Type 2 diabetes mellitus (T2DM) and non-alcoholic fatty liver disease (NAFLD), two common metabolic syndromes, are emerging as global epidemics, whose incidence are rising annually (1,2). T2DM is predominantly characterized by an assembly of hyperglycemia, hyperinsulinemia, insulin resistance and insulin deficiency (3). According to the Diabetes Atlas 9th edition published by the International Diabetes Federation (IDF), 463 million adults aged 20-79 years are suffering from diabetes mellitus worldwide, with the prevalence of diabetes mellitus in that age group being ~9.3%, and the total number of diabetic patients predicted to rise to 700 million (10.9%) by 2045 (1). In total, >90% of the diabetic patients belong to T2DM, as estimated by IDF. NAFLD, currently the most common chronic liver disease, covers a wide disease spectrum, ranging from simple steatosis to non-alcoholic steatohepatitis (NASH), hepatic fibrosis, cirrhosis and hepatocellular carcinoma (HCC), which finally causes liver-associated mortality (4). A meta-analysis on NAFLD epidemiology reported a global prevalence of 25.24% in 2016 (2). In China, the prevalence has risen from 25.4% in 2008-2010 to 32.3% in 2015-2018 (5).

T2DM and NAFLD are closely associated. According to clinical data, the overall incidence of NAFLD is 55.5% among patients with T2DM (6), and NAFLD is an independent risk factor for T2DM, indicating a strong bi-directional relationship between T2DM and NAFLD (7,8). T2DM is a risk factor for progression from simple steatosis to NASH and advanced fibrosis. T2DM is associated with a high morbidity of NASH (9,10). Patients with simple steatosis often have a benign prognosis, whereas NASH can progress to cirrhosis, with patients eventually developing HCC (11,12). The prevalence of HCC is ~5-fold higher when the disease progresses from simple steatosis to NASH, leading to a markedly higher mortality rate (13,14). In addition, the presence of NAFLD in patients with T2DM is highly associated with the incidence of macro- and micro-vascular diabetic complications (15). It is harder to control blood glucose levels in patients with T2DM with NAFLD, compared with patients with only T2DM (16). Both T2DM and NAFLD can be caused by metabolic disorders, and share familiar or even the same risk factors and pathological mechanisms. Although studies have shown that the existing pathogenic mechanisms of T2DM and NAFLD include insulin resistance, hyperglycemia, hepatic lipid accumulation and inflammation (17-19), due to the multifaceted and intricate correlations between these two diseases, the underlying molecular mechanisms require further exploration.

Both T2DM and NAFLD can be largely influenced by dietary structure. The intake of Western diet contributes to the onset and development of T2DM and NAFLD (20-22). A Western diet is mainly characterized by high amounts of saturated fatty acids (such as palmitic acid), simple carbohydrates (corn syrup and fructose), low levels of polyunsaturated fatty acids (PUFAs; n-6 and n-3 PUFAs), and insufficient intake of protein and dietary fibers (22,23). In addition, there is a high intake of n-6 PUFAs (particularly linoleic acid) and a low intake of n-3 PUFAs [such as α-linolenic acid (ALA)] in this dietary pattern among patients with T2DM and NAFLD, which cause a high ratio of n-6/n-3 PUFAs (24). The intake of Western diets results in an increased level of n-6-PUFA-derived arachidonic acid (AA) and subsequent eicosanoid production [particularly 2-series prostaglandins (PGs)], and there is a decreased level of those derivatives from n-3 PUFAs in patients with T2DM and NAFLD (25-27). Decreased n-3 PUFAs, which are partly caused by an impaired ALA desaturation in the liver, can repress fatty acid oxidation and contribute to pro-lipogenic outcome by downregulating peroxisome proliferator-activated receptor-α (PPAR-α); they can also promote lipogenic and glycogenic capacity by upregulating sterol regulatory element-binding protein 1c (SREBP-1c) (24). Furthermore, the downregulation of PPAR-α by n-3 PUFAs depletion activates the nuclear factor-κB (NF-κB) and activating protein 1 in the liver, leading to a pro-inflammatory effect in patients with NAFLD (24). On the other hand, the increased n-6 PUFAs and its derivatives can influence the inflammatory state and disturb glucose and lipid metabolism (28-32). Linoleic acid can alter fatty acid transportation, mitochondrial function, inflammatory responses and oxidative stress by increasing PG release and activating PPAR-γ, interleukin-8 (IL-8) and the NF-κB signaling pathway (30,32). The J2-series PGs can promote adipocyte differentiation by directly activating PPAR-γ (31). Therefore, n-6 PUFAs and n-3 PUFAs exert various vital metabolic effects, and the levels of n-6 and n-3 PUFAs, which can be mediated by similar dietary patterns of T2DM and NAFLD (particularly a Western diet) are important for the pathological development of these two diseases.

