Cellular Protein Markers, Therapeutics, and Drug Delivery Strategies in the Treatment of Diabetes-Associated Liver Fibrosis

Abstract

Liver fibrosis is the excessive accumulation of extracellular matrix due to chronic injuries, such as viral infection, alcohol abuse, high-fat diet, and toxins. Liver fibrosis is reversible before it progresses to cirrhosis and hepatocellular carcinoma. Type 2 diabetes significantly increases the risk of developing various complications including liver diseases. Abundant evidence suggests that type 2 diabetes and liver diseases are bidirectionally associated. Patients with type 2 diabetes experience more severe symptoms and accelerated progression of live diseases. Obesity and insulin resistance resulting from hyperlipidemia and hyperglycemia are regarded as the two major risk factors that link type 2 diabetes and liver fibrosis. This review summarizes possible mechanisms of the association between type 2 diabetes and liver fibrosis. The cellular protein markers that can be used for diagnosis and therapy of type 2 diabetes-associated liver fibrosis are discussed. We also highlight the potential therapeutic agents and their delivery systems that have been investigated for type 2 diabetes-associated liver fibrosis.

1. Introduction

Chronic liver disease is a global health burden that affects 844 million people and accounts for approximately 2 million deaths per year, 1 million due to cirrhosis. The global mortality attributed to liver diseases is approximately 3.5% of all deaths worldwide [1, 2]. There is a high prevalence of liver diseases caused by various risk factors, such as obesity, lifestyle, type 2 diabetes, age, alcohol abuse, and chronic viral hepatitis. Nonalcoholic fatty liver disease (NAFLD) and alcoholic liver diseases are two very common types of chronic liver diseases [3]. Although liver fibrosis is the early stage of chronic liver disease, if left untreated, it will turn into irreversible cirrhosis and hepatocellular carcinoma (HCC), thereby leading to a high death rate [1–4].

More than 400 million and 600 million patients are diagnosed with diabetes and obesity, respectively, around the world, and they are regarded as two of the most prominent risk factors associated with liver fibrosis [2]. Mounting evidence suggests that type 2 diabetes and liver diseases are bidirectionally associated [5]. In addition, type 2 diabetes is also a risk factor for other diseases and complications, such as cardiovascular diseases, renal diseases, and neuropathy [6]. Biochemical parameters, such as hemoglobin A1C (HbA1C), blood pressure, total cholesterol (low-density lipoprotein cholesterol and high-density lipoprotein cholesterol), triglycerides, body mass index (BMI), along with obesity, are biomarkers reflecting a patient’s diet and lifestyle. These biomarkers have demonstrated a profound connection between type 2 diabetes and the development of liver diseases [5, 7]. Overall, 70% of patients who have been diagnosed with chronic liver disease also have complications of type 2 diabetes [8]. Patients with type 2 diabetes may experience more severe symptoms and accelerated progression of liver diseases, such as hepatitis, cirrhosis, HCC, and fulminant hepatic failure, due to their abnormal glucose homeostasis [9]. The risk factors associated with liver fibrosis and type 2 diabetes are illustrated in Figure 1.

Figure 1: Risk factors in diabetes and liver fibrosis.

(A) The common risk factors that affect the association of type 2 diabetes with liver fibrosis. (B) The synergistic effect of type 2 diabetes with other liver fibrosis risk factors.

The liver, the largest organ in the body, is comprised of 60% hepatocytes (parenchymal cells) and about 35% sinusoidal cells (non-parenchymal cells), which can be further divided into liver sinusoidal endothelial cells (44%), hepatic stellate cells (10–25%), hepatic natural killer cells (5%), and Kupffer cells (33%) [10]. The functions of the liver are to produce and secrete bile; regulate the metabolism of carbohydrates, proteins, cholesterol, and lipids; maintain glucose and hormone homeostasis; store vitamins; detoxify toxins; and produce certain immune factors [11, 12].

Liver fibrosis is the excessive accumulation of extracellular matrix (ECM) due to chronic injuries, such as viral infection, alcohol abuse, high-fat diet, and toxins [2, 13]. Liver fibrosis is reversible before it progresses to cirrhosis and HCC [14]. However, to date, there is no standard treatment for liver fibrosis. Once the liver becomes cirrhotic, transplantation is the only treatment, but it is not available or affordable for most patients.

Obesity is currently the most common health issue globally, and approximately 2 billion people are obese or overweight [15]. Obesity is associated with the development of various metabolic complications, especially NAFLD and type 2 diabetes [16]. Insulin, glucose, fatty acyl coenzyme A (CoA), and malonyl-CoA, which are involved in lipid synthesis, oxidation, de novo lipogenesis, triglyceride breakdown, and lipoprotein uptake, maintain lipid homeostasis in the liver [17, 18]. Energy homeostasis is disrupted when excessive fat accumulation in the adipose tissue, liver, and muscle initiates the inflammatory signaling pathway, causing lipotoxic effects and affecting insulin resistance directly or indirectly [19, 20]. Some evidence suggests that obesity-induced inflammation increases the M1/M2 macrophage ratio, leading to the upregulation of pro-inflammatory cytokines, such as tumor necrosis factor alpha (TNFα) and interleukin 6 (IL-6). Obesity also activates the nuclear factor-κB (NF-κB) and c-jun N-terminal kinase (JNK) pathways, which inhibit the insulin receptor by triggering serine/threonine phosphorylation of insulin-receptor substrate (IRS) proteins. Moreover, IL-10 is an anti-inflammatory cytokine that inhibits the TNFα-induced insulin resistance and antagonizes the NF‑κB activation. Obesity-induced inflammation alters the anti-inflammatory M2 polarization state to pro-inflammatory M1 polarization state, which downregulates the generation of the high level of anti-inflammatory cytokines, such as IL-10 and IL4, to maintain insulin sensitivity. [20–22].

