The Biological Clock: A Pivotal Hub in Non-alcoholic Fatty Liver Disease Pathogenesis

Non-alcoholic fatty liver disease (NAFLD) originates from disproportionate triglyceride deposition in hepatocytes, which contributes to more than 5–10% percent of the liver's weight, concurrent to major macrovesicular steatosis, all in the absence of excessive alcohol consumption (Kanwal et al., 2016). Liver steatosis is the hepatic manifestation of metabolic syndrome and represents the primary cause of chronic liver disease in the Western world (Bonora and Targher, 2012; Wong et al., 2015; Kanwal et al., 2016). NAFLD affects at least 30% of the general population, being present in more than 60% of people with obesity and embracing different histological patterns, which range from uncomplicated steatosis to steatohepatitis (NASH) (Bonora and Targher, 2012; Wong et al., 2015; Kanwal et al., 2016). NASH is regarded as the most common likely indication for liver transplantation over the next decade worldwide, and is a prevailing end-stage kidney disease and cardiovascular disease risk factor (Bonora and Targher, 2012; Musso et al., 2014; Wong et al., 2015; Wu et al., 2016). A major, albeit not indispensable, role in NAFLD pathogenesis is played by the reduced secretion of insulin, coupled to insulin insensitivity. Insulin crucially regulates blood glucose levels upholding glycogen synthesis in the fed state and restraining hepatic glucose production during fasting. Insulin resistance is intimately linked to hyperinsulinemia, which boosts hepatic lipogenesis and leads to hepatosteatosis. In turn, the buildup of different lipid species in the liver elicits inflammation or endoplasmic reticulum (ER) stress, turning on numerous protein kinase cascades, which in due course produce insulin signaling disturbances further enhancing insulin resistance (Titchenell et al., 2017).

Augmented hepatic triglycerides levels may be caused by undue free fatty acid (FFA) delivery to the liver, especially after eating, which is related to adipose tissue insulin resistance in people with obesity (Lambert et al., 2014; Titchenell et al., 2017). In addition, increased de novo lipogenesis, driven by hyperinsulinemia and unwarranted carbohydrate intake causing triglycerides overload, may not be balanced by adequate increases in very low density lipoprotein (VLDL) secretion and changes in levels of β-oxidation (Lambert et al., 2014; Titchenell et al., 2017). In recent decades dietary habits have progressively changed, leading to the consumption of low-nutrient, energy-dense foods and beverages containing high levels of sugars and fats. This has also been coupled to decreased physical activity. These factors have triggered the increased prevalence of metabolic diseases, driven by energy imbalance.

Energy homeostasis is maintained by signaling interactions among metabolically sensitive tissues, such as the intestine, pancreas, and adipose tissue. However, the liver is the master hub that regulates conflicting metabolic processes, such as glycolysis/gluconeogenesis and lipogenesis/fatty acid β-oxidation (Corkey and Shirihai, 2012; Paschos et al., 2012). Multiple organs may be implicated in dysmetabolism onset and progression and the molecular mechanisms may entail signal transduction pathways in pancreatic islets, liver, adipose tissue, as well as brain, gut, vasculature, and muscle (Corkey and Shirihai, 2012; Paschos et al., 2012). Metabolic derangements can be initiated in any of these organs and tissues, with blood-borne molecules having metabolic effects in other tissues and organs, in an ongoing process of reciprocated interactions (Corkey and Shirihai, 2012; Paschos et al., 2012). This coordination among specialized tissues to manage essential and dedicated cellular processes under homeostatic as well as stress conditions is maintained through long-distance inter-organ communication networks hardwired by humoral factors, such as hormones (i.e., insulin, adipocytokines, glucocorticoids), peptides, proteins, and the sympathetic and parasympathetic fibers of the autonomic nervous system (ANS; Corkey and Shirihai, 2012; Paschos et al., 2012). A crucial role in the molecular regulation of metabolic tissues is played by metabolites (e.g., glucose, fatty acids, sterols, and bile acids), which are sensed by nuclear receptors and regulate nutrient directed signal transduction pathways (Mazzoccoli et al., 2014). Intestinal nutrient digestion and extraction is a primary process in body homeostasis, which prompts intermediary metabolism. In addition, the intestine secretes enteroendocrine hormones and harbors the gut microbial community, with the latter being increasingly recognized to have considerable importance in the regulation of metabolic processes (Sanz et al., 2015). The temporal dimension is a crucial player in NAFLD pathogenesis considering that the processes involved in hepatosteatosis onset and progression show time-related fluctuations. Altered resonance between geophysical cycles and the array of physiological rhythms bringing about internal timing is gradually more acknowledged as involved in the global NAFLD epidemic.

