Targeting of Secretory Proteins as a Therapeutic Strategy for Treatment of Nonalcoholic Steatohepatitis (NASH).

Nonalcoholic steatohepatitis (NASH) is defined as a progressive form of nonalcoholic fatty liver disease (NAFLD) and is a common chronic liver disease that causes significant worldwide morbidity and mortality, and has no approved pharmacotherapy. Nevertheless, growing understanding of the molecular mechanisms underlying the development and progression of NASH has suggested multiple potential therapeutic targets and strategies to treat this disease. Here, we review this progress, with emphasis on the functional role of secretory proteins in the development and progression of NASH, in addition to the change of expression of various secretory proteins in mouse NASH models and human NASH subjects. We also highlight secretory protein-based therapeutic approaches that influence obesity-associated insulin resistance, liver steatosis, inflammation, and fibrosis, as well as the gut–liver and adipose–liver axes in the treatment of NASH.

Growing evidence shows that secretory proteins are affected by various intracellular events involved in NASH, including endoplasmic reticulum (ER) stress, mitochondrial stress, lipotoxicity, reactive oxygen species (ROS) production, and intercellular events between different liver cell types [3][4][5]. Secretory changes can have either protective or detrimental systemic effects. Here, we briefly describe molecular mechanisms involved in pathogenesis of steatosis, inflammation, or fibrosis, highlighting contributions of each cell type (such as hepatocyte, Kupffer cell (KC), or hepatic stellate cell (HSC)) and crosstalk between these cell types within the liver. We also discuss the in vivo role of key secretory proteins in the development and progression of NASH. Finally, we illustrate therapeutic potential of secretory proteins in NASH.

1.
FFA uptake-Circulating FFAs released from adipose tissues by lipolysis can enter the liver, contributing to the largest amount of hepatic lipid. This process is mediated by cluster of differentiation 36 (CD36), caveolin, and fatty acid transport (FATP), all located in the plasma membrane of hepatocytes [8]. Although the role of hepatic caveolin is less well-studied, FATP isoforms (FATP2 and FATP5) and CD36 have been shown to participate in increased uptake of circulating FFAs and development of steatosis. Knockdown of FATP2, FATP5, or CD36 leads to decrease of hepatic FFA uptake and ameliorates hepatic steatosis in mice [9][10][11], supporting the fundamental importance of FATPs and CD36 to hepatic steatosis.

2.
DNL-DNL is a process to synthesize new fatty acids from acetyl coenzyme A (acetyl-CoA).
In the steatotic, insulin-resistant liver, although insulin-induced suppression of gluconeogenesis is impaired, insulin-stimulated DNL rates increase. This is referred to as selective insulin resistance [12]. Although molecular regulators of this paradox are not fully understood, recent results have suggested that PHLPP2 (pleckstrin homology domain leucine-rich repeat protein phosphatase 2), an Akt Ser473 phosphatase, acts as a novel regulator to terminate insulin-induced DNL, with no impact on early-postprandial gluconeogenesis [13,14]. Degradation of PHLPP2 in obese liver sustains Akt-mediated induction of sterol regulatory element-binding protein 1c (SREBP-1c) and DNL [7]. Consequently, increased DNL has been shown to contribute to the steatosis of NAFLD/NASH patients [15]. 3.
FAO-There have been conflicting reports on the role of FAO in NAFLD/NASH patients [16,17].
Although reduced FAO has been reported to contribute to increased hepatic lipid accumulation [17], increased markers of FAO have also been observed in steatotic livers, a likely compensatory mechanism to alleviate excessive lipid accumulation [16].

