Protein phosphatase 2A (PP2A) is a serine/threonine phosphatase regulating cellular functions and signaling pathways, including ChREBP, apoptosis, and insulin signaling [105], and may be involved in alcohol inhibition of AMPK. A PP2A subunit co-immunoprecipitated with AMPK, alcohol increased PP2A activity in hepatoma cells [106], and the PP2A inhibitor okadaic acid or PP2A siRNA significantly attenuated the inhibitory effect of alcohol on AMPK phosphorylation.

PP2A is regulated by ceramide [107]; in fact, a form of PP2A was originally named the ceramide-activated protein phosphatase [108]. Alcohol treatment of hepatoma cells significantly increased cellular (C16 and C18) ceramide content and increased PP2A activity [106]. Inhibitors of the de novo synthetic pathway and NSMase did not prevent inhibition of AMPK by alcohol; however, the ASMase inhibitor imipramine did prevent this effect. This effect of alcohol was observed in vivo: in mice fed alcohol, there were 30 % increases in hepatic C16 and C18 ceramides, the level of ASMase mRNA was increased by 1.7-fold, and imipramine reduced ASMase activity, ceramide levels, PP2A activity, and triglyceride accumulation [109]. The increase in ASMase activity was not associated with increased levels of TNFα, and imipramine treatment did not restore AMPK activity. More recently, myriocin, an inhibitor of de novo ceramide synthesis, reversed alcoholic steatohepatitis in animals [110].

Alcohol-induced elevation of ceramide may contribute to the induction of steatosis by AMPK-dependent and AMPK-independent pathways.

Adiponectin stimulates fatty acid oxidation, in part through its activation of PPARα and AMPK [38]. Alcohol feeding decreased plasma adiponectin levels [38, 47, 111] and the adiponectin receptors AdipoR2 and AdipoR1 in mice and micropigs, respectively [74, 111]. Treatment of alcohol-fed animals with adiponectin alleviated hepatic steatosis [58]. Conversely, adiponectin knockout mice are highly sensitive to alcohol-induced steatosis, but surprisingly no difference in AMPK activity or PPARα DNA-binding activity was observed [112]. Interestingly, adiponectin can reduce ceramide levels, suggesting another mechanism by which it improves liver metabolism in alcohol-fed animals.

SIRT1 is an NAD⁺-dependent protein deacetylase which senses the cellular redox state as a measure of energy need and can regulate the activity of proteins involved in lipid metabolism. SIRT1 is as a major regulator of AMPK [113]. Liver-specific knockout of SIRT1 results in hepatic steatosis and inflammation [114]. Alcohol feeding and in vitro studies show that metabolism of alcohol reduces SIRT1 activity [115, 116] and mRNA [117]. Mechanisms accounting for this effect include increased levels of miR217 in alcohol-metabolizing cells or liver [118]. This effect could be overcome with resveratrol, which also reversed steatosis [111], or feeding a high-saturated fat diet [115, 116]. Resveratrol increased circulating adiponectin and the activity of SIRT1, AMPK, and PPARγ coactivator-1α (PGC-1α) [111]. Reduction of SIRT1 activity has several downstream effects. Reduced SIRT1 activity leads to increased activity of SREBP-1c and changes the expression of lipin-1 (an Mg⁺⁺-dependent phosphatidic acid phosphohydrolase). This protein is a regulator of fatty acid oxidation and de novo lipogenesis [119]. Alcohol feeding, via SIRT1 inhibition, increases the level of lipin1β, which in turn led to the activation of SREBP-1c. SIRT1 was also reduced in macrophages exposed to acetaldehyde or acetate, and they generated increased amounts of TNFα [120]. Feeding alcohol to SIRT1 knockout animals increased markers of inflammation, oxidative stress, ER stress, cytokine production, and fibrosis, suggesting an important role for SIRT1 in modulating all these effects of alcohol. In samples of liver from patients with alcoholic hepatitis, SIRT1 mRNA was reduced to about 25 % of normal, although the effects on lipin mRNA were more subtle [114]. The effect of alcohol on lipin expression was blocked by activators of AMPK or constitutively active AMPK genes. Thus, SIRT1 (and, by extension, dietary polyphenols such as resveratrol) plays important roles in regulating fat synthesis and oxidation in ALD.