Liver metabolic pathways and bile acid synthesis as well as basic cell functions, such as autophagy, are regulated by circadian processes, which are, in turn, modulated by the gut microbiome. Circadian regulation of patterns of physiological functions as well as hormone secretion, metabolic fluxes, bile acid turnover, autophagic and inflammatory processes, coupled to the behavioral timing of sleep/wake and fasting/feeding cycles are crucial to preserve liver health and circumvent metabolic derangement (Mazzoccoli et al., 2012; Cermakian et al., 2013; Rosselot et al., 2016). Energy regulation is crucial to survival and requires the monitoring of blood-borne nutrient levels, the triggering of food seeking and stimulated insulin secretion to prompt satiety. Inappropriate nutritional signaling triggers suboptimal eating behavior, exemplified by alterations in leptin signaling in the C57BL/KsJ-db/db mouse (Bogdanov et al., 2014), and flawed signaling related to systemic responses to misread cues could set off metabolic derangement (Corkey and Shirihai, 2012).

Glucose levels in the peripheral blood are maintained in tight physiological range, but show fluctuations with 24-h periodicity related to feeding patterns as well as to oscillations in glucoregulatory hormones (Mazzoccoli et al., 2012). Glucagon is secreted by α pancreatic cells and counters the effects induced by insulin. Glucogon shows circadian changes in the rat plasma (Ruiter et al., 2003) and during fasting turns on the CREB/CRTC2 transcriptional complex, which binds to the Arntl promoter and promotes its expression (Sun et al., 2015). Glucagon-like peptide-1 (GLP-1) is an incretin hormone secreted by the intestinal L cells, and operates as a link between the gut and pancreatic β-cells. Notably, GLP-1 is produced with a circadian rhythmicity (Gil-Lozano et al., 2014). Healthy humans show circadian rhythms of insulin sensitivity and insulin secretion, whilst the mouse and human Langherans islet β-cells harbor autonomous and self-sustained circadian oscillators that drive insulin synthesis and release (Sadacca et al., 2011; Eckel-Mahan and Sassone-Corsi, 2013; Pulimeno et al., 2013; Perelis et al., 2015). Experiments performed in mouse models as well as in human islets isolated from the pancreas of brain-dead multi-organ donors show insulin release is rhythmically regulated by peripheral pancreatic β-cell clocks, even in isolated perfused condition (Saini et al., 2016). Accordingly, whole body Arntl^−/− mice show glucose intolerance, hypoinsulinemia, an increased respiratory quotient, reduced fat storage, increased circulating fatty acids, increased ectopic fat formation in liver and muscle, and hypoinsulinemia, whilst mouse liver specific Arntl knock-out induces hypoglycemia in the resting state (Lamia et al., 2008; Marcheva et al., 2010; Shimba et al., 2011). Furthermore, Arntl was found necessary for β-cell adaptative growth, survival, and metabolic adjustment to diet-induced obesity in mice (Rakshit et al., 2016), whilst whole-body Clock mutant mice (Clock^Δ19) are characterized by age-dependent damped feeding rhythms, obesity, hyperphagy, hyperlipidemia, hyperglycemia, hypoinsulinemia, and hepatic steatosis (Rudic et al., 2004; Turek et al., 2005; Marcheva et al., 2010). Notably, the Clock^Δ19 mutant mouse was bred on a C57BL/6J strain background lacking melatonin production, with melatonin limiting basal and glucose-stimulated insulin secretion.