4.
VLDL secretion-Export of lipids from the liver is an important pathway to reduce hepatic lipid accumulation. apoB100 and microsomal triglyceride transfer protein (MTP), key components in hepatic VLDL secretion, are negatively regulated by insulin [18]; however, selective insulin resistance in NAFLD patients allows insulin to stimulate DNL without suppressing VLDL production [19].
Overall, these pathways lead to intracellular lipid accumulation in hepatocytes. The proximal cause for many of these abnormalities is obesity-induced insulin resistance, in response to environmental factors (sedentary lifestyle or nutrient overload such as high-fat, high-carbohydrate, or high-fructose diets) that exacerbate underlying genetic predisposition (such as single nucleotide polymorphisms of patatin-like phospholipase domain containing protein 3 (Pnpla3) or transmembrane 6 superfamily member 2 (Tm6sf2), leading to hepatic lipid accumulation ( Figure 1). In steatoic livers, increased excessive lipids or its metabolites subsequently cause excessive production of reactive oxygen species (ROS) and dysfunction of intracellular organelles (ER/mitochondria). These events culminate in hepatic sublethal or lethal injury (cell death) ( Figure 1).

Inflammation (KCs/Macrophages and Other Immune Cells)
In the setting of NASH, lipotoxicity-induced hepatic sublethal or lethal injury is able to stimulate KCs to release hepatocyte-derived factors such as damage-associated molecular patterns (DAMPs), extracellular vesicles (exosomes), or inflammatory cyotokines [23]. Activated KCs secrete more pro-inflammatory cytokines and chemokines, leading to recruitment of other immune cells (such as macrophages, neutrophils, or natural killer T cells (NKTs)) and activation of HSCs [24] (Figure 1). Toll-like receptors (TLRs) are pattern recognition receptors (PRRs) that act as sensors of the innate immune system together with nucleotide-binding and oligomerization domain (NOD)-like receptors (NLRs). Endotoxin (lipopolysaccharide) and lipid metabolite (ceramide or palmitic acid) activates TLR or NLR signaling in hepatocytes and KCs/macrophages, leading to increased secretion of the inflammasome-activating cytokine IL-1β and pro-inflammatory cytokines/chemokines (TNFα, IL-6, CCL2, CXCL16 or CXCL1/2/8) [24,25]. Growing evidence has suggested that the inflammasome is a critical triggering factor for progression from steatosis to NASH [26,27]. Preclinical animal studies using a small molecule NLR family pyrin domain containing 3 (NLRP3) inhibitor (MCC950) or mice with deletion of Il-1β or Il-1r have revealed that deficiency of the inflammasome results in reduced hepatic inflammation and fibrosis in mice [26,27]. These data suggest that inflammasome could be a potential target for treatment of NAFLD/NASH. Similarly, the chemokine C-C chemokine receptor 2/5 (CCR2/CCR5) inhibitor (cenicriviroc) suppresses recruitment of monocytes and activation of HSCs, and is currently being evaluated in a phase 3 clinical study (NCT03028740) in NASH patients with liver fibrosis. An amine oxidase copper-containing 3 (AOC3) inhibitor (BI1467335/PXS-4728A) is also in a phase 2 clinical trial (NCT03166735) to block infiltration of immune cells in the liver.
Inflammation in other tissues, such as adipose tissue and intestine, may contribute to the development and progression of NAFLD/NASH [28,29]. Inflammation in white adipose tissue (WAT) can induce hepatic inflammation [28]. Neutrophils or macrophages infiltrating in WATs produce pro-inflammatory mediators or cytokines, which contribute to systemic inflammation [30,31]. Further, loss of protective adipokines (such as adiponectin and leptin) secreted from WATs may affect lipid accumulation, inflammation, and fibrosis in the liver [32] (Figure 1). In addition to WAT, decreased brown adipose tissue (BAT) activity has been associated with the development and progression of NAFLD [33,34]. Combined therapy of BAT activation (treatment of β3AR agonist) and caloric restriction synergistically improve NASH in an animal model, although BAT activation alone does not reverse NASH despite alleviation of steatosis [34].
Several HSC-targeted therapies have been tested in NASH patients. Lysyl oxidase-like 2 (LOXL2) is an enzyme that catalyzes collagen cross-linking to remodel the extracellular matrix, leading to development of a monoclonal antibody against LOXL2 (Simtuzumab/GS-6624). However, efficacy of simtuzumab as a monotherapy was minimal [48]. Galectin-3, a lectin derived from KCs/macrophages, plays an important role in transforming growth factor beta (TGFβ)-mediated activation of HSCs [49], leading to phase 3 trial for a galectin-3 inhibitor (GR-MD-02).

