Pathophysiological communication between hepatocytes and non-parenchymal cells in liver injury from NAFLD to liver fibrosis

Abstract

Non-alcoholic fatty liver disease (NAFLD) is a multifactorial disease that encompasses a spectrum of pathological conditions, ranging from simple steatosis (NAFL), nonalcoholic steatohepatitis (NASH), fibrosis/cirrhosis which can further progress to hepatocellular carcinoma and liver failure. The progression of NAFL to NASH and liver fibrosis is closely associated with a series of liver injury resulting from lipotoxicity, oxidative stress, redox imbalance (excessive nitric oxide), ER stress, inflammation and apoptosis that occur sequentially in different liver cells which ultimately leads to the activation of liver regeneration and fibrogenesis, augmenting collagen and extracellular matrix deposition and promoting liver fibrosis and cirrhosis. Type 2 diabetes is a significant risk factor in NAFLD development by accelerating liver damage. Here, we overview recent findings from human study and animal models on the pathophysiological communication among hepatocytes (HCs), Kupffer cells (KCs), hepatic stellate cells (HSCs) and liver sinusoidal endothelial cells (LSECs) during the disease development. The mechanisms of crucial signaling pathways, including Toll-like receptor, TGFβ and hedgehog mediated hepatic injury are also discussed. We further highlight the potentials of precisely targeting hepatic individual cell-type using nanotechnology as therapeutic strategy for the treatment of NASH and liver fibrosis.

1. Introduction

Non-alcoholic fatty liver disease (NAFLD) is a multifactorial disease that is considered as the most common cause of chronic liver disease [1,2]. NAFLD encompasses a spectrum of pathological conditions, ranging from simple steatosis (NAFL), nonalcoholic steatohepatitis (NASH), fibrosis/cirrhosis which can further progress to hepatocellular carcinoma and liver failure [3]. Initially, a “two-hit” hypothesis was established to describe the mechanistic development of NAFLD [4]. The ‘first hit’ usually is the accumulation of lipids, including triglycerides, free fatty acids (FFAs), and small amount of cholesterol in hepatocytes, which induce NAFL and increase the vulnerability of liver to injury caused by the ‘second hits’. Biopsy diagnosis reveals that more than 5% of lipid content by weight is presented in hepatocytes (macrovesicular) in NAFL. The progression from NAFL to liver fibrosis is closely associated with a series of injuries resulting from lipotoxicity, mitochondrial dysfunction, cellular stress and inflammation in the liver [5] that constitutes the “second hit” thought to be required to progress NAFL to NASH. This ‘two-hit’ hypothesis has been further evolved into “three-hit” and “multiple-hit” hypothesis due to the novel findings from recent research. The defective hepatocyte regeneration that is crucial for the replenishment of dead cells has been proposed to be the “third hit” in NAFLD development [6], and the adverse factors, including insulin resistance, inflammatory mediators derived from adipose tissues, dietary factors, lipopolysaccharide (LPS) generated by gut microbiota and genetic and epigenetic factors are the proposed multiple hits that drive the progression of NAFLD to fibrosis [7]. Given the wide range of pathological conditions involved in NAFLD, an international consensus has renamed NAFLD to Metabolic Associated Liver Disease (MAFLD) [8–10]. These liver injuries ultimately leads to the activation of hepatic stellate cells (HSCs), augmenting collagen and extracellular matrix (ECM) production/deposition, thus, promoting liver fibrosis and cirrhosis [11]. A more in-depth histological report in addition to the grading system within each stage of progression may be found in Brunt et al. [12]. Currently, there is no Food and Drug Administration (FDA)-or European Medicines Agency (EMA)-approved antifibrotic treatments available which posts an urgent need for developing anti-NASH and anti-hepatic fibrosis therapeutics [13,14].

Type 2 diabetes (T2D) is a significant risk factor in NAFLD development by driving the progression of liver injury from simple steatosis (NAFL) to NASH and the subsequent liver fibrosis [15]. Clinically, a recent meta-analysis estimated that the global prevalence of NAFLD among patients with T2D is 55.5%, and the prevalence of NASH among patients with T2D is 37.3% [15]. In patients with asymptomatic T2D, the estimated prevalence of advanced fibrosis is around 5%–7%, which leads to high complications of liver cirrhosis and hepatocellular carcinoma [16,17]. Growing bodies of evidence in both human and animal study have contributed to the establishment of the mechanistic link between hepatic steatosis and insulin resistance. Insulin resistance in hepatocytes (HCs) leads to accumulation of lipids (lipotoxicity) which is associated with dysregulated intermediary metabolism, such as enhancing hepatic de novo lipogenesis and augmenting the flow of fatty acids (FAs) from adipocytes to the liver [18], leading to excessive lipid uptake by the liver which favors progression of NAFL to NASH and cirrhosis. Insulin resistance further impairs mitochondrial fatty acid oxidation and promotes fat accumulation, leading to mitochondrial oxidative stress [19]. Activation of inflammatory kinase JNK by ER stress, in turn, aggravates insulin resistance in HCs through blunting the insulin signaling cascade [20]. In addition, glucotoxicity in T2D patients, which refers to metabolic alterations caused by chronically hyperglycemia and impaired branched chain amino acids metabolism, promotes NASH progression via mTOR signalling [21]. In contrast, study on nondiabetic patients with NAFLD demonstrated that increased peripheral insulin resistance, but not the steatosis or obesity, is the critical link with liver fibrosis [22].

The goal of this review is to provide a synopsis on the current literature investigating cellular and molecular mechanisms regulating liver injury at the different disease stages, from NAFL to NASH and liver fibrosis. Pathophysiological communications among HCs, Kupffer cells (KCs), hepatic stellate cells (HSCs) and liver sinusoidal endothelial cells (LSECs) via mediation of Toll-like receptor, hedgehog and TGFβ signaling pathways are further discussed to highlight the potentials of precision medicine via cell-type specific therapeutic strategy for NASH and liver fibrosis.