Moreover, double Cry1 and Cry2 knockout mice show glucose intolerance and chronically elevated levels of circulating corticosterone with augmented glucocorticoid-dependent transactivation in the liver (Lamia et al., 2011), which is coupled to hyperglycemia induced by defective restraint of glucagon-stimulated hepatic gluconeogenesis (Zhang et al., 2010). Accordingly, genome-wide association studies performed in humans found elevated fasting glucose levels and altered response of energy expenditure to weight-loss diets (measured as respiratory quotient and resting metabolic rate) in carriers of CRY2 rs11605924 variant as well as rs10830963G allele of MTNR1B gene, encoding the MT2 melatonin receptor (Mirzaei et al., 2014; Lane et al., 2016; Tuomi et al., 2016). Carriers of MTNR1B risk variant show amplified MT2 receptor expression in pancreatic β-cells, as well as delayed dim light melatonin offset and a more prolonged melatonin secretion. Such alterations increase the risk of food intake coincident with amplified morning melatonin levels, coupled to decreased glucose tolerance (Lane et al., 2016; Tuomi et al., 2016).

Rev-erbs and RORs, working as transcriptional repressors and activators, respectively, control the circadian expression of numerous target genes encoding proteins involved in lipid metabolism. Accordingly, Rev-erbα^−/− mice are hallmarked by elevated VLDL triglyceride levels linked to increased ApoC-III expression in the liver. In contrast, the RORα mutant mouse (RORα^sg/sg, Staggerer mouse), in spite of greater food intake, has decreased body fat, smaller fat cells in brown and white adipose tissue, lower liver triglyceride levels and is not prone to hepatic steatosis (Solt et al., 2011). Whole-body double knockout Rev-erbα and Rev-erbβ mice show a phenotype characterized by altered rhythmic expression of the core clock and lipid homeostatic gene networks leading to altered circadian wheel-running behavior, hyperglycemia, hyperlipidemia and hepatic steatosis (Bugge et al., 2012; Cho et al., 2012). The nuclear receptor corepressors (N-CoR)/Histone Deacetylase 3 (HDAC3) complex is necessary for repression by Rev-erbα, with liver-specific HDAC3 knockout causing deregulated lipid metabolism and hepatic steatosis in the mouse (Feng et al., 2011).

Rev-erbα also impacts on bile acid metabolism through the liver X receptor (LXR or NR1H3), whose transcriptional activity changes upon binding of rhythmically synthesized oxysterols and which temporally interacts with the farnesoid X receptor (FXR or NR1H4), the nuclear receptor capable of binding bile acids and showing circadian rhythmicity in the liver. LXRs and FXRs drive, positively and negatively, respectively, the 24-h periodicity of CYP7A1, the rate-limiting enzyme in bile acid synthesis. The cyclic expression of Cyp7a1 is driven also by Rev-erbα, DBP/E4BP4, and DEC2, contributing to the regulation of amplitude, phase and wave form over the 24-h profile, controlling gene expression by the use of Rev-RORE, DBP/E4BP4-binding elements, and E-boxes, respectively. Liver expression of E4BP4 and the orphan nuclear receptor small heterodimer partner (SHP), both negative regulators of Cyp7a1 expression, are increased in Rev-erbα knock-out mice with a minor impact on the amassing of hepatic fat. This is due to compensation by Rev-erbβ (Duez et al., 2008), whereas total hepatic Rev-erb deficiency significantly increases hepatosteatosis (Bugge et al., 2012). In turn, SHP restrains Npas2 gene transcription and promoter activity, and inhibits Rorγ transactivation and Rorα action, the latter driven by its interaction with Rev-erbα. Npas2 binds rhythmically to the Shp promoter, which drives the circadian expression of Shp with elevating NAD+ levels. Accordingly, Npas2 deficiency causes severe hepatosteatosis in Shp(–/–) mice through derangement of lipoprotein metabolism (Lee et al., 2015; Figure 2).

