Metabolic reprogramming of immune cells in MASH

Abstract

Metabolic dysfunction–associated steatotic liver disease, previously known as NAFLD, affects ~30% of adults worldwide. Approximately 1 out of 6 patients with metabolic dysfunction–associated steatotic liver disease progress to metabolic dysfunction–associated steatohepatitis (MASH) and potentially evolve toward cirrhosis and HCC. This condition imposes a significant burden on global health. Although MASH is primarily a metabolic disorder, immune cells play a crucial role in its development. The activation of immune cells involves alterations in transcription and protein synthesis, which require metabolic adaptation of cellular substrate metabolism, including glucose and lipid metabolism. Moreover, the lipid-rich environment of the liver can affect the metabolism of immune cells. In this review, we delineate the metabolic profiles of immune cells in MASH, focusing on macrophages and T cells, and discuss the impact of targeting cellular metabolism on these cells along with clinical trial outcomes. Macrophages in MASH are characterized by increased glycolysis, compromised oxidative phosphorylation, and augmented lipid uptake, synthesis, and fatty acid oxidation, all of which contribute to their inflammatory phenotype. Lymphocytes, such as T cells and NK cells, exhibit impaired metabolic functions that hamper their immune surveillance capabilities and facilitate the development of HCC. However, research on the metabolic characteristics of other types of immune cells remains limited. With the development of single-cell genomics and animal models of MASH, we anticipate that further studies in this compelling field may shed light on the pathogenesis of MASH and pave the way for novel therapeutic strategies.

INTRODUCTION

Metabolic dysfunction–associated steatotic liver disease (MASLD), previously known as NAFLD, is characterized by fat accumulation exceeding 5% of liver weight. Excess fat in the liver leads to inflammation, resulting in the transition from MASLD to metabolic dysfunction–associated steatohepatitis (MASH). Persistent MASH can lead to fibrosis and a cascade of severe complications, including liver failure and HCC.^1,2 With a rising incidence rate, MASLD affects ~30% of adults worldwide and ~1 out of 6 patients may progress to MASH.³ By 2030, it is estimated that over 300 million individuals in China, over 100 million in the United States, and an estimated 15–20 million in major European countries will suffer from MASLD,⁴ imposing a significant economic burden globally. In recent years, significant progress has been made in the treatment of MASH, with numerous drugs advancing into clinical trials. Resmetirom, a thyroid hormone receptor-β agonist, has been approved for marketing. However, its high cost and limited patient response rate restrict its widespread use. Therefore, there is still a considerable clinical need for drug development targeting MASH.⁵

In MASLD, immune cells in the liver undergo alterations and participate in the initiation and progression of the disease. The traditional “two-hit” hypothesis suggests that lipid deposition and insulin resistance in the liver are involved in the first hit. Subsequent oxidative stress and inflammatory responses mediated by immune cells lead to MASH development. With further investigation, it is evident that the mechanisms underlying MASLD/MASH are more complex, leading to the proposal of the “multiple hits” hypothesis, which suggests that the interplay among various factors is implicated in the development of MASH.¹ Metabolic perturbations in the liver environment, particularly lipid accumulation, have been shown to affect immune cell function.⁶ Metabolic disorders in MASLD/MASH also lead to metabolic reprogramming of immune cells, influencing their phenotype to implicate the development of MASLD/MASH. In addition, some drugs that target liver metabolism can affect immune cell metabolism and contribute to their therapeutic effects. In this review, we discussed intrinsic metabolic changes in immune cells, especially macrophages and T cells, in MASLD/MASH. We also summarized the clinical trial results of current drug candidates targeting cellular metabolism and their effects on immune cells, hoping to provide new insights for the development of MASH therapy.

AN OVERVIEW OF THE LIVER IMMUNE LANDSCAPE

The liver plays an essential role in human metabolism and detoxification, processing more than 2000 L of blood daily, with one-fourth originating from the hepatic artery and the remaining three-fourths from the portal vein. The hepatic lobule is the basic structural and functional unit of the liver. Branches of the hepatic artery and portal vein enter the hepatic lobules at the portal triad and form the hepatic sinusoids. The substances in the blood are metabolized within these sinusoids before draining into the central vein. The sinusoidal endothelium, characterized by its fenestrated structure, supports slow and low-pressure blood flow, facilitating the permeation of blood into the hepatic lobules for extensive metabolic interactions with hepatocytes.^7,8 Portal vein blood is rich in microbial products and food antigens from the gut and can contribute to persistent tissue inflammation. However, the immune microenvironment within the liver is tolerogenic, preventing the excessive activation of immune cells. LSECs serve as the first line of defense and can engulf and process antigen molecules in the sinusoids, presenting them to resident immune cells in the liver. However, the absence of costimulatory molecules such as CD40, CD80, and CD86 leads to CD4⁺ T-cell anergy and promotes Treg differentiation.^9,10 In addition, interactions between endothelial and CD8⁺ T cells can induce a tolerogenic phenotype.^11,12 KCs, which are resident macrophages in the liver, can also engulf cellular debris and microbe-derived molecules and express PD-L1 to suppress T-cell responses.¹³ KCs are also the primary source of IL-10, a cytokine critical for maintaining hepatic homeostasis.¹⁴ Moreover, KCs are sentinel cells and first responders during liver injury¹⁵ As reported, the liver also contains at least 2 additional macrophage populations during homeostasis, including bile duct–associated macrophages and subcapsular macrophages.^16,17 Dendritic cells (DCs), as professional antigen-presenting cells that are localized in the portal or pericentral regions and are divided into 4 subsets: cDC1, cDC2, monocyte-derived DCs, and plasmacytoid DCs.¹⁸ Immature DCs and plasmacytoid DCs play crucial roles in maintaining immune tolerance in the liver. However, under chronic inflammation, immature DCs will differentiate to exacerbate the inflammatory responses.^19,20 In addition, HSCs are located in the liver sinusoidal space of Disse and interact with immune cells to prevent their activation.²¹ These nonparenchymal cells collectively contribute to liver homeostasis. Dysregulated activation of innate and adaptive immune cells, which can precipitate a shift from immune tolerance to autoimmunity in the liver, is a shared pathogenic mechanism across various liver diseases, regardless of their etiology.²²

The liver is abundant with lymphocytes and regarded as an immune organ with lymphoid features.²² T lymphocytes are pivotal for the initiation and resolution of hepatic immune responses. Excessive antigenic load can shift immune tolerance to immune stimulation through interactions between antigen-presenting cells and T cells.²³ The fenestrated endothelium and low pressure allow circulating T cells to recognize underlying antigens and differentiate into effector T cells. Effector T cells adhere to the sinusoidal endothelium and migrate along the sinusoidal wall to identify cognate antigens. After that, they extend cellular protrusions through the fenestrae and connect with the hepatocyte membrane to form a synapse-like process. This connection triggers Fas ligand (FasL), perforin, and granzyme-mediated killing as well as the secretion of proinflammatory cytokines to recruit immune cells.^24,25 Upon the completion of their effector functions, these cells leave the sinusoidal space or undergo apoptosis to prevent excessive activation.²⁶ The liver also harbors tissue-resident T cells that are activated and mature in situ, providing a rapid immune response to threats. Activated tissue-resident T cells persist only in areas where their cognate antigens are detected, curbing unnecessary activation to maintain liver homeostasis.^27,28 Similarly, a subset of NK cells in the liver displays a tissue-resident phenotype, exhibiting heightened cytotoxicity compared with their circulating counterparts.^29,30 However, within the liver, they tend to exhibit a tolerogenic phenotype.^31,32 Interestingly, a subset of tissue-resident NK cells demonstrates antigen-specific activity, known as “memory-like NKs,” which selectively expands and increases IFN-γ production upon secondary cytomegalovirus exposure.³³ NKT cells, a type of innate-like T cell, are abundant in murine livers but scarce in humans, whereas mucosal-associated invariant T cells are rare in mice and abundant in human livers.³⁴ They can recognize bacteria-derived vitamin B metabolites through MHC class I–related protein 1 (MR1), protecting the biliary epithelium from invasion by intestinal microbes.^35,36 Naive B cells and antibody-producing plasma cells are also present in the liver. These plasma cells are primarily located in the portal veins, originating from B cells activated in gut-associated lymphoid tissue, and produce IgA in response to commensal bacteria and oral antigens.^37,38

In summary, the liver harbors diverse populations of immune cells. A subset of these immune cells, predominantly of the myeloid lineage, display a tolerogenic phenotype, preventing inflammation in the liver caused by the intake of antigens from the gastrointestinal tract through the portal vein. Conversely, another group of immune cells, mainly lymphocytes, performs the conventional role of immune surveillance, eliminating infections or mutated cells within the liver. Undoubtedly, dysfunction of these cells contributes to the pathology of MASLD and plays distinct roles at different stages of disease progression.

