates with an increased degree of fibrosis, which may be, at least in part, explained by the fact that fewer cells were available in a given area to be transduced due to space-occupying extracellular fibrous tissue. In the bile duct ligation model, the main extracellular matrix components deposited are basement membrane collagen type IV, laminin and collagen type III, although collagen type I is also increased. These macromolecules form a continuous layer,14 setting a barrier against the free adenoviral diffusion from the sinusoid into the surrounding hepatocytes. This may explain, in part, the differential distribution of ß-gal activity in liver tissue from non-cirrhotic animals (mainly around the central veins) and cirrhotic livers (mostly concentrated in midzonal regions) (compare Figure 2 panels d and g with a; and 3d and g with a). The level of transduction observed in these cirrhotic rat liver models might be significant since it has been shown for other diseases that there is a substantial resolution of the pathology when just a low percentage of targeted cells are successfully transduced1215 (see Figure 6).

Figure 3

Histologic analysis of transduction and fibrosis in livers from normal rats (a–c) and rats treated by bile duct ligation for 2 weeks (d–f) or 4 weeks (g–i). (a, d and g) β-Galactosidase staining 72 h after infusion of 3 × 1011 vp of Ad5βGal via the iliac vein (magnification, ×200). (b, e and h) Staining for collagenous protein using Sirius Red on paraffin sections (magnification, ×200). (c, f and I) Represent the same as b, e and h, respectively, but stained with Masson's trichromic staining (magnification, ×100). Note the increase in collagenous material and distortion of liver architecture which is apparent by 2 weeks (e and f) and complete cirrhosis by 4 weeks (h and i) with characteristic bridging septa of collagen between central veins and portal tracts. Note the increase in collagenous material and bile duct proliferation in e and f, which is even more extensive in h and i.

Figure 6

Proof of concept showing reversion of fibrosis by AdMMP8 in experimental secondary biliary cirrhosis. (a) Liver sections of cirrhotic rats after bile-duct ligation, bile drainage by biliodigestive anastomosis and AdMMP8 or AdGFP adenoviral vector injection. Masson trichromic staining (×100) showing cirrhotic nodules surrounded by thick fibrotic bands (in blue color) even at the end of treatment with AdGFP. On the other hand, animals treated with AdMMP8 show only thin fibrous bands consisting of connective tissue proteins. (b) Percentages of fibrous tissue deposition at every step and after AdGFP or AdMMP8 adenoviral vector infusion. Determinations of fibrous tissue were done in 20 random fields by an automated image analyzer. Values are presented as means ± standard deviation.

The use of adenoviral vectors for gene therapy in liver cirrhosis will partly depend on successful delivery of vector to target tissue while minimizing leakage to extra-hepatic tissues. The distribution of β-gal expression after peripheral vein injection was primarily to the liver in cirrhotic and in normal rats (Figure 4). The analyses of vector expression showed no β-gal staining in brain, heart, lung, testis or ileum in any of the experimental animals. However, β-gal expression was detected at low levels in the spleen and kidney of two of five cirrhotic animals. Previous studies in rodents have shown that approximately 90% of systemic vector is found in the liver of animals with normal function.1617 However, it was important to demonstrate that the distribution would not be significantly different in cirrhotic animals. Recently, and contrary to what is reported here, Nakamura et al18 have demonstrated that adenovirus-mediated LacZ gene expression was preferentially shown in septal cells, rather than hepatocytes in cirrhotic rats. However, they used only one dose of the vector (1.5 × 109 p.f.u.) and delivered it through the tail vein,18 facts that might account for such a difference. On the other hand, a recent study by Nakatani et al11 using 2 × 108 p.f.u./ml of Adex1CalacZ adenovirus as a reporter vector, demonstrated that 80% of cells in cirrhotic livers expressed the LacZ gene as compared with 40% in normal livers. Differences with our work might be accounted for by the fact that they used a different route of administration and cirrhosis was induced with thioacetamide. Concerning these issues, the lack of standardization to establish comparable dosages of vectors makes it extremely difficult to reach comparative conclusions on toxicity and efficacy. Recently, we have addressed this question19 and conclude that it is almost impossible to compare dosages from study to study, as different researchers measure vector titers in different ways. This is true not only for adenoviral vectors, but also for any of the vectors in current use.

