INTRODUCTION
Metabolic dysfunction-associated steatotic liver disease (MASLD) and metabolic dysfunction-associated steatohepatitis (MASH) are major contributors to global morbidity and mortality [1,2]. The renaming of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis to MASLD and MASH reflects a deeper understanding of the metabolic factors driving these conditions [3,4]. Almost all patients with NAFLD meet the diagnostic criteria for MASLD, and therefore the natural history of these two conditions is virtually identical [5]. MASLD affects approximately 30% of the global population, and its prevalence is closely connected to the rising incidences of obesity, type 2 diabetes, and metabolic syndrome [6]. The most robust genetic predictors of MASLD are single nucleotide polymorphisms of the genes encoding patatine-like phospholipase domain containing 3 (PNPLA3) [7], transmembrane 6 superfamily mem-ber 2 (TM6SF2) [8], membrane-bound o-acyltransferase domain containing 7 (MBOAT7) [9], and glucokinase regulator [10]. These genetic polymorphisms affect the metabolism of hepatocytes, thereby contributing to steatosis and disease progression. As MASLD progresses to MASH, monosaccharide, fatty acid, and amino acid metabolism become significantly disrupted (Fig. 1). Although VLDL secretion and β-oxidation are upregulated in an attempt to compensate for early MASLD, this is often insufficient, and the steatosis is exacerbated, such that MASH develops. Insulin resistance (IR) results in an increase in gluconeogenesis, and the impaired β-oxidation and enhanced lipogenesis cause lipid accumulation in the liver [11]. The accumulation of fatty acids also increases reactive oxygen species (ROS) production in mitochondria, which drives oxidative stress and inflammation in the liver. Lipotoxicity in hepatocytes causes the release of extracellular vesicles, which induce inflammation, eventually leading to fibrosis [12,13]. Moreover, the dysfunction of mitochondria owing to amino acid imbalances further exacerbates the liver injury through a deficient oxidative stress response [14-16]. Recent studies have shown that IR results in an upregulation of gluconeogenesis, and glucagon resistance (GR) suppresses amino acid uptake into the liver, leading to hyperaminoacidemia and intrahepatic amino acid depletion, which may also contribute to the pathogenesis of the disease (Fig. 1).
In this review, we first summarize the abnormalities in monosaccharide and fatty acid metabolism that characterize MASLD/MASH, then focus on amino acid metabolism, explaining the mechanisms whereby amino acids play a significant role in disease pathology. Furthermore, we review recent research regarding the relationships between amino acid imbalances and the gut microbiota in MASLD.
MONOSACCHARIDE METABOLISM IN MASLD/MASH
Monosaccharides, and especially glucose and fructose, play a pivotal role in the pathogenesis of MASLD and MASH. Abnormalities in glucose homeostasis are closely linked to IR, a condition that is commonly associated with obesity and type 2 diabetes, which are major risk factors for MASLD/MASH [17]. As IR worsens, hepatic gluconeogenesis and glycogenolysis become unchecked [18], resulting in an increase in hepatic glucose output, which contributes to chronic hyperglycemia. This, in turn, induces a state of glucotoxicity that promotes hepatic inflammation and fibrosis [19].
Selective IR plays a crucial role in MASLD/MASH [20]. Gluconeogenesis typically occurs in periportal (zone 1) hepatocytes, whereas glycolysis and de novo lipogenesis (DNL) predominantly occur in perivenous (zone 3) hepatocytes (Fig. 2). Insulin receptor substrates 1 (IRS1) and 2 (IRS2) are critical mediators of insulin signaling that are expressed according to this metabolic zonation. IRS1 signaling is primarily involved in lipogenesis and glucose metabolism, whereas IRS2 signaling suppresses gluconeogenesis; and a recent study showed that IRS1 expression predominates in zone 3, whereas that of IRS2 predominates in zones 1 and 3 (Fig. 3). In states of diabetes and obesity, IRS2 expression is low in both zones 1 and 3. Conversely, the phosphorylation of IRS1 remains relatively intact; therefore, gluconeogenesis is promoted in zone 1 and lipogenesis is promoted in zone 3, contributing to hepatic steatosis and causing selective IR [21]. The decrease in IRS2 expression increases sterol regulatory element-binding protein 1 (SREBP-1) expression [22] in mice and enhances gluconeogenesis in the livers of patients with MASLD [23]. Exendin-4, a glucagon-like peptide-1 receptor agonist, has been shown to increase IRS-2 phosphorylation and insulin signaling in human hepatocytes [24].