THE ROLE OF IMMUNE CELLS IN MASH

The immune cell repertoire of the liver is reshaped during MASH development, resulting in an unregulated inflammatory response that triggers liver injury and fibrosis.³⁹ Neutrophil accumulation is an early event in MASH, triggering inflammatory cascades and liver damage by releasing reactive oxygen species (ROS), proteases, and neutrophil extracellular traps (NETs).^40,41 NETs are extracellular web-like structures composed of decondensed and extruded neutrophil chromatin, which can activate a variety of immune cells and stimulate macrophages to secrete inflammatory cytokines, playing a significant role in the progression of MASLD toward MASH and fibrosis.^42,43 NETs can also promote Treg differentiation and create a tumorigenic niche in the liver.⁴⁴ Thus, neutrophils serve as one of the initiating factors in the development of MASLD. Depleting neutrophils in the early stages of MASH can limit inflammation and liver damage; however, this protective effect diminishes as MASH advances.⁴⁵

In MASH, the number of KCs decreases due to apoptosis and reduced self-renewal capacity, whereas inflammatory signals lead to increased recruitment of circulating monocytes to the liver.^46–48 Some of these monocytes replenish the KC population, and these monocyte-derived KCs exhibit a higher expression of inflammatory genes than their embryonic counterparts.⁴⁶ Monocytes also differentiate into macrophages and contribute to MASH progression by releasing cytokines. A subset of macrophages induced by a high-fat environment and DAMPs released from steatotic hepatocytes has attracted attention. These macrophages are known by various names, including lipid-associated macrophages (LAMs), NASH-associated macrophages, and scar-associated macrophages, which are characterized by the expression of CD9, triggering receptors expressed on myeloid cells 2 (TREM2), osteopontin (SPP1), and glycoprotein non-metastatic melanoma protein B (GPNMB).^47,49–53 RNA velocity analysis indicates that these subsets originate mainly from monocytes in the circulation.⁴⁹ However, KCs also exhibit a similar phenotype due to reprogramming of liver X receptor (LXR) function by AP-1 and EGR, leading to a loss of their identity and an increase in the expression of Trem2 and Cd9.⁴⁸

The role of TREM2⁺CD9⁺ macrophages in MASH-associated fibrosis remains controversial. Ramachandran et al⁴⁹ identified the role of TREM2⁺CD9⁺ macrophages in MASH-associated fibrosis, revealing that these macrophages express a high number of inflammatory genes and promote the proliferation and collagen production of primary human HSCs in vitro, correlating with the progression of liver fibrosis. Conversely, LAMs promote the resolution of MASH-fibrosis in a TREM2-dependent manner.⁵⁴ As a phagocytic receptor, TREM2 is essential for the efferocytosis of lipid-laden apoptotic hepatocytes to maintain liver immune homeostasis.⁵² TREM2-deficient bone marrow transplantation in mice fed a methionine-choline-deficient diet resulted in increased steatohepatitis and fibrosis.⁵⁵ TREM2-deficient macrophages release exosomes, leading to mitochondrial dysfunction in hepatocytes, thereby promoting MASH progression.⁵⁶ These discrepancies might be attributed to the heterogeneity within TREM2⁺ macrophages. Trem2⁺ macrophages can be divided into 2 subsets in mice: transitional Cx3cr1⁺Ccr2⁺ LAMs and Trem2⁺ Cd63⁺ Cd9⁺ Gpmnb⁺ classical LAMs.⁵⁷ Hepatic crown-like structures (hCLS), characterized by the aggregation of macrophages surrounding dying hepatocytes and lipid droplets, are indicative of the progression of simple steatosis toward steatohepatitis and fibrosis.^58,59 Cx3cr1⁺Ccr2⁺LAMs play a significant role in hCLS formation. Deletion of Ccr2 reduces the recruitment of Cx3cr1⁺Ccr2⁺ LAMs, leading to impaired hCLS formation and exacerbating fibrosis induced by a high-fat, high-sucrose diet.⁵⁷ In addition, TREM2-expressing monocyte-derived macrophages accumulate in both primary and metastatic tumors,^60,61 suggesting a potential role for TREM2⁺ macrophages in the progression of MASLD to HCC.

CDCs, the principal antigen-presenting subset of DCs, serve as a bridge between innate and adaptive immunity. CDCs accumulate within the livers of patients with MASH and exacerbate MASH by activating T cells in draining lymph nodes.⁶² Upon stimulation, conventional CD4⁺ T cells differentiate into T-helper 1 (Th1), Th2, and Th17 subsets, each defined by their production of specific cytokines, and these cytokines contribute to the advancement of MASH.⁶³ However, it is difficult to specifically attribute disease progression to alterations within Th cell populations, as these cytokines are also produced by other cell types. Further studies will require more precise methodologies to elucidate the role of this cell population in MASH. While the roles of Th1 and Th2 in MASH still require further research, the importance of Th17 in MASLD has been extensively investigated. Th17 cells are characterized by the expression of transcription factor RORγt and cytokine IL-17.⁶⁴ Increased Th17 and IL-17 expression in the liver facilitates the progression from simple steatosis to inflammation,⁶⁵ and an increased Th17/Treg ratio in the peripheral blood is indicative of the progression of MASLD to MASH.⁶⁶ However, in the high-fat-high-cholesterol–fed mouse model, an increase in Treg frequency in the MASH liver and the adoptive transfer of Tregs can lead to hepatic steatosis and elevated ALT.⁶⁷ Accumulation of Tregs in MASH also fosters an immunosuppressive microenvironment for oncogenesis.⁴⁴ Therefore, Tregs may play a detrimental role throughout MASH development.

CD8⁺ T cells can produce IFN-γ, TNF, and cytotoxic molecules, such as perforin, to exert cytotoxic effects in an antigen-specific manner. For a long time, their role in MASLD has not received much attention. Dudek and colleagues identified an IL15-induced auto-aggressive resident CD8⁺ T-cell subset characterized by the downregulation of FoxO1 and upregulation of CXCR6. This specific group can induce hepatocyte apoptosis in an antigen-nonspecific manner through Fas/FasL interactions, leading to pathological changes in MASH.⁶⁸ This T-cell subset also contributes to resistance to immunotherapy in MASH-HCC, with high expression of the exhaustion marker PD-1. Notably, anti-PD-1 treatment leads to the expansion of CXCR6⁺PD-1⁺CD8⁺T cells without tumor regression.⁶⁹ However, the CD44⁺CD62L⁻CD69⁺CD8⁺ T-cell subset, which also expresses CXCR6, can induce apoptosis in HSCs and promote the resolution of fibrosis.⁷⁰ Therefore, the impact of this auto-aggressive CD8⁺ T cell on the development of MASH is multifaceted and may require larger single-cell genomic data sets and more refined analyses to elucidate its role.

B cells, another important cell type mediating adaptive immunity, accumulate in the livers of patients with MASH with high levels of lobular inflammation and fibrosis.⁷¹ Alleviation of MASH is observed in B cell–deficient mice, suggesting its important role in disease progression.⁷² Innate-like T cells, including NKT, mucosal-associated invariant T, and γδT cells, mainly exert protective functions by mitigating hepatic steatosis and inflammation.³⁴ The roles of B cells and innate-like T cells in MASLD progression are not yet fully understood and require further research (Figure 1).

FIGURE 1.

The role of immune cells in MASH. In MASH, immune cells from the circulation are recruited into the liver, with neutrophils being the first to respond. They release ROS, NETs, and proinflammatory cytokines. Myeloid cells, including macrophages and DCs, play similar roles. Cytokines released by macrophages can activate HSCs and promote fibrosis, and these macrophages can also surround apoptotic hepatocytes to form hCLSs. DCs, bridging innate and adaptive immunity, are involved in inflammation and activate CD8+ T cells through antigen presentation.CD8+ T cells play a role in immune surveillance to prevent the occurrence of HCC. However, a special subset of auto-aggressive CD8+ T cells can promote the development of MASH and HCC, and mediate resistance to immunotherapy in MASH-HCC. NK cells, which also serve an immune surveillance function, have multifaceted roles in MASH. They contribute to inflammation while also curbing HSC activity and macrophage polarization to prevent fibrosis. The role of Th cells depends on the cytokines they secrete. IL-17, released by Th17, can promote inflammation and fibrosis in MASH. However, the roles of Th1 and Th2 in MASH require further investigation. Abbreviations: DC, dendritic cell; hCLS, hepatic crown-like structure; MASH, metabolic dysfunction–associated steatohepatitis; NET, neutrophil extracellular trap; ROS, reactive oxygen species; Th, T helper. Created in BioRender (BioRender.com).

IMMUNOMETABOLISM IN MASH

Immune cells exhibit unique metabolic profiles, as they adapt to different tissue environments and engage in inflammation or disease. During the progression from MASLD to MASH, the phenotype and function of liver immune cells undergo significant remodeling. Diverse types and subsets of immune cells have specific nutritional demands for their metabolic programs, which lead to distinct metabolic alterations. Therefore, different immune cells encounter unique challenges in the context of MASLD.⁶ Immune cells originate primarily from bone marrow hematopoietic stem cells and are categorized into myeloid and lymphoid lineages. Myeloid cells, derived from myeloid progenitor cells, are predominantly involved in innate immunity, acting as the body’s first line of defense against pathogens and the main inflammatory cells, and play a crucial role in the development of MASH. Lymphoid cells, which arise from lymphoid progenitor cells, constitute the backbone of adaptive immunity, and certain subsets, such as NK cells, also contribute to innate immunity. These lymphocytes can specifically target and eliminate mutated hepatocytes, thereby preventing the development of HCC. Given the distinct origins and roles of myeloid and lymphoid cells, this review separately delineates their metabolic features in MASH, discusses the different effects of metabolism on their effector functions, and explores potential metabolic targets for modulating the MASH immune milieu.

Myeloid cell metabolism in MASH

Glucose metabolism

Immune cells sense environmental and metabolic cues and adjust their metabolic programs in response to them. The liver of patients with MASLD is rich in lipids and cholesterol crystals. Increased intestinal permeability also leads to an influx of microbial metabolites into the liver through the portal vein.^73,74 These products inevitably affect the phenotype of immune cells. Palmitic acid and cholesterol crystals can induce NET production, subsequently polarizing macrophages to a proinflammatory phenotype. Proinflammatory macrophages exhibit elevated glycolysis and reduced oxidative phosphorylation (OXPHOS).^75,76 This metabolic characteristic, known as the Warburg effect, is attributed to increased glucose uptake and expression of glycolytic enzymes in macrophages within the MASH, while the tricarboxylic acid (TCA) cycle and OXPHOS are impaired.