As important derivatives of n-6 PUFAs, 2-series PGs are widely distributed in almost every tissue. In the PG synthesis pathway, four principal bioactive 2-series PGs are generated, including PGD₂, PGE₂, PGF_2α and PGI₂ (33). Clinical and experimental evidence has indicated that 2-series PGs are involved in the initiation and progression of numerous diseases, including diabetes mellitus (34), hypertension (35), obesity (36), fatty liver disease (37), vascular diseases (38), carcinoma (39), inflammatory bowel disease (40), rheumatoid arthritis (41), asthma and allergic diseases (42) and Alzheimer's disease (43). Studies have revealed that 2-series PGs play complex and interlinked roles in mediating metabolic homeostasis and systemic chronic inflammation (34,44-48). Moreover, 2-series PGs have bidirectional effects on insulin secretion and pancreatic β-cell proliferation during hyperglycemia (34). As PPAR-γ modulators, 2-series PGs regulate adipogenesis and lipolysis in lipid metabolism, leading to excessive fat deposit (44-46). In addition, 2-series PGs are involved in immune response by affecting various cytokines and immune cells, such as macrophages and monocytes, under insulin resistance, hyperlipidemic and diabetic status (44,47,48). Of note, the nonsteroidal anti-inflammatory drugs (NSAIDs) and cyclooxygenase-2-selective inhibitors (COXIBs) can interfere with 2-series PG synthesis by inhibiting cyclooxygenases (COXs), and have been widely used in anti-inflammation, analgesia, antiplatelet aggregation and anti-tumorigenesis treatment (49-51). However, the potential application of NSAIDs and COXIBs in the treatment of T2DM and NAFLD requires further investigation.

To the best of our knowledge, 2-series PGs play an important role in the development of T2DM and NAFLD. However, few studies have focused on the therapeutic effect of targeting the 2-series PG pathway in these two metabolic syndromes. Herein, the way in which 2-series PGs exert multifunctional effects on the highly intertwined pathogenesis of T2DM and NAFLD, including insulin resistance, hyperglycemia, hepatic lipid accumulation and inflammation, were systematically reviewed, and it was revealed that targeting the 2-series PG pathway may be an important therapeutic strategy in T2DM and NAFLD treatment.

PGs belong to eicosanoids and have 1-, 2- and 3-series homologues. Each series of PGs is biosynthesized from different PUFAs, including dihomo-γ-linolenic acid (DGLA), AA and eicosapentaenoic acid (EPA) (52). DGLA is catalyzed by COXs to produce 1-series PGs (such as PGE₁, PGG₁ and PGD₁), and can also be converted to AA by the enzyme Δ⁵ desaturase (53). AA is the precursor of multiple important bioactive lipid mediators, including the 2-series PGs (such as PGE₂, PGD₂, PGF_2α and PGI₂) lipoxins, leukotrienes, resolvins, protectins and maresins (54).