Figure 4

Biodistribution of the vector Ad5βGal 72 h after administration to rats in which cirrhosis was induced by bile duct ligation (BDL) (b), or chronic CCl4 treatment (c). Five animals were studied in each model and each received 3 × 1011 total vp injected via the iliac vein. Transduction was detected by β-gal staining. (a) Normal animals.

In this study, we used adenovectors prepared in a Good Manufacturing Practices (GMP) facility and the certified quality of our Ad preparations was assured. Besides, we chose the iliac vein because it was the most efficient method, as compared with tail vein and intraperitoneal injection (data not shown). Also, and this is particularly important in cirrhosis, delivery via more invasive procedures such as to the portal vasculature, would be difficult in cirrhotic patients due to coagulopathy and poor wound healing making a more invasive procedure particularly dangerous. This approach is further supported by a recent report, which showed liver-directed gene transfer in non-human primates via the saphenous vein with nearly the same level of transduction in the liver as compared with portal vein infusion.20

Cirrhotic animals had elevated liver enzymes and these levels increased significantly during the first 48 h after adenoviral administration. Nonetheless, and concomitantly with evidence of significant exposure of the liver to adenovirus, these liver enzymes dropped between 72 to 96 h (Figure 5).

Figure 5

Comparison of HFTs in normal rats (a), with the BDL (b) or CCl4 cirrhotic animals (c), before □, and after ▪, injection of Ad5βGal. Five animals were studied in each group. Blood was drawn from all animals and HFTs were performed. Then, vector dose (3 × 1011 total vp) was injected via the iliac vein and blood analysis was performed 48 and 72 h after comparing with previous HFTs from the same animal.

Although high mortality was observed at the 1012–1013 dose levels, none of the animals died in either experimental model at the 1011 dose level despite efficient transduction at this level.

To determine whether this system could be useful in improving liver cirrhosis in an experimental model, we selected neutrophil collagenase (Matrix Metalloproteinase 8; MMP8) as a potential therapeutic reagent,21 due to the fact that this enzyme preferentially hydrolyzes collagens faster than hepatic fibroblast collagenase. In human cirrhosis, collagen I is the most prominent extracellular protein and its abundant synthesis and deposition is responsible for alterations in hepatic architecture and consequent hemodynamic dysfunctions.

Here, we establish an experimental model of human secondary biliary cirrhosis by ligating and sectioning the common bile duct (BDL) in Wistar rats (n = 10). Liver biopsies were obtained from each rat before BDL ligation and at every step described after ligation. At the end of BDL (4 weeks) all rats had dramatic increases in liver fibrosis as compared with basal fibrosis Figure 6 and were subjected to a bilio-digestive anastomosis to drain the clogged bile. This is a surgical procedure that imitates those performed in humans to alleviate bile-obstructive disorders by discontinuation of the harmful agent and re-establishment of bile flow. However, it is important to state that this procedure is therapeutically effective only in a small number of patients. Rats were left at this stage for 7 days. Hepatic fibrosis index and hepatic functional tests, however, continued to be elevated to similar extents in all animals. Then, five rats were injected via the iliac vein with 3 × 1011 vp/kg of AdMMP-8 along with five rats injected with a 3 × 1011 v/kg of irrelevant adenoviral vector (Ad-GFP) used as controls. The 10 rats were kept under careful surveillance for 10 more days, monitoring for overall health status and then killed. Levels of MMP8 protein expression in AdGFP and AdMMP8-transduced cirrhotic animals were quantified by ELISA using a polyclonal antibody against human neutrophil MMP8 that does not crossreact with MMP-1, -2, -3, -7, -9, -13 or MT1-MMP. Liver extracts from AdMMP8-transduced cirrhotic animals averaged 2.7 ng MMP8/ml. AdGFP-cirrhotic livers showed no detectable levels of recombinant MMP8.