Palmitic acid enhances the expression of glucose transporter 1 (GLUT1) in macrophages.⁷⁷ Consistent with this observation, macrophages surrounding lipid droplets in the hCLS exhibit elevated GLUT1 expression.⁷⁸ Lipopolysaccharide (LPS), a critical bacterial product, also upregulates GLUT1 in proinflammatory macrophages.⁷⁸ The expression of TM4SF5 in hepatocytes within the inflammatory milieu of MASH is induced by CCL2, CCL5, and TGF-β. The interaction between macrophages and TM4SF5⁺ hepatocytes facilitates the translocation of GLUT1 to the macrophage cell surface.⁷⁹ GLUT1 is the principal rate-limiting transporter of glucose. It augments glucose uptake, leading to enhanced glycolysis, pentose phosphate pathway (PPP) activation, and impaired OXPHOS in macrophages.⁷⁸ This metabolic shift may be attributed to a decrease in CoA and B vitamin pantothenate (necessary for CoA generation), resulting in diminished flux of the TCA cycle.⁷⁸ GLUT1 amplifies the NF-κB–mediated proinflammatory gene expression profile, thereby inducing a proinflammatory phenotype in macrophages.⁸⁰ GLUT1-mediated secretion of proinflammatory cytokines is also partly ascribed to the generation of ROS in macrophages.^78,80 The elevated flux of PPP induced by GLUT1 can generate NADPH, which provides substrates for NADPH oxidase activation during the M1 polarization of macrophages. Moreover, the expression of GLUT1 can suppress macrophage autophagy, a process that can inhibit M1 polarization of macrophages to alleviate hepatic inflammation and fibrosis.^81,82

Pyruvate kinase M2 (PKM2) is an isoform of pyruvate kinase, existing in tetrameric and dimeric forms. The tetrameric form of PKM2 primarily exhibits enzymatic activity, whereas the dimeric form translocates to the nucleus to modulate gene transcription.⁸³ In patients and murine models with MASH, instead of simple steatosis, cholesterol crystallization associated with the lipid droplet membrane is present within hepatocytes.^59,84–86 Cholesterol crystals can upregulate GLUT1 and PKM2, thereby promoting glycolysis in macrophages.⁸⁷ Due to increased oxidative stress and cholesterol accumulation, oxidized low-density lipoprotein accumulates in the serum of patients and promotes the progression of MASH.⁸⁸ Oxidized low-density lipoprotein stimulates PKM2 expression. In the early stages, oxidized low-density lipoprotein enhances both the enzymatic activity and nuclear translocation of PKM2. In later stages, it predominantly upregulates the tetrameric form, thereby facilitating aerobic glycolysis in macrophages.⁸⁹ Downregulation of miR-122-5p in KCs is also associated with elevated PKM2 expression.⁹⁰ PKM2 translocates to the nucleus and binds to hypoxia-inducible factor-1α (HIF-1α) to promote the production of proinflammatory cytokines,^88,91 which can facilitate hepatic steatosis through paracrine effects.⁹¹ Given that PKM2 is physiologically abundant only during embryogenesis and is notably absent in healthy liver tissue,⁹² with primary expression in macrophages rather than hepatocytes within the liver,^93,94 PKM2 may serve as a promising target for the surveillance and treatment of MASLD. Blockade of PKM2 can ameliorate steatosis, inflammation, and fibrosis in high-fat diet (HFD)-induced MASH mouse models.^95–97 In addition, PKM2 inhibition may also represent a promising therapeutic strategy for MASH-HCC. Kaplan-Meier analysis indicates that elevated PKM2 expression is significantly correlated with poorer overall survival and disease-specific survival in patients with HCC. Inhibition of PKM2 can suppress aerobic glycolysis of HCC cells, inhibiting tumor growth and metastasis.^98,99 PKM2, secreted by HCC cells through ectosomes, also promotes the differentiation of macrophages into pro-tumor phenotypes.¹⁰⁰ However, considering that PKM2 is primarily expressed in nontumor cells within the liver cancer microenvironment, PKM2 inhibition is more likely to exert antitumor effects through immune cells (Figure 2) (Supplemental Figures, http://links.lww.com/HEP/J789).

FIGURE 2.

The role of PKM2 in MASH-HCC. PKM2 exists in 2 forms: tetrameric and dimeric. Tetrameric PKM2 primarily acts as the rate-limiting enzyme in glycolysis, while dimeric PKM2 can translocate to the nucleus as a transcription factor to regulate gene expression. In HCC, the expression level of PKM2 is significantly higher than in normal tissues, and it is significantly associated with poor OS and DSS in patients with HCC without a history of alcohol consumption and viral infection. Immunohistochemical staining indicates that PKM2 is predominantly expressed in the tumor stroma, suggesting that its influence on the tumor microenvironment might be a primary driver of its tumor-promoting effects. PKM2 facilitates the release of ROS from neutrophils and the activation of Th17 cells. Moreover, PKM2 promotes the proliferation of tumor cells. Ectosomes containing PKM2 secreted by tumor cells can induce macrophages to polarize toward an M2 phenotype. In MASH, PKM2 can induce the M1 phenotype of macrophages to exacerbate inflammation. Therefore, PKM2 may play a detrimental role throughout the development of MASH-HCC. Abbreviations: DSS, disease-specific survival; MASH, metabolic dysfunction–associated steatohepatitis; OS, overall survival; PKM2, pyruvate kinase M2; ROS, reactive oxygen species. Created in BioRender (BioRender.com).

Activation of proinflammatory macrophages leads to the suppression of OXPHOS, redirecting the electron transport chain (ETC) from ATP production toward the generation of ROS.¹⁰¹ Within the MASH environment, ROS can damage DNA and deplete NAD⁺ to suppress the expression of glyceraldehyde 3-phosphate dehydrogenase, a rate-limiting enzyme in glycolysis. However, increased expression of nicotinamide phosphoribosyltransferase in macrophages reverses this effect, restoring their glycolytic activity.¹⁰² The suppression of OXPHOS in proinflammatory macrophages is partly attributed to the increased production of nitric oxide (NO) in M1 macrophages, which inhibits the activity of complexes in the ETC through nitrosylation. This inhibition leads to breakpoints in the TCA cycle of proinflammatory macrophages, specifically isocitrate dehydrogenase (IDH) and succinate dehydrogenase (SDH).¹⁰¹

The initial disruption of the TCA cycle occurs at the IDH, leading to citrate accumulation in macrophages. In proinflammatory macrophages, the expression of the citrate transporter Slc25a1 is upregulated.¹⁰³ This transporter promotes citrate translocation from the mitochondria to the cytoplasm. Subsequently, it is metabolized by ATP-citrate lyase (ACLY) into acetyl-CoA and oxaloacetate. Acetyl-CoA is a pivotal source of acetyl groups for histone acetylation.^103,104 Oxaloacetate is a precursor of NADPH production, which is crucial for the synthesis of NO and ROS. Consequently, cytoplasmic citrate accumulation and the ensuing metabolic reactions support the establishment of a proinflammatory macrophage phenotype.⁷⁵ Targeting Slc25a1 and ACLY can attenuate the secretion of proinflammatory cytokines by macrophages and alleviate liver injury in preclinical models and in patients with MASH.^105,106 However, Palmieri and colleagues have suggested that the accumulation of citrate is due to the inhibition of aconitase rather than the activity of IDH. This model is supported by several observations: (1) While citrate-driven oxidative metabolism is reduced in M1 macrophages, isocitrate-driven respiration remains uncompromised. (2) Although nitric oxide synthase 2 is necessary for alterations in α-ketoglutarate (α-KG) homeostasis, it is not required for the reduction of IDH1 mRNA. (3) Aconitase is a recognized target of NO, and the specific inhibition of aconitase aligns with the lipophilic property of NO.¹⁰⁷ This model suggests that metabolic reprogramming is likely an outcome of macrophage polarization rather than a mediator of this process.¹⁰⁸ SDH, the second metabolic breakpoint in activated macrophages, catalyzes the conversion of succinate to fumarate in the TCA cycle. SDH dysfunction results in succinate accumulation.¹⁰¹ Succinate stabilizes HIF-1α, induces the production of inflammatory cytokines, and enhances the transcription of glycolytic genes.^75,109 Secreted succinate acts as a signaling molecule that activates macrophages and HSCs, aggravating liver damage and fibrosis associated with MASH.^110,111 SDH is also known as complex II of ETC. SDH dysfunction promotes a shift in the mitochondrial respiratory chain, from ATP generation to ROS production.^101,112 Dimethyl malonate inhibits SDH activity, reducing the recruitment of monocytes to the liver, restoring the LAM phenotype in monocyte-derived macrophages, diminishing hCLS, and mitigating hepatic steatosis and injury induced by a high-fat, high-cholesterol (HF-HC) diet in mice.¹¹² Itaconate, an essential intermediate metabolite during M1 polarization of macrophages, originates from cis-aconitate in the TCA cycle through immune response gene 1 (IRG1)-mediated decarboxylation, and inhibits SDH to prevent excessive macrophage activation.¹¹³ However, in patients with MASH, upregulation of β-arrestin in circulating monocytes leads to IRG1 ubiquitination and degradation, enhancing SDH-mediated succinate oxidation to produce ROS and promote MASH progression. Furthermore, β-arrestin mRNA levels correlate positively with the severity of liver steatosis and serum ALT and AST levels in patients with MASLD.¹¹⁴ In conclusion, macrophages in the livers of patients with MASH display a highly glycolytic phenotype, which is both a result of M1 polarization and a factor in establishing a proinflammatory profile of macrophages (Figure 3). Glycolysis provides rapidly generated ATP and metabolic intermediates that are conducive to M1 polarization.⁷⁵ These activated macrophages not only directly exacerbate hepatic lipid accumulation and injury through the secretion of proinflammatory cytokines such as IL-1β and TNF-α^91,95,112 but also amplify inflammatory cascades by releasing chemokines like MCP-1, thereby recruiting additional monocytes/macrophages and forming hCLS.¹¹² Impairment of the mitochondrial ETC in steatotic liver-infiltrating macrophages also contributes to MASH progression through oxidative stress.¹¹² While proinflammatory macrophages induced by metabolic reprogramming might theoretically activate HSCs to promote fibrosis, there is currently no direct experimental evidence confirming a definitive link between macrophage glycolysis and their specific regulatory effects on HSC activation. With the advancing resolution of single-cell sequencing technologies in dissecting macrophage heterogeneity, future research is expected to elucidate the interaction networks between macrophages with distinct metabolic phenotypes (particularly glycolysis-dominant subsets) and other cells in the steatotic liver environment, thereby systematically unraveling the molecular mechanisms underlying the metabolic-immune regulatory axis in MASH progression.

FIGURE 3.