To the best of our knowledge, the 1-series metabolites may be less closely associated with the correlation between T2DM and NAFLD, since a limited number of studies have been conducted. Furthermore, the 3-series PGs (such as PGF_3α and PGE₃) produced by EPA generally have a lower biological activity than their 1- and 2-series homologues (55). Therefore, the present review focused on the 2-series PGs that are the principal PGs derived from AA with a biological significance in T2DM and NAFLD.

The synthesis of 2-series PGs is precisely regulated (Fig. 1). Apart from being produced from DGLA, AA is mainly derived from cellular membrane phospholipids. Membrane phospholipids are esterified by PLA₂s to generate free AA. AA is subsequently converted to PGG₂, followed by a peroxidase reaction that immediately reduces PGG₂ to PGH₂ by the rate-limiting enzymes COXs (56). COXs mainly have two isoforms, COX-1 and COX-2. COX-1 is a constitutive isoform expressed in most tissues and involved in most physiological events, while COX-2 is highly expressed in response to physical, chemical and inflammatory stimuli (57). However, COX-2 is constitutively expressed in several tissues that are not associated with inflammation, such as the brain, kidney, thymus and gut (58,59). Then, PGH₂ is metabolized to PGE₂, PGD₂, PGF_2α and PGI₂ through different PG synthases (56,60). Next, a wide variety of PGs exert their biological functions by binding to their respective G protein-coupled receptors. PGE₂ is converted from PGH₂ by PGE synthase (PGES) and performs pleiotropic effects by binding to four distinct membrane PGE receptors (EP1-4). PGD synthases, including lipocalin-type PGDS (L-PGDS) and hematopoietic PGDS (H-PGDS), catalyze the isomerization of PGH₂ to PGD₂, which binds to PGD receptors (DP1 and DP2). PGI₂, an agonist of PGI receptor (IP), is generated by PGI synthase. PGF_2α is converted by PGF synthase and binds to PGF receptor (FP) (33).

The synthesized 2-series PGs affect various physiological and pathological processes, particularly the important pathogenesis of T2DM and NAFLD, which includes insulin resistance, hyperglycemia, hepatic lipid accumulation and inflammation (Fig. 2) (17-19). In the next section, the role of PGs in these pathogenic mechanisms will be further reviewed (Fig. 3).

Insulin resistance is regarded as a key pathogenic mechanism that accounts for the interplay between T2DM and NAFLD. Insulin resistance is characterized by an impaired insulin sensitivity of liver and peripheral tissues, including skeletal muscle and adipose tissue. It is a crucial contributor to the other related pathogeneses, which includes hyperinsulinemia, hyperglycemia, dyslipidemia, ectopic lipid accumulation (such as in the liver) and inflammation (Fig. 2) (61). More specifically, first, insulin resistance contributes to hyperglycemia. Hepatic insulin resistance markedly increases hepatic glucose production, while peripheral insulin resistance enhances circulating non-esterified fatty acids (NEFAs) and decreases glucose uptake, together leading to elevated glycemia (62-65). Secondly, hepatic de novo lipogenesis, a primary initiation mechanism of liver fat formation, is facilitated by compensatory hyperinsulinemia and increased substrates (such as glucose and NEFAs) under insulin-resistant status in liver (64). Thirdly, insulin resistance is of great significance in the steatosis-to-NASH progression, as it is closely linked to aggravated inflammation, apoptosis and fibrogenesis in the liver (66). As for peripheral insulin resistance, adipose insulin resistance also triggers chronic low-grade inflammation by the release of adipokines and cytokines, which in turn maintains or even exacerbates the development of T2DM and NAFLD (67,68).

Accumulating evidence has revealed the important role of 2-series PGs in the development of insulin resistance (Fig. 3A) (37). Herein, the role of 2-series PGs in both hepatic and peripheral insulin resistance was discussed.