Diabetes mellitus is a type of metabolic disorder in which patients suffer from the hyperglycemic condition. In general, diabetic patients are classified into two types based on the pathogenesis and clinical characteristics. Type 1 diabetes, also known as insulin-dependent diabetes mellitus, is a kind of autoimmune disease resulted from autoantibodies against the islet beta cells of the pancreas. The islet beta cells’ destruction will lead to the deficiency of insulin secretion and chronic hyperglycemia. Type 2 diabetes results from both impaired insulin secretion and decreased insulin sensitivity [23–25]. Cusi et al. reported that the prevalence of NAFLD was low in type 1 diabetic patients (8.8%) but high in type 2 diabetic patients (61.7%−75.6%). Type 2 diabetic patients with NAFLD was highly associated with the markers of insulin resistance, such as higher levels of triglycerides, ALT, and free fatty acid. However, these markers did not show significant difference between type 1 diabetic patients with or without NAFLD [26].

Type 2 diabetes significantly increases the risk of developing various complications because of high blood glucose levels. Strategies for treating type 2 diabetes include reducing sugar intake, increasing the sensitivity of insulin receptors, and supplementing insulin [27]. Hyperlipidemia and hyperglycemia induce islet beta cells to secrete excess insulin, leading to hyperinsulinemia and insulin resistance, which are the reasons why the peripheral tissues cannot utilize glucose, and this condition progresses to type 2 diabetes [28, 29]. Insulin plays a critical role in glucose homeostasis by enhancing adipose and muscle cells to take up glucose, and it inhibits liver glycogenesis and gluconeogenesis [28]. To maintain glucose homeostasis, insulin signaling activates the Akt protein kinase to regulate the glucose transporters (GLUT2 and GLUT4) and hepatic lipogenesis. However, obesity systemically attenuates the sensitivity of the insulin response by activating signaling pathways, such as protein kinase C (PKC), JNK, and IKKβ, to inhibit the phosphorylation of the IRS protein, thereby leading to insulin resistance [28]. In addition, many studies have reported that high-fat diet activates hepatocytes to overexpress forkhead box protein O1 (FOXO1), a transcription factor that is involved in gluconeogenesis and is suppressed by insulin under normal diet conditions. The hypothesis suggested that hyperglycemia and hyperinsulinemia induced by high-fat diets disrupt insulin to inhibit adipocytes lipolysis, thus increasing the production of glycerol and free fatty acids. High-fat diets increase glucose output from the liver and decreases the response of GLUT4 to insulin [28, 30]. Furthermore, insulin and glucose regulate lipid synthesis as well as hepatic de novo lipogenesis via sterol regulatory element binding-protein 1c (SREBP 1c) and carbohydrate response element-binding protein (ChREBP), respectively [31, 32]. Numerous animal studies have demonstrated that insulin resistance leads to overexpression of hepatic SREBP 1c and ChREBP, which induces hepatic steatosis [32].

In this review, we summarize possible mechanisms of the association between type 2 diabetes and liver fibrosis. We discuss the cellular protein markers that can be used for diagnosis and therapy of type 2 diabetes-associated liver fibrosis. We also highlight the potential therapeutic agents and their delivery systems that have been investigated for type 2 diabetes-associated liver fibrosis.

2. Type 2 Diabetes-Associated Liver Diseases

2.1. Type 2 Diabetes-Associated Liver Fibrosis

Liver fibrosis and type 2 diabetes are associated with common risks, such as obesity, inflammation, and insulin resistance. The exact mechanism of diabetes-associated liver fibrosis is still not full elucidated. However, many animal studies and clinical trials have provided evidence that obesity and insulin resistance are the factors that link these two diseases. Most patients diagnosed with type 2 diabetes and liver fibrosis exhibit high levels of blood glucose and free fatty acids, which increase the risk of inflammation and liver enzyme overexpression, thus causing liver injury [33]. Liver fibrosis is a repair mechanism that occurs when the liver is damaged, and chronic inflammation results in the overexpression and deposition of ECM to compensate for the injured cells (Figure 2) [13]. Overconsumption of alcohol, hepatic viral infection, nonalcoholic steatohepatitis, overloaded metal ions, drug induction, and toxins are the leading causes of liver fibrosis [34]. These risk factors induce the hepatic microenvironment to suffer from oxidative stress (e.g., H₂O₂, O₂·⁻, or nitric oxide), excess leptin and TNF-α, and adiponectin deficiency, resulting in activation of Kupffer cells and subsequent damage to the hepatocytes. Activated Kupffer cells release cytokines, such as transforming growth factor-β (TGF-β), platelet-derived growth factor (PDGF), TNF-α, reactive oxygen species (ROS), and interleukins (IL-1 and IL-6), to trigger quiescent HSCs to transform into proliferative myofibroblast-like cells [34–36]. Activated HSCs play a key role in the overproduction and deposition of ECM during liver fibrogenesis [37]. Hyperglycemia and hyperinsulinemia can activate the IRS1-PI3K-Akt pathway to stimulate the proliferation of HSCs and increase the production of fibrotic precursors and matrix [38]. High glucose concentration stimulates ROS and activates the phosphorylation of mitogen-activated protein kinase (MAPK) in HSCs [39]. Insulin induces the activation of HSCs via a PI3K- and ERK-dependent mechanism, while insulin-like growth factor 1 (IGF-1) activates HSCs through a PI3K-dependent but ERK-independent mechanism [40].

Figure 2: Pathophysiology and possible cellular protein biomarkers of liver fibrosis.

Serum proteins such as ALT, AST, and ALP serve as liver function biomarkers. Cytokines released during liver injury are biomarkers for monitoring liver fibrogenesis. Various collagens along with their pro-peptides, glycoproteins, and hyaluronic acid are linked to ECM formation and can be exploited as biomarkers of liver fibrosis. Proteins linked to ECM degradation such as MMPs and TIMPs are also indicative of liver fibrosis.