Autophagy plays a crucial role in NAFLD pathogenesis, with factors modulating autophagy proposed to have therapeutic utility (Lin et al., 2013; Fucho et al., 2014; González-Rodríguez et al., 2014; Jung et al., 2015; Martinez-Lopez and Singh, 2015; Wang et al., 2015). Autophagy is a complex, highly conserved, ATP-dependent cellular process taking place in all eukaryotic cells in response to different stressors, including fasting, changes in cell volume, oxidative stress, damaged protein accumulation, hormonal signals, and xenobiotic exposure (Kaur and Debnath, 2015). Autophagy contributes to the regulation of many biochemical processes, including: remodeling during development and cell differentiation; intracellular protein degradation; amino acid production during stress and nutrient shortage; the replacement of cytomembranes and organelles, such as mitochondria and peroxisomes; and the control of programmed, non-apoptotic cell death (Kaur and Debnath, 2015). Murine studies show autophagic vacuole numbers and autophagy activity, as well as autophagy gene expressions, to show circadian oscillation in the liver, which is driven by the basic leucine zipper transcription factor C/EBPβ and is under circadian regulation by the molecular clockwork (Ma et al., 2011). Timing cues driving transcriptional regulation of autophagy circadian patterns involve both the biological clock and nutritional signals (Ma et al., 2011).

Circadian rhythmicity of autophagy flux and autophagy gene expression is dampened in liver-specific Arntl null mice, implying a clock-related cell-autonomous mechanism. Nutritional cues entrain the biological oscillators of peripheral tissues. This is exemplified by time restricted feeding (Table 2), which resets the phase of peripheral clocks in rodents without affecting the SCN. This suggests a phase shift in the autophagy rhythm that is dependent upon feeding-related biological clock resetting (Atger et al., 2015). Rhythmic autophagy activation could maintain nutrient and energy homeostasis, reshape proteomes, and organelles, and attain temporal compartmentalization of tissue metabolism. Given that the expression of genes involved in glucose metabolism, de novo lipogenesis, cholesterol biosynthesis, and fatty acid β-oxidation oscillate in the liver, coordination of circadian patterns of metabolic cycles with autophagy rhythms could optimize nutrient supply for storage or utilization (Ma et al., 2012).

During glucose starvation, AMPK promotes autophagy by activating Ulk1, in a process involving phosphorylation of Ser 317 and Ser 777 residues, whilst this is restrained when nutrients are abundant via mammalian target of rapamycin (mTOR) (Kim et al., 2011). AMPK directly impacts the functioning of the biological clock, with Arntl negatively modulating mTOR oscillating activity (Khapre et al., 2014). Autophagy is crucial for cellular metabolic homeostasis, with the transcriptional control of autophagy being managed by the nutrient-sensing nuclear receptors peroxisome proliferator-activated receptor-α (PPAR-α) and FXR, which are driven by the biological clock and hepatically induced under nutrient-deprived conditions as well as in the fed state, respectively (Lee et al., 2014; Seok et al., 2014). The FXR transcriptionally represses autophagy in the liver upon feeding, opposing cAMP response-element binding protein (CREB)-mediated activation of autophagy during fasting. Indeed, autophagic genes are hallmarked by FXR binding site enrichment at transcription start sites and CREB binding site enrichment upstream (Lee et al., 2014; Seok et al., 2014).

On the whole, these data pinpoint to the pivotal role played by the molecular clockwork in hepatic steatosis pathogenesis, considering that the processes involved in intermediary metabolism and autophagy are hallmarked by a circadian pattern of variation and their rhytmicity is impacted by feeding–fasting cycles reverberating physiological and behavioral rhythmicity driven by the biological clock.