Hyperglycemia is a hallmark of dysregulated glucose metabolism that contributes to the initiation and progression of T2DM and NAFLD (Fig. 2). When insulin resistance occurs, chronic hyperglycemia can induce insulin release by pancreatic β-cells, thus contributing to hyperinsulinemia (96). Under insulin resistance and hyperinsulinemia, elevated glycemia and circulating NEFAs can cause the deleterious impairment of various organs and tissues, processes that are referred to as glucotoxicity and lipotoxicity, respectively (97,98). In the pancreas, glucotoxicity and lipotoxicity can account for β-cell failure and subsequent insulin secretion deficiency (97-99). In addition, hepatic gluconeogenesis can be facilitated by insulin resistance, most likely contributing to hyperglycemia by increasing the hepatic glucose output. These mechanisms collectively contribute to hyperglycemia during the development of T2DM and NAFLD.

A previous study has demonstrated the close correlation between 2-series PG action and the development of hyperglycemia (Fig. 3B). First, COX-1 and COX-2 participate in the control of glycemia (100). In a 2-week clinical trial of high-dose aspirin treatment among nine patients with T2DM, aspirin treatment was revealed to reduce fasting plasma glucose and improves insulin sensitivity in cases with diabetes (101). In addition, the increased formation of PGs and PG metabolites has been observed in T2DM, including PGE₂, PGI₂ in islet or blood, and 15-keto-dihydro-PGF_2α, 8-iso-PGF_2α in urine (102-106). However, interference with the PGE₂/EP3 signaling pathway through the blockade of the EP3 receptor in mice has been reported to predispose to systemic insulin resistance; in addition, insulin secretion also increases, finally contributing to hyperglycemia (79,107). These observations imply the multifunctional involvement of 2-series PGs in the development of hyperglycemia.

β-cell failure includes β-cell dysfunction and β-cell mass deficiency, which remain the two major causes of hyperglycemic pathogenesis. β-cell dysfunction and apoptosis reduce insulin secretion and deplete β-cell mass, respectively (108). Due to their involvement in inflammation and oxidative stress signaling pathways, PGs mainly act as an initial and deteriorative pathological element for β-cell failure, leading to hyperglycemia.

Glucose-stimulated insulin secretion (GSIS) commonly occurs when β-cells are constantly exposed to high glucose stimulation (109). To a certain extent, PGs act as a potential negative modulator of GSIS. A number of studies have demonstrated that PGE₂ attenuates GSIS. PGE₂ is the predominant E-series PG in islets formed by COX-2, the dominant form of COX in the pancreas (110-112). COX-2 expression is significantly upregulated in pancreatic islets under hyperglycemic conditions (113,114). COX-2-dependent PGE₂ generation is augmented by group X secretory phospholipase A₂ and eventually suppresses GSIS in vitro and in vivo (115). PGE₂ equally inhibits both two phases of GSIS in HIT cells, which is associated with reduced cAMP accumulation mediated by pertussis toxin-sensitive G protein (Gi) (116). Among PGE receptors, EP3 receptor is the most abundant PGE receptor type in islets (112), which is overexpressed in islets from patients with T2DM (117,118). Previous research has indicated that PGE₂ coupling with EP3 receptor is highly associated with a reduction of insulin secretion in terms of β-cell dysfunction. Meng et al (118) revealed that the PGE₂-stimulated gene expression of PG EP3 receptor subtype led to intracellular cAMP reduction, accompanied by a downregulated phosphorylation level of Akt and forkhead box 'Other' (Foxo) in HIT-T15 cells (118). Kimple et al (103) confirmed that this active PGE₂/EP3 receptor pathway in islets depended on coupling to G-proteins of the Gi subfamily in vivo. In addition, EP3 receptor agonists can antagonize glucagon-like peptide-1 (GLP-1) signaling, leading to reduced cAMP production and attenuated GSIS (103). Hence, since GLP-1 treatment is not effective in all patients with T2DM (119), as a non-competitive antagonist of GLP-1 receptor, EP3 receptor may be a potent target for improving the GLP-1 effect in anti-diabetic therapeutics (103,120). Another observation revealed that PGE₂ presents an impotent influence on GSIS suppression in rat islets exposed to epinephrine-induced glucose overload (121). Under hyperglycemic states, crosstalk between PGs and other inflammatory factors has a profound effect on glycemic control. Systemic inflammatory responses are upregulated in T2DM individuals (122), characterized by elevated levels of lipid molecules, including PGs and cytokines such as TNF-α (123), IL-1β (124), IL-6 (125) and IL-8 (126), in correspondence with the decline of their natural antibodies (127). In PG signaling, COX-2 is involved in IL-1β-induced auto-stimulation in islets (111). The COX-2 expression and activity are upregulated by IL-1β-induced NF-κB activation, resulting in a negative effect on GSIS caused by increased PGE₂ via EP3 receptor (112). Recently, the IL-1β/COX-2/PGE₂ pathway loop has been revealed as the underlying mechanism for the onset and progression of diabetes, which leads to β-cell inflammatory impairments by downregulating the expression of β-cell functional genes pancreatic and duodenal homeobox 1, NK6 homeobox 1 and MAF bZIP transcription factor A (124). As previously mentioned, PGE₂ can impact GSIS through different receptors and also by interacting with inflammatory reaction in hyperglycemia.