2.2. Type 2 Diabetes and Various Etiologies of Liver Diseases

The interactions between type 2 diabetes and chronic liver diseases are complicated. Many reports have shown that type 2 diabetes could increase the risk of liver fibrosis when patients are diagnosed with any type of chronic liver disease. Type 2 diabetic patients’ dysregulated liver metabolism and suppressed immune system may cause more severe symptoms than seen in patients without type 2 diabetes. For example, NAFLD is strongly associated with type 2 diabetes, obesity, insulin resistance, and other metabolic syndromes [6, 41]. As estimated, 10–25% of NAFLD patients will progress to nonalcoholic steatohepatitis (NASH), and 10–15% of NASH patients will continue to develop into HCC [42]. The prevalence of NAFLD in diabetic patients is approximately 70%, and it is particularly high in type 2 diabetes [26]. Type 2 diabetes increases the risk of severe NAFLD and promotes NAFLD progression, which is associated with cardiovascular and kidney complications.

Metabolic disorders, such as excessive lipid accumulation, hyperglycemia, and hyperinsulinemia, interfere with the secretion of “hepatokines” and lead to hepatocyte injury, inflammation, and progression of fibrosis through activating the profibrogenic pathway [41]. Hepatic insulin resistance, peripheral insulin resistance, free fatty acids, and high blood glucose level can affect the liver, adipose tissue, and muscle to promote NAFLD progression by dysregulating gluconeogenesis, de novo lipogenesis, and glycogenolysis. Moreover, some studies have suggested that type 2 diabetes is a predictor of liver fibrosis for NAFLD [43]. The family with sequence similarity 3 (FAM3) gene family members (FAM3A, FAM3B, FAM3C, and FAM3D) are involved in glucose and lipid metabolism. They are considered important regulators of NAFLD and type 2 diabetes. Obesity and insulin resistance lead to the imbalance of the FAM3 signaling networks, which contributes to the development of NFALD and type 2 diabetes. [44].

Patients with alcoholic liver disease have a high association with type 2 diabetes [45]. Alcohol consumption increases the risk of type 2 diabetes [46]. For example, heavy alcohol ingestion of more than 270 g/week showed a 2-fold increased risk than those intaking less than 120 g/week [47]. While excessive alcohol ingestion decreases insulin-mediated glucose uptake, chronic alcohol ingestion damages the islet beta cells and leads to pancreatic injury [45].

Hepatitis B virus (HBV) and hepatitis C virus (HCV) are the most common viruses that induce chronic inflammation and colonize both hepatic and extrahepatic sites, such as the kidney and pancreas, and result in other manifestations, such as type 2 diabetes [48, 49]. Patients infected with HBV or HCV are considered to have a higher risk of developing HCC than other liver diseases. Clinical trials have demonstrated that hepatic steatosis is positively associated with the combination of HCV and metabolic factors. HCV infected patients with metabolic derangements showed higher risk of exacerbate steatosis, liver fibrosis, chronic metabolic diseases, and HCC. However, HBV shows less combination effect on increasing the risk of HCC development in patients with type 2 diabetes [50–53]. Despite the unclear interaction between the two diseases, the possible mechanism of HCV infection-induced insulin resistance is related to virus-infected cells, which suppress glucose uptake by hepatocytes and overexpress TNF-α to mediate the glucose transporters. Moreover, the HCV core-induced suppressor of cytokine signaling 3 (SOCS3) elevates the degradation of IRS, and inhibits the PI3K/Akt pathway, thereby inducing the insulin resistance [50, 54]. Alpha-fetoprotein (AFP) is associated with hepatic cancer and liver regeneration. Chronic HCV infection along with hyperinsulinemia, hyperglycemia, insulin resistance, and liver fibrosis promotes AFP production and decreases the therapeutic efficiency of interferon [55]. Therefore, proper control of serum glucose levels in combination with HCV treatment can inhibit the development of type 2 diabetes, cirrhosis, and HCC simultaneously [56, 57].

The presence of diabetes and glucose intolerance in patients with cirrhosis is up to 96% [45]. Patients with cirrhosis and complication of type 2 diabetes are known as “hepatogenous diabetes”, which have normal fasting serum glucose levels but impaired glucose metabolism [58, 59]. Cirrhosis contributes to the development of portal hypertension, reduced insulin clearance, and desensitization of islet beta cells, leading to peripheral hyperinsulinemia [45, 58]. Both type 2 diabetes and hepatogenous diabetes increase the risk of morbidity and mortality of liver cirrhosis [58, 59].

Type 2 diabetes is considered as an emerging risk factor for developing various types of cancer [60]. A meta-analysis of existing literatures found a strong association between diabetes and HCC. Epidemiologic studies showed that type 2 diabetes could increase the risk of HCC by 2.5-fold compared to individuals without type 2 diabetes [61, 62]. NAFLD is also one of the major risk factors for HCC, and its progression is strongly linked to obesity and type 2 diabetes [63, 64].

Various cytokines and hormones are involved in mediating metabolic homeostasis. Leptin, as an appetite suppressive signal, controls energy intake and expenditure via insulin regulation and regulate various signaling pathways, such as JAK/STAT, PI3K/AKT, and MAPK, which are involved in fibrogenesis and the progression of HCC [64, 65]. Adiponectin, produced by adipocytes, has been proven to decelerate the progression of hepatocarcinogenesis by promoting the apoptosis of HCC cells via regulating caspase-3 activation, JNK, AMPK, tumor suppressor TSC1 and mTOR activities [64, 66]. Hyperglycemia is regarded as a potential carcinogen related to the stimulation of pro-inflammatory cytokine release and high serum iron levels, regulated by the glycosylation of hemoglobin, and it synergistically interacts with other risk factors, such as HCV and oxidative stress induced by alcohol [62, 67].

Fujii et al. developed a novel NASH-HCC mouse model by giving low doses of streptozotocin and feeding high-fat diet simultaneously. Streptozotocin induces islet injury, while high-fat diet feeding stimulates oxidative stress and inflammation, leading to the activation of HSCs and NASH-based liver fibrosis. Macrophages activate fibroblasts within Disse’ space in the liver, which accelerates fibrogenesis by upregulation of inflammatory cytokines, such as TNF-α, IL-6, and MCP-1/CCL2. The results showed that NASH-based liver fibrosis is essential for diabetic patients to develop HCC [68]. In addition, the pathologic changes of insulin resistance, obesity, and NAFLD play crucial roles in the dysregulation of cellular proliferation/regeneration/apoptosis via activation of the NF-κB/JNK1 pathway, along with excess stimulation of IRS-1, which can further promote the development of HCC [64, 69, 70]. Moreover, Xia et al. suggested that dysregulation of metabolites, such as methyl glucopyranoside (α and β), 1,5-anhydroglucitol, 2-hydroxystearate, and pyroglutamine might be related to HCC development at high serum glucose levels [71].