The strategic location of the liver between the intestinal tract and the rest of the body makes it a critical organ for clearance of xenobiotics and toxins that enter the portal blood. It is therefore not surprising that the liver has a very high capacity for phase I and II metabolic processes, which is responsible for the well-known “first-pass effect” in xenobiotic metabolism. As early as 1893, Pavlov observed that low levels of intestinally derived “toxins” are normally present in the portal blood and are cleared by the liver [147]. Kupffer cells, the resident hepatic macrophages, are responsible for clearing endotoxin [148] (Fig. 3.4). When high levels of LPS are present in the portal blood, Kupffer cells are activated and can mediate toxic responses in the liver [149]. It has been known since the 1970s that alcohol consumption increases the amount of detectable LPS in the blood [150, 151]. These increases are thought to be caused by both dysbiosis in the GI tract microbiome and a decrease in GI tract barrier function. Importantly, these are highly interdependent functions in that dysbiosis decreases GI tract barrier and altered barrier function can impact the microbiome.

The concept of priming and sensitization is important in ALD. Activation of the inflammatory response is, at least in part, due to increased levels of LPS [164, 165]. However, elevated levels of LPS found in alcoholics and in experimental ALD are relatively low compared to those found in experimental endotoxemia and human sepsis. Furthermore, liver injury caused by alcohol cannot be mimicked by chronic low-dose LPS in the absence of ethanol [166]. This is explained by the fact that inflammatory cells are primed to activation by alcohol administration, e.g., peripheral blood monocytes from patients with alcoholic hepatitis spontaneously produce proinflammatory mediators (e.g., TNFα). In response to LPS, these monocytes produce more proinflammatory mediators than their control counterparts [167]. Hepatocytes also appear to be sensitized to inflammatory stimuli by alcohol administration. Although TNFα is proproliferative in hepatocytes isolated from naïve animals, it is proapoptotic in cells isolated from alcohol-treated animals [168], via cellular “death domain” pathways [169]. HSCs also are sensitized by alcohol administration [170, 171]. The concept of priming and sensitization also means that there may be a series of sequential events in the progression of liver damage due to alcohol exposure. Specifically, the priming of inflammatory cells by alcohol leads to greater cell injury of sensitized hepatocytes; hence, blocking the activation of Kupffer cells [172] or employing knockouts with impaired Kupffer cell responses [173] prevents hepatocyte damage.

The enhanced production of TNFα is considered a key factor in the priming effect of alcohol on macrophage activation by LPS. Alcohol causes changes in the signaling cascade from the LPS receptor partners (CD14/TLR4) to the expression of TNFα. The expression of CD14 on Kupffer cells is upregulated by oxidant-dependent activation of the transcription factor AP-1 by alcohol [174]. Two transcription factors that are critical in TNFα expression are NFκB and EGR-1. Alcohol increases NFκB activation via both ROS-dependent [175] and ROS-independent pathways [176]. Ethanol also increases EGR-1 transcriptional activity via enhancing ERK1/ERK2 signaling [177]. Furthermore, alcohol itself enhances LPS-stimulated ROS production by macrophages via MyD88-independent TLR4 signaling [178]. Alcohol-induced TNFα expression is also regulated by increased mRNA stability. LPS stimulation of p38 mitogen-activated protein kinase contributes to this stabilization of TNFα mRNA. Moreover, studies have shown that at least one mRNA-binding protein, HuR, is also involved in the stabilization of TNFα mRNA stability after chronic exposure to ethanol (see [179] for review). The net effect is an increase in TNFα protein release in alcohol-exposed macrophages after stimulation.

