Because of the shortage of deceased liver grafts, the demand for partial liver grafts from living donors and for the splitting of deceased donor grafts has been increasing worldwide. Such measures are necessary because liver transplantation is the only curative therapy for many patients with end-stage liver disease and/or acute liver failure.1, 2 However, the problems associated with small-for-size liver graft injuries remain major obstacles: small-for-size liver graft injuries are associated not only with high mortality rates and postoperative complications but also with acute rejection.3-5
The precise mechanisms of small-for-size liver graft injuries have not been completely clarified. A critical mechanism is associated with the severe shear stress due to transient portal hypertension just after reperfusion.6 Therefore, therapeutic treatments depend on portal decompression to protect small-for-size liver grafts. Various therapeutic strategies focusing on portal decompression have been reported for animal models.7, 8 Using a small-for-size liver graft model in rats, our previous study showed that a preservation solution including activated protein C significantly increased hepatic microcirculation and decreased portal pressure caused by an increased hepatic level of nitric oxide via up-regulated endothelial nitric oxide synthase expression together with down-regulated inducible nitric oxide synthase expression.9 However, there have been few clinical applications of these strategies.
Clinically, splenic artery ligation or splenectomy (SP) is the most effective surgical method for decreasing portal hypertension.10-12 Previous studies using a rat hepatic ischemia/reperfusion injury (IRI) model have shown that SP ameliorates hepatic IRI via the inhibition of leukocyte infiltration in the liver and via the release of tumor necrosis factor α (TNF-α), the induction of stress protein heme oxygenase 1 (HO-1), and the reduction of superoxide anion release into the hepatic sinusoids.13-15
However, the effects of SP on small-for-size liver graft transplantation have been unclear because, to the best of our knowledge, there have been no experimental data. Accordingly, this study was designed to evaluate the cytoprotective effects of SP on liver transplantation by quantifying survival, liver function, acute inflammatory responses, vascular tone, apoptosis, and liver regeneration after the transplantation of small-for-size liver grafts into rats. The results show that SP prevents excessive portal vein hepatic inflow and eliminates splenic inflammatory cell recruitment into the liver.
For all experiments, we used male Wistar Hannover rats weighing 190 to 240 g (CLEA Japan, Tokyo, Japan). The rats had free access to rat chow and water before the surgical procedures, and they were kept under constant environmental conditions with a 12-hour light-dark cycle. All experiments complied with the guidelines of Guide for the Care and Use of Laboratory Animals (National Institutes of Health publication 86-23, revised 1985).
Model of Small-for-Size Liver Transplantation
Rat partial orthotopic liver transplantation (OLT) without hepatic artery reconstruction was performed as previously described.9 This method was a modified version of the technique described by Kamada and Calne16 and Tanaka et al.17 Briefly, the rats were anesthetized with isoflurane. After systemic heparinization, the donor livers were harvested. The harvested livers were immediately flushed through the portal vein and stored in a histidine tryptophan ketoglutarate solution (Kohler Chemie, Alsbach, Germany) at 2 to 4°C. After the removal of the livers from the recipients, the partial liver grafts (right lobe) were anastomosed to the suprahepatic vena cava with 8-0 continuous sutures. The portal vein and infrahepatic vena cava were anastomosed with the cuff technique. The bile duct was anastomosed with an intraluminal stent. The transplantation procedure lasted less than 45 minutes, during which time the portal vein was clamped for 11 to 14 minutes.
Recipient rats were randomly assigned to 1 of 3 groups to determine the graft preservation time. The grafts were flushed and stored in a histidine tryptophan ketoglutarate solution for 2, 6, or 20 hours. Five rats in each group were used for the survival study. Rats that had lived more than 7 days after partial OLT were considered survivors. The 7-day animal survival rates for 2, 6, and 20 hours of storage were 100% (5/5), 60% (3/5), and 0% (0/5), respectively. On the basis of these results and clinical data for split liver transplantation (mean cold ischemia time = 5.8 ± 2.2 hours),18 the preservation time was determined to be 6 hours.
On the basis of our preliminary study, the liver grafts were assigned to 2 groups: in the control group (n = 26), OLT alone was performed with small-for-size liver grafts, and in the SP group (n = 24), OLT after SP was performed with small-for-size liver grafts.
Liver tissues and blood were sampled 6 and 24 hours after OLT for hepatic protein detection, morphological examinations, and liver function tests. Five rats were included at each time point for the control and SP groups, respectively. Because 2 rats in the control group died within 24 hours after OLT, 2 rats were added to the control group 24 hours after OLT.