Metabolic reprogramming of macrophages in MASH. Macrophages upregulate GLUT1 and PKM2 to promote glucose uptake and glycolysis in MASH, which induces their proinflammatory polarization. Proinflammatory macrophages exhibit increased production of NO, which can disrupt the TCA cycle and impair OXPHOS. The breakpoints of the TCA cycle lead to the accumulation of citrate and succinate. The accumulated citrate can be transported into the cytoplasm to facilitate fatty acid synthesis. Increased fatty acid uptake in macrophages also contributes to fatty acid accumulation. Concurrently, FAO is activated in macrophages. Macrophages also exhibit heightened cholesterol uptake, leading to the formation of cholesterol crystals. This process induces lysosomal swelling and rupture, with a subsequent increase in cholesterol efflux. These metabolic characteristics contribute to the polarization of macrophages toward the M1 phenotype to exacerbate MASH. Abbreviations: FAO, fatty acid oxidation; NO, nitric oxide; OXPHOS, oxidative phosphorylation; TCA cycle, tricarboxylic acid. Created in BioRender(BioRender.com).

However, this is not the sole cause of the proinflammatory phenotype. G-protein–coupled receptor 3 activation can boost glycolysis by facilitating PKM2 nuclear translocation and c-Myc activation, while concurrently inhibiting the expression of inflammatory genes in macrophages.¹¹⁵ Indole, a metabolite of tryptophan metabolism in gut bacteria, enhances the expression of the glycolytic enzyme 6-phosphofructo-2-kinase (PFKB3). However, it suppresses macrophage M1 polarization by inhibiting JNK p46 and NF-κB p65 phosphorylation in a PFKB3-dependent manner, which ameliorates HFD-induced liver steatosis and inflammation in mice.¹¹⁶ Excess fructose intake intensifies the inflammatory response in macrophages but also increases oxidative metabolism and curbs glycolysis, potentially linked to glutaminolysis and the replenishment of TCA cycle intermediates. Fructose-activated macrophages with limited metabolic flexibility are highly sensitive to Hexokinase 2 (HK2) inhibitors. HK2 inhibitors can decrease proinflammatory cytokine secretion and cell death in these macrophages.¹¹⁷ As fructose consumption is associated with the rising incidence of MASLD,^118,119 its influence on macrophage phenotypes requires further research.

Lipid metabolism

The breakpoint of IDH results in citrate accumulation within macrophages, leading to the generation of acetyl-CoA, a pivotal substrate for fatty acid synthesis. An elevated PPP also supplies NADPH for this synthesis. In addition to the accumulation of substrates, proinflammatory macrophages in MASLD undergo transcriptional alterations that promote lipid synthesis. Palmitic acid and LPS can upregulate the expression of fatty acid synthesis enzymes in vitro, including fatty acid synthase (FASN), sterol regulatory element-binding proteins (SREBP), and ACLY within macrophages.¹²⁰ Similarly, in HFD-fed mice, lipid synthesis genes, such as Fasn, Scd1, and Dgat2, are markedly upregulated in KCs.^121,122 Lipid synthesis is essential for the proinflammatory phenotype of macrophages induced by diet. FASN deficiency can alter membrane order and composition, leading to the retention of plasma membrane cholesterol and disruption of Rho GTPase trafficking, which are vital for adhesion, migration, and activation of macrophages.¹²⁰ The SREBP family, comprising SREBP-1a, SREBP-1c, and SREBP-2, acts as a transcription factor that regulates the genes involved in fatty acid and cholesterol synthesis. SREBP cleavage-activating protein (SCAP)-mediated cleavage and activation of SREBP are upregulated in mice fed a high-cholesterol diet, which can activate NF-κB signaling in macrophages, leading to the release of inflammatory cytokines. These cytokines induce lipolysis in adipocytes and stimulate lipid uptake and synthesis in hepatocytes, thereby contributing to the development of lean MASLD.¹²³ SREBP-1a is highly expressed in macrophages and plays a role in environmental sensing. Its activation by LPS promotes lipid synthesis to facilitate cell proliferation and membrane rearrangements, as well as the activation of the NLR family pyrin-containing protein 1 (NLRP1) to initiate inflammation.¹²⁴ However, SREBP-1 also contributes to the resolution of inflammation by promoting the synthesis of anti-inflammatory fatty acids.¹²⁵ Upon feeding with a normal diet, macrophage-specific SREBP-1a knockout mice did not show histological or genetic differences compared to flox mice. However, macrophages in these mice fed an HFD exhibit an inflammatory phenotype, resulting in exacerbated liver injury.¹²⁶ These results underscore the importance of SREBP-1α in macrophages for lipid synthesis and the maintenance of the inflammatory balance within the hepatic microenvironment.

As MASLD progresses to MASH, the influx of lipids exceeds the metabolic capacity of hepatocytes, leading to their accumulation in the liver.¹²⁷ Excess lipids lead to lipotoxicity in hepatocytes and induce proinflammatory and fibrotic responses mediated by macrophages.^1,121 Sphingosine, the most significantly altered active lipid in MASH, has been demonstrated to promote the secretion of inflammatory cytokines by inhibiting HIF-2α in macrophages.¹²⁸ Palmitic acid activates HIF-1α, impairing macrophage autophagy and leading to the proinflammatory phenotype of macrophages through activation of the NF-κB pathway.⁷⁷ Notably, the effect of lipids on macrophages is not confined to the liver. Krenkel and colleagues showed that fatty acids induced a specific subset of macrophages with low S100A8/A9 expression through Toll-like receptor 4 (TLR4) signaling. This subset exhibits decreased chemokine expression for neutrophil recruitment and reduced functions in both metabolism and matrix modulation. The unique phenotype of myeloid cells in MASLD is not only observed in the liver compartment but also imprinted in myeloid bone marrow precursors.¹²⁹

In response to the steatotic environment in the liver, macrophages enhance their capacity for lipid intake, which mitigates lipid deposition in hepatocytes and prevents lipotoxicity-induced liver injury. TREM2, a marker for LAMs, is a lipid-sensing receptor that is involved in lipid clearance. TREM2⁺ macrophages exhibit an enhanced ability to take up lipids and are associated with improved outcomes in MASLD.^14–16 However, the increased intake capacity leads to lipid accumulation within macrophages, polarizing them toward an inflammatory phenotype, which promotes the transition from simple steatosis to MASH.¹³⁰ Macrophage scavenger receptor 1 (MSR1), a scavenger receptor predominantly expressed in KCs, is upregulated in MASLD owing to disturbed lipid metabolism and oxidative stress.¹³¹ Macrophages with elevated MSR1 expression exhibit increased uptake of saturated fatty acids (SFAs), which leads to heightened secretion of proinflammatory cytokines and contributes to the progression of MASLD toward MASH.¹³² With the advancement of MASLD, the expression of MSR1 has increased, indicating its potential as a candidate for disease surveillance and therapeutic intervention.¹³² Members of the fatty acid–binding protein (FABP) family regulate intracellular lipid transportation and metabolism, exhibiting a tightly controlled tissue distribution.¹³³ FABP4, which is predominantly expressed in macrophages and adipose tissue, is upregulated by palmitic acid in a time- and concentration-dependent manner.¹³⁴ FABP4 mediates the absorption of SFA, leading to M1 polarization and the secretion of proinflammatory cytokines by macrophages.¹³⁵ In addition, Fabp4 contributes to palmitic acid–induced macrophage apoptosis through multiple pathways, including the induction of toxic ceramides, ROS, and mitochondrial dysfunction.^134,136

In addition to fatty acids, the uptake of cholesterol by macrophages also increases in MASLD. KCs acquire cholesterol through receptor-mediated endocytosis of circulating lipoproteins, especially LDL. Under normal physiological conditions, nonmodified LDL is internalized through the LDL receptors (LDLR), and then these receptors are downregulated to prevent cellular cholesterol overload.^86,137 However, instead of LDLR, modified LDL, like oxLDL, is taken up by scavenger receptors. Unlike LDLRs, scavenger receptors are not subject to feedback inhibition in response to increased cellular cholesterol levels, and endocytosed oxLDL can inactivate lysosomal enzymes, leading to the accumulation of modified LDL in macrophages and the formation of foam cells.^138,139 The resultant lysosomal cholesterol accumulation in KCs increases hepatic inflammation and promotes MASH development.¹⁴⁰ Genetic ablation of scavenger receptors can reduce lysosomal cholesterol accumulation in macrophages and alleviate inflammation in MASH models.¹⁴¹ Moreover, cholesterol crystals have been demonstrated to activate the NLRP3 inflammasome, an important trigger of inflammation in MASH.¹⁴² In macrophages, phagocytosis of cholesterol crystals leads to lysosomal swelling and the release of cathepsin B, a lysosomal protease that activates the NLRP3 inflammasome. The activated NLRP3 inflammasome further promotes the activation of caspase 1, leading to the maturation of IL-1β and IL-18, which are important mediators of the innate immune response.¹⁴³ Highly oxidized LDLs promote the recruitment of neutrophils by macrophages and trigger NETs release,¹⁴⁴ which can facilitate the progression from simple steatosis to MASH or HCC.^145–147 However, KCs can mitigate the accumulation of intracellular lipids through the upregulation of cholesterol efflux proteins such as ABCA1 and ABCG5.^121,122 This phenomenon may be attributed to the synergistic effects of various intricate mechanisms, given that the total free fatty acid pool, particularly palmitoleate, compromises the cholesterol efflux capacity of macrophages.¹⁴⁸ While this could potentially ameliorate the inflammatory phenotype of KC, it also leads to an influx of cholesterol into the circulation, precipitating cardiovascular-related adverse events, which is one of the primary contributors to unfavorable prognosis in patients with MASLD.^149,150

Mitochondrial dysfunction in the hepatocytes of patients with MASLD has been well documented, encompassing mechanisms such as excessive mitochondrial fission, impaired autophagy, and compromised ETC function.¹⁵¹ However, the understanding of mitochondrial alterations in macrophages of patients with MASLD remains limited. Palmitic acid can induce mitochondrial dysfunction in macrophages,^134,152 promoting the leakage of mitochondrial DNA (mtDNA) in KCs and leading to NLRP3 inflammasome activation,¹⁵² which may also contribute to a decrease in fatty acid oxidation (FAO). Increased FAO is a metabolic characteristic of M2 macrophages.¹⁵³ Inhibition of carnitine palmitoyltransferase 1A (CPT1A), the rate-limiting enzyme of FAO, inhibits the inflammatory response and endoplasmic reticulum (ER) stress responses induced by palmitic acid in macrophages.¹⁵⁴ IL-25, a member of the IL-17 family, is reduced in patients with MASLD, which decreases the expression of lipolytic enzymes, including adipose triglyceride lipase (ATGL), monoacylglycerol lipase (MAGL), and phosphorylated hormone-sensitive lipase (p-HSL) in macrophages, inhibits M2 polarization, and facilitates the development of MASLD.¹⁵⁵ However, lipolysis may also promote an inflammatory macrophage phenotype. MAGL, which links lipid metabolism and inflammation, degrades monoacylglycerol to release free fatty acids that are essential for prostaglandin synthesis.¹⁵⁶ In mice, myeloid-specific ablation of MAGL shifts the macrophage population from the inflammatory Ly6C^hi to the adaptive Ly6C^lo phenotype, limiting hepatic inflammation and fibrosis.¹⁵⁷ MAGL inhibition can also activate the IFNⅠ pathway in macrophages and impair liver regeneration.¹⁵⁸ This implies that the utilization of MAGL inhibitors requires careful evaluation of their potential detrimental impact on hepatic regeneration.