The upregulation of glycolysis, involving increases in the activities of hexokinase and pyruvate kinase (PK), leads to the accumulation of glycolytic intermediates, which can be diverted to DNL in zone 3 hepatocytes in MASLD (Fig. 2) [25]. Simultaneously, higher expression of gluconeogenic enzymes, such as glucose-6-phosphatase (G6PC) and phosphoenolpyruvate carboxykinase (PEPCK), further increases glucose production in zone 1, reinforcing the hyperglycemia [26]. These dual abnormalities in glycolysis and gluconeogenesis create a metabolic environment that is conducive to liver injury and the progression of MASH. The PI3K/AKT/FoxO1 pathway promotes glucose uptake and glycogen synthesis, while inhibiting gluconeogenesis [27]. AKT suppresses the activation of the transcription factor FoxO1, which regulates the expression of the genes encoding gluconeogenic enzymes such as PEPCK and G6PC (Fig. 2) [28]. The interaction between FoxO1 and PGC1a modulates gluconeogenesis via insulin signaling [29]. However, PI3K/AKT/FoxO1 signaling is impaired in IR, leading to a decrease in glucose uptake and an increase in gluconeogenesis [30]. FOXO1 expression and activity are high in patients with MASLD/MASH [31].
Fructose, which also plays a key role in the pathogenesis of MASLD/MASH, is rapidly cleared from the blood by hepatic glucose metabolism. Hepatocytes take up fructose via GLUT2 and GLUT5 transporters and it enters the glycolytic pathway at the level of fructose-1,6-bisphosphate (F-1,6-BP), such that it is rapidly metabolized. Consequently, although normal fasting glucose concentrations in the peripheral blood are approximately 5 mM, fructose circulates at concentrations <0.02 mM under fasting conditions [32]. Excessive fructose intake has been shown to induce MASH in mouse models [33]. It increases DNL, suppresses fatty acid oxidation, induces endoplasmic reticulum (ER) stress, and activates JNK signaling, thereby promoting hepatic inflammation [34] and steatosis [35]. ChREBP is essential for the fructose-associated upregulation of intestinal and hepatic G6PC expression [36]. A restriction of fructose intake has been shown to significantly ameliorate hepatic steatosis in a double-blind, randomized controlled trial [37]. However, systematic reviews and meta-analyses have generated inconclusive findings regarding the relationship between excessive fructose consumption and MASLD [38].
FATTY ACID METABOLISM IN MASLD/MASH
In MASLD and during the early stages of MASH, fatty acid oxidation is often upregulated in compensation for the greater influx of fatty acids. However, as MASH progresses, mitochondrial dysfunction and oxidative stress cause an impairment in fatty acid oxidation [39], leading to lipid accumulation and lipotoxicity. IR and the activation of SREBP-1 promote DNL, further worsening hepatic fat accumulation. The expression of genes encoding key lipogenic proteins, such as acetyl-CoA carboxylase (ACC), fatty acid synthase, elongation of very long-chain fatty acids, and stearoyl-CoA desaturase 1 (SCD1) (Fig. 2) is regulated by transcription factors such as SREBP-1 and ChREBP, and these proteins are considered to be potential therapeutic targets for MASLD/MASH, because they are upregulated in the livers of patients with MASLD/MASH, in association with IR [40,41]. A recent study demonstrated that the inhibition of SREBP-1 suppresses lipogenesis but exacerbates liver injury and carcinogenesis in mice with MASH [42]. However, a partial hepatic SCD1 inhibitor, aramchol, has been shown to slow the pathogenesis in a phase 2b clinical trial [43]. These results suggest that lipogenesis is a physiologic response to the excessive influx of carbohydrates and fatty acids and that inappropriate suppression of lipogenesis may exacerbate lipotoxicity in hepatocytes.
As for SREBP1, the role of ChREBP in MASLD/MASH and whether its suppression or activation is beneficial remains to be determined. Liver-specific inhibition of ChREBP ameliorates hepatic steatosis and IR in mice with MASLD [44], and ChREBP expression is high in hepatocellular carcinomas (HCCs) in patients with MASLD/MASH [45]. In contrast, ChREBP expression in the livers of patients with MASH is high when there is >50% steatosis and has an inverse relationship with IR [40]. HNF4α regulates genes that are essential for gluconeogenesis, bile acid synthesis, cholesterol, and lipid transportation [46]. The liver-specific deletion of the gene disrupts VLDL secretion and increases lipid accumulation by reducing ApoB, MTP, and PPARα expression [47]. Low expression of HNF4α has been demonstrated in MASLD [48]. The secretion of VLDL primarily facilitates the export of lipids from the liver (Fig. 2). Many molecules are involved in the synthesis, transport, and secretion of VLDL in hepatocytes, and their regulation remains underexplored [49]. PNPLA3 and TM6SF2 genetic variants influence the assembly and release of VLDL particles, which transport triglycerides to peripheral tissues [50]. The PNPLA3 I148M variant impairs PUFA mobilization [51], thereby reducing VLDL secretion and increasing hepatic fat accumulation. The TM6SF2 E167K genetic variant is associated with a specific reduction in the hepatic secretion of large, triglyceride-rich VLDL1 [52].
