Lipid peroxidation derived reactive aldehydes in alcoholic liver disease

Abstract

Lipid peroxidation is a known consequence of oxidative stress and is thought to play a key role in numerous disease pathologies, including alcoholic liver disease (ALD). The overaccumulation of lipid peroxidation products during chronic alcohol consumption results in pathogenic lesions on protein, DNA, and lipids throughout the cell. Molecular adducts due to secondary end products of lipid peroxidation impact a host of biochemical processes, including inflammation, antioxidant defense, and metabolism. The aggregate burden of lipid peroxidation which occurs due to chronic alcohol metabolism, including downstream signaling events, contributes to the development and progression of ALD. In this current opinion we highlight recent studies and approaches relating cellular mechanisms of lipid peroxidation to the pathogenesis of alcoholic liver disease.

Introduction: Pathogenesis of alcoholic liver disease

Alcohol abuse is the third most common preventable cause of death in the United States and is a significant source of morbidity and mortality worldwide, accounting for approximately six percent of all global deaths [1, 2]. Alcoholic liver disease (ALD) is a complex disease arising from chronic alcohol consumption, which develops through a number of progressive stages [3]. The initial stage is alcoholic fatty liver (hepatic steatosis), which develops in the vast majority of chronic drinkers due to augmented metabolism, synthesis, transportation, and deposition of fatty acids and triglycerides [4]. The next pathological stage primarily includes alcoholic steatohepatitis (hepatic inflammation and injury), with a subset of patients developing a more severe form of liver damage termed alcoholic hepatitis, with very poor prognosis [5, 6]. Patients demonstrating chronic liver injury may further progress to hepatic fibrosis and cirrhosis, developing end-stage liver disease, and can ultimately lead to the development of hepatocellular carcinoma [6].

The hepatotoxicity of alcohol is largely due to its oxidative metabolism which produces a host of toxic intermediates and by-products [3]. When alcohol enters the liver, it is oxidized to acetaldehyde via hepatic alcohol dehydrogenase (ADH) and cytochrome P450 2E1 (CYP2E1) [7]. The second, and rate-limiting step, is the oxidation of acetaldehyde to acetate via the aldehyde dehydrogenase family of enzymes (ALDH). During periods of excessive alcohol consumption, ALDH detoxification capacity becomes saturated, resulting in the induction of secondary metabolizing pathways, namely CYP2E1. The induction of CYP2E1 leads to the increased generation and accumulation of superoxide and hydrogen peroxide, which can lead to persistent damage of DNA, protein, and lipids [8, 9]. In addition to the generation of radical and non-radical species, many other etiologies are implicated in ALD, including acetaldehyde toxicity [8, 10].

Over the past few decades researchers have elucidated numerous mechanisms contributing to the development of ALD, including oxidative stress [6, 11, 12]. Oxidative stress has recently been redefined as an imbalance between oxidants and antioxidants in favor of the former, leading to a disruption of redox signaling and control and/or molecular damage [9, 13, 14]. The increased generation of radical and non-radical species is known to contribute to enhanced lipid peroxidation (LPO), which includes the overproduction of reactive aldehydes such as 4-hydroxynonenal (4-HNE), malondialdehyde (MDA), and acrolein. The accumulation of these reactive aldehydes leads to depletion of cellular antioxidant defenses and covalent adduction of DNA, lipids, and proteins [8, 15]. Here, we focus on the generation and consequences of alcohol-induced reactive species and consequent lipid peroxidation, focusing on the disruption of redox signaling and control through proteomic and metabolic signaling, thus disrupting normal hepatic function and contributing to the pathogenesis of ALD.

Redox imbalance and lipid peroxidation

The recently revised definition of oxidative stress accounts for the downstream consequences of radical and non-radical species, including a disruption in thiol/disulfide systems such as thioredoxin-1, cysteine/cystine (Cys/CySS), and glutathione (GSH/GSSG) [16-18]. Oxidative stress can impact separate and distinct redox pathways as opposed to the less nuanced global thiol/disulfide balance [13]. The term redox regulation may refer to two distinct functions. In the more traditional sense, it refers to the elimination or generation of reactive oxygen species (ROS) and reactive nitrogen species (RNS) [19]. In a more nuanced context it also applies to the reversible oxidation of cysteine thiols and methionine thioethers of proteins to disulfides or sulfenic acid groups, which contribute to the regulation of biological function. Free radical initiation events leading to the generation of hydrogen peroxide and peroxynitrite from superoxide and nitric oxide pose a significant oxidative burden [20]. The accumulation of these reactive species and resulting generation of lipid peroxidation products disrupts critical cellular signaling and control pathways. Specifically, the thiol containing proteome is highly susceptible to modification and is a critical aspect of redox regulation [16]. Furthermore, these protein thiols are prone to two-electron oxidation by nonradical oxidants like hydrogen peroxide, disulfides, quinones, aldehydes, and lipid hydroperoxides [21].