LAMs can engulf local DAMPs to limit inflammation in hCLS. However, chronic hepatic injury results in the loss of efferocytic function and acquisition of an inflammatory phenotype in LAMs.¹⁵⁹ Studies by Ji and colleagues showed that LAMs exhibited an M1/M2 hybrid phenotype and highly expressed genes involved in lipid metabolism. These cells, dependent on TREM2, display increased lipid uptake and catabolism, as well as reduced fatty acid synthesis. TREM2 deletion increases the accumulation of FAs in macrophage mitochondria, thereby promoting oxidative stress and accelerating the progression of MASLD.⁵³ In summary, macrophages in the MASLD typically exhibit enhanced lipid uptake and synthesis in response to a steatotic environment. The lipids accumulated in macrophages provide substrates for the synthesis of inflammatory mediators such as prostaglandins and function as signaling molecules to activate downstream pathways, driving macrophages to a proinflammatory phenotype.^128,153,160 These macrophages can directly damage hepatocytes through cytokine secretion and stimulate lipolysis in adipocytes, which increases lipid accumulation in the liver and triggers inflammatory and fibrotic responses.¹²⁹ Proinflammatory macrophages also recruit neutrophils and T cells, thereby amplifying the liver inflammatory environment and contributing to the development of MASH.^121,129 Lipid oxidation in macrophages is a complex process. Initially, FAO is upregulated in response to the abundance of substrates. However, subsequent mitochondrial dysfunction results in FAO downregulation, exacerbating lipid accumulation within these cells (Figure 3). Macrophage mitochondrial damage activates the NLRP3 inflammasome, driving IL-1β secretion to exacerbate MASH. It also induces the ROS production in macrophages, resulting in intracellular lipid peroxidation.^152,161 Products of lipid peroxidation, such as 4-hydroxynonenal (HNE), can modify macromolecules like proteins in a process known as lipoxidation. Lipoxidation can induce either a proinflammatory M1 phenotype or a pro-tumorigenic M2 phenotype in macrophages, depending on its concentration.¹⁶² Therefore, lipid peroxidation within macrophages may significantly influence the disease trajectory of patients with MASLD, potentially leading to the progression to fibrosis or HCC.

DCs are sensitive to extracellular fatty acids during the activation of innate immunity. Fatty acid damage impairs antigen processing and presentation capabilities of DCs, leading to compromised antigen-specific cellular and humoral immunity in MASLD mice.¹⁶³ However, high concentrations of fatty acids can also activate TLR signaling in DCs, which is attributed to the inhibition of HK-driven glycolysis and subsequent mitochondrial damage, leading to ROS production.¹⁶⁴

Amino acid metabolism

Under normal conditions, macrophages do not primarily rely on amino acids as their energy source. However, in MASLD, the breakpoints in TCA lead to an enhancement of amino acid metabolic pathways to replenish intermediates of the TCA cycle,¹⁰⁸ which is known as anaplerosis. Glutamine, the most abundant amino acid in the human body, can be broken down into ammonium ions and glutamate by mitochondrial glutaminase. The resulting glutamate can then be converted into α-KG by either glutamate dehydrogenase or transaminases. α-KG, derived from glutamine anaplerosis, is crucial for the differentiation of macrophages into immunosuppressive phenotypes.¹⁶⁵ α-KG limits the activation of the inflammatory macrophage phenotype by inhibiting the HIF-1α and NF-κB signaling and promotes OXPHOS and FAO in macrophages.¹⁶⁶ Glutamine can also produce succinate, exacerbating its accumulation in macrophages to induce an inflammatory phenotype and serve as a substrate for M1 macrophage gluconeogenesis.^167,168 Fructose enhances glutamine catabolism in macrophages, leading to the generation of succinate, fumarate, and malate, which boost the secretion of inflammatory cytokines.¹¹⁷ Nevertheless, the role of macrophage glutamine metabolism in MASLD progression requires further elucidation through in vivo experiments.

Arginine can be converted into NO and citrulline by NOS or into ornithine and urea by arginase (ARG). In classically activated M1 macrophages, inducible NOS (iNOS) is upregulated, whereas ARG is a marker of alternatively activated M2 macrophages.¹⁶⁹ ARG1 is primarily expressed in hepatocytes and is the isoform involved in the urea cycle. ARG1 deficiency is a rare autosomal recessive inherited disorder, with patients exhibiting hyperargininemia, spastic paraparesis, progressive neurological and intellectual impairment, persistent growth retardation, and rare episodes of hyperammonemia.¹⁷⁰ ARG2 is mainly expressed in immune cells such as macrophages and DCs.¹⁷¹ Mice deficient in ARG2 show increased hepatic de novo lipogenesis (DNL) and susceptibility to HFD-induced liver injury. The depletion of macrophages using liposome-encapsulated clodronate normalized the expression of DNL-related genes, reversing hepatic steatosis and fibrosis in Arg2^−/− mice.¹⁷² This finding suggests that macrophages can regulate hepatocyte metabolism and promote lipid accumulation in hepatocytes, challenging the 2-hit hypothesis that steatosis precedes inflammation in MASLD development.¹⁷³ However, in atherosclerosis models, Arg2 can induce the M1 phenotype in macrophages,^174,175 indicating the tissue-specific role of ARG in macrophage polarization. Interestingly, iNOS, a marker of proinflammatory macrophages, promotes lipid accumulation but limits liver inflammation and fibrosis,¹⁷⁶ which is possibly related to its inhibition of HSC activation.¹⁷⁷ NO derived from arginine metabolism can inhibit rate-limiting enzymes to form a breakpoint in the TCA cycle.^101,108 In patients with liver fibrosis, CD14⁺TREM1⁺iNOS⁺ macrophages accumulate in the gut. These macrophages can increase the intestinal barrier permeability by releasing IL-6 and NO, leading to the translocation of bacteria and their products.¹⁷⁸ This process induces excessive activation of immune cells and exacerbates inflammation in the liver and progression of MASH.¹⁷⁹ The distinct functions of ARG and iNOS across various organs underscore the importance of designing an appropriate administration method to achieve precise targeting and to avoid adverse effects when interfering with arginine metabolism to intervene in the progression of MASLD.

Compared with healthy controls, glutamate, serine, and glycine levels are decreased, whereas proline, hydroxyproline, and tyrosine levels are elevated in patients with MASLD/MASH.⁷⁶ The impact of altered amino acid availability on macrophage polarization requires further study. Recent studies have indicated that E2F2 is downregulated during the progression of MASH, leading to upregulation of the amino acid transporter SLC7A5. This upregulation facilitates the transport of leucine into macrophages to activate mTORC1, which inhibits autophagy to impair mitochondrial function, causing macrophages to shift toward glycolysis and develop a proinflammatory phenotype.¹⁸⁰

Collectively, dysregulated amino acid metabolism in MASLD modulates macrophage polarization through its metabolic intermediates and crosstalk with glucose and lipid metabolic pathways.^{166–168,172} These reprogrammed macrophages exhibit dual pathogenic effects: (1) aggravating hepatic inflammatory responses, and (2) directly stimulating DNL in hepatocytes. This mechanistic insight provides experimental validation for the multiple-hit theory of MASLD pathogenesis.^{166–168,172,173} Nevertheless, the specific steatogenic mediators released by macrophages and the downstream signaling events in hepatocytes require systematic investigation through integrated multi-omics strategies and cell-specific knockout models.

Lymphoid cell metabolism in MASH

T cell

The role of lymphocyte metabolism in the progression of MASLD has not been extensively studied, and the immunometabolism of T cells in fatty liver disease has emerged as a field of significant interest. T-cell activation is regulated by nutrient availability in the environment.¹⁸¹ Excessive intracellular lipid accumulation may impair T-cell function through lipid peroxidation.¹⁸² In the MASH mouse model, hepatic lipid accumulation leads to impaired glycolysis and FAO in CD8⁺ T cells, resulting in mitochondrial dysfunction that compromises their mobility and cytotoxicity.^183,184 In CD4⁺ T cells, linoleic acid can activate PPARα to upregulate CPT, thereby promoting the transport of fatty acids into the mitochondria. However, in the context of MASLD, due to impaired mitochondrial ETCs, the accumulation of fatty acids within the mitochondria leads to increased ROS production and apoptosis of CD4⁺ T cells.¹⁸⁵ Mitochondrial dysfunction in T cells within MASLD promotes the development of HCC,^183–185 and impairs the response of MASH-HCC to anti-PD-1 therapy. Interestingly, auto-aggressive CXCR6⁺ CD8⁺ T cells appear to evade the adverse effects of MASH background, exhibiting enhanced motility.¹⁸³ This may be attributed to the high expression of PD-1 on these cells,⁶⁸ which can promote intracellular lipolysis and β-oxidation of endogenous fatty acids, thereby adapting to the metabolic disruptions caused by MASLD.¹⁸⁶

