Hepatology highlights



Controversy exists as to the extent and importance of hepatocyte apoptosis in cholestatic disease. It has been shown previously that genetic Fas deficiency (1pr mice) protects mice from cholestatic injury. Gujral et al. explored the role of apoptosis and Fas in bile duct–ligated mice. Three days after bile duct ligation, marked oncotic necrosis, inflammation, and elevated alanine aminotransferase were seen in wild-type mice (Fig. 1) and were markedly attenuated in 1pr mice. However, no increase in apoptosis was observed, but Fas deficiency was accompanied by diminished necrosis (Fig. 1), neutrophil infiltration, chlorotyrosine staining (marker of neutrophil oxidant stress), and proinflammatory gene expression. Large doses of pancaspase inhibitor did not affect injury in the wild-type mice. Moreover, no protection from Fas deficiency was observed in another mouse strain inherently prone to less inflammation and injury. Thus, despite the convincing lack of apoptosis of hepatocytes following bile duct ligation, Fas deficiency attenuated injury, which seems to be linked to decreased inflammatory signaling. However, the authors have ruled out apoptosis as the inflammatory signal. Precisely how Fas participates in the neutrophilic infiltration remains an open question. The neutrophil infiltration and chlorotyrosine staining correlated closely with the extent of necrosis, suggesting a major role for inflammation in the injury. (See HEPATOLOGY 2004;40:998–1007.).

Illustration 1.

Inhibition of NF-κB: a Two-Edged Sword

Nuclear factor κB (NF-κB) is a key transcription factor that promotes cytokine gene expression in macrophages (proinflammation) and survival gene expression in hepatocytes (antiapoptosis). Using an in vivo liver inflammation model induced by anti-CD40, which promotes macrophage activation, Kimura et al. assessed the impact of globally inhibiting NF-κB by treatment with adenovirus containing an inhibitor κB (IκB) superrepressor (Ad5IkB) versus control (Ad5LacZ). Anti-CD40 induced cytokines, liver inflammation, and modest apoptosis. Inhibition of NF-κB suppressed cytokines and inflammation but lead to marked enhanced apoptosis, which was nevertheless dependent on both interferon-gamma and tumor necrosis factor α, since immunoneutralization of either cytokine prevented the enhancement of hepatocyte apoptosis in Ad5IkB-treated mice (Fig. 2). This work demonstrates the danger of treatment strategies to inhibit NF-κB-induced inflammation, since sensitization of hepatocytes to tumor necrosis factor–induced apoptosis may worsen liver injury. (See HEPATOLOGY 2004;40:1180–1189.)

Illustration 2.

The Best Defense is a Good Offense

Considerable evidence suggests that the resident liver innate immune system, especially natural killer (NK) / natural killer T (NKT) cells, plays an important role in preventing tumor cell metastases in the liver. Activation of the liver innate immune system, while enhancing tumor surveillance, can cause hepatitis. Miyagi et al. used a nonhepatotoxic versus hepatotoxic dose of the innate immune activator, concanavalin A (Con A), to assess the number of hepatic tumor foci after intrasplenic injection of colon cancer cells. The nonhepatic dose of Con A activated NK and NKT cells comparably to the hepatotoxic dose. In interferon (IFN)-γ−/− mice, Con A activated NKT cells but not NK cells, indicating that IFN-γ production by NKT cells leads to the activation of NK cells. Tumor foci and liver weight were decreased in tumor bearing mice after treatment with the low, nontoxic dose of Con A (Fig. 3). However, selective depletion of NK cells or inhibition of their activation (IFN-γ−/−) abrogated the protective effect of Con A. Finally, the authors demonstrated that activated NK cells could kill colon cancer cells in vitro. Thus, this study provides a novel strategy for combating hepatic metastases by careful modulation of the innate liver immune system through NK cell activation while avoiding hepatotoxicity. (See HEPATOLOGY 2004;40:1190–1196.)