Importantly, a statistically significant decrease in fibrosis (43.9%; P < 0.05) was noted in rats injected with AdMMP8, but no decrease in the percentage fibrosis index was observed in the five animals injected with AdGFP. Also, an important decrease in total and direct bilirubins was noted, although ALT and AST did not diminished at the same extent. Ascitis decreased in AdMMP8-injected animals as well. Furthermore, extrahepatic cholestasis due to prolonged obstruction of bile flow in bile duct-ligated rats, resulted in important morphological and biochemical changes. They included an extensive proliferation of bile ducts in enlarged and dilated portal tracts, with inflammation and necrosis. Nonetheless, a salient improvement in these pathological changes was observed at the end of AdMMP8-treatment as determined by observations made by two different pathologists blinded to the study. It is clear then that the resulting extracellular turnover is facilitating recovery of hepatic lobule function.

The level of transduction achieved in the two rat cirrhosis models is encouraging for the further development of gene therapy approaches for human cirrhosis. Of particular importance, is the demonstration that cirrhotic animals can tolerate doses of adenoviral vector necessary to achieve significant transduction efficiency in the absence of life-threatening liver toxicity. Although adenovectors may not be ideal for treatment of human liver disease, they may provide important tools for the evaluation of gene therapy strategies in rodent models of cirrhosis.

Materials and methods

Experimental models of liver cirrhosis

Wistar rats were housed and cared for according to National Institutes of Health guidelines in the animal facility of University of Guadalajara. For the first experimental model of cirrhosis, five rats per group were used. Rats weighing 150–200 g were anesthetized with ethylether and the common bile duct was exposed and ligated to induce fibrosis by total biliary obstruction (BDL rats) according to the method of Lee et al.22 Briefly, animals underwent a 2-cm upper-midline abdominal incision below the xifoides appendix, the extrapancreatic common bile duct was identified and double ligated with 4/0 silk (Ethicon; Johnson and Johnson, Mexico City, Mexico) and transected between the ligatures. Subsequently, the abdomen was closed in one layer with continuous 4/0 silk. These rats were kept for 4 weeks and they were given free access to food and water throughout the experimental period. Five rats were sham operated at the same time and used as controls. For the experiments which aimed to show proof of concept on fibrosis revertion, at least five rats were used for experiment. That is, 10 rats were ligated as described before for 4 weeks and a liver biopsy was obtained from each animal before and after the surgical procedure. Then, a biliodigestive anastomosis was performed on each animal to drain the clogged bile and re-establish bile flow in order to eliminate the injurious fibrogenic cause. Animals were left at this stage for 7 days and liver biopsy was also taken. At this point, five animals received either 3 × 1011 vp/kg of AdMMP8 or AdGFP and 10 days later were killed obtaining the fourth and final liver biopsy for each animal. Fibrosis index in multiple liver biopsies from animals at different stages were then carried out. Similarly, blood was obtained from each rat to carry out hepatic functional tests.

The second experimental model consisted of animals undergoing chronic intoxication with CCl4.23 Briefly, animals weighing 50–80 g received three doses a week via i.p. of a mixture 1:6 of CCl4-mineral oil for the first week, the 2nd week the ratio was 1:5, 3rd week 1:4, and 4th–8th week the ratio was 1:3. Control rats were pair fed and injected similarly with vehicle only.