The generation of excess radical and nonradical species facilitates hepatic injury through a variety of mechanisms in ALD, including mitochondrial damage, activating hepatic stellate cells resulting in transformation into myofibroblasts, activating periportal fibroblasts and fibrocytes, remodeling of the extracellular matrix (ECM), stimulating immune responses to ROS, and by exacerbating inflammation through activation of the nuclear factor NF-κB [8, 22-36]. The major sources of normal physiological oxidants in the human body are the mitochondrial respiratory chain and the cytochrome P450 mixed-function oxidases. Redox imbalances during the initiation and progression of ALD arise due to many factors. The metabolism of alcohol by ALD and ALDH both result in the formation of NADH, which increases respiratory chain activity and superoxide radical production [22, 37]. Chronic alcohol consumption results in oxidative damage to mitochondrial respiratory complexes, eventually resulting in further ROS production [38]. Alcohol has been shown to induce p66^SHC, a protein known to generate ROS through electron transfer from cytochrome c, causing mitochondrial damage in mice [39, 40]. Inducible CYP2E1 not only reduces O₂ to the superoxide radical and hydrogen peroxide, but also produces the hydroxyl radical in the presence of iron catalysts, which are increased in response to alcohol consumption [41, 42]. Alcohol has been shown to facilitate xanthine dehydrogenase conversion to xanthine oxidase, resulting in superoxide radical and hydrogen peroxide production [43]. NADPH oxidase enzymes (NOX) and inducible nitric oxide synthase (iNOS) are activated in hepatocytes in response to alcohol ingestion, resulting in superoxide and nitric oxide formation, the combination of which produces peroxynitrite, which is a powerful oxidant leading to mitochondrial damage in ALD [44-49]. Apart from hepatocytes, infiltrating monocytes and lymphocytes also contribute to oxidative stress. Activated Kupffer cells produce TNFα in response to free radical-activated NF-κB, contributing to injury [6, 25-30, 50-57]. Increased iron levels and Fenton chemistry also contribute to Kupffer cell activation [58, 59]. Animal studies involving treatment with antioxidant compounds, depletion of endogenous antioxidants, and both overexpression and knockout of endogenous antioxidant genes (thioredoxin, SOD, catalase, glutathione peroxidase-1, metallothionein, sulfiredoxin, peroxiredoxin-1, and Nrf2) have provided evidence for the causal role of oxidative stress in ALD [60-72]. Additionally, the effect of alcohol on the antioxidant glutathione continues to be debated as a critical factor in redox imbalances [61, 73-76]. Interestingly, the efficacy of antioxidant therapy in humans remains uncertain, which supports the idea that two-electron non-radical oxidants may present an equal if not more significant role than radical species in terms of redox imbalance [35, 65-71, 77-85].

Lipid Peroxidation (LPO) is initiated by the free radical or non-radical abstraction of hydrogen from polyunsaturated fatty acids (PUFAs), resulting in decreased membrane fluidity, increased membrane permeability, and loss of membrane-bound protein function [86, 87]. The initiation step generates a carbon-centered lipid radical, which reacts with oxygen to form a lipid peroxyl radical. This lipid peroxyl radical abstracts another electron from a nearby lipid to generate a new lipid radical and lipid hydroperoxide, termed the propagation step. Importantly, this propagation step can result in hundreds of lipid peroxidation products (LPP) generated from a single initiation event [88]. The termination step of lipid peroxidation involves the self-quenching of the lipid peroxy radical or through antioxidants like vitamin E, vitamin C, SOD, catalase, and peroxidase. Hydroperoxide degradation occurs through two-electron reduction, carried out primarily by glutathione peroxidases and selenoprotein P, or through one-electron reduction, which leads to continued lipid peroxidation [86]. When a lipid peroxyl radical is not neutralized it can fragment and decompose to form secondary products, including Di-aldehydes (MDA), α,β-unsaturated aldehydes [acrolein, 4-hydroxynonenal (HNE), 4-hydroxyhexenal (HHE), 4-oxononenal (4-ONE)], and acetaldehyde. These reactive aldehydes, in particular, play a central role in signaling, autophagy, senescence, the cell cycle and proliferation, and apoptosis [89].