NAFLD may progress to NASH, hallmarked by steatosis, necroinflammation, and fibrosis, with varying degrees of severity. This can then progress to cirrhosis, portal hypertension and hepatocellular carcinoma (HCC). NASH may also arise independently of simple hepatic steatosis, driven by a diverse array of pathogenic events (Yilmaz, 2012). The progression of NAFLD to NASH, fibrosis and HCC is a pattern more commonly evident in obese humans and which is paralleled in wild-type mice challenged with chronic jet lag (Kettner et al., 2016). Hepatic circadian homeostasis regulation prevents this progression to HCC, including from the dysregulation of nuclear receptors involved in cholesterol/bile acid turnover and xenobiotic metabolism (Kettner et al., 2016). All the mechanisms involved in NAFLD pathogenesis and its evolution to NASH are handled by the molecular clockwork and include endoplasmic reticulum (ER) stress, lipotoxicity, insulin resistance, mitochondrial dysfunction, oxidative stress, adipose tissue dysfunction, deranged control of innate immunity, and cytokine secretion as well as alterations in the gut–liver axis.

NAFLD, without hyperglycemia or insulin resistance, may arise from a variety of different mechanisms, including triglyceride secretion hindrance, subcellular lipid hijacking, decreased lipolysis, increased lipogenesis, gluconeogenesis faults, and fatty acid β-oxidation interference (Sun and Lazar, 2013). Excessive lipogenesis weakens gluconeogenesis, redirecting precursors, and decreasing glycogen storage with resultant hypoglycemia (Sun and Lazar, 2013). In contrast, gluconeogenesis hindrance forces intermediates into lipogenesis and promotes hepatic steatosis. Impediments in lipid β-oxidation lead to triglyceride accumulation, whilst also obstructing gluconeogenesis due to associated shortages in energy and cofactors (Sun and Lazar, 2013).

Hepatic steatosis with systemic insulin hypersensitivity is also associated with FGF21 transactivation by the aryl hydrocarbon receptor (AHR), indicating that the AHR-FGF21 endocrine signaling pathway is a critical environmental transducer that integrates chemical and nutrient signaling in the liver (Lu et al., 2015). The FGF21 promoter contains several putative dioxin response elements (DREs). The AHR-ARNT heterodimer can bind to specific DREs that have overlapping binding sequences for PPARα, as well as for the carbohydrate response element-binding protein (ChREBP), and the cAMP response element-binding protein, hepatocyte specific (CREBH) (Cheng et al., 2014; Girer et al., 2016). The AHR, a member of the basis-helix-loop-helix, PAS domain (bHLH-PAS) family, acts as a ligand-activated transcription factor. The AHR interacts with the biological clock and binds xenobiotics, chiefly planar aromatic hydrocarbons and polyphenols, as well as endobiotics, including endogenous reactive by-products of normal metabolism such as indigoids, heme metabolites, eicosanoids, and tryptophan derivatives (Anderson et al., 2013). The interplay between the AHR/ARNT complex and the circadian network is also important to the increasingly recognized role played by these compounds, also called endocrine-disrupting or metabolism-disrupting chemicals, in NAFLD pathogenesis (Foulds et al., 2017). Such factors act on endocrine and metabolic organs through the AHR-dependent signaling pathway, among other nuclear hormone receptors, with their early-life exposure proposed to prime hepatosteatosis development and progression in adulthood (Foulds et al., 2017).

Experiments performed in AHR liver-specific and inducible transgenic (TG) mice [tetracycline responsive element (TetRE)–constitutively activated human AHR (CA-AHR) TG mice] fed an obesogenic diet showed that AHR activation worsens hepatic steatosis, but inhibits diet-induced obesity and systemic insulin resistance. The metabolic effects of the AHR are mediated through its direct transcriptional target FGF21 (Lu et al., 2015). The metabolic phenotypes shown in CA-AHR TG mice fed chow or a high fat diet could also be related to decreased expression of the circadian genes Clock, Per1, and Per2 with derangement of the biological clock contributing to the dissociation between fatty liver and insulin resistance (Vinciguerra and Mazzoccoli, 2016). The metabolic and protective effects of FGF21 were also shown in liver-specific SIRT1 knockout mice and their wild-type littermates when fasted as well as following SIRT1 activation or fasting mimicking chemical agents (Li et al., 2014). FGF1 also shows circadian expression and positive metabolic effects. FGF1 is regulated by the lipid sensor PPARγ and is crucial for adipose tissue function. Likewise FGF15/19, driven by FXR, may also have a significant role in these processes, including from its impact on bile acid metabolism (Nies et al., 2016).