2. Biocommunication among liver cells

There are two major cell types in the liver, parenchymal cells and non-parenchymal cells. Parenchymal HCs are the primary component of the liver, occupying 60% of the total cell number and 80% of the total liver volume. They are radially located between the liver capillaries and sinusoids in the lobule (Fig. 1). HCs are the predominant functional units of liver that play pivotal role in energy metabolism, detoxification and protein synthesis owing to the large number of mitochondria, rough endoplasmic reticulum (ER), Golgi apparatus and free ribosomes. HCs possess high regeneration capabilities and play a major role in proliferation after liver injury [23,24]. Sinusoidal cells are non-parenchymal cells, constituting 35% of the total cell number and 17% of the total volume of the liver, which include liver sinusoidal endothelial cells (LSECs), Kupffer cells (KCs), hepatic stellate cells (HSCs) [25,26]. LSECs are the second most abundant cell type of the liver after HCs, forming the sinusoidal wall and represent the interface between the blood and KCs on the one side and HCs and HSCs on the other side. They are not only physical barriers, but also contribute to numerous physiological functions, including metabolite transportation, inflammation, and angiogenesis [27]. HSCs are pericytes resided in the perisinusoidal space (space of Disse) and are surrounded by HCs and LSECs and secrete laminin, proteoglycans, and type IV collagen to form basement membrane [28]. HSCs are predominant cell type primarily responsible for liver fibrosis. Hepatocytes loss, inflammation and metabolic alterations stimulate quiescent HSCs trans-differentiation into proliferative, migratory, and contractile myofibroblast-like cells (activated HSCs) [29]. KCs, the specialized macrophages of the liver, also residing in sinusoids, are vital to hepatic and systematic response to pathogens. KCs have emerged as an essential regulator in liver injury and wound healing (Fig. 1) [30].

Fig. 1.

Physiological interaction of hepatocytes (HCs), hepatic stellate cells (HSCs), Kupffer cells (KCs) and liver sinusoidal endothelial cells (LSECs) in the liver.

In the liver, the space of Disse is filled with permeable connective tissue which favours biological exchange between portal blood flow from gastrointestinal tract and HCs. Here, bio-communication takes place among different neighboring cell types, including HCs, biliary epithelial cells, hepatic progenitor cells (HPSc), KCs, bone marrow-derived macrophages, HSCs, LSECs, infiltrating immune cells, and nerve cells via soluble mediators and cytokines [31]. The lipotoxicity induced by the excessive accumulation of lipids in HCs upon metabolic dysregulation, promotes the occurrence of oxidative and ER stress, metabolic inflammation, hepatocyte ballooning, apoptosis and cell death. These events may, ultimately, be followed by the occurrence of fibrogenic activation leading to the initiation and progression of fibrosis through complex processes involving nearby cells, mainly including LSECs, HSCs and KCs (Fig. 2), as it will be presented in this review.

Fig. 2.

The progression of NAFL to NASH and liver fibrosis is associated with a series of liver injury resulting from lipotoxicity, oxidative stress, nitric oxide, ER stress, inflammation and apoptosis that occur sequentially in different liver cells which ultimately leads to the activation of liver regeneration and fibrogenesis, augmenting collagen and extracellular matrix deposition and promoting liver fibrosis and cirrhosis. Toll-like receptor, TGFβ and hedgehog signalling are key pathways mediated hepatic injury.

3. Lipotoxicity and oxidative stress take a Toll on liver cells in the progression of NAFL to NASH (Fig. 2)

3.1. HCs: The primary target of lipotoxicity and oxidative stress in NAFLD

3.1.1. Lipotoxic lipid species

Hepatic steatosis is initiated by increased influx of lipids to the liver or decreased lipid disposal. The main sources of FA are plasma free fatty acid (FFA) from high adipose tissue lipolysis, increased de novo lipogenesis [32], dietary FAs, reduced fatty acid oxidation as well as decreased secretion of very-low-density lipoproteins (VLDL). HCs cope with the excessive lipids by enhancing synthesis of triglycerides and storage as lipid droplets, resulting in lipid accumulation and steatosis [5,33]. The ability of triglyceride synthesis to compensate for increased hepatic FA exposure appears to determine whether or not lipotoxicity results [34]. For example, studies of mouse models with NASH showed that despite reducing steatosis, the inhibition of liver triglyceride synthesis increased hepatic FFA accumulation and the severity of liver injury and fibrosis. The extent of triglyceride accumulation has been the basis for grading the severity of steatosis in NAFLD. However, triglycerides per se are not hepatotoxic [34]. It is the accumulation of some other types of lipids, including FAs, diacylglycerol, oxysterols, cholesterol, and phospholipids that cause HC injury [35]. The demonstration of lipotoxicity induced by lipids other than triglycerides led to the development of therapeutic strategies aiming to block the hepatic accumulation of lipotoxic lipids [36]. Lipotoxicity, therefore, initiates NASH development and is a new therapeutic target.

3.1.2. Mechanisms of hepatic lipotoxicity in HCs

Abnormal accumulation of lipotoxic lipids cause liver damages via the following mechanisms: (i) increased FFAs overwhelms the capacity of mitochondrial and peroxisomal fatty acid oxidation, resulting in overproduction of reactive oxygen species (ROS) that may be immediately toxic or deplete antioxidants, rendering HCs more vulnerable to other adverse factors that generate oxidative stress [37]. (ii) Accumulation of FAs within mitochondria further compromises mitochondrial respiration [35] that makes mitochondria more vulnerable to other insults, e.g., inflammatory cytokines, leading to the release of mitochondrial factors that promote apoptosis [35]. (iii) Accumulation of FAs can exacerbate insulin resistance and hyperinsulinemia [38], resulting in further hepatic lipid accumulation and promoting inflammatory [39] and fibrogenic responses. (iv) Lipotoxicity can disturb a variety of cellular signaling. For example, FAs interact with or modify other molecules, such as transcription factors (hepatocyte nuclear factor-4 alpha), Toll-like receptors (TLRs), nuclear factor-κB (NF-κB) and JNK pathways [35], leading to global changes of metabolic signaling in hepatocytes [40]. (v) In addition to FAs, other types of bioactive lipid molecules, such as oxysterols, diacylglycerol, cholesterol, and phospholipids also module metabolic signaling pathways and compromise cell viability [41]. Prolonged lipotoxicity may induce oxidative and ER stress [42], impair autophagy [43] and promote metabolic inflammation that potentiates liver cell injury and cell death.

The progression of NAFL towards NASH is characterized by increased hepatic steatosis, inflammation, and accompanied with specific histological signatures, such as hepatocellular ballooning observed in NASH and liver fibrosis development (Fig. 2) [44]. In patients with NASH, HCs become more susceptible to damage and incapable of resisting injury due to increased accumulation of lipid toxic metabolites, triglycerides, and ROS [20,44]. At the initial progression of NAFL towards NASH, HCs typically undergo cell cycle arrest and express apoptosis markers like caspases and Fas receptors [45,46]. However, ballooned HCs are resistant to apoptosis due to the activation of Hedgehog signaling that inhibits expression of caspases, but accompanied with the activation of pro-inflammatory and pro-fibrogenetic factors [45,47,48]. Thus, necrosis (uncontrolled cell death) overtakes the HCs as disease progresses.

