3. Genetic Signature of Glucose and Lipid Metabolism in NAFLD
In the last decades, it clearly emerged that IR is a key player in NAFLD pathogenesis
[36],
[37]. In particular, IR strongly predicts the severity of hepatic fibrosis
[36], the main determinant of NAFLD prognosis
[38], and advanced fibrosis often occurs in NAFLD patients with T2D, even independently of inflammation and NASH
[39],
[40].Therefore, genetic variants that suppress the activation of insulin signaling may induce fibrosis in NAFLD
[41]. The rs1801278 (G972R) loss-of-function mutation in insulin receptor substrate (
IRS1) and the gain-of-function one in the ectonucleotide pyrophosphatase/phosphodiesterase1 (
ENPP1) 121Q genes were both related to dyslipidemia, obesity and hepatic fibrosis
[41]. On the contrary, the rs2954021 variant in tribbles homolog1 (
TRIB1), involved in the modulation of hepatic glycogen storage, affected plasma glucose, TG and cholesterol levels
[42].
Similarly, other variations in genes governing hepatic lipid handling and release predispose to fatty liver. For instance, variants within Apolipoprotein B (
APOB), involved in VLDL organization and secretion, have been associated with a protection against cardiovascular complications, due to the lowering of circulating lipoproteins and in turn, they favor severe hepatic fat depot formation, that may foster the progression of liver injury up to HCC
[43],
[44]. Moreover, even microsomal triglyceride transfer protein (
MTTP) inherited alterations may prompt VLDL retention
[45].
In addition, two common promoter variants in the apolipoprotein C3 (APOC3) (
APOC3 T-455C and C-482T), a component of chylomicrons, VLDL and HDL cholesterol particles, may predispose to steatosis in Indians, but not in other ethnic groups, supporting the notion that genetic factors modulating TG metabolism outside of the liver are less implicated in the onset of progressive NAFLD
[46],
[47],
[48].
Likewise, Dongiovanni et al.
[49], elucidated that the proprotein convertase subtilisin/kexin type 7 (
PCSK7) rs236918 G > C variant affects fasting lipids and hepatic injury in a large cohort of NAFLD subjects, coupling atherogenic dyslipidemia with NASH and hepatic fibrosis. By stratifying patients according to the presence of the PNPLA3 p.I148M allele, the
PCSK7 rs236918 polymorphism was associated with advanced steatosis. Indeed, in hepatoma cells carrying the p.I148M allele in homozygosity,
PCSK7 genetic deficiency decreased the expression of genes involved in DNL, inflammation and fibrosis, even after FFA supplementation
[49]. Furthermore, Huang et al.
[50], revealed a correlation between the
PCSK7 at-risk allele, hyperinsulinemia and homeostatic model assessment for IR (HOMA-IR) after a high-carbohydrate challenge.
PCSK7 is strongly implicated in lipid homeostasis, since it is localized in a genomic region close to the gene cluster
APOA5/APOA4/APOC3/APOA1, involved in lipoprotein metabolism regulation. In keeping with these findings,
Pcsk7−/− mice fed HFD are characterized by elevated plasma apolipoprotein concentrations and enhanced lipoprotein lipase (Lpl) adipose tissue activity
[51].
Even more, aberrancies in another member of the proprotein convertase subtilisin/kexin family, PCSK9, have been widely associated with hereditary hypercholesterolemia
[52], severe fat deposition
[53] and cardiovascular abnormalities
[54], due to its impact on LDL uptake. PCSK9 is a nutrient sensor, and it is strongly influenced by nutritional
status. Indeed, its expression declines in mice after 24 h of fasting. On the contrary, PCSK9 mRNA levels are renewed through SREBP-1c and DNL activation upon high carbohydrate refeeding or insulin stimulation
[55]. Loss-of-function mutations in
PCSK9 diminish plasma LDL cholesterol, without inducing steatosis
[56]. For example, the
PCSK9 rs11591147 (p.R46L) loss-of-function variant blunted LDL levels and protected against NAFLD, NASH and fibrosis, irrespectively of confounders
[57]. Conversely, gain-of-function alterations of
PCSK9 as the rs7552841 variant lead to familial hypercholesterolemia and enhance CAD risk.