3. Cellular protein markers

3.1. Biomarkers of liver function

Serum markers such as alanine aminotransferases (ALT), aspartate aminotransferases (AST), and alkaline phosphatase (ALP) are indicative of liver injury, and they are routinely used to access the liver function and damage (Figure 2). Liver fibrosis is the outcome of chronic liver injury, and these serum markers could indicate the degree of hepatocellular damage. Serum ALT and AST levels have been widely used by clinicians to estimate hepatocyte integrity [72], and elevated ALT levels are positively correlated with fibrosis [73]. Similarly, abnormalities in the liver caused by insulin resistance such as lipolysis and altered triglyceride storage also leads to hepatocyte injury. Elevated levels of ALT and AST were found to be positively associated with the incidence of type 2 diabetes [74]. Likewise, ALT and AST levels were found to be significantly higher in type 2 diabetic patients with NAFLD as compared to type 2 diabetic patients without NAFLD [75]. However, several studies have shown that ALT is not a reliable marker of fibrosis because many patients with advanced stages of fibrosis may still have normal levels of ALT [76]. Like ALT, elevated AST and ALP may also suggest liver damage. In the advanced stages of liver fibrosis, AST and ALP levels are correlated with the progression of fibrosis [77, 78]. An increase in the levels of liver function biomarkers may indicate the presence of liver fibrosis or identify patients at higher risk for developing type 2 diabetes.

3.2. Biomarkers linked to extracellular matrix formation

Collagens are the primary connective tissues that replace normal liver cells during wound healing. These collagens, along with the pro-peptides of collagens are being extensively studied to see if they can be exploited as reliable biomarkers for liver fibrosis (Figure 2). Among them, type I collagen was found to be significantly elevated in fibrotic liver. In an experimental rat liver fibrosis model induced by bile duct ligation, the type I collagen turnover product, N-terminal pro-peptide of collagen type I (PINP) was found to be correlated with the progression of liver fibrosis [79].

Type III collagen is also found in abundance in fibrotic liver, and the two common markers for the detection of type III collagen are N-terminal pro-peptide of collagen type III (PIIINP) and PRO-C3. PIIINP levels were found to be elevated in alcoholic liver fibrosis, and the serum levels of PIIINP was associated with the different fibrosis stages [80]. Similarly, serum levels of PIIINP were found to correlate with the stages of liver fibrosis in children with NAFLD [81]. The problem with PIIINP is that it is not specific to liver disease. Serum PIIINP levels have been found to be increased in other diseases, such as pancreatitis and rheumatoid arthritis [82, 83]. Another study found that PIIINP levels were not significantly different in type 2 diabetic patients with and without NAFLD [84]. PRO-C3 is another biomarker that detects the formation of type III collagen. Several studies have validated the use of PRO-C3 in the detection of liver fibrosis [85–87]. A recent study carried out in patients suffering from type 2 diabetes showed that PRO-C3 can be used to detect liver fibrosis and distinguish the different stages of fibrosis. [88].

The 7S domain of type IV collagen (P4NP 7S) is a potential biomarker that can distinguish mild fibrosis from advanced fibrosis. In a study conducted in patients with NASH, serum levels of P4NP 7S were found to be markedly elevated in patients with advanced fibrosis as compared to mild fibrosis [89]. Another study demonstrated that P4NP 7S levels can serve as a good diagnostic marker to distinguish advanced fibrosis in NAFLD patients with type 2 diabetes as compared to NAFLD subjects without diabetes [90]. Serum levels of Chitinase 3-like 1 glycoprotein (YKL-40) were also found to be elevated in liver fibrosis. A study by Kumagai et al. reported elevated YKL-40 serum levels in NAFLD patients with liver fibrosis as well as in patients with HCV infection [89]. Similarly, another study by Yan et al. demonstrated that serum YKL-40 levels correlated with the different stages of fibrosis in patients with hepatitis B infection [91]. Similarly, plasma YKL-40 level has been known to be positively correlated with insulin resistance, and elevated levels of plasma YKL-40 were also observed in type 2 diabetic patients [92]. Serum YKL-40 level is, therefore, a useful noninvasive marker. Laminin, a noncollagenous glycoprotein is another biomarker that is synthesized in the liver. Serum laminin levels could distinguish between fibrotic and nonfibrotic cases in patients with NAFLD [93]. The laminin plasma levels in conjunction with hyaluronic acid levels was exploited as a marker for predicting the stages and severity of liver fibrosis [94].

Microfibrillar-associated protein 4 (MFAP4) is a component of the ECM that plays a role in tissue remodeling. Serum levels of MFAP4 are upregulated and correlated with the severity of liver fibrosis in patients with alcoholic liver disease [95]. In a cohort of patients suffering from hepatitis C, plasma levels of MFAP4 were used to distinguish between mild/moderate and severe liver fibrosis [96]. Hyaluronic acid (HA), a glycosaminoglycan polymer which plays an important role in ECM formation, is another widely studied biomarker for the identification of liver diseases. In the liver, HA is synthesized by HSCs, whereas sinusoidal endothelial cells participate in its degradation [97]. During liver injury, serum levels of HA have been found to be elevated. Serum HA can be used to differentiate different stages of liver fibrosis in patients with hepatitis C. Additionally, serum hyaluronic acid concentrations are indicators of liver fibrosis in NAFLD, alcoholic liver disease, and primary biliary cirrhosis, but cannot differentiate the different stages of liver fibrosis [98]. Elevated HA levels in the serum and muscle were also detected in humans with type 2 diabetes. Increased levels of HA were found to be associated with the inflammatory aspects of type 2 diabetes,[99] which supports its use as a biomarker in type 2 diabetes.

