Hiromi Iwagaki MD, Department of Gastroenterological Surgery, Transplant, and Surgical Oncology, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata, Okayama 700-8558, Japan. Tel.: +81 86 235 7255; fax: +81 86 221 8775; e-mail: firstname.lastname@example.org
Signal transducer and activator of transcription-3 (STAT3) is one of the most important transcription factors for liver regeneration. This study was designed to examine the effects of constitutively activated STAT3 (STAT3-C) on post-transplant liver injury and regeneration in a rat 20% partial liver transplant (PLTx) model by ex vivo adenoviral gene transfer. Adenovirus encoding the STAT3-C gene was introduced intraportally into liver grafts and clamped for 30 min during cold preservation. After orthotopic PLTx, liver graft/body weights and serum biochemistry were monitored, and both a histological study and DNA binding assay were performed. STAT3-C protein expression and its binding to DNA in the liver graft were confirmed by Western blotting and electrophoretic mobility shift assay (EMSA), respectively. This treatment modality promoted post-Tx liver regeneration effectively and rapidly. The serum levels of alanine aminotransferase/aspartate aminotransferase (AST/ALT) and bilirubin decreased in rats with STAT3-C. However, albumin (a marker of liver function) did not. Ex vivo gene transfer of STAT3-C to liver grafts reduced post-Tx injury and promoted liver regeneration. Thus, the activation of STAT3 in the liver graft may be a potentially effective clinical strategy for improving the outcome of small-for-size liver transplantation.
The first successful living-donor liver transplantation (LDLT) in an adult patient was reported by Hashikura et al.  in 1994. Since then, increasing numbers of adult-to-adult LDLT have been performed. As a result, this operation has now become an established standard surgical method for the treatment of late-stage liver diseases . At present, a small-for-size liver graft to the recipient remains a major obstacle to successful LDLT, because the left lobe is usually transplanted in order to reduce surgical stress to the donors. In small-for-size liver transplant (Tx), the reduction of postischemic oxidative injury seems to be more important than in whole liver Tx, and a rapid recovery of the liver mass and function is essential to avoid primary nonfunction of the graft.
Postischemic oxidative injury is one of the critical causes of injury in reperfused tissue , and it is still a major issue in organ transplantation. Accumulating data suggest that oxidative stress plays a critical role in inducing postischemic tissue injury [4–7]. Intracellular production of reactive oxygen species (ROS) has been reported at a very early phase of reperfusion , which consequently activates redox-dependent molecules such as nuclear factor (NF)-κB, and induces apoptosis [9–11]. Therefore, suppressing oxidative stress and the redox-sensitive pro-apoptotic pathway in postischemic tissue may be essential to prevent ischemia/reperfusion (I/R)-induced liver injury.
Signal transducer and activator of transcription-3 is an important transcription factor in hepatic parenchymal and nonparenchymal cells which mediates mainly mitogenic signals after binding to various ligands including the interleukin-6 (IL-6) family, epidermal growth factor (EGF) and platelet-derived growth factor (PDGF), and by a number of receptor- and nonreceptor-tyrosine kinases [12–16]. STAT3 was originally recognized as a molecule that was capable of selectively interacting with an enhancer element in the promoter of acute-phase response genes following stimulation by IL-6 [17, 18]. Its activation requires either phosphorylation on tyrosine by cytokine receptor-associated kinases (Jak), growth factor receptor-tyrosine kinases, or nonreceptor-tyrosine kinases. Once activated, STAT3 forms homo- or hetero-dimers, translocates to the nucleus, and finally binds to the specific promoters of its target genes to induce gene expression . The physiological functions of STAT3 have been extensively studied, and it is clear that it plays a crucial role in both organ development and cell proliferation . Studies using liver-specific conditional knockout mice have also indicated that STAT3 is an important initiator of hepatocellular proliferation [20–22]. Other than cell proliferation, anti-apoptotic and anti-oxidant properties of STAT3 have recently been reported in a variety of cells including hepatocytes. STAT3 activation protects hepatocytes against Fas-induced apoptosis and hypoxia/reoxygenation (H/R)-induced damage through the up-regulation of Mn-SOD, redox factor-1 (Ref-1), and anti-apoptotic genes (Bcl-2/-xL) . It also protects fibroblasts against apoptosis induced by serum-withdrawal or UV , and protects cardiomyocytes against H/R-induced cellular damage . Therefore, the activation of STAT3 in liver grafts may be a strategy for both preventing injury and promoting regeneration after partial liver transplant (PLTx). In the present study, we performed ex vivo adenoviral gene transfer of STAT3 to small size liver grafts (20%), and examined whether the constitutive activation of STAT3 can thereby reduce liver injury, while promoting liver regeneration after small-size liver Tx.