In normal physiology, lipid and carbohydrate metabolism in the liver are circadianly coordinated with behavioral cycles, via the biological clock driving the proportionate rhythmic flow of metabolic intermediates between lipid synthesis and glucose production (Sun and Lazar, 2013). Lipids synthesized in the liver or deriving from other sources are included in lipid droplets before secretion or β-oxidation. Lipotoxicity ensues when lipids overload, thereby overwhelming the subcellular lipid sequestration capacity. In diurnal organisms, intermediary metabolites fueled by carbohydrate catabolism are incorporated into lipids in the fed state coinciding with daytime wakefulness and activity, whereas intermediates are redirected toward glucose production, a process propelled by lipid β-oxidation, in the fasted state associated with night-time sleep and rest (Sun and Lazar, 2013).

The expression of genes enriching lipid metabolic pathways is largely regulated by PPARα, a ligand-activated transcription factor of the steroid hormone receptor superfamily expressed mostly in tissues with high fatty acid catabolism, such as the liver, heart, and muscle. PPARα drives expression of genes encoding enzymes engaged in lipoprotein and lipid metabolism, through fatty acid uptake, binding, and activation [fatty acid transport protein (FATP), fatty acid translocase (FAT/CD36), liver cytosolic fatty acid-binding protein (L-FABP), acyl-CoA synthetase (ACS), carnitine palmitoyltransferase I and II (CPT-1 and CPT-II)], mitochondrial β-oxidation [very long chain acyl-CoA dehydrogenase (VLCAD), long chain acyl-CoA dehydrogenase (LCAD), medium chain acyl-CoA dehydrogenase (MCAD), short chain acyl-CoA dehydrogenase (SCAD)], peroxisomal β-oxidation [acyl-CoA oxidase (ACOX), bifunctional enzyme (HD), 3-ketoacyl thiolase (Thiolase)], microsomal ω-hydroxylation [Cytochrome P450 4A1 and 4A6 (CYP4A1 and CYP4A6)], hydrolysis of plasma triglycerides [lipoprotein lipase (LPL), apolipoprotein C-III (ApoC-III)], fatty acid synthesis [acetyl-CoA carboxylase (ACC), fatty acid synthase (FAS)], liver ketogenesis [HMG-CoA synthase (HMGCS)], fatty acid interconversion [tearoyl-CoA desaturase (SCD)], HDL metabolism [apolipoprotein A-I and A-II (ApoA-I and ApoA-II), ATP-binding cassette transporter 1 (ABCA1)], electron transport chain [uncoupling protein 1, 2, and 3 (UCP1, UCP2, and UCP3)] (Yoon, 2009; Mazzoccoli et al., 2012, 2014).

PPARα can bind endogenous ligands (fatty acids and fatty acid-derived compounds such as modified fatty acids, conjugated fatty acids, oxidized phospholipids, and fatty acid-derived eicosanoids (for example 8-S-hydroxytetraenoic acid and leukotriene) and synthetic ligands (carbaprostacyclin, non-steroidal anti-inflammatory drugs, pirinixic acid, phthalate ester plasticizers, and hypolipidemic drug fibrates, including fenofibrate, gemfibrozil, clofibrate, and ciprofibrate; Yoon, 2009). PPARα expression has a 24-h rhythmicity, driven by Arntl (Canaple et al., 2006), which heterodimerizes with Clock and transactivates PPARα target genes (Inoue et al., 2005). In turn, PPARα acts to regulate Arntl expression, directly binding PPAR response elements (PPRE) located at 1519 and 45 base pairs upstream of the transcription initiation site in the Arntl promoter. Accordingly, fenofibrates are capable of prompting Arntl expression in Rat-1 fibroblasts in vitro and in the mouse liver in vivo (Canaple et al., 2006).