Considering other PGs, PGI₂ also plays a pivotal role in the protection of β-cell function and survival via IP receptor signaling. The IP receptor/cAMP/PKA/nephrin signaling pathway participates in the preservation of β-cell function and mass in vitro and in vivo. IP receptor agonism augments insulin release in pancreatic β-cells and promotes the viability of MIN6 β-cells as a consequence of intracellular cAMP increase, PKA activation and subsequent nephrin phosphorylation (104). Consistently, a study revealed that selexipag, a prodrug form of IP receptor agonist, exerted a similar improvement on GSIS and β-cell mass in diabetic mice (104).

Deficient β-cell mass is recognized as another essential event that results in elevated glycemia in T2DM progression (128,129). Apart from delaying the GSIS process, COX-2/PGE₂ signaling also plays a role in the regulation of β-cell proliferation and apoptosis. In a model of transgenic mice overexpressing COX-2 and microsomal prostaglandin E synthase 1 (mPGES-1), increased PGE₂ appeared to be associated with a significant reduction in the number of β-cells and further caused severe hyperglycemia (130). A different study concluded that the blockade of EP3 receptor and activation of EP4 receptor enhanced human β-cell proliferation and survival ex vivo, suggesting a reciprocal effect of different EP receptors on the mediation of β-cell failure in T2DM (117). Furthermore, an EP3 receptor antagonist improved β-cell proliferation partly by enhancing phospholipase C-γ1 activity in young mouse islets rather than in old ones, while the EP4 receptor was activated to exert the same protective effect in human β-cells only with combination of EP3 inhibition. In terms of promotion of β-cell survival, forkhead box protein M1, a critical β-cell proliferation factor, is upregulated by EP3 antagonist and EP4 agonist in islets from obese T2DM individuals (117). However, EP4 has further been revealed to be involved in PKA signaling activation through a G_S-coupled mechanism in the survival of mouse β-cells, which is proposed to facilitate the phosphorylation of eukaryotic initiation factor 4E and PKC-ε in a putative downstream mechanism (117). In addition, α-subunit of the heterotrimeric G_z protein (Gα_z), a member of the Gα_i family, may couple to EP3 in pancreatic β-cells (131). The global deletion of Gα_z can block the PGE₂/EP3 pathway, which subsequently results in a robust increase in β-cell mass and augments GSIS by cAMP upregulation in mice with both insulin resistance and glucose intolerance (120). In addition, EP3 receptor knockout in HFD-fed mice was revealed to promote β-cell proliferation (79), which is consistent with the aforementioned Gα_z-null data. Notably, in human islets from patients with T2DM and MIN6 β-cells, palmitate can upregulate the expression levels of COX-2 and EP3 receptor, which initiates β-cell apoptosis through the COX-2/PGE₂/EP3 pathway (132). These results demonstrated that the PGE₂ pathway can inhibit proliferation and induce apoptosis in β-cells exposed to glucotoxicity and lipotoxicity.