Ten rats from the control group and 10 rats from the SP group were used for the survival study. Rats that had lived more than 7 days after transplantation were considered survivors.
Measurement of Portal Pressure and Hepatic Microcirculation
Four rats from the control group and 4 rats from the SP group were used for the measurement of the portal pressure and hepatic microcirculation. The ileocolic vein was cannulated with a 24-gauge catheter (Terumo, Tokyo, Japan) to measure the mean portal pressure as described previously.19 This catheter was connected via a pressure transducer (BSM-3201, Nihon Kohden Corporation, Shinjuku-ku, Tokyo, Japan). The hepatic microcirculation was measured with a laser Doppler flowmeter (BRL-100, Bio Research Center, Nagoya, Japan) for 60 minutes of reperfusion as described previously.20 The results were expressed as percentages of the predonation level (the whole size).
Measurement of Serum Transaminase Levels
Liver injury was quantified by the measurement of serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels with the Wako test for transaminases (Wako, Osaka, Japan) according to the manufacturer's instructions.
Myeloperoxidase (MPO) Assay
The activity of MPO was used as an index of hepatic leukocyte accumulation.21 Frozen tissues were homogenized in iced 0.5% hexadecyltrimethylammonium bromide (Sigma-Aldrich, St. Louis, MO) and a 50-mmol potassium phosphate buffer solution (pH 5.6; Sigma-Aldrich). After centrifugation, the supernatants were mixed with a solution of hydrogen peroxide sodium acetate and tetramethylbenzidine (Sigma-Aldrich). The change in absorbance was measured spectrophotometrically at 655 nm. One unit of MPO activity was defined as the quantity of the enzyme degrading 1 μmol of peroxide/minute/g of tissue at 25°C.
Measurement of Hepatic TNF-α and Interleukin-6 (IL-6) Protein Levels
Liver samples were homogenized in an extraction buffer [50 mmol/L trishydroxymethylaminomethane (pH 7.2), 150 mmol/L sodium chloride, and Triton X-100] and a protease inhibitor cocktail. The homogenates were shaken on ice for 90 minutes and centrifuged at 3000g and 4°C for 15 minutes. The hepatic levels of TNF-α and IL-6 were determined with commercially available enzyme-linked immunosorbent assays (BD Biosciences, San Diego, CA) according to the manufacturer's instructions.
Measurement of Plasma Endothelin 1 (ET-1) Levels
Plasma ET-1 levels were measured with an assay kit (Cayman Chemical, Ann Arbor, MI) according to the manufacturer's instructions.
Liver specimens were fixed in a 10% buffered formalin solution and embedded in paraffin. Sections were then prepared and stained with hematoxylin and eosin. The histological severity of hepatic IRI was graded with a modification of the Suzuki criteria.22 In accordance with this classification, sinusoidal congestion, hepatocyte necrosis, and ballooning degeneration were graded from 0 to 4. Sections exhibiting necrosis, congestion, or centrilobular ballooning were given a score of 0, whereas those with severe congestion and ballooning degeneration and 60% lobular necrosis were given a score of 4.
Paraffin-embedded tissue sections of livers and spleens were deparaffinized and rehydrated, and this was followed by a proteinase K treatment (Chemicon, Temecula, CA). Next, the sections were incubated in 0.3% hydrogen peroxide in methanol to block endogenous peroxidase activity. A primary antibody against ED-1 (CD68, Chemicon) and HO-1 (Stressgen, Victoria, Canada) was added at an optimal dilution. The secondary antibody and the ABC reagents were applied according to the manufacturer's instructions (Vectastain ABC kit, Vector, Burlingame, CA). Color development was induced by incubation with a 3,3-diaminobenzidine substrate (Vector). The neutrophils were stained with a naphthol AS-D chloroacetate esterase staining kit (Sigma-Aldrich) according to the manufacturer's instructions. Using 20 high-power fields (HPFs) per section (×200), we evaluated the results by counting labeled cells in triplicate.
Measurement of Serum Hyaluronic Acid (HA) Levels
Serum HA levels were measured as a sinusoidal cell function with an assay kit (Fujirebio, Tokyo, Japan) according to the manufacturer's instructions.
Measurement of Caspase-3 and Caspase-8 Activity
Caspase-3 and caspase-8 activity were measured with an assay kit (Clontech Laboratories, Mountain View, CA) according to the manufacturer's instructions.