Certain hypoglycemic or hypolipidemic agents have been demonstrated to ameliorate MASH through their effects on T-cell function. Metformin, a widely used antidiabetic agent, restores the responsiveness of MASH-HCC to anti-PD-1 therapy.¹⁸³ Sodium-glucose co-transporter 2 (SGLT2) inhibitors have been shown to augment the expression of 3-hydroxybutyrate dehydrogenase 1 (BDH1), thereby stimulating ketogenesis within T cells. β-hydroxybutyrate, a pivotal ketone body, exerts an inhibitory effect on CD8⁺ T-cell activation by suppressing the transcription factor interferon regulatory factor 4 (IRF4). Administration of SGLT2 inhibitors has been associated with a reduction in the infiltration of hepatic macrophages and auto-aggressive CXCR6⁺CD8⁺ T cells, consequently ameliorating the histopathological and biochemical characteristics of MASH.¹⁸⁷ Excessive hepatic cholesterol accumulation is implicated in the disruption of OXPHOS and amino acid catabolism in T cells, thereby inhibiting their self-renewal and effector functions and promoting the aggregation of myeloid-derived suppressor cells in the liver.¹⁸⁴ The over-accumulation of cholesterol within NKT cells can also precipitate intracellular lipid peroxidation, thereby diminishing their cytotoxic capabilities and undermining immune surveillance in the hepatic compartment.¹⁸⁸ These mechanisms reveal that excess cholesterol fosters the immunosuppressive milieu within the hepatic compartment. Statins, a class of drugs widely used to lower cholesterol levels, can reduce the incidence of HCC associated with MASH.^189,190 Squalene epoxidase (SQLE) inhibitor terbinafine can rescue the efficacy of anti-PD-1 therapy in MASH-HCC by inhibiting cholesterol biosynthesis.¹⁸⁴ Given the extensive clinical trial data and long-term safety profiles of these agents, large-scale clinical cohort studies examining the impact of various antidiabetic and hypolipidemic regimens on MASH progression are required.

The hepatic composition of distinct fatty acids exerts differential effects on T-cell phenotypes. An increased C16:1n7/C16:0 ratio in MASH is associated with elevated CD25⁺CD45⁺CD4⁺ T cells and reduced PD-1⁺CD4⁺ T cells, indicating a state of excessive immune activation.¹⁹¹ In the early phase of MASLD, NETs induced by fatty acids can enhance OXPHOS in naïve T cells in a TLR4-dependent manner, leading to their differentiation into Tregs and promoting HCC development.⁴⁴ Furthermore, NETs can activate Th17 cells.¹⁹² DNL plays a pivotal role in Th17 polarization, with acetyl-CoA carboxylase 1 (ACC1) serving as the key enzyme in this pathway. The upregulation of ACC1 can enhance the production of IL-17A, thereby mitigating liver inflammation, steatosis, hepatocyte ballooning, and fibrosis.¹⁹³ A distinctive Th17 cell subpopulation is present in the MASLD milieu, which is characterized by increased expression of CXCR3 and proinflammatory cytokines. Recruited by CXCL9/10 within the MASH microenvironment, these cells exhibit upregulation of the glucose transporter SLC2A3 and key glycolytic enzymes compared with other Th17 subpopulations. In MASH mouse models, the specific knockout of PKM2 in T cells results in the reduced expression of Th17-dependent cytokines and macrophage accumulation within the liver, consequently diminishing liver damage.¹⁹⁴ In addition to mitochondrial metabolism, peroxisome-mediated FAO also influences the phenotype of CD4⁺ T cells. Acyl-CoA oxidase 1 (ACOX1), a rate-limiting enzyme in this process, induces the polarization of naïve T cells toward the Th17 lineage, promoting the secretion of proinflammatory cytokines and the recruitment of myeloid cells. HFD can accelerate Acox1-dependent peroxisomal β-oxidation-driven hepatocyte injury.¹⁹⁵ In summary, impaired metabolism of CD8⁺ and CD4⁺ T cells in MASH disrupts immune surveillance, enabling mutated hepatocytes to evade clearance. This defect promotes the development of MASH-HCC and contributes to tumor resistance to immunotherapy.^183–185 Intriguingly, auto-aggressive CXCR6⁺CD8⁺ T cells evade such metabolic constraints, sustaining activation and exacerbating liver injury in MASH through FasL/Fas pathway–mediated hepatocyte apoptosis.^68,183 Concurrently, steatotic liver microenvironment drives the infiltration and expansion of Th17 cells. These cells amplify inflammatory damage through the secretion of proinflammatory cytokines, such as IL-17A and IFNγ.^193–195 Notably, IL-17A is implicated in the transition from MASH to HCC, suggesting dual roles of Th17 in early inflammation and late-stage carcinogenesis¹⁹⁶ (Figure 4). Therefore, targeting T-cell metabolism may represent a promising therapeutic strategy for MASH-HCC and restoring its sensitivity to immunotherapy.

FIGURE 4.

Immune cell metabolism in MASH. We utilized the Scmetabolism package to investigate the metabolic profiles of immune cells in the GSE202379 data set. From the periphery to the center, they represent different immune cells in samples of healthy controls, MASLD, MASH without fibrosis, MASH with fibrosis, and end stage. In MASH, neutrophils exhibit activation of glycolysis and the PPP, which promotes the generation of ROS. Macrophages also display similar metabolic characteristics, which facilitate their proinflammatory phenotype. However, in the end stage, glycolysis and PPP in macrophages and neutrophils are downregulated, indicating metabolic dysfunction. Regarding lymphocytes, both T and NK cells show mitochondrial dysfunction and impaired metabolism, leading to compromised immune surveillance. Besides, a subset of CD4+ T cells exhibit increased glycolysis, DNL, and peroxisome-mediated FAO, promoting their differentiation into proinflammatory Th17 cells. In MASH, all of these immune cells exhibit active lipid metabolism, which may be an adaptive response to the environment. Abbreviations: DNL, de novo lipogenesis; FAO, fatty acid oxidation; MASLD, metabolic dysfunction–associated steatotic liver disease; MASH, metabolic dysfunction–associated steatohepatitis; PPP, pentose phosphate pathway; ROS, reactive oxygen species. Created in BioRender (BioRender.com).

NK cell

The potential role of NK cells in the pathogenesis of NAFLD has attracted increasing attention, but the field is still in its infancy, especially because most previous studies have not distinguished between NK cells and ILC1 or liver-resident NK cells.¹⁹⁷ Hepatic NK cells play a dual role in MASLD: they can enhance the inflammatory response in MASH, but they can also inhibit macrophage polarization and kill activated HSC to suppress liver fibrosis.^198–200 Typically, the activation of NK cells leads to an increase in glycolysis and OXPHOS, which provide energy and ribose for cell proliferation.²⁰¹ Within the context of MASH, persistent high-fat stimulation induces downregulation of uncoupling protein 1 (UCP1) in NK cells, promoting mitochondrial damage and autophagy, which impairs the metabolism and function of NK cells and facilitates the progression of MASH toward fibrosis.²⁰² In a high-fat milieu, the downregulation of mTOR signaling can inhibit glycolysis in NK cells, thereby compromising their cytotoxic capabilities and IFN-γ secretion.²⁰³ A study by Cuff and colleagues indicated that NK cells in MASH displayed a cytotoxic phenotype, which was partly attributed to the suppressive effects of TGF-β on glycolysis and in the steatotic liver. However, enhanced triglyceride (TG) metabolism in NK cells may be an adaptive response to the environment. The impairment of NK cell metabolism and function, on the one hand, prevents the transition from simple steatosis to MASH, but on the other hand, it also increases the risk of MASH-HCC.²⁰⁴

Advances in single-cell and spatial omics technologies have enabled precise examination of the metabolic profiles of immune cells in the livers of patients with MASLD. We analyzed the GSE202379 data set, which includes 4 healthy samples, 3 samples of simple steatosis, 3 samples of MASH, 4 samples of MASH with cirrhosis, and 6 samples of end-stage disease.²⁰⁵ Based on the cell annotations provided by the authors and using the scMetabolism package, we analyzed the metabolic conditions of macrophages, neutrophils, and lymphocytes. Consistent with the literature, macrophages in MASH predominantly exhibited inflammatory characteristics, marked by increased glycolysis and PPP activity.⁷⁶ However, macrophages with high OXPHOS levels should not be discounted. On one hand, the M1 and M2 phenotypes are merely a rough binary classification, and macrophage phenotypes actually exist on a spectrum.²⁰⁶ On the other hand, MASH tends to progress to liver cancer, which is predominantly populated by M2 macrophages. The shift from the M1 to M2 phenotype in macrophages is a crucial mechanism contributing to the development of MASH-HCC.²⁰⁷ In addition, some scholars have posited that alterations in immune cell metabolism are more likely a consequence of phenotypic changes than the cause,¹⁰⁸ with the metabolites produced potentially amplifying these phenotypes. Neutrophils also show similar metabolic profiles despite the lack of research on neutrophil metabolism in MASH. Existing evidence suggests that neutrophil glycolysis can enhance the release of ROS and NETs, which are crucial for MASH progression.^208,209 Fatty acid metabolism is notably activated across all 3 immune cell types, with a marked increase observed in lymphocytes, considering their relatively low glucose metabolism activity. As previously discussed, distinct lymphocyte subsets in MASH exhibit marked heterogeneity in their metabolic signatures and pathophysiological contributions. Therefore, studies with expanded sample sizes are required to thoroughly investigate the metabolic characteristics of these subsets and the immunometabolic crosstalk driving disease progression, thereby advancing our understanding of their roles in MASH pathogenesis (Figure 4).

THE ROLE OF METABOLIC TARGETS IN IMMUNE CELLS

Targeting hepatocyte metabolism is a key therapeutic strategy for MASH, which includes reducing lipid synthesis, increasing FAO, and improving insulin sensitivity. These liver metabolism-targeting drugs also exert an effect on immune cells to synergistically alleviate MASH. In this section, we review the current state of clinical trials for these medications and their impact on immune cells, examining the responses of patients with MASH to these drug candidates from a unique perspective (Table 1).

TABLE 1.