Saturated fatty acids (SFAs) are closely linked to the pathogenesis of MASLD and MASH, mainly through their roles in promoting hepatic steatosis, lipotoxicity, and inflammation [53]. Palmitic acid, in particular, has been shown to have lipotoxic effects and trigger ER stress [54], mitochon-drial dysfunction, and apoptosis in hepatocytes [55]. Hepatocytes under ER stress secrete exosomes, which induce macrophage migration and trigger inflammation [13,56]. Recent studies have also highlighted the role of SFAs as promoters of fibrosis in MASH. Palmitic acid has been shown to activate hepatic stellate cells (HSCs), which are responsible for collagen deposition and fibrosis in the liver [57]. In MASLD, excessive dietary intake of SFAs, combined with greater endogenous monounsaturated fatty acid (MUFA) synthesis, increases the serum MUFA concentration and lipid accumulation in hepatocytes [58]. This is primarily driven by IR, which promotes lipogenesis via the activation of SREBP-1 [59]. Interestingly, despite the association between MUFA accumulation and hepatic steatosis, previous studies have shown that diets rich in oleic acid, such as the Mediterranean diet, may protect against the progression to MASH [60]. Although their accumulation contributes to the overall lipid burden in the liver, exacerbating disease progression, MUFAs are less lipotoxic than SFAs and protect against the palmitate-induced lipotoxicity triggered by triglyceride accumulation in hepatocytes [61,62]. Furthermore, MUFAs reduce the expression of pro-inflammatory cytokines, potentially mitigating the inflammatory component of MASH [63].
The two main classes of polyunsaturated fatty acids (PUFAs) are omega-6 and omega-3 fatty acids. The former has pro-inflammatory effects and promotes the progression of MASLD, whereas the latter has anti-inflammatory effects and ameliorates this condition [64]. Clinical studies have demonstrated that omega-3 PUFA supplementation reduces the liver enzyme activities, hepatic steatosis, and plasma lipid concentrations in patients with MASLD [65], thereby reducing the risk of coronary events [66]. However, conflicting data have been obtained regarding the efficacy of omega-3 PUFAs [67,68]. Therefore, further studies are needed to assess their efficacy [69]. These findings imply that metabolic flexibility, the ability to switch between glucose and fatty acid utilization, is impaired in MASLD/MASH. The contributing factors include IR, mitochondrial dysfunction, and a disruption of nutrient signaling. Whether this inflexibility is a cause or consequence remains to be determined, but both imbalances in and the chronic excessive intake of carbohydrates, lipids, and amino acids, as well as genetic factors that affect metabolic flexibility, contribute to the pathogenesis of MASLD. Furthermore, the resulting vicious cycle is thought to promote progression to MASH.
AMINO ACID METABOLISM IN MASLD/MASH
Amino acids are critical mediators of metabolic homeostasis, as well as being components of proteins. As substrates for gluconeogenesis, precursors for neurotransmitters, and regulators of cellular signaling pathways, amino acids are closely associated with energy balance in the liver. The glucogenic amino acids serve as substrates for gluconeogenesis in MASLD and MASH [70]. In the liver, they play pivotal roles in nitrogen metabolism, ureagenesis, and mitochondrial function, orchestrating metabolic adaptations to nutritional and pathologic states. Notably, enzymes such as aspartate aminotransferase (AST) and alanine aminotransferase (ALT), which are involved in amino acid metabolism, are widely used as biomarkers of liver injury, regardless of its etiology. Nevertheless, the mechanisms by which amino acid metabolism affects disease pathogenesis are not as well understood as those for glucose and fatty acid metabolism. However, given the central role of amino acid metabolism in hepatic physiology, the dysregulation of amino acid metabolism has emerged as a key feature of MASLD/MASH.