Illustration 3.

Acetaminophen: a Killer With Many Weapons

Controversy exists as to the relative contribution of necrosis versus apoptosis in acetaminophen (APAP) hepatotoxicity. Kon et al. examined this issue in cultured mouse hepatocytes. The mitochondrial permeability transition (MPT) inhibitor, cyclosporin A (CsA) prevented early but not late MPT and cell killing in the presence of 10 mmol/L APAP. Treatment with the combination of fructose and glycine maintained cell ATP and prevented necrotic killing but did not prevent MPT while switching to increased caspase dependent apoptosis (Fig. 4). These findings indicate that APAP induces an early regulated MPT (CsA inhibitable) and late unregulated MPT (not CsA inhibitable). Maintenance of adenosine triphosphate (ATP) levels lead to a switch from necrosis to apoptosis in response to APAP-induced MPT. Since glycogen acts similarly to fructose in sustaining glycolytic ATP generation, the findings suggest that the mechanism of fasting-induced potentiation of APAP toxicity may be at least partly due to glycogen depletion and lessened ability to sustain ATP. The presumption is that switching from necrotic to apoptotic cell death may influence the severity of liver failure in vivo. Necrosis releases intracellular contents, which may promote inflammation and collateral damage. It remains to be seen whether switching to apoptosis will have an overall beneficial effect in vivo on the extent of organ damage or survival. (See HEPATOLOGY 2004;40:1170–1179.).

Illustration 4.

ACE Clubs the Liver

Angiotensin has been found to exert a number of interesting proinflammatory and profibrotic effects in liver independent of blood pressure control. Guo et al. explored the role of the renin-angiotensin system in the rat model of hepatic ischemia-reperfusion (I/R) injury in which the left and median lobes are clamped for one hour. Captopril (angiotensin converting enzyme inhibitor) or losartan (AT-1 receptor antagonist) pretreatment markedly protected against reperfusion injury as reflected in serum alanine aminotransferase, histology (Fig. 5), neutrophil infiltration, and 4-hydroxynonenal staining. Protection was accompanied by inhibition of intercellular adhesion molecule-1 and tumor necrosis factor expression in the liver. Interestingly angiotensinogen expression increased 4-fold after I/R. More work is required to elucidate the target cells of angiotensin in hepatic inflammation and the role of improved perfusion versus other mechanisms in the protective effect. However, this work provides experimental support for assessing the role of the renin-angiotensin system in various forms of liver injury and the potential therapeutic benefits of pharmacological blockade of this system. Indeed, a preliminary report in our journal shows a beneficial effect of such blockade in patients with nonalcoholic steatohepatitis. (See HEPATOLOGY 2004;40:583–589.)

Illustration 5.

A StAR is Born in the Liver

Cholesterol is converted into bile acids by a microsomal neutral pathway regulated by CYP7A1 and a mitochondrial acidic pathway initiated by CYP27A1. Since the former but not the latter pathway is greatly influenced by the level of expression of the respective CYP, it is possible that cholesterol trafficking to mitochondria regulates the acidic pathway. In previous in vitro studies overexpression of steroidogenic acute regulatory protein (StAR) increased the acidic pathway for conversion of cholesterol to bile acids. Ren et al. now have extended this work to in vivo studies. StAR was overexpressed in rats and mice by infection with recombinant adenovirus containing cytomegalovirus-StAR and lead to a 2-fold increase in rate of bile acid synthesis. StAR has been previously shown to mobilize cholesterol from the outer to the inner mitochondrial membrane in adrenal cells. The present studies suggest that cholesterol delivery to CYP27A1 in the mitochondrial inner membrane is rate limiting for acidic pathway of bile acid synthesis. StAR expression is low in normal liver and not upregulated by biliary diversion. Thus, it remains to be seen if and how StAR is regulated or if other proteins also perform this function in hepatocytes. (See HEPATOLOGY 2004;40:910–917.)