Another member of the nuclear receptor ligand-activated transcription factors superfamily, PPARγ, is driven by the biological clock and is a molecular target for thiazolidinediones. PPARγ contributes to the control of adipogenesis, as well as glucose and lipid metabolism, and can bind prostaglandins and eicosanoids. PPARγ positively impacts on Arntl expression by binding its promoter at the same PPRE site, at −1,519 bp, which is involved in the response to PPARα activators (Wang et al., 2008). Experiments performed in Rat-1 fibroblasts in vitro and in the mouse liver have also shown that Arntl expression is controlled by the transcriptional repressor, and member of the 3 zinc finger family of Krüppel-like transcription factors (KLF), KLF10 (also known as TGFβ Inducible Early Gene-1, TIEG1). KLF10 senses intracellular glucose levels and joins two juxtaposed GC boxes near the Arntl gene transcription initiation site, operating in addition to RORE-dependent repression by Rev-erbα (Hirota et al., 2002, 2010). Arntl drives the 24-h periodicity of KLF10 expression, which in the mouse liver regulates the circadian variation of genes involved in lipogenesis, gluconeogenesis, and glycolysis (Guillaumond et al., 2010) The reciprocal control between PPARα/Arntl, PPARγ/Arntl, and KLF10/Arntl allows fine tuning of metabolic fluxes and determines “acceleration and braking,” thereby redirecting intermediary metabolites through the different metabolic pathways in line with the biochemical and behavioral rhythms related to feeding/fasting and waking/resting cycles, driven by the circadian timing system. Nutrient excess or feeding mistiming may overwhelm the liver handling capacity and cause imbalance between lipid and glucose metabolism, with ensuing hepatosteatosis.

The increasing worldwide epidemic of NAFLD and NASH, the association with multiorgan pathology and the lack of valuable pharmacological strategies, motivate research into ground-breaking pathophysiological mechanisms, leading to innovative therapeutic approaches for this pathology. The clinical and histological phenotypes of NAFLD and NASH are comparable worldwide, while the epidemiology and demographic characteristics vary in different countries around the world and proper categorization of NAFLD in line with comprehensive classifiers could afford more effective therapeutic and preventive strategies. Insulin resistance is associated with being overweight, and is a central pathogenetic mechanism in the majority of NAFLD cases in the Western world, whereas the disease is to a lesser extent associated with obesity and insulin resistance in the Eastern world (Loomba and Sanyal, 2013).

NAFLD, the most common hepatic disease in the Western world, is caused by deranged hepatic free fatty acid trafficking and metabolism and may proceed to NASH, leading to cirrhosis, portal hypertension and eventually liver cancer. The pathophysiology of NAFLD includes alterations in nutrient fluxes, bile acids turnover, nuclear receptors ligand binding, gastro-intestinal insulinotropic hormone secretion, microbiota-gut barrier interaction, autophagy, and the immune/inflammatory cascade. These processes are managed by the biological clock and disruption of their circadian synchronization and phase resonance with behavioral alternation of feeding/waking and fasting/resting provokes immune-metabolic dysregulation (Reinke and Asher, 2016; Tahara and Shibata, 2016). Chronic metabolic stress, related to ongoing nutrient surplus or mistimed feeding, overwhelms the fine-tuning of biochemical mechanisms, so that lipid influx engulfs the adipose tissue and causes the amassing of damaging lipid species in the liver and muscle. Lipotoxicity interferes with metabolic and immune/inflammatory signaling pathways and causes a pathological process defined as “metaflammation” that underlies NAFLD to NASH progression. Therapeutic approaches including time-related dietary patterns (chrononutrition) and life habits changes, as well as addressing the molecular clockwork and the circadian timing system are under evaluation. Further investigation of these factors may provide innovative strategies for the treatment of metabolic disorders and hepatosteatosis (Table 3). A more comprehensive knowledge of the molecular signaling pathways involved in NAFLD pathogenesis, as well as in NAFLD to NASH progression, could open the way to ground-breaking science-based strategies for prevention and therapy, such as pharmaceuticals directly targeting the biological clock or acting through nuclear receptors, bile acid signaling or gut microbiota modulation.

GM, SD, and TM: manuscript design, writing, revision; TM: bioinformatics analysis and figures drawing; GM, SD, and TM: approval of the final version of the manuscript.