3.1.3. Mitochondrial dysfunction and oxidative stress in HCs

Healthy hepatocytes contain remarkably high number of mitochondria. Generally, mitochondria convert ~2% of consumed O₂ into ROS and are the major site of cellular ROS formation [49,50]. With increased levels of lipids and ROS in HCs during early stages of NAFLD, mitochondria initially undergo adaptive remodeling as a protective mechanism [51,52], characterized by morphological changes through mitochondrial fission and fusion processes along with variations in energy expenditure and gene expression. However, as NAFLD progresses toward NASH, damaged HCs display maladaptive remodeling characterized by reduced ATP production, increased oxidative stress [53], enlarged mitochondria (3–10 μm in diameter) with reduced number of cristae and presence of crystalline bodies [54,55]. In patients with NAFLD, aberrant lipid metabolism enhances the damage of mitochondrial DNA and mitochondrial respiratory chain (MRC) proteins [56] as well as activates JNK bound to MRC complexes, leading to excessive ROS production [52,57]. Consistently, progression of NASH was evident by increased ROS level and reduced catalase activity which compromised detoxification and enhanced mitochondrial damage in HCs [58]. Moreover, accumulation of cholesterol further leads to a loss of reduced glutathione, an anti-oxidant defense mechanism, leading to inhibition of FA β-oxidation and exacerbation of lipotoxicity [59]. The damaged HCs further induce release of mitochondria-derived damage associated molecular patterns (DAMPs) through necrosis which, in turn, activates the NLRP3 inflammasome pathway [60] and induces metabolic inflammation in the liver.

3.1.4. Nitric oxide in HCs

Nitric Oxide (NO), the most investigated member of the reactive nitrogen species (RNS) family, mediates diverse functions in liver pathophysiology [61]. NO is synthetized enzymatically from L-arginine by three NO synthase isoforms, endothelial (eNOS), inducible (iNOS), and neuronal NOS (nNOS) [62]. iNOS is primarily expressed in HCs, but also in LSECs, HSCs and cholangiocytes [63]. NO can either play a protective role during insulin resistance-induced HCs death and liver generation [64] or partially promotes liver steatosis and NASH progression [40], indicating the importance of fine-tuning of NO metabolism. Constant excessive production of NO creates nitrosative stress inside HCs, resulting in hepatic mass damage mostly via nitrosylation of thiol residues. Proinflammatory cytokines IL-1β and TNFα induce iNOS mediated NO production in HCs has been observed during NAFLD progression [65]. Clinically, both iNOS and nitrotyrosine levels are significantly higher in NASH patients, and a major role for NO⁰ during NASH-associated fibrogenesis has been suggested. In addition, RNS is synergistic with ROS to produce more damage to HCs either through generation of ⁰OH or ⁰NO₂ radicals or peroxynitrite (ONOO⁻) radical. Each chemical reaction involves excessive generation of ${\mathrm{O}}_2^{0-}$ and ^oNO radical and contributes to the pathology seen in NAFLD and fibrosis [61].

3.2. Redox imbalance activates HSCs and contribute to fibrosis

Redox imbalance due to excessive reactive oxygen and nitrogen species (ROS and RNS, respectively) produced either as byproduct of metabolic process or from other enzymatic sources also play a major role in HSCs activation. Besides HSCs, damaged HCs, KCs, or infiltrating lymphocytes also produces ROS/RNS to stimulate HSCs activation and proliferation as well as increased production of ECM [66]. Recent studies demonstrate a strong link between immune reaction and ROS imbalance that leads to advanced fibrosis in NAFLD patients. In a mouse model of NASH [67], augmented NO metabolites in serum correlated with the severity of autoimmune hepatitis and fibrosis [68].

NO is involved in nitrosative stress in liver fibrosis which are associated with HSCs activation, including contractility, proliferation and ECM deposition. Endothelins (ET) and NO are well known modulators of HSC contractility and relaxation. NO inhibitors, N-nitro-L-arginine methyl ester (L-NAME) and endothelin-1, were shown to attenuate HSCs contractility [69]. A study in CCl₄ induce-liver fibrosis in rats further showed that induction of NO synthesis by L-arginine partially protected the liver from ECM collagen deposition, reduced several markers of liver injury and adverse hepatic remodeling. In contrast, administration of NO inhibitors L-NAME or aminoguanidine increased collagen deposition and liver damage [70], suggesting that endogenous NO protects the liver from injury, probably by decreasing the toxicity of free radicals and/or by attenuating the activity of HSCs. Moreover, administration of NO antagonists induces neutrophil infiltration in the hepatic parenchyma and exacerbates liver injury due to the increased ECM collagen accumulation, whereas increased NO protects HCs from injury by reducing the adhesion of neutrophils [69]. NO was also reported to prevent ROS mediated HSC proliferation. Together, this evidence suggests that upregulation of NO protects liver from collagen deposition and proliferation, thereby, liver fibrosis. The most likely mechanism(s) of action involves scavenging reactive free radicals and decreasing HSC activity.

3.3. Excessive NO in LSECs induced fibrosis associated capillarization

The healthy liver contains LSECs that are very fenestrated in which the holes that go through the cells are organized in sieve plates and allow solutes in the blood plasma, including chylomicron remnants and lipoproteins, to perfuse through to the space of Disse. A key feature of LSECs is that they are responsible for approximately 45% of all the endocytic activity of the liver [71,72], due to the rich profile of scavenger receptors expressed on the cell surface, primarily MR/CD206/SR-E3, SR-H1/2 (Stabilins), and FcγRIIb/CD32b as well as other minor receptors such as SR-A and SR-B [73].