3.3. Biomarkers linked to extracellular matrix degradation

Matrix metalloproteinases (MMPs), along with their inhibitory proteins, tissue inhibitors of metalloproteinases (TIMPs), regulate the activity of the ECM in a healthy liver. MMPs have been found to increase in liver diseases. One study found that serum MMP1 levels were not correlated with the fibrosis stages in patients with NASH but were elevated in the initial fibrosis stages, suggesting that MMP1 could be used as a potential biomarker for monitoring the progression of NASH [100]. Another study performed in patients with stage C liver cirrhosis found serum levels of MMP2, MMP8, and MMP9 to be elevated as compared to healthy controls [101]. Likewise, plasma levels of MMP2 were also found to be associated with the different stages of liver fibrosis in chronic hepatitis B patients with cirrhosis [102]. A study by Anna Kerola and colleagues demonstrated that MMP7 could serve as a serum biomarker in liver fibrosis in patients with biliary atresia. An increase in MMP7 gene expression and protein levels in the liver correlated with the progression of liver fibrosis. Similarly, there was a positive correlation between MMP7 gene and protein expression with serum MMP7 levels [103]. Plasma levels of MMP2 and MMP9 are also found to be elevated in type 2 diabetic patients compared to healthy patients, which suggests abnormal activity in the ECM metabolism. [104, 105].

Plasma levels of TIMP1 have also been found to be elevated during liver fibrosis. A study performed in biopsy-confirmed NAFLD patients found that serum levels of TIMP1 were significantly higher in advanced stages of fibrosis and could serve as a suitable marker to differentiate NAFLD patients with and without liver fibrosis [106]. Similarly, serum TIMP1 levels were also found to be elevated in chronic hepatitis B patients with significant liver fibrosis [107]. In type 2 diabetic patients, plasma levels of TIMP1 and TIMP2 are significantly elevated compared to non-diabetic patients [104, 108]. Even though MMPs and TIMPs levels are found to be altered in a wide range of diseases, they are indicative of some form of liver injury and can be exploited as potential biomarkers.

3.4. Cytokines as biomarkers

Cytokine production in a healthy liver is absent or minimal. By contrast, cytokine production is increased in various liver diseases, including liver fibrosis, making them suitable candidates as biomarkers [109]. A clinical study in patients with chronic hepatitis B, which was carried out to study the diagnostic capabilities of various cytokines, reported that IL-2R and TGF-α were independent predictors of liver fibrosis. Similarly, CXCL-10 and IL-8 were also correlated with liver fibrosis [110]. Another study carried out in patients with NAFLD reported that higher levels of IL-8 were associated with significant liver fibrosis [111]. IL-6 is another candidate that may have prognostic capabilities in the detection of liver fibrosis. In a study performed in an HIV-infected population with alcohol abuse problems, the serum level of IL-6 was significantly elevated in those with liver fibrosis. Additionally, Shoji et al. demonstrated the feasibility of using IL-34 to predict liver fibrosis in NAFLD patients. IL-34 levels were elevated as the liver fibrosis progressed. The authors also developed a predictive model using serum concentrations of IL-36, P4NP 7S, and age to predict liver fibrosis in NAFLD patients [112]. Similarly, inflammatory cytokines are being studied for their role in predicting the onset of type 2 diabetes. An investigation carried out by Spranger and colleagues demonstrated that combined effects of cytokines could be useful in the prediction of type 2 diabetes. Subjects having detectable levels of IL-1β and elevated levels of IL-6 were found to have a higher risk of developing type 2 diabetes [113].

Serum concentrations of TGF-β1 were higher in patients with chronic hepatitis C and were associated with the advancement of liver fibrosis, making it a good marker for the prediction of liver fibrosis [114]. The degradation product of latency-associated peptide (LAP), which is a pro-peptide of TGFβ, has also been outlined as a novel blood biomarker for liver fibrosis in mice. Plasma levels of L⁵⁹LAP-DP (a degradation product of LAP) were found to be increased in the early stages of liver fibrosis in CCl₄- and BDL-induced liver fibrosis models [115]. Serum and urinary levels of TGF-β1 were significantly elevated in patients with type 2 diabetes as compared to healthy controls [116]. Platelet-derived growth factor BB (PDGF-BB) is another protein that could serve as a potential marker. Serum levels of PDGF-BB were found to decrease in chronic hepatitis B patients and were negatively correlated with the advancement of liver fibrosis [117]. Likewise, urinary levels of PDGF-BB are elevated in type 2 diabetic patients as compared to healthy controls [118]. Detection of connective tissue growth factor (CTGF) serum levels has also shown some promise in detecting liver fibrosis. In a study by Zhang et al., serum levels of CTGF indicated the presence of liver fibrosis. CTGF levels in the serum could also indicate the stages of liver fibrosis [119]. Similarly, another study conducted in patients with chronic hepatitis C infection showed that CTGF levels in the serum were significantly higher than those in the control group and were also associated with the histological stages of liver fibrosis [120].

4. Therapeutic Agents and Drug Delivery Systems

The treatment of diabetes-associated liver fibrosis can be classified into several strategies based on pharmacology and pathogenesis. First, cure the primary disease, such as remove or reduce viral infections, metabolic disorders, alcohol abuse, and autoimmune causes. Targeting HSCs is considered the most efficient strategy in the treatment of liver fibrosis because activated HSCs are the key contributors during liver fibrogenesis [37]. Based on the role of HSCs in liver fibrogenesis, treatment strategies could be developed into various aspects, such as inhibition of HSC activation; suppression of the proliferation, motility, contraction, and pro-inflammatory responses of activated HSCs; degradation of accumulated ECM; induction of apoptosis of activated HSCs; and reduction of HSC-induced angiogenesis [121]. Accordingly, antifibrotic therapeutics including anti-inflammatory agents, antioxidants, herbal medicines, TGF-β antagonists, inhibitors of signal transduction (TGF-β/Smad, MAPK, and NF-κB signaling), and blockers of cell-matrix interactions have been investigated [121]. Cannabinoid receptor type 1 antagonists, angiotensin receptor blockers, and angiotensin converting enzyme inhibitors, have shown the potential to alleviate liver fibrogenesis and insulin resistance [122–125]. Obeticholic acid, a selective farnesoid X receptor (FXR) agonist, has been investigated to attenuate fibrotic progression in patients with NAFLD and type 2 diabetes (NCT00501592) [126]. FXR is involved in the regulation of inflammatory biomarkers, fibrotic biomarkers, and lipid metabolism [122, 126–128]. Strategies of regulating fibrogenic signaling pathways using siRNAs and miRNAs were recently summarized in our previous review [129]. To date, antidiabetic drugs and nanoparticle-based therapeutics are two promising approaches for the treatment of diabetes-associated liver fibrosis [130].