Amino acid utilization and imbalance in MASLD/MASH
Under normal aerobic conditions, FAAs are metabolized and enter the TCA cycle in the mitochondria, resulting in the synthesis of NADH, which is essential for oxidative phosphorylation and ATP synthesis (Fig. 4). Recent research has demonstrated that amino acids, rather than glucose, are the principal substrates for the TCA cycle in the mitochondria of the liver [71]. Therefore, the intracellular and extracellular FAA concentrations are closely associated with mitochondrial function. The mechanisms by which the concentrations of individual FAAs are sensed and how they influence energy metabolism remain largely unclear; however, they are expected to involve sirtuin 1 (SIRT1) and AMP-activated protein kinase (AMPK), which are activated by high NAD+/NADH and AMP/ATP ratios, respectively. The changes in circulating FAA concentrations that characterize MASLD/MASH are well known. The circulating concentrations of branched-chain amino acids (BCAAs), such as leucine, isoleucine, and valine, as well as those of glutamate and tyrosine, are often high in MASLD/MASH [72,73]. Conversely, the concentrations of amino acids such as glycine and serine are frequently low. In contrast, little is known about the changes in FAA concentrations that occur in the liver during the progression of MASLD. A recent study showed that some glucogenic FAAs and NAD+ are present at low concentrations in the livers of mice with MASH, and this is accompanied by mitochondrial dysfunction in hepatocytes [73]. The differences in amino acid concentrations between the intrahepatic and extrahepatic compartments are thought to be the result of GR and IR, which reduce amino acid uptake into hepatocytes and promote gluconeogenesis from glucogenic amino acids, respectively [74,75]. The fasting plasma concentration of FAAs significantly correlates with the plasma glucagon concentration of patients with MASLD, independently of glycemic control [76]. The amino acid imbalance that characterizes MASLD is further modified during the progression of chronic liver disease owing to the reduction in hepatocyte number [77,78]. In early MASLD, the concentrations of BCAAs are high, but these rapidly decline with the onset of liver cirrhosis. Aromatic amino acids (AAAs), tyrosine, and phenylalanine are also present at high concentrations during the early stage, but further increase as the disease progresses. These changes in intracellular and extracellular FAA concentrations that occur in MASLD largely reflect the con-sequences of metabolic dysfunction. However, an increasing quantity of evidence suggests that these changes also have effects on metabolism.
Glutamine and glutamate
Glutamine and glutamate metabolism is clearly demarcated according to the zonation of the hepatic lobule (Fig. 3) [79]. In zone 1, glutaminase (GLS) in mitochondria converts glutamine to glutamate and ammonia, which is detoxified by the urea cycle. GLS exists in two isoforms, GLS1 and GLS2. In the liver, GLS2 is mainly expressed in hepatocytes, whereas GLS1 is expressed at low levels in hepatocytes, but at high levels in activated HSCs and cancer cells. Glutamate enters the TCA cycle for ATP production, and gluconeogenesis and FFA oxidation occur in zone 1. Lactate and glucogenic amino acids are the primary substrates for gluconeogenesis (Fig. 1). However, recent studies have demonstrated that glucagon increases GLS2 expression and activity, thereby promoting the hydrolysis of glutamine to glutamate and α-ketoglutarate (Fig. 4), which are used for gluconeogenesis [80]. This is consistent with the zonation of gluconeogenesis, because GLS2-expressing cells are predominantly localized to the regions where gluconeogenesis occurs.
Zone 3, which is characterized by lower oxygen levels, is specialized for anabolic and detoxification processes, including cytochrome P450 (CYP)-mediated metabolism. Glutamine synthetase (GS), which is expressed in the cytosol of hepatocytes in zone 3, detoxifies residual ammonia by synthesizing glutamine [81], ensures systemic nitrogen balance [82], and is regulated by Wnt/β-catenin signaling [83]. Glycolysis predominates in this zone, and it efficiently generates ATP under hypoxic conditions. A previous study showed that the expression of the GLS2 gene is low, whereas that of the GLUL gene, which encodes the GS protein, is high in MASH [84]. These metabolic abnormalities and hepatocyte injury increase the glutamate/glutamine ratio in the peripheral blood of humans and mice with MASH [85]. Chronic hypoxia-induced HIF-2α activation increases fibrosis and accelerates the progression of MASH via GLS1-induced glutaminolysis in HSCs [86]. A recent study showed that GLS1 activity in HSCs is high in MASH associated with the progression of fibrosis [85], and GLS1 inhibition reduces steatosis in MASH [87]. Finally, during the progression of MASH, structural alterations in the hepatic lobule further dysregulate zone-specific functions [88]. Thus, data regarding the differences in GLS and GS expression, as well as the zone-specific enhancement of gluconeogenesis and glycolysis, are not consistent, and these abnormalities require further investigation.
Glycine, serine, threonine, and methionine; one-carbon metabolism
Glycine, serine, and threonine play significant roles in one-carbon metabolism, which primarily comprises the folate and methionine cycles (Fig. 5). The liver is the leading site of one-carbon metabolism, which is connected to hepatic lipid and phospholipid homeostasis and has been implicated in the pathophysiology of MASLD [89]. One-carbon metabolism plays roles in several vital cellular processes: i) DNA synthesis and repair, by providing purines and thymidylate for nucleotide production; ii) DNA and histone methylation via SAMe [90,91]; iii) antioxidant defense, by recycling homocysteine as cysteine for glutathione synthesis; and iv) energy metabolism, by providing the reducing equivalent NADPH and ATP [92]. This pathway is crucial for cell growth, differentiation, the stress response, and lipogenesis using NADPH [93]. Glutathione is a tripeptide of glycine, glutamate, and cysteine, and is synthesized by GCLC, which is transcriptionally regulated by nuclear factor erythroid 2-related factor 2 (Nrf2) [94]. Glutathione neutralizes the ROS produced in mitochondria, contributing to protection of mitochondrial function.