Healthy LSECs secrete a constant and constitutive level of NO via eNOS in response to normal stimuli such as shear stress by the blood and vascular endothelial growth factor (VEGF) secretion. NO produced by eNOS maintains liver homeostasis and, importantly, maintains the quiescence of HSCs and KCs [74]. During early events of simple steatosis, the LSECs remain largely untouched. This is the period in which the HCs are accumulating lipid droplets and begin to suffer from lipotoxicity. LSECs do not store lipids, although they may suffer from the effects of lipotoxicity. The induction of iNOS generates an increase of NO which interacts with ROS to form ${\mathrm{N}\mathrm{O}}_2^{+}$, HNO₂, NO₂ radicals, and ONOO⁻ [75]. The first response to these radicals is an alteration of phenotype in which the fenestrae begin to disappear in a process called defenestration or capillarization presumably due to the nitrosylation of cellular components [75]. Although the precise stimuli for the induction of capillarization is not known and may be the result of multiple inputs, the onset of capillarization decreases LSEC permeability which affects lipoprotein secretion from HCs and the de novo synthesis of lipids within the HCs [76,77]. Recent investigations into how resilient LSECs are to the induction of capillarization are beginning to suggest that the pro-inflammatory phenotype of LSECs occurs while the fenestrae are preserved. Kus et al reported that, in C57BL/6 mice fed a high-fat diet (60% Kcal of fat) for 2 to 20 weeks, the sieve plates were preserved throughout the time course showing that both fenestrae diameter and porosity (measured by electron and atomic force microscopy) did not decrease relative to the mice fed on control diets [78]. These studies need to be reinforced and repeated in a larger scope to ensure that these observations are not within the narrow confines of the animal model or diet. It is also important to investigate whether this phenotype occurs in patients with NAFLD.

The progression of NAFLD to steatohepatitis is promoted by NADPH oxidase, NOX1. In human patients and in mice fed a high-fat/high cholesterol diet, NOX1 expression is increased and with concomitant higher levels of alanine aminotransferase (ALT) and cleaved caspase-3, markers of liver damage, which were attenuated in NOX1 knockout mice fed on the same diet. NOX1 upregulation also increased the levels of nitrotyrosine adducts within the sinusoids, together with ballooning HCs and contracting HSCs, resulting in the reduction of blood flow through the sinusoids which is associated with liver fibrosis development [79].

4. ER stress and insulin resistance augments HC injury

The unfolded protein response (UPR) is a signalling system emanating from the ER in response to disruption of ER homeostasis [80]. Under physiological conditions, the UPR is critical for the development and function of highly active secretory cells, including hepatocytes and cells in exocrine and endocrine glands. The endoplasmic reticulum (ER) plays a major role in proteins synthesis. An organized quality control system consisting of molecular chaperones and folding proteins ensures that the ER lumen is an optimal environment for proper protein folding and development [80]. In states of physiological stress, such as an increased demand for protein synthesis or the disruption of protein chaperones in the ER lumen (e.g., owing to pharmacological treatment), protein misfolding can occur. In mammalian cells, the UPR is governed by signaling cascades initiated by three ER-localized protein sensors: the dsRNA-dependent protein kinase (PKR)-like ER kinase (PERK); inositol-requiring enzyme 1α (IRE1α), and activating transcription factor 6 (ATF6) [81]. Diverse functions have been ascribed to the mammalian UPR, including a role in the hepatic lipogenic pathway [80].

Insulin resistance, the hallmark of type 2 diabetes, with its associated hyperlipidaemia and metabolic inflammation is a risk factor for NAFLD and the subsequent fibrosis. Ample evidence from animal models and human patients with insulin resistance (either with or without Type 2 diabetes) displays similar lipid abnormalities which include: hypertriglyceridemia, low plasma concentrations of HDL, and the appearance of small, dense LDL [82]. A key underlying pathological mechanism of this profile appears to be hepatic VLDL overproduction. The link between insulin resistance and increased VLDL secretion has been thought to be derived from increased delivery of FAs to the liver secondary to increased lipolysis within adipose tissue, increased hepatic lipogenesis, increased levels and activity of microsomal triglyceride transfer protein, and loss of insulin’s ability to direct apolipoprotein B (ApoB) toward degradation [82]. It is now clear that, in response to increased lipid delivery to hepatocytes, significant amount of ApoB is synthesized and accumulated in the ER which coupled with the accumulated lipids in HCs induces chronic metabolic imbalance and disturbs the metabolic homeostasis of ER and compromises its functionality [83]. Overproduction of ApoB in the ER of the injured HCs leads to increased production of misfolded proteins in the ER lumen, and induces ER stress. Although the hepatic UPR may be triggered to alleviate ER protein load and reduce HC damage [81]. However, this response can be overwhelmed in the long term, leading to HC cell death.

In addition, the induced ER stress can further activate inflammatory kinases JNK and NF-κB, leading to the activation of metabolic inflammation and hepatic insulin resistance [81,84]. In the insulin resistance state, hepatic de novo lipogenesis and VLDL assembly are further augmented due to the activation of SREBP-1c, SREBP-2 and their downstream genes involved in lipid synthesis. Thus, the lipid overload-induced ER stress, lipotoxicity and insulin resistance enter a vicious cycle, which eventually induce HC apoptosis and cell death [85]. Clinically, augmented ER stress is clearly observed in biopsies of patients with different stages of liver fibrosis as well as in animal models of fibrosis induced by CCl₄ or high-fat diet [86].

5. Crosstalk among different liver cell types via toll-like receptor signaling drives NAFL progression to NASH and fibrosis

5.1. Damaged HCs activates inflammation via KCs and Toll-like receptors (TLRs)

Inflammation is critically involved in the transition from NAFL to NASH in the liver. Generally, the optimal triglyceride deposition as lipid droplets in HCs is hepatoprotective as it attenuates lipotoxicity by reducing extracellular FFAs [47]. Otherwise, incorrect lipid droplet deposition or mobilization and protein expression can enhance HC lipotoxicity, leading to NAFLD progression. HCs express membrane and cytoplasmic pattern recognition receptors (PRRs), like TLR-2, -4, -5, and -9 and nucleotide-binding oligomerization domain (NOD) -1 and -2, etc., which easily sense their respective pathogens and ligands/metabolites, leading to increased local inflammation and the development of injury and insulin resistance [87]. Role of TLRs in HCs is diverse. Deficiency of TLR-2, TLR-4, or TLR-9 receptors in HCs attenuated hepatic inflammation, insulin resistance, oxidative stress and hepatic steatosis [88,89]. Conversely, deficiency of TLR-5 receptors strongly impaired bacterial clearance by the liver and aggravated NAFLD development from steatosis to NASH and fibrosis upon HFD or MCD diet challenge [90].