4.1. Antidiabetic drugs

Dysregulation of glucose homeostasis, insulin resistance, and obesity are regarded as the main factors that lead to diabetes-associated liver fibrosis. Studies have shown that impaired insulin response results in homeostatic imbalance and leads to high levels of free fatty acids and glucose in the blood. Therefore, controlling blood glucose levels and improving insulin sensitivity are the potential strategies to mitigate liver fibrosis progression in diabetic patients [131, 132]. Several studies reported that insulin therapy applied to diabetes-associated liver fibrosis is not an ideal option compared to other anti-diabetic drugs [132, 133]. Bodyweight control is one of the indexes to evaluate the treatment efficiency for obese and NAFLD patients. However, insulin therapy may increase the body weight of type 2 diabetic patients [134]. Furthermore, insulin, as a growth hormone, could increase cell proliferation and the production of IGF-1. The oncogenic effects of insulin may also potentially increase the risk of cancer progression [135]. Anti-diabetic drugs, including biguanides, sulfonylureas, sodium-glucose cotransporter 2 (SGLT2) inhibitors, peroxisome proliferator-activated receptors (PPAR) agonists, α-glucosidase inhibitors (α-GIs), glucagon-like peptide1 receptor agonists (GLP-1RAs), and dipeptidyl peptidase-4 (DPP-4) inhibitors, were found to efficiently control hyperglycemia, reduce body weight, and attenuate the progression of liver fibrosis in animal studies and clinical trials [132]. For example, pioglitazone has been investigated in clinical trials (NCT00994682 and NCT01002547) for the treatment of patients with NASH and type 2 diabetes [136, 137]. The mechanisms of antidiabetic drugs applied to liver fibrosis treatment are summarized in Figure 3.

Figure 3. The mechanisms of antidiabetic drugs used for liver fibrosis treatment.

The black arrows indicate the interactions among insulin resistance, hyperglycemia/hyperinsulinemia, and de novo lipogenesis induced by high fat diet (HFD). The blue arrows indicate the effect of antidiabetic drugs on the attenuation of liver fibrosis.

4.1.1. Metformin

Metformin is a biguanide antidiabetic agent used as the first-line drug for type 2 diabetes. Metformin controls serum glucose levels by lowering the production of hepatic glucose, reducing glucose absorption from the intestinal, and enhancing peripheral glucose utilization. It improves insulin sensitivity without altering the secretion of insulin [138]. Several studies have demonstrated that metformin attenuates the progression of liver fibrosis and HSC activation by reducing oxidative stress, inflammation, and endothelial NO synthase (eNOS) [139, 140]. Metformin can regulate glucose homeostasis via gluconeogenesis, suppress the activity of TGF-β1, reduce the activation of HSCs, and inhibit the succinate receptor (G-protein coupled receptor 91) signaling [141]. It was suggested that metformin could directly suppress the expression and phosphorylation of the TGF-β1/Smad3 pathway [142]. Al‐Hashem et al. demonstrated that metformin inhibits the mammalian target of rapamycin (mTOR)-hypoxia‐inducible factor‐1 alpha (HIF‐1α) axis, which plays an important role in liver fibrogenesis. Metformin reduces the production of profibrogenic biomarkers (α‐smooth muscle actin and TIMP‐1) and liver injury enzymes. It inhibited inflammation in a thioacetamide-induced liver fibrosis rat model [143]. Moreover, Li et al. showed that metformin suppresses the expression of VEGF secreted by HSCs and decreases the activation of HSCs by activating adenosine monophosphate-activated protein kinase (AMPK) [144].

4.1.2. SGLT2 inhibitors

SGLT2 inhibitors are a second-line treatment for type 2 diabetes by inhibiting the kidneys from reabsorbing glucose from the proximal tubular cells [145]. Goto et al. investigated the effect of the SGLT2 inhibitor tofogliflozin in a medaka NASH model. The results showed that tofogliflozin specifically targets the Sglt2 protein expressed in medaka kidneys, ameliorates the progression of liver fibrosis, reduces the fat accumulation, and improves insulin resistance [146]. In another study, the SGLT2 inhibitor dapagliflozin ameliorates type 2 diabetes-induced liver and renal fibrosis in a db/db mouse model [147]. Similar antifibrotic effects were observed in ipragliflozin-treated choline-deficient L-amino acid-defined (CDAA) diet-fed rats, where it prevents TG and lipid droplets from accumulating in the liver [148]. SGLT2 inhibitors also regulate lipid synthesis-related gene expressions, such as downregulation of the expression of the fatty acid synthase (FAS), stearoyl-CoA desaturase (SCD1), and SREBP 1c. However, Nishimura et al. suggested that ipragliflozin alleviated liver fibrogenesis by improving insulin resistance in obese diabetic Otsuka Long-Evans Tokushima fatty rats [38]. Jojima et al. demonstrated that canagliflozin attenuates the progression of NASH and NASH-related HCC. Canagliflozin inhibits the SGLT2 in liver tumors and suppresses the proliferation of HepG2 cells by reducing the expression of cyclin D and cyclin-dependent protein 4. Canagliflozin also induces HepG2 cell apoptosis through the activation of caspase 3 in diabetes and NASH-related HCC [149]. Clinical trial (NCT02279407) has demonstrated that dapagliflozin could improve blood glucose levels, reduce body weight, and lower the injury biomarkers in patients with NAFLD and type 2 diabetes [150]