The sensitization effect of alcohol on hepatocytes represents an increased sensitivity of parenchymal cells to cytotoxic killing. TNFα-induced signaling via TNFR1 appears to play a key role. In normal hepatocytes, TNFα is a mitogen; however, alcohol causes changes in intracellular signaling that switch this mitogenic response to a cytotoxic response. There is a critical role of CYP2E1 induction in this process [180]. Using the model of CYP2E1 induction by pyrazole, they have shown that sensitization was reduced by half in CYP2E1 knockout mice. Sensitization was associated with the activation of p38 and JNK kinases, and liver injury was blocked with inhibitors of these kinases. Liver mitochondria from animals treated with pyrazole and TNFα underwent increased swelling in response to calcium, and cyclosporine A, an inhibitor of the mitochondrial permeability transition, reduced liver injury. Pyrazole increased the levels of IkB, which reduces the activation of NFκB by TNFα. NFκB is thought to mediate cell-protective effects of TNFα; this inhibition would tilt the effect of TNFα toward cell injury. The sustained activation of JNK (which is an apoptotic signal) coupled with the inhibition of NFκB activation are proposed to mediate the sensitization effect of alcohol (see [181] for review). Administration of N-acetylcysteine to mice treated with pyrazole and LPS or TNFα ameliorated in the changes in p38, JNK, and activation of NFκB. Further, pyrazole plus TNFα administration caused less liver injury in NOS2 (iNOS) knockout animals than in controls, indicating that peroxynitrite radicals contributed to sensitization of the liver. Thus, there is strong evidence from animal studies that oxidative stress, in particular that resulting from induction of CYP2E1, is very important in response of the liver to increased levels of LPS and TNFα.

MicroRNAs are ubiquitous regulators of gene expression, which interact with mRNAs and can either block translation or stabilize the mRNA. Levels of a host of miRNAs have been reported to be altered in alcohol-fed animals or in cultured cells (reviewed by McDaniel et al. [182]). Some miRNAs are increased (miR21, miR34a, miR155, and miR320) and some are decreased (miR122, miR181a, miR199a, and miR200a) by alcohol exposure. Recent studies have indicated that these miRNAs are likely involved in many of the pathological responses to alcohol already discussed, including the modulation of SIRT1 (miR34a) and of intestinal permeability (miR122), apoptosis [183], and the sensitization of Kupffer cells to LPS (miR155). These RNAs may be targetable for therapy of ALD and may also provide serum biomarkers for different stages of the disease.

The two major death responses available to the cell are necrosis and apoptosis (Fig. 3.4). Necrosis is generally an acute response to traumatic stress and is characterized by frank autolysis of the cell and, in its most severe form, is independent of any intracellular signaling process. Apoptosis is a controlled ATP-dependent response to intrinsic and/or extrinsic stimuli. In contrast to necrosis, cellular fragments are released as blebs, and there is no release of intracellular contents into the extracellular space. As will be discussed below, necrosis is considered more proinflammatory by inducing the inflammasome. Both apoptosis and necrosis are validated histologies in fatty liver diseases, including ALD. However, the relative contribution of apoptosis histologically to ALD is lower than would be expected by in vitro studies or in studies with inhibitors of apoptosis (e.g., [184]). It is now understood that apoptosis and necrosis are not distinct events, but rather part of a continuum of cellular responses to cytopathic stress. In this, continuum may extend to very early effects of alcohol, such as ER stress. ER stress was reported to stimulate the association of IRF3 with an adaptor protein (stimulator of interferon genes or STING) and the activation of IRF3, which is proapoptotic. This effect was independent of the development of inflammation [185]. An apoptotic response may convert to necrosis, if cellular energy stores are insufficient to complete the apoptotic process (i.e., “necroptosis”) [186]. The metabolic stress caused by alcohol exposure that is described above (Sections “Metabolism of Ethanol and Its Role in Oxidative Stress” and “Effects of Alcohol Consumption on Liver Fat Homeostasis”) may thereby cause necroptosis [187].

Inflammation is induced not only by pathogen-associated molecular patterns (“PAMPs”), such as LPS (see section “Contribution of cell death to inflammation”), but also by molecules released from dead or dying cells (damage-associated molecular patterns (DAMPs)) (Fig. 3.4). These danger signals interact with the toll-like receptor (TLR) family of pattern recognition receptors, which thereby activate intracellular sensor molecules, the NOD-like receptors (NLRs), to subsequently activate caspases. The net result is the proteolytic activation and release of IL-1β, which induces the innate immune response [188]. As apoptosis does not release intracellular contents into the extracellular space, it is assumed to be less likely to induce the inflammasome response. However, it should be emphasized that apoptosis is not completely benign. Apoptotic bodies have been shown to activate the innate immune response, as well as to stimulate HSCs in the liver [189]. Another emerging area of intercellular communication in response to stress is the release of exosomes. Exosomes are small vesicles of extracellular membrane released by cells (e.g., hepatocytes) in response to stimuli. These exosomes can serve as DAMPs to stimulate the inflammasome response [190].