NASH-driven hepatic inflammation and fibrosis in HCs involve various forms of lytic death, including apoptosis, necrosis, necroptosis, pyroptosis, and ferroptosis which participated in the release of DAMPs and strong inflammatory responses [47,91]. Apoptotic HCs can directly initiate inflammation through KCs/macrophage activation after engulfment of apoptotic bodies which is further evidenced by increased levels of caspase generated keratin-8 fragments, a circulating surrogate biomarker of HC apoptosis, with NASH and fibrosis resolution [92]. Clinically, increased accumulation of senescent HCs are observed in fibrosis stage and T2D [93].

5.2. TLR signaling regulates transition KCs between M2 and M1 during NASH

5.2.1. Stimuli regulate switch of KCs from M2 to M1 state

KCs are the macrophages in the liver and were first identified by Carl Wilhelm von Kupffer in 1876 [94]. Gut-derived antigens drain through the portal vein and arrive in the liver with KCs as the first line of defense and tolerance of foreign molecules. KCs have several pattern recognition receptors such as the mannose receptor, Toll-like receptors, scavenger receptors, and NOD-like receptors that recognize DAMPs and pathogen-associated molecular patters (PAMPs).

Under normal conditions, macrophages are tolerant and trend in the M2 state which suppressess inflammation by IL-4, IL-10, and IL-13 cytokines [30,95]. In contrast, macrophages in the M1 state are often key players in chronic inflammatory diseases such as inflammatory bowel disease (IBD) [96] and atherosclerosis [97]. During the development of NAFLD, multiple factors participate in driving the transition of KCs from M2 to M1 state. These include (i) the decreased blood flow and partial occlusion of the sinusoids places a strain on the LSECs, sending them into a pro-inflammatory phenotype; (ii) HSC activation also secretes a number of pro-inflammatory factors that stimulate the activation of KCs; (iii) Intestinal leakage during NASH and a co-morbidity of obesity increases the levels of bacterial-derived lipopolysaccharides (LPS) in the blood. LPS interacts with TLR-4 resulting in increased secretion of TNFα, IL-1β and IL-6 which promote inflammation [95]; (iv) Higher circulating FFAs from adipose tissue and absorbed from dietary intake activate inflammasomes to further increase the chronic inflammatory condition of NAFLD [98]. After these stimuli increase inflammation, KCs begin to trend to the M1 phenotype from the M2 state established in NAFL which may be usher in the next phase of chronic inflammation-NASH. The balance of M1/M2 states may be a key determinant for the progression of chronic liver inflammation into fibrosis. This has been tested in an experimental setting in which blocking IL-10 with antibodies promoted the M1 phenotype. The promotion of the M2 phenotype under these conditions destroyed M1 KCs, thus ameliorating NAFLD [99].

5.2.2. Activation of TLR-9 signaling keeps KCs in M1 state

TLR-9 is a significant DAMP that is involved with many chronic liver diseases. During HC ballooning and necrosis, mitochondria and other internal organelles are exposed to KCs and other immune cells which can be released into blood circulation. Plasma from mice and patients with NASH contains high levels of mitochondrial DNA (mtDNA) and intact mitochondria which can activate TLR9 [89]. Elevated oxidized mtDNA was also detected in liver biopsies from patients with NASH [51]. Development of NASH in mice fed a high fat diet (HFD) required TLR9 on lysozyme-expressing cells which can be blocked or ameliorated by a clinically applicable TLR-9 antagonist (IRS954) when administered to mice during or before NASH development [89]. Specifically, IRS954 was administered to mice after 8 weeks of HFD administration and efficacy was assessed at week 12. The results showed a significant reduction in steatosis, inflammation, HC ballooning, ALT and the expression of IL1β, IL-6 and TNFα inflammatory cytokines. In addition, knocking out expression of TLR-9 in mouse models diminishes the NASH phenotype. Without TLR-9, KCs trended strongly to the M2 phenotype and were very resistant to M1 activation [100]. These and other lines of evidence strongly suggest that TLR-9 is a major factor in determining the M1 phenotype in KCs to perpetuate NASH going forward into fibrosis.

5.3. TLRs mediate activation of HSCs and contribute to liver fibrosis

Altered lipid metabolism in HCs results in accumulation of free cholesterol and the associated lipotoxicity which is thought to activate resident KCs and trigger secretion of inflammatory cytokines, e.g., TNFα and IL-6, to induce HSC activation [101]. It is also suggested that accumulated free cholesterol in HSCs may directly activate inflammation in the cells in a TLR-4 dependent mechanism, and promote liver fibrosis. Activated human HSCs abundantly express TLR-4 and CD14 which respond to gut bacterial endotoxin LPS and promote release of pro-inflammatory cytokines [102]. Similarly, activated mouse HSCs are enriched in TLR-2, TLR-4, and TLR-9 expression and promote expression of inflammatory cytokines and adhesion molecules, including NF-κB, JNK, TGFβ, IL-6, IL-8, CCL2 (MCP-1), MIP-2, intercellular cell adhesion molecule 1 (ICAM1), vascular cell adhesion molecule 1 (VCAM1) and E-selectin, thus, stimulate pro-fibotic gene expression in HSCs [103]. Although TLR-4 is expressed ubiquitously in liver, studies have demonstrated that TLR-4 expressed on HSCs is crucial for liver fibrosis [88]. For instance, the TLR-4 specific ligand, gut bacterial endotoxin LPS, can activate TLR-4 signaling and the subsequent release of various chemokines and adhesion molecules which attract macrophage migration into the liver and directly activates HSCs as well as the development of liver fibrosis. Resident KCs in liver can also produce TGFβ and activate TLR4 signaling in HSCs that may also promote liver fibrosis through chronic inflammation [102]. In the CCl₄ or bile duct ligation induced liver fibrosis animal models, release of DAMP molecules from damaged HCs interacted with TLR-9 and participated in HSCs activation and liver fibrosis while knockdown of TLR-9 reduced fibrosis resolution prominently [102].

In case of unresolved chronic inflammation, HSCs acquire a senescent phenotype over time characterized by permanent cell cycle arrest. Compared to the young and dynamic stage, senescent HSCs produce more proinflammatory cytokines (e.g., IL-6, IL-8, IL-11), less ECM components and more matrix metalloproteinases (MMPs) proteins, thus favoring their immune clearance and attenuating fibrosis progression [104]. The senescent HSCs can encode cytokines or ligands/molecules that potentiate natural killer (NK) cell function. For instance, senescent HSCs express NK cell receptor ligands, MICA, ULBP2 and CD58, on the cell surface which facilitate their removal by hepatic resident NK cells through immunosurveillance mechanism [104].