Lipid peroxidation and Alcoholic Liver Disease

Recent studies using murine models of ALD indicate that when combined with a diet rich in polyunsaturated fatty acids (PUFAs), chronic alcohol consumption results in enhanced lipid peroxidation and significant liver injury [90-93]. Importantly, substitution of PUFAs with saturated fatty acids significantly reduced accumulation of LPP as well as ameliorated hepatic injury. It is thought that the increase in dietary polyunsaturated fat provides a pool of lipids that are readily peroxidizable, serving as a source for the production of reactive aldehydes. Numerous key studies demonstrate an enhanced production of reactive aldehydes in ALD. First, using immunohistochemistry (IHC), increased periportal post-translational modification by 4-HNE, MDA, 4-ONE and acrolein occurs following chronic alcohol consumption [94-98]. This is further supported by data demonstrating co-localization of 4-HNE with the lipid vesicle protein adipophilin (Perilipin 2, PLIN2) on the outer coat of lipid vesicles in a murine model of ALD [99]. Reactive aldehyde accumulation has been reported in cytosolic, microsomal, mitochondrial, and nuclear fractions following chronic alcohol consumption, impacting biochemical processes across multiple cellular compartments [97, 98, 100-103].

Lipid peroxidation occurs in patients with ALD, making a compelling argument for a causal role in ALD. Studies have shown increases in LPP and protein adducts in blood, urine, and tissues of ALD patients, including 4-HNE, 8-OHdG, 8-isoprostane, and MDA [104-113]. Studies in both mouse and human liver tissue have also demonstrated antibody production against oxidized phospholipids, MDA, and 4-HNE, which lead to TNFα production and hepatic injury [114-119]. Alcohol-induced LPP adducts disrupt protein function, promote fibrosis, and induce humoral and cell-mediated immune responses [114-118, 120-128]. Indeed, many of the same elevated LPP detected in humans have been identified in animal models of ALD, including MDA, 4-HNE, and F₂-isoprostanes [11, 31]. Animal models have also demonstrated lipid peroxidation-derived 4-HNE stimulation of stellate cells resulting in collagen production [129]. Intragastric infusion models of ALD have demonstrated a correlation with increased LPP, LPP-protein adducts, and lipid radicals [61, 68, 79, 130]. An increase in mitochondrial LPP-protein adducts in chronic alcohol models, specifically carbonylated proteins, parallels mitochondrial dysregulation resulting from chronic alcohol metabolism [73, 94, 119, 131, 132].

A central question concerning the appearance of aldehyde-modified proteins in the liver due to chronic alcohol consumption revolves around the association of these molecular lesions with development of histopathology or impairment of hepatocellular function. Recently, the contribution of reactive aldehydes to alcohol-induced injury has been explored using the 10-day chronic-plus-binge model of alcohol consumption, where supplementation with the acrolein scavenger hydralazine significantly decreased the accumulation of acrolein adducts and hepatic injury [133]. In addition, supplementation with the herb Danshen, resulted in an increase in PPAR activation and GSTA4 expression, resulting in the increased clearance of 4-HNE and reducing oxidative injury [134]. A recent report demonstrated that PPAR knockout increased 4-HNE accumulation when compared to alcohol-fed WT mice, providing evidence that both GSTA4 and β-oxidation are contributing factors in the mitigation of lipid peroxidation [94, 119].