Hepatic dysfunctions may be caused even by rare mutations of lysosomal acid lipase (
LIPA) gene, that induce lysosomal acid lipase (LAL) defects. LAL participates to the hydrolysis of cholesteryl esters, TG and LDL into free cholesterol and fatty acids. Its functional aberrancy fosters un-hydrolyzed compounds accumulation into the hepatocytes, whereby favoring atherogenic dyslipidemia, hepatic steatosis and severe fibrosis
[58],
[59]. In turn, LAL restoration using recombinant sebelipase alpha administration in patients for up to 5 years may improves liver enzymes, hepatic features of NAFLD and circulating lipids in patients (clinical trial NCT01488097)
[60],
[61].
Finally, even variants that alter FFA fluxes into the liver or their catabolism, such as the rs56225452 in fatty acid transport proteins (
FATP5) or the rs13412852 in Lipin1 (
LPIN1), may leverage IR and steatosis
[62],
[63].
4. Genetics of Lipid Droplets
LD accumulation in the liver is the primary hallmark of NAFLD. Compelling evidence indicates that LDs should not be considered as just ‘innocent bystander’, but conversely, they participate to multiple processes that lead to NASH. LDs are enormously dynamic, modifying their localization, size, lipid and protein composition in response to environmental stimuli and energy demand. Hence, they are engaged not only in energy expenditure to produce ATP but also in signaling pathways, acting as hubs that integrate metabolic and inflammatory processes.
Genetic risk factors may play a crucial role as modifiers of lipid composition and LD dimensions, whereby causing the alteration of the expression of LD-associated proteins, which regulates lipid storage. Thus, together with the above-mentioned p.I148M PNPLA3 and
HSD17B13 rs72613567 variants, many other polymorphisms in genes implicated in LD handling have been recently associated with NAFLD. Among them, Perilipin-2 (PLIN2) rs35568725 (Ser251Pro) variant has been associated with IR and atherosclerosis, in two population studies. In particular, PLIN2 protein participates to the formation, stability and trafficking of LDs and in VLDL lipidation. The Ser251Pro mutation induces smaller, but more numerous LDs in hepatocytes, resembling microvesicular steatosis and conveying the risk of NASH in NAFLD patients
[64]. The increasing number of small LDs was not translated into enhanced cellular capacity to store fat, but more so into high number of LD-associated proteins on the surface area. Thus, we could speculate that the presence of this variant may enhanced the risk related to PNPLA3 p.I148M carriage in patients. Even more,
PLIN2 variant has been associated with reduced circulating TG and VLDL
[65].
Similarly, the common noncoding polymorphism, rs884164 in another LD-associated gene,
PLIN5, causes a down-regulation of PLIN5 expression, a protein that facilitates the association between LDs and mitochondria. The recruitment of mitochondria to LDs during conditions of high substrate availability may favor lipid catabolism. Thus, PLIN5 hampered expression is associated with a poorer outcome following myocardial ischemia and
PLIN5 deficiency is related to increased oxidative stress in cardiomyocytes
[66],
[67].
Since the degradation of cellular lipids is mediated by a selective autophagic process, named lipophagy, an impairment in this mechanism induced by genetic defects may dampen lipid β-oxidations, accelerating their accumulation. For instance, the rs10065172 variant in the autophagy-related
IRGM gene may increase the risk of developing steatosis and
IRGM knockdown inhibits autophagic flux and increases LD content in HepG2 cells
[68]. IRGM is generally localized on endosomes/lysosomes, while in HFD-fed mice it co-localizes with ATGL/PNPLA2 at LD surface, where it recruits autophagic mediators, such as LC3B, in attempt to counteract to steatosis development. Hence, IRGM overexpression protects against hepatic lipid storage
[69].
5. Advanced Liver Injuries and Genetic Variants
A series of stressful triggers may precipitate fatty liver up to NASH and severe fibrosis. Among them are included oxidative stress fostered by reactive oxygen species (ROS) overproduction, intracellular organelle derangement, i.e., ER and mitochondrial abnormalities and dysfunctions, innate immune inflammation and pro-inflammatory cytokine and chemokine release
[70], intestinal high permeability and gut-derived harmful by-products (due to leaky gut)
[71], and HSCs activation to myofibroblasts
[72].