Materials and methods
Animal models and gene delivery
A cDNA construct of STAT3-C (the constitutively active form of STAT3), which was kindly provided by Dr James E. Darnell (Rockefeller University, New York, USA), was made by substituting cysteine residues for A661 and N663 of murine STAT3 and then tagged with FLAG. This renders the STAT3-C molecule capable of dimerizing without causing the phosphorylation of Y705 . A replication-deficient adenoviral vector encoding STAT3-C (AxCAS3-C) was constructed as previously described . AdLacZ, an adenovirus encoding inert bacterial β-galactosidase, was used as a control vector. All viruses were amplified in HEK 293 cells, purified on double cesium gradients, and then plaque titered.
C57BL/6 mice (male, 20–25g) were used to examine the effect of STAT3-C on liver mass and hepatocyte proliferation. AxCAS3-C was injected intravenously into the mice (1 × 107 to 5 × 108 pfu/body). The liver/body weight ratios were evaluated chronologically and liver histology [hematoxylin and eosin (H & E)] was also examined. The mice were followed up to 2 months after injection.
Rat partial hepatectomy and ex vivo gene transfer was performed as previously described [26, 27]. Briefly, adult male Lewis rats weighing 230–250 g were used as donors and recipients. The rats were subjected to a mid-ventral laparotomy and an approximately 80% liver resection (left, middle lobe and caudate lobe). The remaining liver lobe (right lobe) was then used as the graft. The 20% liver grafts were washed with ice-cold physiological saline (PS), and immediately infused with PS, AdLacZ (1 × 108 pfu) or AxCAS3-C (1 × 108 pfu) through the portal vein (total volume, 0.6 ml). The IVC and portal vein (PV) of the liver graft were then clamped and preserved in ice-cold PS for 30 min. Finally, the liver graft was flushed with ice-cold PS to remove any intra-vascular adenovirus in order to avoid transferring the gene into the recipient's other organs prior to transplantation.
Orthotopic liver transplantation of the partial liver grafts was performed with revascularization but without hepatic artery reconstruction, as previously described [26, 27]. In brief, the donor liver was flushed via the portal vein with ice-cold PS, and the suprahepatic vena cava was anastomosed by 7–0 continuous sutures. The portal vein and infrahepatic vena cava were then anastomosed by the cuff technique. More than 95% of the rats survived this operative procedure. Five rats per group were sacrificed 1, 3 and 5 days after PLTx.
The animals were maintained under standard conditions and were treated according to the Guidelines for the Care and Use of Laboratory Animals of the Okayama University Graduate School of Medicine and Dentistry.
Analysis of STAT3 protein expression and its DNA binding
For Western blotting, 30 μg of whole liver protein extract were separated by 10% SDS–PAGE and transferred to a nitrocellulose membrane. Antibodies against STAT3, Tubulin (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and FLAG (Sigma-Aldrich Co., St. Louis, MO, USA) were used as the primary antibodies. The STAT3 DNA binding activity was assayed using 32P-labeled SIE (sis-inducible element)-m67 oligonucleotide as a probe (5′-actgGGATTTTTCCCGTAAATGGTC-3′). Next, nuclear protein extract (5 μg) was reacted with 105 cpm of 32P-labeled STAT3-binding consensus oligonucleotides for 30 min in a binding buffer (10 mm tris, pH 7.4; 80 mm KCl; 5% glycerol; 1 mm DTT; 0.25 μg dIdC) at room temperature. The incubation mixtures were run on a 5% polyacrylamide gel at 4 °C, and autoradiographed.
For X-gal staining in order to detect the expression of the LacZ gene, frozen sections (4-μm thick) were prepared and fixed in 1.5% glutaraldehyde in cold PBS. They were then incubated with X-gal solution (0.5 mg/ml) at 37 °C and mounted.
Analysis of regeneration and liver injury
In order to examine both liver regeneration and hepatocyte proliferation after liver transplantation, the liver graft was excised 72 h after PLTx and the ratio of liver weight to body was estimated. Formalin-fixed, paraffin-embedded specimens were prepared and stained with H & E. The liver specimens were histo-pathologically examined and the number of mitoses was counted to assess the degree of hepatocyte proliferation. The number of hepatocytes mitoses was counted in three grafts per group, and 3 × 100 hepatocytes were counted for each rat. The liver specimens were also immunostained with anti-proliferating cell nuclear antigen (PCNA) (Santa Cruz Biotechnology, SantaCruz, CA, USA).