Clinical trials of drug candidates targeting cellular metabolism

Agents	Target	Intervention	Disease	Outcome measure	Phase	Clinical trail
Lanifibranor (IVA337)	Pan-PPAR agonist	Mono	MASH	Liver biopsy	2b	NCT03008070
			MASH	Liver biopsy	3	NCT04849728
Pioglitazone	PPARγagonist	Mono	MASH	Liver biopsy	2	NCT00013598
			MASH	Liver biopsy	2	NCT00062764
			MASLD with IGT or T2DM	Blood parameters and NMRS	2	NCT00633282
Canagliflozin	SGLT2i	Comparator: Pioglitazone	MASH with T2DM	Blood parameters	Not applicable	NCT05513729
			MASLD with T2DM	Blood parameters	Not applicable	NCT05422092
Dapagliflozin	SGLT2i	Mono	MASH	Liver biopsy	3	NCT03723252
			MASLD	Not Provided	3	NCT05308160
			MASLD with T2DM	Blood parameters	4	NCT05459701
Empagliflozin	SGLT2i	Mono	MASLD	CAP	Not applicable	NCT05694923
		Mono	MASH	Blood parameters	3	NCT05605158
Licogliflozin	SGLT1/2i	+tropifexor	MASH-fibrosis	Liver biopsy	2	NCT04065841
Dapagliflozin/Empagliflozin	SGLT2i	+T2DM standard treatment	MASLD with T2DM	Blood parameters and VCTE	3	NCT06117137
		+/−metformin and/or DPP4i	MASLD with T2DM	Fibroscan	4	NCT04976283
Not provided	SGLT2i	Mono	MASLD	Fibroscan	Not applicable	NCT06739486
PF-05221304 (Clesacostat)	ACCi	Mono	MASLD	Rate of Secretion of VLDL and TG	1	NCT04395950
		+ PF-06865571 (Ervogastat)	MASH	Liver biopsy	2	NCT04321031
TVB-2640 (Denifanstat)	FASNi	Mono	MASH	Liver biopsy	2	NCT04906421
			MASH	MRI	2	NCT03938246
Aramchol	SCD-1i	Mono	MASLD	NMRS	2	NCT01094158

Abbreviations: ACCi, acetyl-CoA carboxylase inhibitor; CAP, controlled attenuation parameter; FASNi, fatty acid synthase inhibitor; IGT, impaired glucose tolerance; MASH, metabolic dysfunction–associated steatohepatitis; MASLD, metabolic dysfunction–associated steatotic liver disease; NMRS, nuclear magnetic resonance spectroscopy; SGLT2i, sodium-glucose co-transporter 2 inhibitor; T2DM, type 2 diabetes mellitus; TG, triglyceride; VCTE, vibration-controlled transient elastography.

PPAR agonist

Peroxisome proliferator–activated receptors (PPARs) are a family of nuclear transcription factors widely distributed in human tissues. They are categorized into 3 main subtypes, PPARα, PPARβ/δ, and PPARγ, and play a pivotal role in metabolic regulation and the development of MASH-fibrosis.²¹⁰

The PPAR family exerts multifaceted regulatory effects on hepatic metabolism through isoform-specific mechanisms. PPARα enhances fatty acid uptake, FAO, and ketogenesis in hepatocytes, while its activation promotes cholesterol efflux and suppresses proinflammatory responses in macrophages.^211,212 Intriguingly, PPARα upregulates CPT, the rate-limiting enzyme of FAO, and induces mitochondrial ROS-mediated apoptosis in CD4⁺ T cells, which contributes to the decreased HCC incidence in murine steatohepatitis models following PPARα activation.¹⁸⁵ PPARα agonist also attenuates hepatic fibrosis progression in MASH models.^213,214 This effect probably attribute to PPARα-mediated modulation of hepatocellular lipid metabolism, especially given the relatively low expression of PPARα in hepatic macrophages.²¹⁵ PPARβ/δ can induce FAO and suppress fatty acid synthesis to alleviate TG accumulation in the liver.^211,212 This isoform upregulates the lipid metabolism-related genes (CD36, perilipin-2, etc.) to promote anti-inflammatory polarization of macrophages.^214,216 However, it shows no significant impact on intracellular cholesterol/triglyceride levels of macrophage,²¹⁶ suggesting that its specific mechanisms regulating macrophage metabolism and immune functions require further elucidation. PPARγ demonstrates tissue-specific duality: in adipose tissue, it reduces FA release through enhanced lipid storage, indirectly decreasing hepatic lipid uptake and steatosis.^212,217 In hepatocytes, it accelerates MASH progression by upregulating lipogenesis and lipid droplet formation genes.²¹⁸ PPARγ potentiates macrophage anti-inflammatory polarization by increasing cholesterol efflux, OXPHOS, and glutamine metabolism,^219,220 and PPARγ-mediated fatty acid uptake is essential for CD4⁺ T-cell activation.²²¹ However, the pathophysiological contribution of this metabolic adaptation to MASH progression remains mechanistically undefined.

Mechanistic insights into PPAR isoform-specific functions have propelled the development of pan-PPAR agonists in clinical trials, aiming to harness synergistic metabolic-immune benefits while navigating tissue-specific risks. The effectiveness of the pan-PPAR agonist lanifibranor has been confirmed in phase II clinical trials.²²² A systematic review of clinical trials for MASH indicated that lanifibranor can ameliorate the histological characteristics and fibrosis associated with MASH.^223,224 Lanifibranor can also reduce the occurrence of cardiovascular events in patients with MASH.²²⁵ However, the phase III trial of PPARα/β agonist elabibranor cast doubt on this approach. Interim analysis revealed no significant difference in the improvement of MASH scores compared to placebo. This may be attributed to elabibranor’s lack of PPARγ activation, as multiple clinical trials using pioglitazone have underscored the importance of PPARγ in resolving MASH.²²⁶ PPARγ can also prevent HSC activation.²²⁷ Furthermore, lanifibranor can upregulate the expression of macrophage lipid metabolism genes such as CD36 and TREM2, suppressing the proinflammatory polarization of macrophages induced by fatty acids. Lanifibranor’s rapid and specific inhibition of the accumulation of macrophages in the liver, which promotes the resolution of liver fibrosis, indicates that this cell type may be the primary target of PPAR agonists for alleviating MASH.²¹⁴

SGLT-2 inhibitors

SGLT inhibitors (SGLT-2i) are a kind of medication approved for the treatment of patients with type 2 diabetes mellitus. These drugs disrupt renal glucose reabsorption, leading to glycosuria and weight reduction, and exert positive effects on body weight, glucose metabolism, and cardiovascular and renal health. SGLT-2i can also stimulate lipolysis, FAO, and ketogenesis, all of which are intimately linked to the pathogenesis of MASLD. Consequently, SGLT-2i are of considerable interest for patients with MASLD and have been investigated in both preclinical and clinical studies for MASLD.²²⁸ SGLT-2i can inhibit macrophage glycolysis to induce an anti-inflammatory phenotype and promote CD8⁺ T-cell ketogenesis to reduce their secretion of cytotoxic cytokines.^187,229 These mechanisms collectively contribute to the protective effects of SGLT-2i in MASH mouse models. Diabetes mellitus is a risk factor for MASLD.²³⁰ SGLT-2i can reduce the incidence of MASLD in patients with type 2 diabetes.²³¹ Numerous studies have reported that SGLT2i can reverse hepatic steatosis and reduce plasma aminotransferase levels in patients with type 2 diabetes.^232–237 However, the role of SGLT2 in alleviating MASH-fibrosis remains controversial. Some studies using noninvasive methods to assess the degree of fibrosis suggest that SGLT2i have no significant effect on fibrosis improvement.^238–240 However, the accuracy of these imaging techniques and algorithms for noninvasive indices of fibrosis requires further consideration. In an Asian open-label clinical trial employing liver biopsy, the gold standard for fibrosis assessment, 3 patients with MASH treated with empagliflozin for 6 months showed fibrosis regression among 5 patients at the F1 fibrosis stage.²⁴¹ A systematic review confirmed that SGLT-2i can improve liver fibrosis in patients with MASH. However, individual differences among patients are worth considering.²⁴² Larger-scale studies are needed to explore whether SGLT-2i has a beneficial effect on MASH-fibrosis. Among the various SGLT-2i, dapagliflozin has entered phase III trials. Dapagliflozin targets PFKFB3 to inhibit glycolysis in macrophages, thereby inducing M2 polarization and improving steatosis and fibrosis in MASH models.²⁴³ This highlights the significant synergistic effect of SGLT-2i in modulating both immune and hepatocyte metabolism to ameliorate liver injury.

De novo lipogenesis inhibitors

Elevated hepatic DNL is a critical contributor to MASLD, and targeting the rate-limiting enzymes in this pathway represents a strategic approach to treating MASH. ACC emerges as a compelling target for MASH therapy, given its pivotal role in catalyzing the transformation of acetyl-CoA into malonyl-CoA, a fundamental step in DNL. T cells depend on ACC-mediated DNL to generate membrane phospholipids essential for Th17 differentiation.^193,244 Pharmacological inhibition of ACC suppresses Th17 polarization and reduces IL-17 production to ameliorate MASH pathology.¹⁹³ Moreover, ACC inhibition reduces HSC activation by decreasing glycolysis and mitochondrial respiration.²⁴⁵ Intriguingly, ACC ablation induces a hyperglycolytic bioenergetic state in macrophages yet paradoxically compromises their metabolic flexibility by blunting LPS-induced glycolytic augmentation, thereby impeding proinflammatory polarization and the release of proinflammatory cytokines.²⁴⁶ ACC inhibitors such as GS-0976 and MK-4074 can significantly reduce hepatic steatosis. However, hyperlipidemia, a common side effect of ACC1/2 inhibitors, has restricted their broader clinical use.²⁴⁷ Currently, various strategies are available to avoid this side effect. Diacylglycerol acyltransferase 2 (DGAT2) controls TG synthesis. Calle et al²⁴⁸ demonstrated that co-administration of DGAT2 inhibitors could mitigate hypertriglyceridemia induced by ACC inhibitors. In patients with MASH with fibrosis, co-administration of fenofibrate reduced the increase in LDL-apoB caused by ACC inhibitors.²⁴⁹ Phosphorylation is essential for ACC activation. Inhibition of the interaction between 12-lipoxygenase (ALOX12) and ACC1 can suppress ACC1 phosphorylation and improve MASH without elevating blood TG levels.²⁵⁰ However, the efficacy of these strategies requires validation in large-scale clinical trials.