Detection of Apoptosis
Apoptosis was detected with the terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick-end labeling (TUNEL) method. Paraffin-embedded tissue sections were deparaffinized, and the dehydrated sections were treated with a peroxidase in situ cell death detection kit (Roche Diagnostics, Temecula, CA) according to the manufacturer's instructions. The results were scored semiquantitatively through the averaging of the number of TUNEL-positive cells per field; there were 6 fields per tissue sample.
Evaluation of Liver Regeneration
Rats were treated with 50 mg/kg bromodeoxyuridine (BrdU; Sigma, St. Louis, MO) intraperitoneally 1 hour before sacrifice. Sections with a thickness of 5 μm were prepared from formalin-fixed and paraffin-embedded liver samples. BrdU incorporation was evaluated immunohistochemically with an anti-BrdU antibody (Dako, Glostrup, Denmark). Using 20 HPFs per section (×200), we evaluated the results by counting labeled cells in triplicate. The regeneration rate was calculated as follows:
The data were expressed as medians and ranges (minimum to maximum). Differences between the distributions of the control and SP groups were determined with the Mann-Whitney U test with Bonferroni multiple comparison tests. For the survival study, a Kaplan-Meier log-rank analysis was performed. All differences were considered statistically significant with P values < 0.05.
SP Prolonged Animal Survival and Ameliorated Hepatocellular Injury by Improving Hepatic Microcirculation and Decreasing the Initial Portal Pressure
The SP group showed significantly increased 7-day animal survival in comparison with the control group [100% (10/10) versus 50% (5/10), P < 0.05]. The hepatic microcirculation was markedly improved in the SP group after reperfusion (P < 0.05; Fig. 1A). The portal pressure was significantly reduced in the SP group after reperfusion in comparison with the control group (P < 0.05; Fig. 1B). Serum levels of AST and ALT were markedly decreased in the SP group in comparison with the control group 6 and 24 hours after OLT (Fig. 2A,B). According to Suzuki's histological classification with hematoxylin and eosin staining (Fig. 2C-F), the control group showed moderate to severe hepatocyte vacuolization with disruption of the lobular architecture and moderate sinusoidal congestion; the Suzuki scores were 5 (range = 3-7) and 6 (range = 5-6) at 6 and 24 hours, respectively. In contrast, the SP group showed minimal hepatocyte vacuolization and sinusoidal congestion 6 and 24 hours after OLT; the Suzuki scores were 2 (range = 2-2) and 4 (range = 3-5) at 6 and 24 hours, respectively (P < 0.05).
SP Prevented the Infiltration of Neutrophils and Macrophages by the Direct Elimination of Neutrophils and Macrophages in the Spleen
The numbers of naphthol AS-D–positive cells (Fig. 3A-C) and ED-1–positive cells (Fig. 3D,E) were decreased in comparison with the numbers in naive animals 6 and 24 hours after OLT. By the same token, the MPO activity of the spleen tissues was significantly reduced in comparison with the MPO activity in the naive animals 6 and 24 hours after OLT (P < 0.05; Fig. 3G). The SP group showed significantly decreased intrahepatic infiltration of naphthol AS-D–positive cells (P < 0.05; Fig. 4A-E) and ED-1–positive cells (P < 0.05; Fig. 4F-J) in comparison with the control group 6 and 24 hours after OLT. The MPO activity of the liver tissues was significantly reduced in the SP group versus the control group 6 and 24 hours after OLT (P < 0.05; Fig. 4K). The SP group showed in comparison with the control group significantly decreased hepatic expression of TNF-α [18.6 pg/mg of protein (range = 13.6-20.5 pg/mg of protein) versus 27.4 pg/mg of protein (range = 23.8-31.0 pg/mg of protein), P < 0.05] and IL-6 [218 pg/mg of protein (range = 170-248 pg/mg of protein) versus 282 pg/mg of protein (range = 239-335 pg/mg of protein), P < 0.05] 6 hours after OLT.
SP Reduced Endothelial Cell Injury and Plasma ET-1 Levels and Increased Hepatic Expression of HO-1
The SP group showed markedly decreased serum HA levels in comparison with the control group 6 [231 pg/mL (range = 140-378 pg/mL) versus 532 pg/mL (range = 415-704 pg/mL)] and 24 hours after OLT [598 pg/mL (range = 581-698 pg/mL) versus 891 pg/mL (range = 793-1030 pg/mL), P < 0.05].