Interleukin 28 (
IL28) gene codifies for the interferon λ3/λ4 (IFNL3/4), and the rs12979860 CC variant has been associated with interferon λ3 over-production
[73]. The latter mediates the clearance of hepatitis C virus, and more aggressive NASH and fibrosis in NAFLD patients
[74],
[75]. In particular, it has been yielded a genetic model to predict significant fibrosis, named FibroGENE, that includes the rs12979860 variant, age, gender and the routinely assessed clinical and biochemical parameters
[76]. The rs12979860 is in linkage disequilibrium with the another variant, the
IFNL4 rs368234815 TT > δG. Patients carrying the rs368234815 TT allele are predisposed to develop higher degree of lobular inflammation and fibrosis compared to non-carriers
[77]. Contrasting findings have been observed in carriers of the rs3480 A > G variant in the fibronectin type III domain-containing protein 5 (
FNDC5) gene, encoding irisin, a myokine, that intervenes in HSCs activation and collagen deposition
[78],
[79]. The minor G allele predisposes to elevated extents of steatosis, likely by modulating irisin expression
[79]. Likewise, the rs2228603 polymorphism in Neurocan, the rs12137855 variation in lysophospholipase-like 1 (
LYPLAL1) and the rs10883437 SNP close to the carboxypeptidase n subunit 1 (
CPN1) have been coupled to severe NAFLD
[80],
[81].
Alongside, mounting evidence indicates that the gut-derived fibroblast growth factor (FGF) 19, engaged in lipid and carbohydrate metabolism in response to nutritional
status through the binding to its hepatic receptor, fibroblast growth factor receptor 4 (FGFR4), is involved in metabolic diseases and NAFLD
[82]. Dongiovanni and Crudele et al.
[83], demonstrated that the rs17618244 G > A variant in the β-Klotho (
KLB) gene, encoding the hepatic co-receptor of FGFR4, dampened KLB plasma levels, leading to inflammation, ballooning, fibrosis and to the over-expression of genes involved in lipotoxicity in overweight NAFLD pediatric patients
[83]. Furthermore, KLB complexing with others FGFRs also binds the hormone FGF21, released from the liver and adipose tissue. In detail, FGF21 is implicated in glucose and TG uptake by white and brown adipose tissue, through the interaction with FGFR1
[84]. However, FGF21 circulating levels are paradoxically increased in obese patients and in those with NAFLD, as a protective response to KLB down-regulation and to NAFLD-induced adverse effects, e.g., lipotoxicity, oxidative and ER stress
[85],
[86]. Thus, KLB/FGF19/FGF21 pathway may represent a druggable target in NAFLD patients through the rescue of KLB levels.
Concerning the development of fibrosis, the alternative splicing of the Krueppel-like factor 6 (
KLF6) gene, that is expressed by the HSCs during their transdifferentiation, associates with mild NAFLD and reduced fibrosis
[87]. Conversely, variants in
HFE and
TMPRSS6 genes likely by predisposing to hepatic iron depot formation are correlated with more severe fibrosis in NAFLD patients
[88].
The rs4374383 non-coding variant in the macrophage c-mer tyrosine kinase (
MERTK), a tyrosine kinase that initiates the removal of dying cells by phagocytes and that mediates HSCs activation, protects against fibrosis in both NAFLD and in viral hepatitis C, eliciting MERTK down-modulation
[89],
[90]. Consistently, it has been stated that MerTK cleavage in hepatic macrophages is reduced during the transition from simple steatosis to NASH, promoting transforming growth factor β (TGF-β) release and HSCs activation
[91]. Novel insights into the role of MERTK in metabolic processes, has been brilliantly proposed by Nicolás-Ávila and colleagues
[92], which demonstrated that macrophages may actively entrap materials, including dysfunctional mitochondria ejected from injured cardiomyocytes through dedicated membranous particles enriched in phosphatidyl-serine (PS), with the purpose to maintain the global tissue homeostasis. This peculiar process occurs in MERTK-dependent manner, and it is driven by the cardiomyocytes’ autophagy machinery, prompted by cardiac stress. Thus, MERTK depletion abolished the removal of the exhausted mitochondria, hindered autophagic processes and resulted in the inflammasome and autophagy arrest, ultimately compromising mitochondrial fitness. Thus, this novel non-canonical route for the extrusion of cellular waste, including abnormal mitochondria and other organelles into the extracellular space, then scavenged by resident macrophages, may pave the way to potential translational implications on the study of other tissues characterized by high mitochondrial biomass and energy demand, in both healthy and disease status.
Finally, the susceptibility to fibrogenesis and carcinogenesis is also influenced by cellular senescence and cell cycle arrest. Therefore, the rs762623 in cyclin dependent kinase inhibitor 1A (
CDKI1A) which encodes the cellular senescence marker p21, was significantly associated with disease progression in NAFLD
[93]. Likewise, telomerase reverse transcriptase (
TERT) gene loss-of-function mutations associated with familial cirrhosis and accelerated HCC
[94]. Similarly, the rs599839 A > G variant, which causes the overexpression of the oncogene Proline And Serine Rich Coiled-Coil 1 (
PSRC1), has been associated with enhanced HCC risk in NAFLD patients, irrespectively of fibrosis severity, and with poor prognosis and advanced tumor stage
[95]. Even more, the Neurotensin (
NTS) rs1800832 variant predisposes to cirrhosis and HCC in NAFLD patients likely by affecting NTS protein activity
[96].