6. TGFβ signaling promotes liver fibrosis

6.1. TGFβ1 secreted by HCs and HSCs induces liver fibrosis

TGFβ1 belongs to the TGFβ superfamily of related proteins including growth differentiation factors (GDFs), bone morphogenetic proteins (BMPs) and activins [105]. It is a pleotropic cytokine and potent inducer of fibrosis in several organs like heart, lung, liver etc. In liver, TGFβ1 broadly regulates liver pathophysiology sequentially, encompassing an early induction of HCs apoptosis, differentiation of activated HSCs to myofibroblasts during liver fibrosis to cirrhosis and finally to hepatocellular carcinoma (HCC) [106]. Initially, the profibrogenic effect of TGFβ was recognized in HSCs activation and collagen deposition; later it was also identified in HCs and hepatic progenitor cells of liver (Fig. 2) [107].

TGFβ1 is a ligand for the TGFβ1 receptor (TGFβ1R) on the cell surface and propagates signaling through the SMAD family of proteins. The TGFβ1R is phosphorylated on serine and threonine residues and promote the activation of SMAD2 and SMAD3 kinase activity which may further bind to SMAD4 to form a SMAD2/3/4 complex [108]. The complex can then translocate to the nucleus and activate genes/pathways involved in liver fibrosis, including MAPK, PI3K, NF-κB, NOX4 and connective tissue growth factor (CTGF) pathways [109]. TGFβ1 is kept in check by BMP-7, a molecule that interacts with other receptors and signals through SMAD1/5/8 mediation. Both SMAD1/5/8 and SMAD2/3 complexes compete for SMAD4 to reach a balance in the cells. Continual insult in the liver may favor SMAD2/3 formation with SMAD4, leading to the increase of SMAD2/3/4 complex and the associated fibrogenesis. SMAD6 and SMAD7 are inhibitory in nature and can counteract SMAD2/3 activity. In fact, a recent report suggests that SMAD3 is the primary driver for the promotion of fibrosis as the knock-out of SMAD2 enhanced SMAD3 activity even though both are complexed together in the competition of SMAD4 binding [110]. SMADs are regulated by multiple mechanisms including SUMOlyation (ubiquitination), phosphorylation, and poly-ADP ribosylation which may be complex in the setting of the multi-factorial inputs of NAFLD that lead up to liver fibrosis [111].

Several years ago, with regard to the general functions of TGFβ1, Boris Hinz developed a hypothesis that describes the mechanism of the extracellular matrix and the pro-fibrotic TGFβ1 cooperating to regulate the remodeling activities of stromal cells. Matrix straining and stiffening decrease the threshold for TGFβ1 activation in the ECM by increasing the mechanical resistance to cell pulling [112]. In the liver, activation of TGFβ1 by KCs overcompensates in the injury process (e.g., HC ballooning and lipotoxicity) and scar development begins with HSC activation and the deposition of collagen in the Space of Disse [113]. Activated HSCs or myofibroblasts continually secrete TGFβ1 along with TNFα to recruit more macrophages (M1) and leukocytes to the liver in a positive feedback loop that further increases inflammation and stimulates hasty mechanisms for healing, leading to augmented collagen deposition [114,115]. Knockdown of transient receptor potential melastatin7 (TRPM7) attenuates TGFβ1-induced expression of myofibroblast markers, increases the ratio of MMPs/tissue inhibitor of metalloproteinases (TIMPs), and decrease SMAD2/SMAD3-associated collagen production [116]. Migration and proliferation of activated HSCs are hallmark of liver fibrosis. TGFβ induces HSCs migration through guanosine triphosphate (Rho GTPase) signaling activation [117] and proliferation by inducing platelet-derived growth factor receptor-beta expression and oxidative and activating Ca²⁺/calmodulindependent protein kinase II (CaMKII) [116]. Therapeutic strategies, e.g., blocking circulating TGFβ1, antagonizing its receptors, or blocking its activation at the cell surface, have been challenging due to adverse effects of systemic TGFβ antagonism, including induction of inflammation and cancer [116,118].

6.2. BMP-9 mediates signaling between HSCs and LSECs in liver regeneration

BMP-9 is a member of the TGFβ1 family that is closely involved with liver regeneration and fibrosis. This paracrine hormone is constitutively secreted by HSCs which is upregulated when HSCs are activated during liver injury or disease [119]. BMP-9 binds to activin receptor-like kinase 1 (ALK1) receptors on the surface of LSECs and transduce their signals through the SMAD system to active gene expression [120]. BMP-9 has been described as a profibrotic factor since its increased secretion during HSC activation promotes fibrosis. It has been reported that using antibodies against BMP-9 neutralizes the upgraded BMP-9 in a mouse model and attenuate liver fibrosis, suggesting that this molecule may serve as a therapeutic target of liver fibrosis[121].

Emerging evidence have demonstrated that function of BMP-9 in LSECs is associated with genetic background. For example, the most prominent strain of mice used in liver injury models is the C57BL/6 due to its propensity to overeat and develop NAFLD on HFD diets. Knocking out of BMP-9 in BALB/c mice did not result in a significant phenotype. In contrast, mouse in the 129/Ola background is more prone to vascular alterations and its LSECs are more likely to be impacted than other hepatic cell types [122]. This is due to the defectiveness of LSECs in the 129/Ola strain, especially in the male mice. LSECs from the 129/Ola BMP-9 KO mice were capillarized and had lower overall expression of terminally differentiated markers such as the Stabilin receptors, mannose receptor, Ehd3 endocytic receptor and Gata4 that regulates fenestration [123,124]. Addition of BMP-9 partially restored some but not all of these markers in LSECs isolated from 129/Ola BMP-9 KO livers [124]. Collectively, these data implicate that BMP-9 is a regulator of liver homeostasis and is required for liver regeneration. Low constitutive levels of BMP-9 are optimal for adequate response to mild insults and may prevent liver injury to some degree due to the liver’s unique regenerative properties [125,126]. However, BMP-9 is detrimental at high levels, a situation found in a setting of inflammation or steatohepatitis, and adversely affects LSEC morphology and function as observed in the 129/Ola mouse model studies [124,127].

7. Hedgehog signaling in NASH and liver fibrosis

Hedgehog (Hh) is a classic morphogen secreted by ligand-producing cells and diffuses into the extracellular space to regulate Hh-responsive target cells[128]. Mammals have three different Hh ligands: Sonic (Shh), Indian (Ihh) and Desert (Dhh) hedgehog. The Hh signalling pathway is a highly complex signalling cascade with many players and intricate regulation. However, it can be simplified into four major elements: i) the ligand Hh; ii) the receptor Patched (Patch [PTCH1]); iii) the signal transducer G protein-coupled receptor Smoothened (Smo); and iv) the effector transcription factor, Gli (GLI1) [129].