The SP group showed significantly decreased plasma ET-1 levels in comparison with the control group only 6 hours after OLT [0.94 pg/mL (range = 0.90-1.16 pg/mL) versus 1.39 pg/mL (range = 1.20-1.72 pg/mL), P < 0.05]. The SP group showed significantly increased hepatic expression of HO-1 in comparison with the control group 6 and 24 hours after OLT (P < 0.05; Fig. 5A-E).
SP Promoted Antiapoptotic Actions and Improved Hepatic Regeneration
Six hours after OLT, liver tissues of the SP group showed significantly decreased caspase-3 and caspase-8 activity and fewer TUNEL-positive cells in comparison with the control group (P < 0.05; Fig. 6A-E). Twenty-four hours after OLT, the SP group showed a significantly increased hepatic regeneration rate and a significant increase in intrahepatic BrdU-positive cells in comparison with the control group (P < 0.05; Fig. 7A-D).
The major causes of small-for-size liver graft failure in the early phase after OLT are (1) hepatic sinusoidal damage caused by transient portal hypertension due to hemodynamic forces, (2) severe inflammatory responses triggered by shear stress, and (3) subsequent hepatic microcirculatory disorders due to an imbalance of vascular mediators, a higher metabolic burden for regeneration, and the up-regulation of apoptotic signals.7, 23-26 Therefore, we generally perform SP when the portal vein pressure is 20 mm Hg or higher after living donor liver transplantation (LDLT) on the basis of previous data.27 Ogura et al.28 reported that patients with a portal pressure < 15 mm Hg after LDLT demonstrated better 2-year survival than patients with a portal pressure of 15 mm Hg or higher, and they concluded that intentional portal vein pressure control (<15 mm Hg) by SP seemed to be a key for successful LDLT.28 SP is a well-known technique for controlling portal pressure after LDLT. However, the precise mechanisms of SP with respect to small-for-size graft injuries have not been reported. To the best of our knowledge, this is the first report demonstrating cytoprotective effects of SP on small-for-size liver transplantation in an experimental model. In the present study, SP exerted cytoprotective effects on small-for-size liver transplantation in rats. SP (1) attenuated liver damage and improved hepatic microcirculation while decreasing portal pressure, (2) prevented the activation/accumulation of inflammatory cells and proinflammatory cytokine expression, (3) attenuated sinusoidal endothelial injury, (4) increased hepatic HO-1 expression and decreased plasma ET-1 levels, (5) exerted antiapoptotic effects, and (6) improved liver regeneration.
Generally, the splenic vein flow accounts for 20% to 40% of the total portal vein flow; thus, the present study demonstrates that SP immediately decreases portal pressure after reperfusion. Interestingly, SP increases hepatic microcirculation. Small-for-size liver graft injuries are probably related to microcirculatory disorders due to an imbalance of vasoconstricting and vasorelaxing mediators caused by excessive shear stress.24 Additionally, sinusoidal endothelial cells (SECs) are important in regulating hepatic vascular resistance, and SEC injuries contribute to increased hepatic vascular resistance and lead to high portal pressure.25 In particular, ET-1 derived from SECs exerts a powerful vasoconstricting effect, and it is a key mediator controlling vascular tone. In the present study, SP prevented SEC injury (as evidenced by HA levels) and decreased plasma ET-1 levels. However, the expression of hepatic nitric oxide, which induces a strong vasorelaxing effect, was comparable in the 2 groups after partial OLT (data not shown). Additionally, in the present study, SP significantly increased the hepatic expression of HO-1 (heat shock protein 32). HO-1 is responsible for the breakdown of heme into equimolar amounts of biliverdin, iron, and carbon monoxide. Therefore, HO-1 plays an important role in the production of carbon monoxide, which serves as a vasodilator in the hepatic sinusoids.29, 30 The previous study using a rat IRI model speculated about the mechanism by which HO-1 is highly expressed in the spleen; this includes high levels of heme released from hemoglobin, so this function of the spleen might be replaced by the liver after SP.14 Accordingly, SP appears to prevent hepatic microcirculatory disorders not only by directly decreasing the portal vein inflow but also by inducing vasodilation through decreased plasma ET-1 levels and increased hepatic HO-1 expression; this in turn attenuates liver damage and prolongs animal survival. Additionally, previous studies have demonstrated that early graft function after OLT is correlated with the initial hepatic microcirculation just after reperfusion,31, 32 and this also supports the findings of the present study.