Adenoviral vectors

The Ad5βGal vector used here is a first generation E1- and E3-deleted replication-defective adenovirus vector previously described.24 The vector was produced at the Baylor College of Medicine Gene Vector Laboratory under Good Laboratory Practice conditions, checked for general sterility and was free of contaminants such as endotoxin, mycoplasma, in vitro adventitious virus and replication-competent adenovirus (RCA). Thus, the vector was titered (1 × 1013 vp/ml from lot No. 01496AbgalA) and characterized as described24 and had a vector particle (v.p.) to infectious unit (IU) ratio of ≤10 which makes it suitable for controlled trials. Escalating doses were injected via iliac vein diluted in 250 μl of 10 mM Tris-HCl, pH 8.0, 2 mM, MgCl2 and 4% sucrose. All rats used in both experimental cirrhosis models weighed approximately 250 g at the time of Ad5βGal injection. For the experiments to check for fibrosis revertion, AdMMP8 was generated as described before by our group.21 Briefly, pAdHM2-MMP8 was transfected into 293 cells. pAdHM2-MMP8 was constructed by an in vitro ligation method of the MMP-8 cDNA in a replication-defective dE1, dE3 adenoviral plasmid.25 This and AdGFP were also titered and characterized as described above.

In situ β-gal assay in whole tissue preparations

The β-gal activity was determined 72 h after Ad5βGal injection. To monitor the degree of transduction of different organs, liver, brain, heart, ileum, testis, lung, kidney and spleen were obtained and immediately cut in thin slices using a sterile scalpel. These manipulations were quickly performed on ice and the tissue slices were placed in 24-well tissue culture plates and washed three times with cold PBS and fixed for 15 min with 10% buffered-formalin at 4°C. After fixation, tissue slices were washed twice with cold PBS and X-gal staining was performed. All reactions were carried out at pH 8.5 to avoid endogenous galactosidase activity. X-gal reaction solutions contained 1 mg/ml X-gal, 100 mM K3Fe(CN)6, 100 mM K4Fe(CN)6, 1 M MgCl2, 2% NP-40 and 1% sodium desoxicholate. Exposure of tissues was performed in the dark at 4°C for 16–18 h and then tissues were carefully washed with PBS and preserved in 10% buffered-formalin. For histological analyses, tissues were embedded in paraffin, sectioned and mounted on glass slides.26

Evaluation of β-gal activity in tissue section preparations

Tissues were cut in 5–6-mm3 squares and embedded in Tissue-Tek OCT compound (Miles, Elkhart, IN, USA), frozen at –30°C and cut in a cryostat to obtain 8-μm thin sections. These sections were placed on methacrylxypropyl-trimethoxisilane-treated glass slides and fixed with 10% buffered-formalin at room temperature for 15–30 min. After this, sections were processed as described above. Slides were subsequently washed in PBS, counterstained with Neutral Red and coverslipped with organic mounting media after standard dehydration in alcohol.

The average number of transduced cells was obtained by evaluating 50 different microscopic areas per rat liver at a magnification of ×200 by computer-assisted morphometric analysis using a Leica Quantimet Q570 image processor (Cambridge Instruments, Cambridge, MA, USA).

Histological staining for fibrous tissue and immunohistochemistry

To monitor the evolution and reversion of the fibrotic process taking place in the liver of cirrhotic animals, liver tissue blocks were fixed in 10% buffered-paraformaldehyde and embedded in paraffin. Sections (10 μm) were cut and stained with Picrosirius red solution at pH 4.0. Sirius red is a specific dye for collagenous proteins.27 The same tissue sections were counterstained with Fast green dye which has affinity for non-collagenous protein. Parallel tissue sections were stained with Masson's trichromic staining to visualize collagen deposits and characteristic proliferation of bile ducts in the bile duct-ligated rats. Evaluation of tissue area occupied by fibrosis was performed using the same image analysis procedure described above.

For immunohistochemistry, liver sections were mounted in silane covered slides, deparaffinized and endogenous activity of peroxidase was quenched with 3% H2O2 in absolute methanol. Liver sections were incubated overnight at room temperature with mouse antibodies against human αvβ5 integrin (GIBCO-BRL, Rockville, MD, USA) diluted 1/600 in PBS. Bound antibodies were detected with peroxidase-labeled rabbit polyclonal antibodies against mouse immunoglobulins and diaminobenzidine, and counterstained with hematoxylin. For quantification, 10 random fields of intralobular and periportal areas were evaluated at ×320 magnification. Immunohistochemical-positive and -negative cells were counted by an automated image analyzer (Qwin; Leica).