4.1.3. PPAR agonists

PPARs can be classified into three isoforms, including PPARα, PPARδ(β), and PPARγ. They are distributed in different tissues, such as adipose tissue and hepatocytes, and are responsible for various biological functions, such as controlling inflammatory pathways and regulating lipid metabolism, gluconeogenesis, and ketogenesis [151]. PPARγ prevents quiescent HSCs from transforming to activated HSCs and decreases the proliferation of HSCs [152]. Selective PPAR agonists, such as fenofibrate (a PPARα agonist), have shown reduction effect on hepatocyte ballooning. Some PPARδ agonists attenuate inflammation and liver fibrosis in NASH animal models [153]. Elafibranor targets both PPARα and PPARδ and subsequently alleviates liver fibrosis and steatohepatitis in animals and patients with NASH [154]. PPARγ agonists (i.e., pioglitazone and rosiglitazone) can improve insulin resistance and reduce inflammation, steatosis, and fibrosis in the liver. Pan-PPAR agonists are new modulators designed to act on all PPAR isoforms [155]. Studies have shown that pan-PPAR agonists, such as IVA337 and lanifibranor, can provide better antifibrotic effects than selective PPAR agonists due to the well-balanced activation of the three isomers [153, 156].

4.1.4. αGIs

αGIs inhibit the glucosidase enzymes to convert starch into a simple glucose form, which reduces the carbohydrate absorbed from the small intestine [157]. Patients treated with miglitol, a typical αGI, in a 12-month long-term study showed decreased BMI, ALT, HbA1c, and inflammatory cytokines. However, no significant improvement in liver fibrosis or hepatocyte ballooning was observed in NASH patients [158].

4.1.5. GLP-1RAs

The endogenous hormone GLP-1, secreted from Langerhans cells, plays a critical role in serum glucose regulation by stimulating insulin secretion and suppressing glucagon secretion. However, the DPP-4 enzyme rapidly degrades GLP-1 and causes its malfunction [159]. To prolong the half-life of GLP-1, GLP-1RAs were developed to mimic the function of natural hormones for treating type 2 diabetes. Most animal and clinical studies have suggested that GLP-1RAs affect serum transaminase levels; improve lipid metabolism; activate AMPK, PPAR-α, and hepatic insulin signaling pathways; protect the liver from steatosis via modulating the GLP receptor on hepatocytes; mitigate hepatocytes necrosis; decrease the release of inflammatory cytokines (CRP, TNF-α, and IL-6); and inhibit the inflammatory pathways (IKKβ/NF-κB pathway and JNK) to alleviate NAFLD-induced liver fibrosis in patients with type 2 diabetes [42, 160, 161]. Exenatide, a GLP-1Ra, showed a hepatic-protective effect in patients with NAFLD and type 2 diabetes [162].

4.1.6. DPP-4 inhibitors

DPP-4 inhibitor, a specific protease inhibitor, could prolong the half-life of GLP-1 [159]. In addition to lowering hyperglycemia and improving insulin resistance, linagliptin, a DPP-4 inhibitor, reduced the secretion of fibrotic biomarkers (collagen, α-SMA, and TGF-β1) and inflammatory factors (TNF-α, IL-6, and NF-κB) in a diabetic rat model induced by high-fat diet and streptozotocin. Linagliptin downregulates the NF-κB signaling pathway and activates the IRS-1/PI3K/Akt pathway to inhibit the proliferation of HSCs [163]. A six-month clinical trial of gliptin therapy showed a significant reduction in body weight, hepatic fat content, and myocardial fat content in female patients [164]. However, a randomized controlled trial reported that the administration of DPP-4 inhibitors alone did not show any benefit compared to placebo in type 2 diabetic patients with NAFLD [165]. Kawakubo et al. reported that anagliptin can attenuate liver inflammation, fibrosis, and carcinogenesis in MC4R-KO mice, but has no effect on the inflammation and fibrosis of adipose tissue [166]. To enhance the antifibrotic effect, Kaya et al. combined anagliptin and oleanolic acid, a Takeda G protein-coupled receptor 5 (TGR5) agonist, to treat porcine serum-induced liver fibrosis in a rat model. The antifibrotic effect of the two drugs determined by qRT-PCR showed that the combined treatment could synergistically reduce the expression of the profibrogenic markers (i.e., Acta2, Col1a1, Fn1, and Ctgf) and inhibit the activation/proliferation of HSCs [167].

4.2. Nanoparticle platform

Nanoparticle is a rapidly growing and promising platform for a great variety of therapeutic agents from small molecules to macromolecules. Numerous studies have demonstrated the efficacy and advantages of nanoparticles, such as enhanced tissue-specificity, increased stability, decreased toxicity, and improved therapeutic index [168, 169].

4.2.1. Nanoparticles for Diabetes-Associated Renal, Retinal, and Cardiac Fibrosis

Diabetes-related fibrotic complications, including hepatic, renal, retinal, and cardiac fibrosis, have attracted substantial public health attention, and nanoparticles have shown great potential in treating these complications. For example, nanoparticle-based anti-inflammatory agents or antioxidants derived from natural sources, such as curcumin and aged garlic extract, effectively alleviate diabetic cardiomyopathy by increasing the insulin concentration and decreasing glucose levels in the blood [170].