Serum samples were drawn at various time-points after PLTx and used to measure alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin, and albumin levels using a standard automatic analyzer (Type 7150; Hitachi, Japan).
Cell Death Detection-ELISA Plus (Roche Diagnostics Corp., Basel, Switzerland) was used for the quantitative evaluation of apoptotic cells in liver tissue. Liver specimens were prepared before and 48 h after LTx. Lysates of liver tissue were directly applied for this assay (40 mg of protein), which was performed according to the manufacturer's recommendations. Five micrograms of liver tissue was also used for Bcl-2 protein detection (anti-Bcl2, Santa Cruz Biotechnology, SantaCruz, CA, USA).
The results are expressed as means ± SEM. Significant differences between two or more groups were identified by the unpaired Students’t-test or the Fisher test, respectively. P-values of less than 0.05 were considered statistically significant.
STAT3-C increases liver mass and hepatocyte proliferation
Intravenous injection of AxCAS3-C into mice caused an enlargement of the liver in a dose-dependent manner without any tumor formation (Fig. 1a). Lower viral titers resulted in some mitogenic foci in liver tissue with more hepatocytes showing mitosis or large nuclei with increased amounts of hyperchromatism. The expression of STAT3-C peaked 3 days after injection, remained strongly expressed for at least 2 weeks, and diminished thereafter and was no longer present within 2 months (Fig. 1b). The liver mass increased in size and recovered in parallel to the amount of STAT3-C protein in the liver tissue.
Successful gene transfer and protein expression of STAT3-C in liver grafts using an ex vivo adenoviral gene transfer method.
Ex vivo gene transfer and orthotopic PLTx were performed as described in Materials and methods. The presence of protein in the liver graft and other organs in the recipient after PLTx were analyzed using X-gal staining in the LacZ group. Ex vivo gene transfer of LacZ resulted in protein expression almost exclusively in the liver tissue. Three days after PLTx, about 30–40% of the hepatocytes stained positive for X-galactosidase. Other than in the liver, almost no positive cells were observed except in the spleen, where some lymphocytes in germinal centers were X-gal-positive (data not shown).
Western blotting confirmed that ex vivo transfer of the STAT3-C gene resulted in a significant increase in FLAG-tagged STAT3-C protein in the liver tissue 72 h after PLTx, whereas AdLacZ treatment had no effect on STAT3 protein levels (Fig. 2a). EMSA revealed that STAT3 DNA binding in the liver tissue increased only transiently and weakly 24 h after PLTx in the LacZ group, but strongly and persistently increased up to 72 h in the STAT3-C group (Fig. 2b).
Rapid increase of liver graft weight after PLTx
Macroscopically, liver grafts over-expressing STAT3-C showed a more significant enlargement than controls (Fig. 3a). The ratio of the liver weight to body weight documented that STAT3-C had a significant effect on increasing liver weight (P-value, 0.014). Hepatocyte proliferation was assessed by counting mitotic cells under the microscope 72 h after PLTx. A few mitotic hepatocytes were observed in LacZ liver grafts at this time, but there were many more in the STAT3-C liver (P-value, 0.038, Fig. 3b). Staining of liver tissue for PCNA also revealed that over-expression of STAT3-C increased the number of hepatocytes in G1 to S phase (Fig. 3c). These data indicate that rapid liver regeneration in STAT3-C rats can be attributed to accelerated mitosis caused by STAT3-C.
Improvement of liver recovery and function after injury in STAT3-C rats
To examine the effect of over-expressed STAT3-C on liver graft injury induced by ischemia and/or reperfusion, we measured serum levels of AST/ALT 3 days after PLTx. Over-expression of STAT3-C significantly reduced the increase of AST/ALT in comparison to the LacZ group (P-values, 0.045 and 0.039 in AST and ALT, respectively, Fig. 4a). Serum bilirubin levels were measured 3 days after PLTx to assess hepatic metabolic function. STAT3-C rats were found to have significantly lower levels of bilirubin (P-value, 0.047, Fig. 4a). Serum levels of albumin were also measured to assess liver protein synthesis; levels of this protein were maintained after PLTx (P value, 0.044 at 72 h, Fig. 4b).
In order to examine the mechanism of STAT3-C protection, we analyzed apoptotic cell death in liver tissue (Fig. 4c). This was slightly but significantly reduced in STAT3-C rats compared to LacZ controls 2 days after LTx (P value, 0.034). Consistent with this, Bcl-2 protein was increased in STAT3-C but not LacZ liver 48 h after LTx.