Quantification of MMP8 secretion by ELISA

Liver homogenates were obtained as previously described8 and total MMP8 was quantified using a Biotrak ELISA assay according to the manufacturer's instructions. Anti-human MMP8 Biotrak ELISA system was obtained from Amersham Pharmacia Biotech (Piscataway, NJ, USA).

Hepatic function tests (HFTs)

Blood was drawn from animals before and after adenovirus administration and serum transaminases ALT, AST, bilirubins and alkaline phosphatase were determinated in an automated Sincron-7 machine at Hospital Civil de Guadalajara.

Statistical analyses

Results are expressed as mean ± s.d. Student's t test was used. P < 0.05 was considered to indicate a significant difference between groups.


  1. 1

    Olaso E., Friedman S.L. . Molecular regulation of hepatic fibrogenesis J Hepatol 1998 29: 836 836

    CAS  Article  Google Scholar 

  2. 2

    Armendariz-Borunda J., Seyer J.M., Kang A.H., Raghow R. . Regulation of TGFβ gene expression in rat liver intoxicated with carbon tetrachloride FASEB J 1990 4: 215 215

    CAS  Article  Google Scholar 

  3. 3

    Terao R. et al. Supression of proliferation cholangitis in a rat model with direct adenovirus-mediated retinoblastoma gene transfer to the biliary tract Hepatology 1998 28: 605 605

    CAS  Article  Google Scholar 

  4. 4

    Lusky M. et al. In vitro and in vivo biology of recombinant adenovirus vectors with E1, E1/E2A, or E1/E4 deleted J Virol 1998 72: 2022 2022

    CAS  PubMed  PubMed Central  Google Scholar 

  5. 5

    Worgall S., Wolff G., Falck-Pedersen E., Crystal R.G. . Innate immune mechanisms dominate elimination of adenoviral vectors following in vivo administration Hum Gene Ther 1997 8: 37 37

    CAS  Article  Google Scholar 

  6. 6

    Drazan K.E. et al. Hepatic function is preserved following liver-directed, adenovirus-mediated gene transfer J Surg Res 1995 59: 299 299

    CAS  Article  Google Scholar 

  7. 7

    Lieber A. et al. The role of Kupffer cell activation and viral gene expression in early liver toxicity after infusion of recombinant adenovirus vectors J Virol 1997 71: 8798 8798

    CAS  PubMed  PubMed Central  Google Scholar 

  8. 8

    Salgado S. et al. Liver cirrhosis is reverted by urokinase-type plasminogen activator gene therapy Mol Ther 2000 2: 545 545

    CAS  Article  Google Scholar 

  9. 9

    Rudolph K.L. et al. Inhibition of experimental liver cirrhosis in mice by telomerase gene delivery Science 2000 287: 1253 1253

    CAS  Article  Google Scholar 

  10. 10

    Hecht N. et al. Hyper-IL-6 gene therapy reverses fulminant hepatic failure Mol Ther 2001 3: 683 683

    CAS  Article  Google Scholar 

  11. 11

    Nakatani T. et al. Assessment of efficiency and safety of adenovirus-mediated gene transfer into normal and damaged murine livers Gut 2000 47: 563 563

    CAS  Article  Google Scholar 

  12. 12

    Ramm G.A. et al. Contribution of hepatic parenchymal and nonparnchymal cells to hepatic fibrogenesis in biliary atresia Am J Pathol 1998 153: 527 527

    CAS  Article  Google Scholar 

  13. 13

    Kossakowska A.E. et al. Altered balance between matrix metalloproteinases and their inhibitors in experimental biliary fibrosis Am J Pathol 1998 53: 1895 1895