Gold nanoparticles alone exhibit an angiogenic effect by inducing reorganization of VEGFR2 on endothelial cells. VEGF is a key stimulator for choroidal neovascularization (CNV), which is subsequently followed by subretinal fibrosis. Gold nanoparticles were, therefore, used to inhibit the VEGF-induced Akt/endothelial nitric oxide synthase pathway in choroid-retina endothelial (RF/6A) cells and inhibit cell migration [171]. Zhao et al. developed two nanoparticle delivery systems to treat and prevent the development of diabetic cardiomyopathy. One formulation is basic fibroblast growth factor (bFGF)-loaded liposome (bFGF-lip), and the other one is acidic fibroblast growth factor (aFGF)-loaded Poloxamer 188-grafted heparin copolymer nanoparticle. Both delivery systems were used in combination with the ultrasound-targeted microbubble destruction (UTMD) technique and demonstrated efficacy in reducing collagen expression, activating the PI3K/AKT signaling pathway, reducing myocardial cell apoptosis, increasing the myocardial microvascular density, and attenuating myocardial fibrosis. [172, 173]. Moreover, antago-miR155 gold nanoparticle was applied to regulate the macrophage M1/M2 ratio and reduce inflammatory factors in ovariectomized diabetic mice with diabetic cardiomyopathy [174]. The diabetes-induced renal disorder is a complicated disease, and crocetin-loaded PLGA nanoparticles (CT-PLGA-NPs) were used to treat rats with renal nephropathy induced by streptozotocin. The CT-PLGA-NP-treated group demonstrated relatively normal blood glucose levels and insulin concentrations. CT-PLGA-NPs downregulated the expression of fibrotic and inflammatory molecules, including TGF-β1, Type IV collagen, NF-κB, VEGF, TNF-α, IL-6, IL-1β, and MCP-1 [175].

4.2.2. Nanoparticles for Liver fibrosis

The applications of nanoparticles in treating diabetes-associated liver fibrosis are still limited due to the ambiguous mechanism and lack of direct evidence to support the relationship between the two diseases. Based on the current studies on diabetes-associated liver fibrosis, nanoparticle platforms provide the delivery advantages to alleviate the fibrosis progression. The anti-inflammatory agent curcumin is regarded as a natural antioxidant with free radical scavenging activity. Moreover, curcumin could prevent cytokine-induced damage to β-cells, inhibit hepatic gluconeogenesis, and increase GLP-1 secretion to stabilize the serum glucose and insulin concentrations. Curcumin-loaded nanoparticles could decrease the expression of pro-inflammatory cytokines (IL-6, TNF-α, and IL-1β) and passively target the inflammatory liver tissue through the enhanced permeability and retention (EPR) effect, which provides a higher therapeutic efficacy than free curcumin. Curcumin-loaded PLA-PEG copolymer nanoparticles protect hepatic cells from injury by upregulating PPARγ expression to inhibit the activation of the NF-κB pathway and decrease COX-2- and TGF-β-induced diabetic hepatopathy [176]. Berberine suppresses the body weight and maintains the homeostasis of blood glucose, insulin, ALT, and TG. The berberine solid lipid nanoparticles (BBR-SLNs) were administered orally to treat hepatosteatosis in a db/db mouse model. The results showed that berberine nanoparticles were able to accumulate in the liver and showed liver protective effect to alleviate hepatosteatosis through inhibiting the expressions of SREBP1c, SCD1, and FAS. Meanwhile, the lipolytic gene carnitine palmitoyltransferase-1 (CPT1) was upregulated in the liver treated with BBR-SLNs. [177].

In addition to delivering natural anti-inflammatory compounds, Dogra et al. demonstrated that zinc oxide nanoparticles (ZnO NPs) alone can improve the physiological homeostasis during obesity. ZnO Nps reverse dysregulated hepatic lipogenesis and thereby ameliorating the development of hepatosteatosis and insulin resistance in high-fat diet induced obese mice. [178]. The hepatocyte Notch activity is aberrant in both type 2 diabetes and NASH. Richter et al., therefore, developed a PLGA nanoparticle loaded with a γ-secretase inhibitor (GSI), dibenzazepine, to block the Notch signaling pathway in the liver. GSI prevents translocation of the Notch intracellular domain (NICD) into the nucleus, thereby reducing the hepatic glucose production-related gene transcription. Administration of GSI NPs reverses the diet-induced glucose tolerance and suppresses the inflammation and liver fibrosis in the NASH-provoking diet mouse model without inducing gastrointestinal toxicity[179].

Genome editing is the newest and the most promising approach for treating genetic diseases. Clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated protein 9 (Cas9) is the most advanced gene-editing tool to specifically edit target genes with high efficiency and accuracy. DPP-4 inhibitors are one of the major strategies applied to the treatment of type 2 diabetes. DPP-4 enzyme could rapidly degrade GLP-1, a pivotal hormone stimulating insulin secretion, and regulate fat accumulation in the liver. Therefore, suppression of DPP-4 enzyme expression or inhibition of the DPP-4 enzyme is one of the potential approaches to alleviate liver fibrosis and impairment in diabetes- or obesity-related NAFLD. For example, Cho et al. developed a recombinant Cas9 nuclease complexed with sgRNA targeting the DPP-4 gene (Cas9-RNP) and delivered it using lecithin-based liposomal nanoparticles to downregulate DPP-4 enzyme, thereby upregulating the GLP-1 serum concentration in a db/db mouse model [180].

5. Conclusion and future perspectives

Emerging diabetes-associated liver fibrosis is a complex metabolic disorder affected by various risk factors, and it can have serious complications. Type 2 diabetes and liver fibrosis usually develop chronically and silently. Obesity and insulin resistance resulting from hyperlipidemia and hyperglycemia are regarded as the two major risk factors that link type 2 diabetes and liver fibrosis. In general, exercise, diet control, and a healthy lifestyle are the easiest and most common approaches to prevent type 2 diabetes and liver fibrosis by avoiding overnutrition and insulin dysfunction.

Developing a standard treatment for liver fibrosis is currently an urgent need. The overexpressed biomarkers involved in the pathways of inflammation, formation and degradation of ECM, de novo lipogenesis, glycogenesis, and gluconeogenesis, are potential targets for treating metabolic disorders. Treatment strategies for diabetes-associated liver fibrosis, such as antidiabetic agents and nanomedicines, are anticipated as prospective candidates to mitigate the progression of liver fibrosis. Anti-diabetic agents can target specific receptors regulating insulin sensitivity and glucose homeostasis in tissues, which ameliorates liver fibrosis. Similarly, drug delivery systems can promote the therapeutic efficiency of antioxidants, anti-inflammatory agents, and antifibrotic agents to suppress the profibrotic pathways. Despite the complexity of the association between type 2 diabetes and liver fibrosis, the development of targeted nanomedicines has the great promise to prevent liver impairment and its subsequent pathologies.