In addition, glycine, serine, and threonine metabolism are closely linked to glucose metabolism. Serine and glycine are not essential, because they can be synthesized from 3-phosphoglycerate, an intermediate in glycolysis, by phosphoglycerate dehydrogenase [95]. In mammals, threonine is essential and is primarily catabolized to α-ketobutyrate, which is further converted to succinyl-CoA, which can be used for gluconeogenesis, because mammals lack threonine aldolase, which directly converts threonine to glycine. In contrast, certain gut bacteria can metabolize threonine to glycine, acetate, propionate, or butyrate through the expression of threonine aldolase or dehydratase, and these products influence host metabolism [96]. In MASLD/MASH, the concentrations of glycine and serine are often low [73,97], which impairs these critical metabolic pathways and contributes to the increases in oxidative stress and inflammation [98]. Recent studies have shown that a high-glycine, serine, and threonine diet slows the pathogenesis of MASLD [73]. In particular, glycine administration ameliorates oxidative stress via glutathione synthesis or an effect on innate immunity [98,99].
The methionine cycle is also dysregulated in MASLD/MASH. Methionine is converted to S-adenosylmethionine (SAMe), a major methyl donor that is crucial for the methylation of DNA, histones, other proteins, and phospholipids, thereby regulating gene expression and cellular function [100]. SAMe is transported into mitochondria by SLC25A26, and mitochondrial SAM is involved in the stability of electron transport chain complex I [101]. In human obesity, the serum concentration of SAMe is high and correlates with abdominal adiposity, fat mass, and energy intake, suggesting that it is synthesized in the liver in larger amounts [102,103]. MAT1 is principally expressed in the liver, and the absence of MAT1A induces hepatic steatosis and makes the liver more susceptible to injury [104]. HNF4α regulates the expression of MAT1 and enzymes regulating other sulfur-containing amino acids, such as cysteine and taurine, in liver cells [105].
In models of MASLD/MASH, methionine metabolism is often impaired, leading to the accumulation of homocysteine, a toxic intermediate [106]. High homocysteine concentrations are associated with increases in oxidative stress, inflammation, and fibrosis, which contribute to the progression of MASLD to MASH [107]. In addition, studies using the methionine- and choline-deficient (MCD) or methionine-restricted diet in mice have shown that these diets suppress the release of VLDL through mechanisms involving SAMe–phosphatidylcholine metabolism, further exacerbating liver injury and steatosis [14,108].
Branched-chain amino acids
The expression pattern of BCAT, the rate-limiting enzyme in BCAA metabolism, varies according to the tissue [109]. BCAAs, including leucine, isoleucine, and valine, bypass first-pass liver metabolism, owing to the low BCAT2 activity of hepatocytes [109,110]. Instead, they are transaminated in extrahepatic tissues, such as skeletal muscle and adipose tissue, and the resulting branched-chain keto acids (BCKAs) are recirculated to the liver for oxidation (Fig. 6), where BCKA dehydrogenase (BCKDH) activity is high [111,112]. BCKDH regulates BCAA metabolism in the liver and its activity is influenced by bioactive molecules such as insulin, lipids, sugars, and TCA cycle metabolites. Obese, insulin resistant, and hyperinsulinemic animal models are characterized by low hepatic BCKDH activity and high BCKDK activity [113]. Furthermore, recent studies have shown that fructose metabolism in the liver may also regulate systemic amino acid metabolism. Specifically, fructose-mediated activation of ChREBP increases BCKDK activity and reduces protein phosphatase, Mg2+/Mn2+-dependent 1K (PPM1K) activity [114]. The ketogenic amino acid leucine has been recently revealed to complement the main pathway of DNL from pyruvate-derived acetyl-CoA through its conversion to acetyl-CoA via acetoacetate [115].
The concentrations of BCAAs are often high in the peripheral blood (Fig. 6) of patients with MASLD, and high BCAA concentrations are associated with heterogeneity of the size of lipid droplets in hepatocytes [116] and contribute to the development of diabetes and hepatic steatosis [117,118]. Furthermore, several studies have shown that excessive dietary BCAA administration induces MASLD by stimulating lipogenesis, whereas reducing circulating BCAA concentrations, except those of leucine, increases the healthspan of mice [119,120]. Thiazolidinedione, an insulin-sensitizing agent, improves the liver function and ameliorates the steatosis, in association with lower plasma BCAA concentrations, in patients with MASLD and type 2 diabetes [121,122]. A recent study showed that high plasma BCAA concentrations present during the early stages of MASLD rapidly decrease as the disease progresses to MASH and cirrhosis, as in the case of other etiologies [77]. This indicates that even if plasma BCAA concentrations decrease, MASH pathology is not ameliorated, suggesting that the circulating BCAA concentration is more likely to be a consequence of MASH pathology than a cause and that increases in BCAA concentrations alone are insufficient to induce MASH.