6. Mitochondrial Dysfunctions: The Tipping Point in the Switching from Simple Steatosis to Steatohepatitis
Growing evidence pinpoints the critical role of organelle abnormalities in the switching from fatty liver towards NASH. Mitochondrial anomalies are closely entangled into the pathogenesis of NAFLD so much so that it has been considered as a mitochondrial disease
[97]. During early stages of NAFLD, mitochondrial activity and biomass is adapted in response to IR and to fat accumulation. However, sustained mitochondrial oxidative flux hesitates in exasperated ROS production, triggering phospholipid lipoperoxidation, cellular stress and mitochondrial DNA damage, tissue inflammation and cell death which may precipitate the progression to NASH and more advanced liver injuries
[98].
In this context, the knowledge of genetically determined mitochondrial dysregulations may be determinant to predict the course of the disease. Indeed, common polymorphisms in genes regulating mitochondrial homeostasis have been associated with NAFLD and to its progressive forms. For instance, the rs4880 C47T variant in the superoxide dismutase 2 (
SOD2) gene, encoding the antioxidant enzyme manganese superoxide dismutase, results in a Valine to Alanine substitution in the signal region addressing the protein to the mitochondrial matrix, where it exerts its function, and the T allele has been related to increased enzymatic activity. Thus, an higher frequency of
SOD2 T/T genotype in biopsy-proven NASH patients compared to healthy controls has been reported
[99]. This variant has been further associated with severe fibrosis in NAFLD patients, as a proof of concept that mitochondria-derived oxidative stress is required for fibrosing NASH onset
[100].
Alongside, the homozygosity for the −866 G > A mutation in the promoter region of the uncoupling protein 2 (
UCP2) gene protects against NASH, whereby enhancing hepatic UCP2 expression
[101]. The latter is implicated in the regulation of mitochondrial lipid efflux and oxidative metabolism and its hepatic expression increased in NASH patients causing a proton leak and a reduction of redox pressure on the mitochondrial respiratory chain, protecting the hepatic tissue against liver damage worsening
[102].
Conversely, a non-coding variant in the promoter (−55C > T, rs1800849) of another member of the UCP family, the uncoupling protein 3 (
UCP3) gene has been correlated with low insulin sensitivity, IR, reduced adiponectin secretion, moderate-severe hepatic steatosis and inflammation in obese NAFLD individuals
[103]. UCP3 is a mitochondrial proton transporter that protects against fatty acid-mediated oxidative stress, uncoupling the oxidative phosphorylation by increasing the proton leak of the inner mitochondrial membrane.
Sirtuins (SIRTs) are a family of nicotinamide adenine dinucleotide (NAD
+)-dependent deacetylases embroiled in cellular metabolism. There are 7 distinct SIRTs in mammals (SIRT1–7), which share the catalytic core domain, but they have different subcellular localizations. Indeed, SIRT1, SIRT6 and SIRT7 are mainly localized into the nuclei, SIRT2 is primarily found into the cytoplasm, while SIRT3, SIRT4 and SIRT5 have a mitochondrial distribution
[104]. SIRTs along with UCPs may modulate oxidative stress thereby influencing the risk of subclinical atherosclerosis and cardiovascular complications. Indeed, it has been demonstrated that the
SIRT6 rs107251 and the
SIRT5 rs12216101 were associated with an elevated susceptibility to carotid plaques formation, whereas carriers of the T allele of
UCP5 rs5977238 had a lower risk, in 1018 stroke-free subjects from the Northern Manhattan Study (NOMAS)
[105]. Even though cardiovascular abnormalities are recurrent in NAFLD patients, the implication of SIRTs genetic variations in this context remains to be fully elucidated.
More recently, a novel common missense variant (rs2642438 A165T) in the mitochondrial amidoxime-reducing component 1
(MARC1) gene has been identified.