Of the 3 hedgehog ligands, Shh has the greatest impact in fatty liver development and progression. Shh regulates a plethora of biological processes in cells expressing Shh receptors (patched 1), including proliferation, differentiation and viability as well as adult liver regeneration [130]. Shh in healthy adult livers is not detectable. However, damaged HCs quickly induces expression of Shh. For example, ER stress inducers stimulate Shh expression in HCs [131]. More importantly, liver injury induced by FFA-associated lipotoxicity also enhances Shh expression [131]. In contrast, HCs with deleted caspase-2 are protected from apoptosis and do not induce Shh expression when challenged with FFA, palmitic acid [132]. Activation of Shh has been linked to the outcomes of patients with NASH [48]. Ballooning HCs produce Shh [133] which can be released either directly into the ECM or incorporated into exosomes, the small membrane vesicles that transduce stress signal to distant cells. Increased secretion of exosomes from apoptotic HCs regulate other HCs, as well as KCs, HSCs and LSECs [134]. Non-lipid particle-associated Shh binds to matrix proteins, i.e., lypican 3, to increase its concentration near the Shh producing cells in livers of patients with NASH. Shh also associates with lipoproteins which allows the ligand to initiate signalling in distant cells [135]. Activation of Shh stimulates Shh-responsive cells to produce other factors that regulate wound healing. For example, Shh signalling induces expression of TGFβ, CTGF, amphiregulin, jagged, and Wnt ligands by HSCs [136]. Shh stimulates LSECs to produce VEGF [137] and induces ductal cells to express osteopontin and chemokines that recruit various types of immune cells to the injured liver [138]. These immune cells further secret diverse cytokines, e.g., IL-1β, IL-6, TNFα and INFγ to exacerbate hepatic inflammation and liver injury. Shh also regulates macrophage polarization, thereby modulating the local balance of inflammatory, anti-inflammatory, and fibrogenic cytokines [138]. Through this process, apoptotic HCs induce a liver regenerative response to replace the dying cells [134]. Inhibition in hedgehog signalling compromises liver regeneration and replacement for HC loss [139]. However, activation of hedgehog signalling also promotes development of NASH-associated cirrhosis [140], HCC [141]. Therefore, hedgehog signalling must be tightly regulated to assure the appropriate programming for hepatic wound healing and regeneration.

Similar to Shh, Ihh is upregulated in animal models of NAFLD consuming a HFD, primarily through the inflammation component of this disease [142]. Increased Ihh secretion in the fatty liver promotes fibrosis through transdifferentiation of HSCs mediated by Myc and TGF-β2 signaling. This results in the release of Wnt5a from HSCs which may result in HCC. Once HCC develops, it is poorly differentiated and also secretes Wnt5a to act in an autocrine manner to further tumorigenesis [143]. Dhh is restricted to the nervous system and to testis and plays no known role in NAFLD development or progression [144].

8. Perspective therapeutic strategy of precision medicine for NASH and liver fibrosis treatment

The current therapeutic strategy for patients diagnosed with mild forms of NALFD is diet and life style modification, such as body weight control and adequate physical activity. With more severe forms of NASH, there are also other disease modalities to tackle. In diabetes associated NAFLD, insulin sensitizers, including metformin, thiazolidinediones, glucagon-like peptide-1 receptor agonists, and the dipeptidyl peptidase IV inhibitors, have been used in NASH management alongside with dietary control and behavior changes which alleviate the progression of NASH in most cases [145].

For several decades, physicians and researchers have been developing drug carriers that will increase the drug delivery to liver for targeting NASH and liver fibrosis and decrease the side effects of drug metabolism. Nanoparticle-based drug delivery approach has been shown to be the promising remedies for this purpose. Of all the classes of nanoparticles available to treat fatty liver and other hepatic diseases, liposome (LNP) based formulations are the most successful due to their high stability, low toxicity and scale-up production methods [146]. The advantage of liposomes is that they have the ability to encapsulate both hydrophobic and hydrophilic drugs and the hydrophilic membrane shell may be modified by chemical moieties for aiming the liposome to target a specific liver cell-type (Fig. 3) [147]. The lipid composition of the LNP is of particular importance as it determines the half-life in circulation, including how the LNP interacts with targeted cells and how the LNP escapes the endosomal pathway to deliver the cargo into cellular cytoplasm [148,149]. For example, SB431542 is a TGFβ inhibitor with low water solubility and poor bioavailability as a stand-alone drug. When encapsulated by an oleate-based liposome, it had relatively good uptake and efficacy in liver with few side effects in other organs when used in an experimental CCl₄-induced liver fibrosis mouse model [150]. Specific HC targeting may be accomplished through the asialoglycoprotein receptor (ASGPR) which is only expressed on HCs that recognize galactose or N-acetylgalactosamine ligands with high affinity [151,152]. LSECs may be specifically targeted with hyaluronic acid (HA)-based liposomes as the HARE/Stabilin-2 receptor specifically recognizes HA [153,154]. HSCs may be targeted using liposomes with PDGFR-β-targeting antibodies for specific binding to HSC cell surfaces [155]. KCs and other macrophages may be targeted with liposomes containing phosphatidylserine (PS) lipids (Fig. 3). Since both KCs and LSECs express Stabilin-1 which will bind with PS, and even Stabilin-2 expressed by LSECs will bind with PS [156], PS may not be a very good molecule to exclusively target KCs. However, if the liposome is greater than 200 nm in diameter, it will favor KC uptake over LSEC internalization through phagocytosis [157]. More importantly, even though liposomes of many different types tend to accumulate in the liver, targeting specific liver cell type will increase efficacy of the pharmaceutical cargos and decrease off-target effects, and therefore reduce side effects.

Fig. 3.

Liposomes have the ability to encapsulate hydrophobic and hydrophilic drugs. The membrane shell may be modified by chemical moieties for aiming the liposome to target a specific liver cell-type. HC targeting may be accomplished through the asialoglycoprotein receptor HSCs may be targeted using liposomes with PDGFR-b-targeting antibodies for specific binding to HSC cell surfaces. LSECs may be specifically targeted with hyaluronic acid-based liposomes and KCs and other macrophages may be targeted with liposomes containing phosphatidylserine lipids.