Amino acids and the gut microbiota in MASLD
Recently, a study of a healthy human cohort using machine-learning algorithms demonstrated that diet and the microbiota are the best predictors of serum metabolite concentrations [123]. Although BCAAs cannot be synthesized in the human body, Prevotella copri and Bacteroides vulgatus can synthesize BCAAs in the gut, which contribute to IR (Fig. 6) [124]. Consistent with this, individuals with IR exhibit high serum BCAA concentrations, which are associated with a larger microbial capacity for BCAA production [124]. In addition, obesity and IR alter the gut microbiome and serum amino acid profile, and smaller numbers of Bacteroides thetaiotaomicron are associated with high circulating glutamate, BCAA, and AAA concentrations. Furthermore, a restoration of the B. thetaiotaomicron population through weight loss or direct administration reduces the plasma glutamate, BCAA, and AAA concentrations, thereby ameliorating obesity-related metabolic dysfunction [125]. In these previous studies, the nature of the relationship between the changes in the gut microbiota and circulating amino acid concentrations was unclear, although carbohydrate metabolism may directly contribute to IR [126], and a substantial body of evidence suggests that high plasma BCAA concentrations exacerbate IR [117,120,127]. However, in a recent study, the modification of gut microbial amino acid metabolism through the CRISPR-mediated gene editing of Clostridium sporogenes, Bacteroides ovatus, and Clostridium senegalense was shown to alter the circulating concentrations of BCAAs, tryptophan, arginine, and histidine in the host. Specifically, knocking out the BCAA transaminase gene BCAT in C. sporogenes increased the serum BCAA concentrations, but improved the glucose tolerance of the host [128]. These findings demonstrate that microbial BCAA metabolism is the cause of changes in the circulating amino acid concentrations in the host. However, the finding that high BCAA concentrations are associated with superior glucose tolerance differs from those of previous studies. C. sporogenes converts tryptophan to indolepropionic acid (IPA), which strengthens the gut barrier, whereas a lack of IPA increases gut permeability and activation of the immune system. Peptostreptococcus anaerobius and Clostridium cadaveris also produce IPA, implying a broader role of the microbiota [129]. Another study showed that Akkermansia muciniphila promotes liver regeneration in MASLD by increasing the activity of the TCA cycle. This increases glutamine metabolism, supplying α-ketoglutarate to the TCA cycle and supporting energy production [130]. With respect to the effects of amino acid intake, a high-protein diet has been reported to promote lipogenesis and be a risk factor for MASLD [71], whereas a glycine-based treatment increases fatty acid oxidation, supports glutathione synthesis, and improves the composition of the gut microbiota by reducing the Clostridium sensu stricto population and increasing that of Alistipes, thereby restoring diversity [131].
To summarize the results of these previous studies, the metabolic abnormalities that characterize MASLD/MASH are both consequences and causes of disease progression. They affect systemic metabolism, including in the liver, via mediators such as metabolites, humoral factors, the local microenvironment, and the gut microbiota. During the early stages of the disease, the peripheral blood profile likely reflects metabolic compensation occurring within the liver. In contrast, hepatic decompensation resulting from hepatocyte loss causes systemic abnormalities in metabolism during the more advanced stages, with the breakdown of hepatic homeostasis ultimately becoming detectable using peripheral blood indices. Furthermore, the resulting vicious cycle is thought to promote the progression of the disease to MASH.
MECHANISMS OF THE SENSING AND REGULATION OF ENERGY METABOLISM BY AMINO ACIDS
Amino acids, as the third metabolic driver, are important in the pathology of MASLD/MASH. However, many aspects of their role remain unclear, compared with glucose and fatty acid metabolism. Various factors directly or indirectly modulate amino acid metabolism in the liver, and the homeostasis is precisely controlled through the exchange of metabolites with skeletal muscle and adipose tissue. In this section, we describe the key regulators of amino acid metabolism, which are summarized in Table 1.