MARC1, also known as
MTARC1 or
MOSC1, encodes the mitochondrial amidoxime reducing component 1, a molybdenum-containing enzyme that regulates endogenous nitric oxide levels and biosynthesis, catalyzing the conversion of nitrite to produce nitric oxide. The A165T variant is located at the N-terminal domain which anchors the protein to the outer membrane of the mitochondria. The threonine to alanine aminoacidic substitution results in a truncating protein making the rs2642438 a loss-of-function mutation. The A165T variant has been associated with protection against all-cause cirrhosis, reduced hepatic fat content and lower levels of liver enzymes
[106]. Specifically, in patients affected by alcohol-related cirrhosis
MARC1 and heterogeneous nuclear ribonucleoprotein U like 1 gene (
HNRNPUL1) variations has been emerged as risk modifiers of liver damage, in a GWAS of samples from the United Kingdom Biobank
[107]. Afterwards, Luukkonen and collaborators
[108] investigated the effect of the rs2642438 variant on the severity of NAFLD and they demonstrated that patients carrying the A165T allele had markedly lower prevalence of inflammation and fibrosis, compared to non-carriers. This effect seems to be due to the precise lipid signature that describes A165T allele carriers, displaying increased levels of hepatic polyunsaturated-PC similarly to carriers of the
HSD17B13 rs72613567 variant and opposite to what the same authors observed in
PNPLA3 p.I148M carriers. According to these observations, the wt forms of
MARC1 are related to higher levels of sphingomyelins (i.e., C20:2), Lyso-PC (C14:0 and C15:0) and PC (C34:1 and C40:2) compared to patients carrying the A165T allele, thereby confirming the presence of a distinctive metabolomic pattern by using comprehensive metabolomics data from two population-based studies, including 9135 participants from the Fenland study and 9902 participants from the EPIC-Norfolk cohort
[109]. Collectively, these observations pointed out
MARC1 as a potential pharmacologic target for liver diseases without affecting cardiovascular outcomes
[110], although further investigations are needed to clarify its function and its role in oxidative stress regulation. A schematic over-view of the main genetic risk factors involved in NAFLD onset and progression is represented in
Figure 1 and in
Table 1.
Figure 1. Impact of genetics in NAFLD pathogenesis and progression towards advanced liver damage. Schematic illustration of the most relevant inherited variations involved in progressive NAFLD, shedding light into their functional effects. PNPLA3, localized at the LD surface in hepatocytes, catalyzes TG hydrolysis. The p.148M variant enhances hepatic TG content upon mutant protein accumulation, hampering TG turnover and dismissal. TM6SF2 is implicated in VLDL formation in ER and release, whereas MBOAT7 transfers arachidonoyl-CoA to Lyso-PI, maintaining membrane fluidity. Their variations dampen VLDL secretion and membrane dynamism, respectively. Viceversa, genetic variants in HSD17B13 and PPP1R3B may exert a protective effect against NAFLD. Heritable variations may also influence glucose and insulin signaling, FFA uptake, fat deposition and VLDL turnover, precipitating fatty liver. In addition, IR and elevated FFAs derived from adipose tissue lipolysis exacerbate fat depot formation induced by genetic modifiers, even activating DNL. Recently, common SNPs in modulators of mitochondrial (MT) function have been proposed as active players in the switching from steatosis to NASH and fibrosis, further corroborating the role of organelle abnormalities in these processes. Furthermore, variants in genes regulating inflammatory response and HSCs activation may precipitate fatty liver to worsened conditions. Finally, genetically determined perturbations in circulating lipids may trigger cardiovascular comorbidities. Dotted lines refer to influx and efflux processes into the hepatocyte, whereas solid lines refer to cell activation or to the transition from simple steatosis up to cirrhosis-HCC.
Table 1. Schematic list of the main inherited variations related to the histological hallmarks of NAFLD.