Currently, no nanoparticle or liposomal formulation has been approved for the treatment of NALFD and liver fibrosis. However, there are some nanoparticle formulations in clinical trials. For example, a natural mutation in patatin-like phospholipase domain-containing 3 (PNPLA3) protein at I148M reduces it’s enzymatically activity in remodeling lipid droplets during normal lipid droplet catabolism [158]. An antisense oligonucleotide (ASO) targeting the mRNA of PNPLA3-I148M improved overall NASH outcomes in a mouse model and this ASO is currently in phase I human trials in people with this same mutation in PNPLA3 (Ionis & AstraZeneca) [159]. Similarly, targeting the 17-β hydroxysteroid dehydrogenase 13 (HSD17B13) using RNAi silencing (Alnylam and Regeneron) has being developed for the treatment of NASH. Although HSD17B13 is not a very well-characterized enzyme, it is known to be localized on lipid droplets similar to PNPLA3 and a number of polymorphisms have been linked to human NASH [160,161]. Both ASO and RNAi-based strategies use liposomal encapsulation as well as naked nucleic acid strategies targeting galactose to the ASGP receptor for efficient uptake in HCs are also under development [162,163].

9. Concluding remarks and future perspectives

The fundamental metabolic perturbations that trigger NAFL and its progression to NASH and liver fibrosis in diabetic subjects are insulin resistance which is centrally involved in the pathogenesis of lipotoxic induced tissue injury, a pathological condition in which accumulation of harmful lipids (e.g, FFAs) in HCs perturbs cellular organelle function, causing a series of liver damage by oxidative stress, excessive nitric oxide, ER stress, inflammatory cytokines and apoptosis that occur sequentially in the neighbouring parenchymal cells (HCs) and non-parenchymal cells (e.g., HSCs, KCs and LSECs), which induce a hepatic regeneration and wound-healing response to replace dead HCs. The repair process is intricate and involves activation of multiple signalling pathways, including redox signalling, oxidative stress, neurohumoral signaling, pro-inflammatory cascades, growth factors, senescence and programming cell death, which are deleterious when deregulated.

Deregulated wound-healing responses worsen clinical outcomes by promoting development of fibrosis, cirrhosis and liver cancer.

Currently, there is no approved drugs for NASH and liver fibrosis treatment. NASH is a heterogeneous disease with multiple adverse factors contributing to the disseise development. Ideal drugs for both NASH and fibrosis should be able to improve the liver microenvironment and tissue histology as well as its complications such as cardiovascular disease. Reversal of NASH without worsening of fibrosis has been a primary focus of the novel therapeutics in clinical trials. Unfortunately, the outcomes from clinical trials showed no treatment was able to significantly improve histological status of NASH and the resolution of fibrosis. Nevertheless, novel therapeutic strategy for NASH and fibrosis has been under developing which takes into consideration of insulin sensitization, anti-inflammatory, antifibrotic and drug delivery approaches. Future research should focus on combination therapies should shed light on the prevention and treatment of NASH and liver fibrosis.

Abbreviations:

ALK1

Activin receptor-like kinase 1

ALT

Alanine aminotransferase

ApoB

apolipoprotein B

ASGPR

Asialoglycoprotein receptor

ASO

Antisense oligonucleotide

ATP

Adenosine triphosphate

BMPs

Bone morphogenetic proteins

CaMKII

Ca²⁺/calmodulindependent protein kinase I

CCl₄

Carbon tetrachloride

CTGF

Connective tissue growth factor

DAMPs

Damage associated molecular patterns

ECM

extracellular matrix

EMA

European Medicines Agency

ER

Endoplasmic Reticulum

ET-1

Endothelin-1

FDA

Food and Drug Administration

FA

Fatty acid

FFAs

Free fatty acids

GDFs

Growth differentiation factors

HA

Hyaluronic acid

HCC

Hepatocellular carcinoma

HCs

Hepatocytes

HFD

High fat diet

Hh

Hedgehog

HNF-4α

Hepatocyte nuclear factor-4 alpha

HPSc

Hepatic progenitor cells

HSCs

Hepatic stellate cells

HSD17B13

17-β hydroxysteroid dehydrogenase 13

HIF-1α

Hypoxia-inducible factor-1alpha

ICAM1

Intercellular cell adhesion molecule 1

IL-1β

Interleukin-1beta

IL-6

Interleukins-6

INFγ

Interferon gamma

JNK

c-Jun N-terminal kinase

KCs

Kupffer cells

L-NAME

N-nitro-L-arginine methyl ester

LNP

Liposome based nanoparticles

LSECs

Liver sinusoidal endothelial cells

MAFLD

Metabolic associated fatty liver disease

MAPK

Mitogen activated protein kinase

MCD

Methionine-choline-deficient

MCP-1

Monocyte chemoattractant protein-1

MICA

Major Histocompatibility Complex (MHC) Class I Chain-Related gene A

MIP-2

macrophage inflammatory protein-2

MMPs

Matrix metalloproteinases

MRC

mitochondrial respiratory chain

NAFLD

Non-alcoholic fatty liver disease

NASH

Non-alcoholic steatohepatitis

NF-κB

Nuclear factor-kappaB

NK

Natural killer

NLRP3

NACHT, LRR and PYD domains-containing protein 3

NO

Nitric oxide

NOD

Nucleotide-binding oligomerization domain

NOS

Nitric oxide synthase

NOX1

NADPH oxidase1

PAMPs

Pathogen-associated molecular patterns

PDGFR-β

Platelet derived growth factor receptor-beta

PI3K

Phosphatidyl inositol-3 kinase

PNPLA3

Patatin-like phospholipase domain-containing 3

PS

Phosphatidylserine

RNS

Reactive nitrogen species

ROS

Reactive oxygen species

Shh

Sonic hedgehog

Smo

Smoothened

SREBP-1c

Sterol regulatory element binding protein-1c

T2D

Type 2 diabetes

TGF-β

Transforming growth factor-beta

TIMPs

Tissue inhibitor of metalloproteinases

TLRs

Toll-like receptors

TNFα

Tumor necrosis factor – alpha

TRPM7

Transient receptor potential melastatin7

ULBP2

UL16 binding protein 2

UPR

Unfolded protein response

VCAM1

Vascular cell adhesion molecule 1

VEGF

Vascular endothelial growth factor

VLDL

Very low density lipoprotein

Wnt/β-catenin/Wnt signalling

Wingless and Int-1 signalling