The spleen contains 15% of the fixed-tissue macrophages in the body and generates proinflammatory cytokines in response to inflammatory stimulation.33 It is thought that splenic monocytes/macrophages are an important part of the mononuclear phagocytic system.34 Previous studies employing rat hepatic IRI models demonstrated that SP attenuated not only liver damage but also damage to remote organs, including the lungs, kidneys, and intestines.15, 35 The authors of these studies concluded that splenic monocytes/macrophages were directly removed from the body via SP, so the number of mononuclear phagocytic system cells contributing to remote organ injuries was reduced. In the present study, we found that the number of inflammatory cells (neutrophils and monocytes/macrophages) in the spleen after OLT was decreased in comparison with the number in naive animals. In contrast, the number of inflammatory cells in the liver was increased after OLT. Therefore, we speculate that splenic inflammatory cells, which produce inflammatory cytokines, immediately accumulate in the graft liver after OLT, and this in turn accelerates liver damage. Accordingly, it is thought that SP before OLT is the most effective method for preventing splenic inflammatory cell recruitment into the liver.
Apoptosis has been identified as a key mechanism in hepatic IRI.36 Liang et al.37 found more apoptotic cells in small-for-size grafts versus whole or half-size grafts after OLT in rats.37 In the present study, SP significantly reduced the activation of caspase-8 and caspase-3, and this resulted in a decreased number of TUNEL-positive cells, which is consistent with antiapoptotic effects.
A large number of genes are involved in liver regeneration, but the essential circuitry required for the process may be categorized into 3 networks: cytokine, growth factor, and metabolic.38 Several previous studies using a rat partial hepatectomy model demonstrated that SP enhanced liver regeneration, released from the spleen, transforming growth factor β1 exerted a growth inhibitory effect on liver regeneration,39 and SP enhanced liver regeneration because of increased levels of hepatocyte growth factor.40 Similarly, the present study demonstrated that SP significantly increased the liver regeneration ratio and the BrdU uptake of hepatocytes 24 hours after OLT. However, the underlying mechanism of liver regeneration induced by SP was not sufficiently explored in this study. Recently, Lesurtel et al.41 reported that platelet-derived serotonin was involved in the initiation of liver regeneration in a mouse partial hepatectomy model. Using a mouse small-for-size liver graft transplantation model, Tian et al.42 demonstrated that serotonin improved graft failure through a 5-hydroxytryptamine receptor 2B pathway by preserving the hepatic microcirculation, which in turn facilitated liver regeneration. Thus, in the present study, the platelet count may have been clinically increased after SP, and we speculate that the increase in platelet-derived serotonin after SP had a critical effect on liver regeneration.
Clinically, however, SP has been thought to increase perioperative complication rates and increase risks of posttransplant sepsis; thus, simple ligation of the splenic artery, which is a less radical way of reducing portal flow in comparison with SP, seems to be much safer. SP or splenic artery ligation is clinically performed to prevent excess shear stress. In the past, SP has been considered an invasive technique in comparison with ligation of the splenic artery. However, SP has recently become a relatively safe and easy procedure with the development of various surgical devices, such as LigaSure (Valleylab, Boulder, CO) and the Endo GIA stapler (United States Surgical Corp., Norwalk, CT). With respect to portal pressure, it has been reported that SP clinically reduces excessive portal hypertension in comparison with splenic artery ligation in adult living related donor liver transplantation.43 Additionally, Morinaga et al.44 reported that SP significantly improved liver regeneration with a reduction of plasma transforming growth factor β1 levels in comparison with splenic artery ligation in a dimethylnitrosamine-induced cirrhosis rat model. This report indicates that it is impossible to prevent the expression of cytokines from the spleen after splenic artery ligation. SP not only decreases the portal pressure but also eliminates the inhibitory factors of liver regeneration.
However, SP has a negative impact because it increases the perioperative risk of sepsis after transplantation. Additionally, it has been reported that SP is a major risk factor for the development of opportunistic pneumonia after liver transplantation.45 On the other hand, it has been reported that SP does not have a negative impact on the risk of sepsis after transplantation.28 Recently, the concept of an overwhelming post-SP infection has been established: to combat overwhelming post-SP infections, various guidelines recommend medical treatment with the pneumococcal polysaccharide vaccine and/or a prophylactic antibiotic (sulfamethoxazole/trimethoprim).
In conclusion, SP prevents excessive portal vein hepatic inflow and eliminates splenic inflammatory cell recruitment into the liver, and this in turn inhibits hepatocellular apoptosis and improves liver regeneration. However, the precise cytoprotective mechanism induced by SP is not yet clear. Further studies are required to investigate other vascular mediators, apoptotic factors, and inhibitors of liver regeneration. These studies might shed light on therapeutic strategies for OLT with small-for-size grafts.