Variant
|
Gene
|
Global MAF
|
Function
|
Effect
|
Impact
|
Phenotype
|
rs738409 C > G
|
PNPLA3
|
0.26 (G)
|
Lipid remodeling
|
p.I148M
|
Loss-of-function
|
↑ NAFLD, NASH, fibrosis, HCC
|
rs58542926 C > T
|
TM6SF2
|
0.07 (T)
|
VLDL secretion
|
p.E167K
|
Loss-of-function
|
↑ NAFLD, NASH, fibrosis
|
rs641738 C > T
|
TMC4/
MBOAT7
|
0.37 (T)
|
Lipid remodeling
|
p.G17E
|
Loss-of-function
|
↑ NAFLD, NASH, fibrosis, HCC
|
rs1260326 C > T
|
GCKR
|
0.29 (T)
|
Regulation of DNL
|
p.P446L
|
Loss-of-function
|
↑ NAFLD, NASH, fibrosis
|
rs72613567 T > TA
|
HSD17B13
|
0.18 (TA)
|
Lipid remodeling
|
Truncated protein
|
Loss-of-function
|
↓ NASH, fibrosis, HCC
|
rs4841132 G > A
|
PPP1R3B
|
0.09 (A)
|
Glycogen synthesis
|
Non-coding
|
Gain-of-function
|
↓ NAFLD, fibrosis, HCC
|
rs1801278 C > T
|
IRS1
|
0.05 (T)
|
Insulin signaling
|
p.G972R
|
Loss-of-function
|
↑ Fibrosis
|
rs1044498 A > C
|
ENPP1
|
0.34 (C)
|
Insulin signaling
|
p.K121Q
|
Gain-of-function
|
↑ Fibrosis
|
rs2954021 G > A
|
TRIB1
|
0.45 (A)
|
Regulation of DNL
|
Non-coding
|
Gain-of-function
|
↑ NAFLD
|
rs12137855 C > T
|
LYPLAL1
|
0.16 (T)
|
Lipid metabolism
|
Intronic
|
Loss-of-function
|
↑ NAFLD
|
Several
|
APOB
|
NA
|
VLDL secretion
|
Protein change
|
Loss-of-function
|
↑ NAFLD NASH, fibrosis, HCC
|
Several
|
MTTP
|
NA
|
VLDL secretion
|
Protein change
|
Loss-of-function
|
↑NAFLD
|
rs236918 G > C
|
PCSK7
|
0.26 (C)
|
Membrane transferrin receptor shedding and regulation of circulating lipids
|
Intronic
|
Gain-of-function
|
↑ NASH, fibrosis
|
Several
|
PCSK9
|
NA
|
LDL uptake
|
Protein change
|
Loss-of-function
|
No evidence of association with steatosis
|
Several
|
LIPA
|
NA
|
Lipid remodeling
|
Protein change
|
LAL deficiency
|
↑ NAFLD, NASH, fibrosis
|
rs56225452 G > A
|
FATP5
|
0.16 (A)
|
FFAs uptake
|
Non-coding
|
Gain-of-function
|
↑ NASH, fibrosis
|
rs13412852 C > T
|
LPIN1
|
0.21 (T)
|
Lipid metabolism
|
Intronic
|
Not Defined
|
↓ NASH, fibrosis
|
rs35568725 A > G
|
PLIN2
|
0.02 (G)
|
Lipid remodeling
|
p.S251P
|
Loss-of-function
|
↑ NAFLD, NASH, IR, atherosclerosis
|
rs884164 A > G
|
PLIN5
|
0.19 (G)
|
Lipid remodeling
|
Non-coding
|
Loss-of-function
|
↑ oxidative stress
|
rs17618244 G > A
|
KLB
|
0.15 (A)
|
FGF19/FGFR4 pathway
|
p.R728Q
|
Loss-of-function
|
↓ NASH, fibrosis
|
rs4374383 G > A
|
MERTK
|
0.45 (A)
|
Innate immunity
|
Intronic
|
Loss-of-function
|
↓ Fibrosis
|
rs3750861 G > A
|
KLF6
|
0.07 (A)
|
HSCs activation
|
Splice variant IVS1-27G
|
Loss-of-function
|
↓ Fibrosis
|
Several
|
TERT
|
NA
|
Telomere maintenance
|
Protein change
|
Loss-of-function
|
↑ Fibrosis, HCC
|
rs12979860 C > T
|
IL28B
|
0.36 (T)
|
Innate immunity
|
Alternative IFNL3/4
transcription
|
Loss-of-function
|
↓ NASH, Fibrosis
|
rs3480
A > G
|
FNDC5
|
0.42 (G)
|
HSCs activation
|
Non-coding
|
Loss-of-function
|
↓ Fibrosis
|
rs4880
C > T
|
SOD2
|
0.33 (T)
|
Mitochondrial antioxidant
|
p.A16V
|
Loss-of-function
|
↑ Fibrosis
|
rs695366 G > A
|
UCP2
|
0.26 (A)
|
Mitochondrial lipid metabolism Oxphos
|
−866 promoter variant
|
Gain-of-function
|
↓ NASH, fibrosis
|
rs2642438 G > A
|
MARC1
|
0.19 (A)
|
Mitochondrial detoxification
|
p.A165T
|
Loss-of-function
|
↓ NAFLD, NASH, fibrosis
|
MAF: minor allele frequency.