Mechanistic target of rapamycin, general control nonderepressible 2, and ATF4
Mechanistic target of rapamycin (mTOR) and general control nonderepressible 2 (GCN2) are critical amino acid sensors that regulate energy metabolism in the liver (Fig. 7). mTORC1, which is activated by amino acids such as leucine, via sestrin 2 [132]; arginine, via CASTOR1 [133,134]; and methionine, via SAMTOR [135]; promotes anabolic processes, such as fatty acid and triglyceride synthesis, through transcription factors such as SREBP-1 [136,137], thereby linking nutrient abundance with energy storage. mTORC1 also influences peroxisome proliferator-activated receptor alpha (PPARα) expression [138], which regulates fatty acid oxidation, thereby helping the liver maintain energy balance during periods of nutrient abundance. SAMTOR, which senses intracellular SAMe, activates mTORC1 via GATOR1 [135], providing a mechanism whereby methionine and homocysteine have effects through mTORC1. A recent study has shown that in hepatic lobules, both mTORC1 activity and that of its suppressor sestrin, exhibit zonal expression, with sestrin expression decreasing from zone 1 to zone 3 and mTORC1 activity increasing (Fig. 3) [139]. A previous study demonstrated high mTORC1 activity in a mouse model of MASLD [140], but the activity of liver mTORC1 in human MASLD/MASH has yet to be determined. In contrast, GCN2 responds to amino acid starvation by sensing uncharged tRNAs, and transcription factors such as ATF4 are activated [141,142], modifying the expression of multiple amino acid transporters, such as SLC3A2, SLC7A5 (LAT1), and SLC7A11 (xCT) [143]. Leucine deficiency activates ATF4, which causes an increase in FGF21 production, specifically in zone 1 [139], and the inhibition of Mat1a reduces hepatic SAMe level, thereby increasing FGF21 secretion via NRF2 (Fig. 7) [144].
FGF21, a hormone secreted by the liver, balances the metabolism of sugars, lipids, and proteins, enhances insulin sensitivity, protects hepatocytes against oxidative and ER stress, and suppresses inflammation. In this phase 2b trial, treatment with the FGF21 analogue pegozafermin led to an amelioration of the fibrosis [145]. ATF4 is a key transcription factor in the ER stress response and cooperates with Nrf2, an oxidative stress sensor, to protect hepatocytes against lipotoxicity in MASLD/MASH [146,147]. Recent studies have demonstrated associations between these transcription factors and amino acid metabolism [148]. Nrf2 promotes purine nucleotide synthesis and glutamine metabolism in proliferating cells and increases the expression of metabolic genes via PI3K/AKT signaling [149]. Nrf2 expression in the liver increases in MASLD/MASH in response to oxidative stress [150], and lipotoxicity-induced cellular stress is important in disease progression. However, contradictory findings have been made regarding the expression of ATF4 in the livers of patients with MASLD/MASH [151]. The keap1–Nrf2 system prevents the onset of diabetes mellitus [152] and slows the progression of MASH by restricting methionine [14,153]. The gene encoding the glutamate/cystine transporter xCT is a target of Nrf2/ATF4 and is known to be upregulated in specific cancers [154]. It has also been associated with HCC [155], but another recent study has shown that ATF4 suppresses hepatocarcinogenesis via xCT [156]. Furthermore, the expression of xCT in hepatocytes has not been confirmed [157]. Taken together, mTOR increases the activity of anabolic pathways under nutrient-rich conditions, and GCN2/ATF4 activates catabolic processes during nutrient deprivation, thereby maintaining energy homeostasis in the liver. mTOR is considered to play an essential role in the pathology of MASLD [158], but knowledge regarding GCN2 is still limited. Thus, further research is required to elucidate the mechanisms in more detail.
AMP-activated protein kinase, sirtuin 1, and peroxisome proliferator-activated receptors
AMPK, SIRT1, and PPARs are key interacting regulators of cellular metabolism. AMPK and SIRT1 expression is regulated by the AMP/ATP ratio and the NAD+/NADH ratio, respectively (Fig. 7). AMPK and mTOR are mutually inhibitory [159,160], and recent studies have revealed that amino acids regulate AMPK activity, as well as that of mTOR [161]. AMPK increases glucose uptake by promoting the translocation of GLUT4 to the cell membrane in muscle cells [162] and inhibits hepatic glucose production by reducing the activities of gluconeogenic enzymes, such as PEPCK and G6Pase (Fig. 2) [163]. AMPK inhibits ACC and HMG-CoA reductase, thereby reducing fatty acid and cholesterol synthesis, while promoting fatty acid oxidation [164,165]. In a recent study, cryo-electron microscopy was used to reveal the detailed structure of the inactive form of AMPK bound to ATP [166]. AMPK activity is often low in MASLD/MASH, which contributes to metabolic dysregulation, including greater lipogenesis and less fatty acid oxidation; and hepatic AMPK activation ameliorates systemic IR, leading to reductions in steatosis and inflammation in obese mice [167,168]. A recent study showed that an AMPK agonist slows the pathogenesis of MASH by reducing the activity of the proapoptotic enzyme caspase-6 [169]. Furthermore, AMPK is activated by metformin, the most prescribed diabetes treatment [170]. However, in clinical trials, the effects of AMPK activation on MASLD remain unclear [171].