Based on the available studies of 2-series PG-mediated glycemic control, the PGE₂ and PGI₂ signaling pathways are considered crucial pathogenic contributors in the regulation of β-cell function and proliferation, although PGE₂ exerts completely different effects on hyperglycemia through EP3 and EP4 receptors. Furthermore, PGs, mainly PGF_2α and PGI₂, can enhance circulating blood glucose by accelerating gluconeogenesis and glycogenolysis through insulin signaling (described in Hepatic insulin resistance). Enhanced glycemia can be predisposed to excessive glucose accumulation in the liver which can be converted to lipid formation. Therefore, targeting the 2-series PG pathway can be a promising therapeutic strategy for the protection and recovery of both β-cell and liver abnormalities.

Elevated circulating lipid contents (such as cholesterol, TG and NEFAs) are a characteristic of T2DM and NAFLD (Fig. 2) (133-135). Lipotoxicity alone or in combination with glucotoxicity is highly associated with the impairment of insulin sensitivity in various organs. Previous clinical studies have revealed that lipid infusion contributes to hepatic insulin resistance (136). In addition, the accumulation of hepatic lipid contents, particularly TGs, is an initiation of liver steatosis. Liver steatosis is the first hit in the progression of NAFLD, whose onset is due to insulin resistance (19,137). The direct contributors to excess lipid storage in the liver include increased circulating NEFAs, accelerated de novo lipogenesis, overloaded dietary fat and inadequate lipid oxidation (64,65,134,138). In turn, hepatic steatosis induces subacute intrahepatic inflammation through the NF-κB pathway as a pathogenic mechanism for exacerbated hepatic and systematic insulin resistance both in NAFLD and T2DM (19,139).

PGs markedly contribute to the dysregulation of the lipid metabolism in hepatic lipid accumulation (Fig. 3C). PGE₂ acts synergistically with insulin in the pathogenesis of hepatic steatosis, but their roles remain discordant and controversial. PGE₂ decreases the activity of lipogenic enzymes in primary hepatocytes in vitro through sustained ERK1/2 activation, thereby attenuating insulin-dependent phosphorylation of Akt kinase. This finally abrogates insulin signaling and further alleviates SREBP-1c pathway in hepatic de novo lipogenesis (140). Furthermore, short-term blockade of PGE₂ signaling by EP3 antagonist in mice with diet-induced obesity caused a significant reduction of TG content in skeletal muscle and slightly increased hepatic TGs (107). As a result, it can be hypothesized that PGE₂ elicits preservation against hepatic steatosis. However, other observations vary from this hypothesis. A previous study has indicated that extracellular PGD₂, PGE₂ and PGF_2α diminish the secretion of very low-density lipoprotein (VLDL)-apolipoprotein B (apoB) to promote steatosis in primary hepatocytes (141). The reduction of VLDL-apoB is correlated with decreased TG transportation and impaired cellular TG recycling, which finally results in a reduced TG output. In addition, only PGE₂ can completely antagonize the IL-6-induced secretion of VLDL-apoB in hepatocytes (141). Furthermore, PGE₂ acts synergistically with insulin and enhances the incorporation of glucose into TGs in hepatocytes. PGE₂ and insulin synergistically inhibit lipolysis, mitochondrial β-oxidation and VLDL synthesis, which are mediated by PGE₂-dependent suppression of adipose TG lipase, carnitine palmitoyltransferase-1 and apoB-mediated lipidation, respectively (140). Moreover, apoB and microsomal transfer protein are downregulated by PPAR-γ-coactivator-1α and PCK1 in insulin-dependent and PGE₂-dependent manners. In combination, these events contribute to a reduced TG breakdown and increased fat droplets in hepatocytes (140). In terms of NAFLD development in vivo, under HFD feeding, increased COX-2 activity and PGE₂ concentration also results in hepatic steatosis in mice mostly through NF-κB activation and lipid peroxidation enhancement. An aggravation of insulin resistance also appears with increased levels of serum alanine aminotransferase and total hepatic fatty acids (142). Another putative mechanism of hepatic steatosis formation involves CD36-mediated PG levels in the liver. Although the expression level of CD36 was 5-fold higher in hepatic steatosis liver than in normal liver, the global deletion of CD36 in ob/ob mice aggravated hepatic lipid accumulation by significantly suppressing the outputs of VLDL, apoB and TGs by increasing hepatic PGD₂, PGE₂ and PGF_2α (143). Based on these experiments, PGs including PGD₂, PGE₂ and PGF_2α may accelerate the initiation and progression of hepatic steatosis.