These data therefore suggest that STAT3 protects the liver from injury while also helping to preserve its function as reflected by serum levels of AST/ALT and bilirubin/albumin, respectively.
Here, we examined the effectiveness of an ex vivo gene transfer method in a rat model of PLTx. Adenoviral transfer of the target gene via the portal vein during liver preservation seemed to be an ideal method for gene delivery to the graft, which would be safe for both donor and recipient. We showed that the STAT3-C gene could be introduced into liver grafts in this way and that it remained functionally active for at least 72 h after PLTx, in terms of protein expression and DNA binding. To our knowledge, this is the first report describing the efficacy of STAT3-C exclusively over-expressed in the liver following ex vivo adenoviral gene transfer, for promoting liver regeneration as well as for preventing postischemic liver injury.
It has been reported that STAT3 may be involved in tumor cell progression [28, 29]. Therefore, tumorigenicity of STAT3-C, even if the protein is only transiently expressed in liver tissue by adenoviral gene transfer, must be a serious concern. However, we observed that STAT3-C transfer resulted in a reversibly enlarged liver mass by inducing dose-dependent hepatocyte proliferation without tumor formation in a mouse model. In addition, most of this over-expressed STAT3-C protein had disappeared within 2 months. We have also confirmed that this ex vivo method employing adenovirus can deliver genes exclusively to the liver graft in recipient rats after PLTx, with the exception of some lymphocytes in the spleen. These findings on the ex vivo adenoviral gene transfer method support the effectiveness of this approach as a clinical strategy.
Signal transducer and activator of transcription-3 is generally accepted to be located downstream of the IL-6 signal transduction pathway. It binds with high affinity to the c-fos promoter and is thus responsible for its activation during mitogenic stimulation [30–34]. Furthermore, the rapid activation of the STAT3 transcription complex after partial hepatectomy has already been shown , and it is also recognized as an early factor crucially involved in liver regeneration . In the present study, STAT3 functioned as a strong mitogenic transcription factor, according to expectations. It also reduced post-Tx liver injury (assessed by serum AST/ALT levels), and maintained liver function (assessed by serum bilirubin/albumin levels). STAT3 possesses anti-apoptotic properties in various cells, according to some previous reports. Studies on the pro-survival actions of STAT3, as a potential oncoprotein, have been especially focused on tumor progression [28, 29]. As a result, the anti-apoptotic and anti-oxidant properties of STAT3 have been closely studied over the past few years. Some unique roles of STAT3, other than mitogenicity, are now widely recognized in nontumor cells as well as in tumor cells [37–39].
It is noteworthy that both extensive apoptosis and necrosis were observed after a simple partial hepatectomy in IL-6 knockout mice with defective STAT3 activation . In addition, liver regeneration following d-galactosamine-induced liver failure was significantly accelerated by prior administration of IL-6 and its soluble receptor . These findings may suggest that hepatocellular apoptosis plays a potential role in liver regeneration, although its pathophysiological relevance remains controversial. Survivin, Bcl-xL, and Bcl-2, which have recently been reported to act as anti-apoptotic molecules in some tumor cells, are also induced by activated STAT3 in primary cultured hepatocytes . STAT3 may play an important role in liver regeneration by preventing apoptosis and/or necrosis, thereby contributing to underlying mechanisms of liver regeneration. It is also known that STAT3 targets anti-oxidant genes, such as Mn-SOD and Ref-1. The induction of Mn-SOD by STAT3 has been shown in hepatocytes  and cardiomyocytes , and is possibly involved in redox-dependent signal regulation in the cell. STAT3 also transcriptionally up-regulates Ref-1, which regulates DNA binding of redox-sensitive transcription factors such as NF-κB and activator-protein-1 (AP-1). These factors are certainly involved in the initiation of liver regeneration. Ref-1 also leads to the inactivation of caspase-3, which would have an anti-apoptotic effect in the liver . STAT3 may also play a role in protection against postischemic oxidative liver injury. Additional studies are required to establish the role of anti-apoptotic/anti-oxidative properties of STAT3 in this experimental model of PLTx, as we do not have any data on these target molecules thus far.
In conclusion, the results of the present study provide the first evidence that STAT3-C introduced into liver grafts by adenoviral vectors not only protected them from I/R-induced injury, but also accelerated their regeneration in a PLTx model. Although the precise mechanisms of these effects of STAT3-C on liver grafts require further examination, STAT3 delivery, by reducing injury and promoting regeneration, may be a potentially effective clinical strategy for improving the outcome of small-for-size liver transplantation.