    Article  Google Scholar 

  14. 14

    Desmouliere A. et al. Extracellular matrix deposition, lysyl oxidase expression and myofibroblastic differentiation during the initial stages of cholestatic fibrosis in rat Lab Invest 1997 76: 765 765

    CAS  PubMed  Google Scholar 

  15. 15

    Zern M.A., Kresina T.F. . Hepatic drug delivery and gene therapy Hepatology 1997 25: 484 484

    CAS  Article  Google Scholar 

  16. 16

    de Roos W.K. et al. Isolated-organ perfusion for local gene delivery: efficient adenovirus-mediated gene transfer into the liver Gene Therapy 1997 71: 55 55

    Article  Google Scholar 

  17. 17

    Ferry N., Heard J.M. . Liver-directed gene transfer vectors Hum Gene Ther 1998 9: 1975 1975

    CAS  Article  Google Scholar 

  18. 18

    Nakamura T., Akiyoshi H., Saito I., Sato K. . Adenovirus-mediated gene expression in the septal cells of cirrhotic rat livers J Hepatol 1999 30: 101 101

    CAS  Article  Google Scholar 

  19. 19

    Aguilar L.K. et al. A prescription for gene therapy Mol Ther 2000 1: 385 385

    CAS  Article  Google Scholar 

  20. 20

    Sullivan D.E. et al. Liver-directed gene transfer in non-human primates Hum Gene Ther 1997 8: 1195 1195

    CAS  Article  Google Scholar 

  21. 21

    Siller-Lopez F. et al. Truncated active matrix metalloproteinase-8 gene expression in HepG2 cells is active against native type I collagen J Hepatol 2000 33: 758 758

    CAS  Article  Google Scholar 

  22. 22

    Lee S.S. et al. Hemodynamic characterization of chronic bile duct-ligated rats: effect of pentobarbital sodium Am J Physiol 1986 251: 176 176

    Google Scholar 

  23. 23

    Armendariz-Borunda J., Katayama K., Seyer J.M. . Transcriptional mechanisms of type I collagen gene expression are differentially regulated by IL-1β TNF-α and TGFβ in Ito cells J Biol Chem 1992 267: 14316 14316

    CAS  PubMed  Google Scholar 

  24. 24

    Nyberg-Hoffman C. et al. Sensitivity and reproducibility in adenoviral infectious titer determination Nat Med 1997 3: 808 808

    CAS  Article  Google Scholar 

  25. 25

    Mizuguchi H., Kay M.A. . Efficient construction of a recombinant adenovirus vector by an improved in vitro ligation method Hum Gene Ther 198 9: 2577 2577

    Article  Google Scholar 

  26. 26

    Weiss D.J., Liggitt D., Clark J.G. . In situ histochemical detection of β-galactosidase activity in lung: assessment of X-gal reagent in distinguishing lacZ gene expression and endogenus β-galactosidase activity Hum Gene Ther 1997 8: 1545 1545

    CAS  Article  Google Scholar 

  27. 27

    Lopez-De Leon A., Rojkind M. . A simple micromethod for collagen and total protein determination in formalin-fixed paraffin-ambedded sections J Histochem Cytochem 1985 33: 737 737

    CAS  Article  Google Scholar 

Download references


This work was supported in part by a grant from CONACyT No. 28832-M to Juan Armendariz-Borunda. The authors are indebted to Dr Guillermo Grijalva and Jose M Vera for their invaluable assistance.

Author information



Corresponding author

Correspondence to J Armendariz-Borunda.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Garcia-Bañuelos, J., Siller-Lopez, F., Miranda, A. et al. Cirrhotic rat livers with extensive fibrosis can be safely transduced with clinical-grade adenoviral vectors. Evidence of cirrhosis reversion. Gene Ther 9, 127–134 (2002).

Download citation


  • cirrhosis
  • gene transfer
  • safety
  • fibrosis reversion

Further reading


Quick links