FASN, a rate-limiting enzyme downstream of ACC in DNL, synthesizes acetoacetyl-CoA for cholesterol production to maintain lipid rafts, which is essential for TLR4 membrane translocation, driving macrophage M1 polarization.^251,252 Notably, FASN shares functional parallels with ACC in promoting Th17 differentiation.²⁵³ These findings indicate that fatty acid synthesis might mediate inflammatory immune responses by structurally remodeling immune cell membranes. FASN can also suppress activation of the STING pathway in macrophages, mitigating liver damage induced by sepsis.²⁵⁴ TVB-2640, a FASN inhibitor, can effectively inhibit liver fat synthesis in patients with metabolic abnormalities.²⁵⁵ In a randomized placebo-controlled trial, TVB-2640 significantly reduced liver fat content, inflammation, and fibrosis markers without elevating plasma TG levels.²⁵⁶ Denifanstat is an orally administered small-molecule FASN inhibitor, and a multicenter, double-blind, randomized, placebo-controlled phase Ⅱb clinical trial showed that it can alleviate MASH and fibrosis in ~30% of patients. It also decreases plasma LDL-C levels and increases the ratio of beneficial polyunsaturated fatty acids (PUFAs), potentially offering new metabolic health benefits for patients with MASH.²⁵⁷ Given that MASH represents a hepatic manifestation of metabolic syndrome, denifanstat may be a promising drug candidate. Moreover, recent studies have demonstrated that FASN inhibition can enhance the expression of MHC-I molecules in hepatocytes, promoting antigen presentation and stimulating the immune surveillance function of CD8⁺ T cells. Consequently, FASN inhibitors could serve as potent treatments for MASH-HCC, particularly when co-administered with immune checkpoint inhibitors.²⁵⁸

Stearoyl-CoA desaturase-1 (SCD-1) converts SFAs into monounsaturated fatty acids. The inhibition of SCD-1 can reduce fatty acid synthesis, increase FAO, and decrease the storage of TG and fatty acids in the liver. The SCD-1 inhibitor aramchol has shown promise in mitigating fibrosis in patients with MASH.^259,260 Interestingly, while aramchol decreased liver fat content, the difference is not statistically significant compared with the placebo group.^260,261 Given the tight link between fibrosis and the activation of immune cells, aramchol may exert its protective effects by modulating immune cell activity. The fact that the improvement of MASH-fibrosis by aramchol is abolished in patients with MASLD with HIV also supports this view.²⁶² SCD-1 inhibition can boost ATGL-mediated hydrolysis of TGs and phospholipids, leading to the release of docosahexaenoic acid (DHA) in CD4⁺ T cells, which promotes their differentiation into Tregs.²⁶³ SCD-1 inhibition also enhances macrophage cholesterol efflux, potentially suppressing their proinflammatory activation.²⁶⁴ In CD8⁺ T cells, SCD-1 inhibition restrains oleic acid and esterified cholesterol production by inhibiting acetyl-CoA acetyltransferase 1 (ACAT1). This enhances the release of cytotoxic cytokines like IFN-γ and TNF-α, bolstering the antitumor effects of CD8⁺ T cells. However, whether this mechanism exacerbates MASH remains unclear.²⁶⁵ Overall, the impact of SCD-1 inhibition on immune cell phenotypes in fatty liver conditions remains poorly understood and requires further research.

CONCLUSION AND PERSPECTIVE

Immunometabolism is a thriving field that has been extensively explored in cancer. However, research on MASH remains limited.²⁶⁶ This is largely attributed to the challenges in achieving specific gene knockouts within the immune cells in mice. However, as a disease characterized by metabolic disorders and high involvement of immune cells, metabolic changes in immune cells in MASH are of considerable importance. This review elucidates the metabolic reprogramming of immune cells in MASH, with a primary focus on macrophages and T cells. In MASH, hepatic macrophages undergo metabolic reprogramming characterized by enhanced glycolysis, disrupted lipid homeostasis, and dysregulated amino acid metabolism, collectively triggering the release of proinflammatory factors and chemokine-mediated immune cell recruitment.^6,76 CD8⁺ T cells exhibit mitochondrial dysfunction, leading to impaired immune surveillance and facilitation of hepatocarcinogenesis.^183,184 Conversely, CXCR6⁺CD8⁺ T cells evade metabolic constraints, exacerbating hepatocyte apoptosis through FasL/Fas-mediated cytotoxicity.^68,183 Th17 cells, driven by the steatotic hepatic microenvironment, amplify inflammatory damage through secretion of proinflammatory cytokines such as IL-17A and IFNγ.^193–195 Despite these advances, research on the metabolic features of other immune cells, including neutrophils and DCs, remains limited. Single-cell sequencing provides a powerful technological approach to solving the problem. However, current research faces 2 major obstacles. First, existing sample sizes remain insufficient for the comprehensive characterization of immune cell metabolic heterogeneity. The scarcity of high-quality specimens stems from the clinical nature of MASH: early-stage patients rarely undergo liver biopsy, while late-stage samples are often complicated by comorbidities. Second, technical disparities across research platforms hinder effective cross-data set integration. Addressing these limitations requires establishing standardized specimen collection and processing protocols, coupled with multicenter collaborations to promote open sharing of biological samples and omics data, thereby laying the foundation for the mechanistic dissection of immunometabolic regulation in MASH pathogenesis.

Current research primarily focuses on the impact of immune cell metabolic reprogramming on their functional phenotypes. However, there remains a lack of direct evidence supporting the specific mechanisms by which metabolically activated immune cells drive the progression of MASH.^6,76 For instance, although it is known that M1 macrophages can promote liver injury and fibrosis through the secretion of proinflammatory cytokines, recruit neutrophils to amplify inflammatory responses, or activate HSCs via cell-cell contact,²⁶⁷ whether these pathological effects are directly driven by specific metabolic alterations has not yet been causally demonstrated through genetic or pharmacological approaches. Crucially, it remains an unresolved problem how metabolically reprogrammed immune cells dynamically regulate their interactions with hepatocytes and other NPCs, and whether the downstream signaling pathways of these interactions exhibit metabolic dependency. Therefore, future research should integrate spatial multi-omics technologies, conditional gene knockout models, and real-time metabolic imaging approaches to precisely track the characteristics of metabolically reprogrammed immune cells during specific pathological stages (simple steatosis → inflammation → fibrosis).

Insulin resistance is a pivotal driver of hepatic metabolic dysfunction, characterized by reduced responsiveness of peripheral tissues to physiological insulin levels, leading to compensatory hyperinsulinemia. Studies demonstrate that insulin resistance exacerbates hepatic lipid imbalance through dual mechanisms: on one hand, it enhances adipose tissue lipolysis, resulting in excessive free fatty acid delivery to the liver; on the other hand, it amplifies hepatic lipid synthesis, further aggravating lipid accumulation.²⁶⁸ Notably, hyperinsulinemia not only disrupts hepatic metabolism but also reprograms immune cell function. In diabetic models, insulin signaling suppresses macrophage apoptosis.^269,270 It can also upregulate CD36 and scavenger receptor A expression to promote oxLDL uptake, downregulate ABCA1 to inhibit cholesterol efflux,^270,271 and reduce ATGL-mediated TG hydrolysis,²⁷² collectively driving macrophage lipid accumulation and proinflammatory phenotype polarization. Simultaneously, insulin enhances glucose and glutamine uptake in T cells, activating glycolysis and OXPHOS to promote proinflammatory cytokine secretion (eg, IFN-γ, IL-17A).^273,274 These mechanisms synergistically fuel MASH progression, underscoring the necessity for precise modulation of insulin signaling in therapeutic strategies. Similarly, leptin, an anti-lipogenic hormone, exerts protective effects in hepatocytes by suppressing lipid synthesis and promoting lipid mobilization. However, supra-physiological leptin concentrations under pathological conditions are closely associated with inflammation and fibrosis.^275,276 In vitro studies have demonstrated that leptin can modulate the activation, proliferation, maturation, and production of inflammatory mediators across diverse immune cell populations.²⁷⁷ The interplay between leptin and the mTOR pathway also indicates its critical role in immunometabolism.²⁷⁸ Leptin enhances glucose uptake and glycolysis in T cells, thereby fueling the differentiation of CD4⁺ T cells into Th1 and Th17 subsets and their subsequent proinflammatory cytokine production.^279,280 Nevertheless, systematic investigations into leptin-driven immunometabolic regulation within the MASH microenvironment remain limited, hindering comprehensive risk assessment of leptin-based therapies.

Although current studies have extensively investigated the direct metabolic effects of hormones in hepatocytes, their role in immunometabolic crosstalk and mechanistic contributions to MASH pathogenesis remain inadequately defined. Given the pivotal role of immune cells in MASH pathogenesis, a systematic investigation into these mechanisms is imperative to prevent unintended consequences of hormone-modulating therapies. The approval of resmetirom represents a pivotal advancement in MASH treatment. As a hepatocyte-specific thyroid hormone receptor-β agonist, this agent selectively enhances hepatic FAO, while sparing thyroid hormone receptor-α receptors abundantly expressed in immune cells (macrophages, neutrophils) and extrahepatic tissues (cardiac muscle, bone).^281,282 This spatial precision effectively uncouples therapeutic metabolic effects from off-target inflammatory responses and systemic toxicity, establishing a framework for refining hormone-based interventions.

A variety of drugs that target hepatocyte metabolism are undergoing clinical trials designed to inhibit hepatic lipid synthesis, promote lipid degradation, and prevent lipid accumulation in the liver.⁵ These lipids can induce immune cell activation and subsequent oxidative stress, exacerbating liver damage. By inhibiting lipid accumulation, these drugs eliminate the primary stimulus for inflammation. These drug candidates also affect immune cell metabolism. Our studies suggest that the inhibition of these targets can suppress immune cell metabolism and impair their activation. This synergistic effect underscores the potential of metabolic targeting in MASH therapy.

In this review, we outline the metabolic processes within immune cells, predominantly macrophages and T cells, and discuss clinical trial results for metabolic targets and their effects on immune cells. With the development of single-cell technologies and the diversification of transgenic mice, we anticipate that the immunometabolism field in MASH will evolve rapidly and will bridge current knowledge gaps between in vitro metabolic observations and clinical disease phenotypes.

Supplementary Material

hep-84-1962-s001.pdf ^{(180.4KB, pdf)}