Growth Differentiation Factor (GDF15) Agonists
GDF15 belongs to the transforming growth factor β (TGFβ) superfamily, and has been identified as a secretory protein with predominant expression in liver, placenta, and macrophages [62]. However, GDF15 expression has also been reported in adipose tissue, muscle, lung, kidney, and heart [62]. Growing evidence has suggested that GDF15 is induced in the livers of mice or human subjects with various liver injuries including NAFLD/NASH, hepatic viral/bacterial infections, and HCC [63][64][65]. For example, GDF15 expression is increased in the livers of mice fed a NASH-provoking diet (methionine-choline-deficient (MCD) or amylin liver NASH model (AMLN)) and in human subjects with NASH or advanced fibrosis [66]. GDF15 expression is also increased in the livers of mice with alcohol feeding or human subjects with alcoholic steatohepatitis (ASH) [67], suggesting that GDF15 may be a biomarker for common liver diseases. Interestingly, treatment with recombinant GDF15 or genetic overexpression of GDF15 resulted in improvement of inflammatory and fibrotic features in mice fed an ASH or NASH diet [66,67]. Consistent with this data, GDF15 has been recently reported to exert an anti-fibrotic action in other tissues such as kidney and lung [68,69]. Despite these preclinical studies, clinical trials of GDF15 for NASH have not been performed (Table 1).

Fibroblast Growth Factor 15 (FGF15)/FGF19 Agonists
FGF15/FGF19 (FGF15 in mice and FGF19 in humans) is a hormone produced in the intestine that plays a crucial role in the regulation of bile acid metabolism in the liver [70]. Feeding rapidly induces the release of bile acids stored in the gallbladder to help digestion and absorption of fats. In the late-postprandial state, small intestine enterocyte-derived FGF15/FGF19 enters the liver via the portal vein, where it binds its receptor (FGFR4) in hepatocytes to repress de novo bile acid synthesis through suppression of cholesterol 7a-hydroxylase (CYP7A1), a rate-limiting enzyme for conversion of cholesterols to bile acids [70]. Intriguingly, FGF15/FGF19 also has potent effects on lipid or glucose metabolism [71,72]. FGF15/FGF19 stimulates glycogen synthesis and suppresses gluconeogenesis, similar to the action of insulin in the liver [71]; however, in contrast to insulin, FGF15/FGF19 decreases hepatic TG accumulation [72]. In addition, FGF15/FGF19 may enhance insulin sensitivity by multiple actions, including increased β-oxidation, reduced lipogenesis, or diminished lipotoxicity [73,74]. Thus, FGF15/FGF19 is a promising therapeutic molecule for treatment of obesity-related metabolic deterioration. Intriguingly, serum FGF19 level is reduced in patients with NAFLD/NASH, and its action in the liver is also impaired in these subjects [75,76]. Early phase 2 studies (NCT02443116) of aldafermin (NGM282, a nontumorigenic FGF19 variant) showed efficacy in reducing hepatic steatosis and hepatic inflammation/fibrosis in patients with NASH [77]. Further clinical studies (NCT03912532; NCT04210245) are currently underway to evaluate the efficacy of FGF15/FGF19 as a NASH therapeutic (Table 1).