Conversely, beraprost sodium, a PGI₂ analog, was revealed to be effective in ameliorating metabolic disturbances in obesity and obesity-associated T2DM. Various manifestations were revealed to be improved by PGI₂ analog treatment, including hepatic steatosis, adipose hypertrophy, glucose intolerance, hyperglycemia, hyperinsulinemia and other related complications, such as pancreatic fibrosis and nephropathy (144). This suggests that PGI₂ can be beneficial to the treatment of obesity-associated T2DM and NAFLD.

Since the mediation of PGs on hepatic lipid accumulation is ambiguous, the precise mechanism requires further study. Considering the findings of the aforementioned studies, it can be hypothesized that various PGs (PGD₂, PGE₂ and PGF_2α) promote hepatic lipid accumulation, mostly through facilitating TG storage and inhibiting TG output by repressing lipolysis, fatty acid oxidation and VLDL synthesis. In addition, under insulin resistance, PGs can increase de novo lipogenesis and promote the development of hepatic steatosis. As hepatic lipid accumulation is the initial step of NAFLD as well as a risk factor for T2DM, the inhibition of the 2-series PG pathway may be a potential option for treating NAFLD.

Systemic chronic inflammation is a health-damaging phenotype that plays a central role in multiple metabolic syndromes, including T2DM and NAFLD (Fig. 2) (145,146). 2-Series PGs have multifunctional effects on the promotion and resolution of inflammation following the occurrence of insulin resistance, β-cell failure and hepatic steatosis (Fig. 3D) (37,71,78,79,82,112,142). PGs, TNF-α, IL-1β and IL-6 have been recognized as the major inflammatory mediators in T2DM and NAFLD (102,145-148). There are multifaceted interactions between PGs and other inflammatory molecules or cells in the pathogenic mechanisms of T2DM and NAFLD. As aforementioned, COX-2-derived PGs disrupt insulin signaling by activating the STAT3/SOCS3 signaling pathway in the liver or interacting with TNF-α and ILs in WAT (37,71,78), whereas, in obesity, PGD₂ and PGE₂ mediate macrophage polarization and infiltration with downregulated TNF-α, MCP-1 and IL-6 in WAT, leading to an improvement in peripheral insulin resistance (79,82). In turn, when peripheral insulin resistance occurs, adipokines and cytokines are released from dysfunctional adipose tissues and subsequently induce inflammation, which is associated with β-cell failure and hepatic steatosis. COX-2-derived PGE₂ contributes to β-cell dysfunction in GSIS via IL-1β-induced NF-κB activation (112), and hepatic lipid accumulation via NF-κB activation (142). NF-κB-mediated inflammation is important in the pathogenesis of T2DM and NAFLD. PGs may be involved in inflammatory processes mostly through NF-κB pathway activation with or without coaction with other inflammatory factors. The inhibitor κB kinase β (IKK-β)/NF-κB pathway plays a critical role in chronic hepatic inflammation, leading to insulin resistance and steatohepatitis, in which TNF-α and IL-1β are also involved (19,139).