Significant progress has been made in the identification of proteins modified by LPP in ALD. Early proteomic approaches used to identify modified proteins in ALD used 2-dimensional electrophoresis followed by protein identification [135, 136]. Interestingly, the majority of identified proteins were involved in either protein folding and oxidative stress responses. In a recent study using 4-HNE antibody conjugated co-immunoprecipitation, elevated mitochondrial adducts were identified in a 5-week murine model of ALD [97, 103]. Advances in biotin hydrazide chemistry and in the sensitivity of mass spectrometry have permitted a more in-depth proteomic approach to identify less abundant proteins modified due to alcohol metabolism. To date, global proteomic approaches applied to various models of ALD have identified over 2000 proteins that undergo adduction in either murine models or in human hepatic tissue isolated from patients with end-stage ALD [94, 97, 113, 137]. Additionally, recent reports have identified which amino acids are modified by reactive aldehydes on over 70 proteins [94, 97, 113, 137]. Utilizing GSTA4-4 knockout mice it was determined that protein adduction by LPP was increased in mitochondrial fractions in pathways regulating oxidative stress, fatty acid metabolism, and amino acid metabolism, supporting the contribution of GSTA4-4 in protecting mitochondria from reactive aldehydes [119, 137]. Another report demonstrated that lipid peroxidation derived protein adduct formation is increased in tissue obtained from end-stage alcoholics [113]. Not surprisingly, modification of proteins regulating oxidative stress, metabolic, and cytoskeletal processes was increased [119]. Protein-specific approaches have also been utilized in models of ALD to identify low abundant molecular targets including L-FABP, GRP78, PDI, PTEN, AMPK, Akt, and Sirtuin 3 in the liver after chronic alcohol treatment using co-immunoprecipitation techniques and antibodies directed against each protein [95, 135, 138-142]. In each report, protein adduction was identified to contribute to altered cellular signaling. A recent cutting-edge approach utilized an aldehyde reactive probe to enrich 4-HNE modified proteins for identification and label-free quantitation by LC-MS/MS [143]. Further scrutinizing protein targets, the authors applied a peptide-specific assay to reveal distinct reactivity profiles for 4-HNE protein adduction while identifying 31 protein targets with a total of 61 modification sites in an early-stage model of ALD. In aggregate, these findings support the hypothesis that alcohol-induced production of reactive aldehydes exacerbates hepatocyte injury by directly impacting cellular metabolism and anti-oxidant responses through protein-specific modifications.

Concluding Remarks

The production and accumulation of hepatocellular LPPs is a pathogenic hallmark of alcohol metabolism. These reactive species have been found to contribute to inflammation, metabolic perturbations, and redox signaling. While current studies continue to investigate the underlying mechanisms of LPP toxicity in ALD, further research in needed to develop specific assays that can identify and quantify these reactive aldehydes and their reactivity profiles. Antioxidants continue to lack effectiveness as ALD therapies due to a delicate balance that exists in redox regulatory signaling and control. Ultimately, an overwhelming amount of research supports the restriction of alcohol consumption as a means to prevent ALD, which coincides with minimizing the production and accumulation of LPPs.

Figure 1.

Illustrative diagram showing the involvement of hepatocytes, hepatic stellate cells, and Kupffer cells in ROS/RNS and LPP mediated pathophysiology of alcoholic liver disease.

In hepatocytes, ethanol ingestion results in the activation of nitric oxide synthase (NOS) and NAD(P)H oxidase (NOX), which produce nitric oxide (NO) and the superoxide radical (O₂^•−) respectively. NO and O₂^•− react to form peroxynitrite (ONOO⁻), which causes mitochondrial damage through further production of reactive oxygen species (ROS) and lipid peroxidation products (LPP), which can form adducts with proteins, lipids, and DNA. Adduct formation then contributes to hepatic injury characteristic of ALD (inflammation, steatosis, and fibrosis). Ethanol mediated activation of NOX can also occur in hepatic stellate cells and Kupffer cells, which results in ROS/RNS production and the generation of LPP, leading to the differentiation of hepatic stellate cells into myofibroblasts and initiating inflammation through NF-κβ (nuclear factor kappa-light-chain-enhancer of activated B cells) activation in Kupffer cells. Lipid peroxidation products, like 4-hydroxy-2-nonenal (4-HNE) and malondialdehyde (MDA), as well as lipid hydroperoxides (LOOH) can form adducts within hepatic stellate cells and Kupffer cells. Cytochrome P450 2E1 (CYP2E1) induction by alcohol results in O₂^•− production, and potentially hydroxyl radical (OH^•) production from the reaction of hydrogen peroxide (H₂O₂) and iron (Fe), in all three cell types. ROS and RNS in hepatocytes and Kupffer cells stimulate interleukin-1 beta (IL-1β) production in the inflammasome, leading to inflammation and cell death. In Kupffer cells and hepatic stellate cells, ROS and RNS production results in activation of matrix metalloproteinases, increased deposition of extracellular matrix (ECM) proteins, and resultant fibrosis.

Abbreviations

ALD

alcoholic liver disease

CYP2E1

cytochrome P450 2E1

ADH

alcohol dehydrogenase

ALDH2

aldehyde dehydrogenase 2

NAD⁺

oxidized nicotinamide adenine dinucleotide

NADH

reduced nicotinamide adenine dinucleotide

ROS

reactive oxygen species

LC-MS/MS

liquid chromatography tandem mass spectrometry.