FGF21 Agonists
FGF21 was identified as a secretory protein that enhances insulin action in adipocytes [78]. Growing evidence has suggested that FGF21 acts as a hormone that is able to regulate glucose or lipid metabolism in response to environmental stimuli or nutrient stresses [79]. FGF21 exerts beneficial effects on obesity and related metabolic diseases via multiple actions, such as enhancements of insulin-mediated glucose uptake and β-oxidation/thermogenesis, as well as amelioration of ER stress [78,80,81]. There results suggest the potential of FGF21 as a drug candidate to treat obesity-related type 2 diabetes. In a recent clinical study (NCT02097277) of pegbelfermin (BMS-986036, polyethylene glycol-attached, PEGylated FGF21), however, little effect on Hb1Ac or body weight was observed in obese patients with type 2 diabetes [82]. Nonetheless, FGF21 remains an attractive therapeutic target for treatment of NASH, as numerous preclinical studies have suggested that FGF21 alleviates steatohepatitis and fibrosis in NASH diet-fed mice via decrease of hepatic lipotoxicity, increase of β-oxidation, inhibition of fibrogenesis, or activation of an adiponectin-IL17A axis [83][84][85]. In several clinical studies (NCT02097277 or NCT02413372) for human NASH subjects, 12 weeks or 16 weeks of pegbelfermin led to alleviation of steatosis and improvements of metabolic parameters (HDL, TG, adiponectin) and fibrosis biomarkers [82,86]. Ph2b clinical studies (NCT03486899 and NCT03486912) of pegbelfermin in NASH patients with advanced fibrosis are currently underway (Table 1). In addition, phase 1/2 clinical trials of other FGF21 agonists (NCT03060538; agonistic anti-FGFR1/KLB antibody (BFKB8488A), NCT03298464; anti-FGFR1c/KLB antibody (MK-3655/NGM313), NCT03976401; an engineered Fc-FGF21 (AKR-001) and NCT04048135; a glycopegylated FGF21 (BIO89-100)) are underway or completed in patients with NAFLD or NASH (Table 1).

Mitochondrial Open Reading Frame of the 12S rRNA Type-c (MOTS-c) Agonists
MOTS-c, a naturally occurring mitochondrial peptide, has been identified as a regulator of metabolic homeostasis. Treatment with MOTS-c alleviates high-fat diet (HFD)-induced steatosis, obesity, and insulin resistance in mice [87]. The metabolic beneficial effect of MOTS-c is likely due to increased lipid catabolism and enhanced energy expenditure [87], but MOTS-c is also able to inhibit lipolysis in adipocyte in an insulin-independent manner, which contributes to reduced release of fatty acids from adipocytes and improvement of hepatic steatosis. Interestingly, treatment with MOTS-c resulted in improvements of serum alanine aminotransferase (ALT) level, hepatic TG content, and NAFLD activity score (NAS) in the stelic animal model (STAM) animal model, suggesting therapeutic potential of MOTS-c for treatment of NASH. A phase 1 clinical study (NCT03998514) of MOTS-c analogs (CB4211) in patients with NAFLD/NASH is currently underway (Table 1). Considering the role in energy metabolism of mitochondrial-derived peptides (MDPs) such as Humanin or small Humanin-like peptides 2 (SHLP2) [88,89], mitochondria-based therapeutics might be potential strategies for treatment of various human metabolic diseases including NAFLD/NASH. Further studies are needed to identify more novel MDPs and evaluate their functional role in pathogenesis of NAFLD/NASH and the possibility of their therapeutic application.

Conclusions
Despite the significant clinical impact of NAFLD/NASH, no therapeutic drugs are approved for its treatment. Most therapeutic strategies have focused on monotherapy targeting one among four main pathogenic processes (steatosis, hepatic sublethal/lethal injury, or inflammation or fibrosis). As discussed above, intracellular events (hepatic lipid accumulation or ER/mitochondria/ROS stress), inter-cell liver communication (between parenchymal and non-parenchymal cells), and inter-organ communications (adipose-liver axis or gut-liver axis) all participate in the development and progression of NAFLD/NASH. Given the value of a "multiple-hit parallel model" in the pathogenesis of NASH, strategies targeting two or more pathogenic processes in NASH may be more appropriate. Considering the action of secretory proteins in different liver cell types and in distant organs, therapeutic strategy with secretory proteins may bring about novel NASH therapeutic opportunities. In addition, combined therapy with agents targeting single pathogenic processes (e.g., FXR agonist (tropifexor) and CCR2/5 antagonist (cenicriviroc)) could be coupled with novel entries (e.g., GLP1-IgG Fc-FGF21) to produce an efficacious therapeutic approach for treatment of NASH. Finally, combined therapeutic approaches based on chemicals, biologics, antisense, or microbiome might also be attractive. Further preclinical and clinical testing allow for more valuable approaches for treatment of NAFLD/NASH. Author Contributions: K.K. and K.H.K. wrote the manuscript. All authors have read and agreed to the published version of the manuscript.