SIRT1 regulates lipid homeostasis by increasing Foxo1 and PPARα activity (Fig. 7). The hepatocyte-specific deletion of SIRT1 impairs PPARα signaling and reduces fatty acid oxidation, whereas the overexpression of SIRT1 induces the expression of PPARα targets [172].
PPARs, PPARα, PPARγ, and PPARδ play distinct roles in lipid metabolism, and there have been numerous basic studies of their effects [173,174]. Hepatocyte-specific deletion of Pparα promotes steatosis and inflammation in MASLD mice [175]. Hepatic PPARγ regulates adipogenesis and lipid storage in adipose tissue and improves insulin sensitivity in mice [176]. In MASLD/MASH, low PPARα activity leads to a decrease in fatty acid oxidation and an increase in lipid accumulation [177,178]. PPARα activation increases the expression of fatty acid oxidation-related genes in adipocytes [179]. A phase 2 clinical trial showed that the selective PPARα agonist pemafibrate did not reduce liver fat content, but significantly reduced liver stiffness [180]. Pioglitazone, a thiazolidinedione, ameliorates fibrosis in patients with MASH, regardless of whether or not they have diabetes [181,182]. PPARδ/β-overexpression increases insulin sensitivity, improves the lipid profile, and reduces obesity [183]. In addition, the pan-PPAR agonist lanifibranor achieved a pathological effect in a phase 2b MASH trial [184], thereby improving cardiometabolic health [185]. However, the administration of the selective PPARδ agonist seladelpar was discontinued during a phase II clinical trial. Furthermore, there are differences between the human and animal findings regarding PPARs, emphasizing the need for further research. PPARs, and especially PPARα, regulate amino acid metabolism in the liver. PPARα influences the expression of several proteins that are involved in the transamination and deamination of amino acids, such as AST, glutaminase, and glutamine synthetase [186]. Finally, the PPARα agonist WY14,643 increases the plasma concentrations of 12 of the 22 amino acids, including glucogenic and some ketogenic amino acids, but not arginine or BCAAs [187].
CLINICAL TRIALS OF DRUGS FOR THE TREATMENT OF MASLD/MASH THAT TARGET GLUCOSE AND LIPID METABOLISM
A global race is underway to develop therapies for MASLD/MASH, with numerous clinical trials of drugs targeting various mechanisms, including anti-inflammatory, anti-fibrotic, hormonal, bile acid, and metabolic pathways, being conducted. Of these, therapies targeting glucose and lipid metabolism are generating significant interest, owing to their potential impacts (Table 2) [43,145,171,180,184,188-197]. There are currently no therapies that directly target amino acid metabolism. Glucose, fatty acid, and amino acid metabolism are closely interconnected, and drugs targeting glucose metabolism and lipogenesis would also be expected to affect amino acid metabolism. Furthermore, plasma FAA concentrations are likely to be biomarkers of glucagon receptor agonist activity [188,189,198], because glucagon increases the intracellular transport of amino acids. Resmetirom, which recently yielded encouraging results in a phase 3 clinical trial [190], likely affects hepatic amino acid metabolism by improving mitochondrial function and increasing fatty acid oxidation [199]. However, current therapies for advanced MASH remain inadequate, and longitudinal studies will be crucial to assess the long-term effects of metabolic interventions on disease progression and the associated complications.
SUMMARY
Recent advancements in analytical techniques have led to the identification of over 4,000 serum metabolites [200]. Thus, a better understanding of metabolism at each stage of disease progression, achieved through multi-omics approaches, is crucial for the development of personalized treatment strategies for MASLD/MASH. Future research should focus on both inter-organ communication and metabolic interactions within the liver, not only in hepatocytes, but also in immune cells, stellate cells, and vascular endothelial cells. Furthermore, the gut–liver axis should be studied with respect to the effects of gut microbiota-derived metabolites on hepatic metabolism. In addition, investigations of the roles of disruptions of the circadian rhythm in metabolic dysregulation may provide new insight into disease progression and therapeutic targets. With recent advances in genetic analysis, single-cell spatial transcriptomics now offers new oppor tunit ies to improve understanding of the metabolic regulatory mechanisms in MASLD/MASH at an unprecedented level of resolution. Future therapeutic approaches should prioritize combination strategies targeting multiple metabolic pathways and metabolic reprogramming to restore homeostasis. Addressing the gaps in research in this area should facilitate precision medicine and improve the therapeutic outcomes of patients with MASLD/MASH.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