Hepatic inflammation is a landmark in the development of NASH, which is also triggered by progressed insulin resistance and other injurious stimuli, such as glucotoxicity and lipotoxicity (149-152). Progressively, NASH-related hepatic fibrosis and cirrhosis become long-term manifestations of NAFLD. Hepatic fibrosis is characterized by high-density extracellular matrix protein deposition (153). Both NASH and liver fibrosis can be exacerbated in NAFLD with comorbidity of T2DM (154,155). T2DM-promoted NASH is attributed to peripheral insulin resistance, intrahepatic lipotoxicity and M1 macrophage recruitment in adipose tissue (156-158). The activated M1 macrophages secrete pro-inflammatory cytokines, including MCP-1, TNF-α and IL-1β, which induce systemic inflammation. These cytokines are further delivered to the liver and cause steatohepatitis (159). Therefore, insulin resistance, hyperglycemia, hyperinsulinemia and hyperlipidemia are key factors for pro-inflammatory status in hepatic inflammation, in which PGs are involved as inflammatory mediators. In addition, the upregulation of transforming growth factor-β (TGF-β) and connective tissue growth factor in T2DM can lead to NAFLD-related fibrosis progression (160,161).

PGs are correlated with the progression from hepatic steatosis to NASH. The upregulated expression of COX-2 and mPGES-1, the key enzymes of PGE₂ synthesis, is closely associated with NASH activity score in human liver from patients with NASH (162). Lipidomics profiling was performed in a clinical cohort that attempted to describe the hepatic inflammatory characteristic of NASH. As a result, the plasma PGE₂ level was revealed to be elevated only in patients with NASH, while the level of 13,14-dihydro-15-keto-PGD₂, a metabolite degraded from PGD₂, was found to be remarkably higher in the NASH group, compared with the simple steatosis or control groups (163). As a consequence, it is reasonable to suggest that PGs, particularly PGE₂, may aggravate the course of NAFLD.

However, there are some discrepancies in the impact of PGs and COX-2 activity on the development of NASH under dietary nutritious stress. An in vivo study revealed that hepatocyte-specific COX-2 transgenic mice (hCOX-2-Tg) with an increased level of PGE₂ improved intrahepatic steatosis, ballooning and inflammation (164). This was partially achieved by decreasing the plasma levels of pro-inflammatory cytokines (such as IL-1β, IL-6, TNF-α and MCP-1), and inhibiting macrophage recruitment and infiltration in steatohepatitis liver induced by a methionine- and choline-deficient diet (MCDD) (164). In addition, there are ameliorations of augmented oxidative stress and apoptosis in liver samples with NASH (164). Similarly, under a NASH diet, hepatic PGE₂ production derived from mPGES-1 is increased to potently inhibit monocyte-derived macrophage infiltration, which is associated with PGE₂-induced suppression of TNF-α-triggered responses in hepatocytes. These responses consist of pro-inflammatory cytokine IL-1β production and hepatocyte apoptosis (162). These results suggest a combined action of PGs and other inflammatory factors in NASH development. Furthermore, the blockade of L-PGDS in PGD₂ signaling rapidly accelerates non-alcoholic simple steatosis to severe steatohepatitis in nutrition overload or normal conditions (165). This progression to NASH is also accompanied by enhanced lipogenic gene expression (such as SREBP-1c and liver X receptor α) and deranged metabolic features, including progressed insulin resistance and increased fasting glucose, insulin and lipid levels in the blood (165). With regards to the PGI₂/IP pathway, under MCDD conditions, IP-receptor-knockout (IP-KO) mice had accelerated progression to steatohepatitis, with greater iron deposition due to marked oxidative stress. PGI₂-IP signaling prevents the development of NASH in anti-inflammatory response, as evidenced by the suppressed expression of MCP-1 and TNF-α in lipopolysaccharide-stimulated Kupffer cells in vitro. Consistently, the Kupffer cell-induced expression levels of MCP-1 and TNF-α were progressively increased in IP-KO mice, and the oxidative stress-induced hepatic iron deposition was reduced in the MCDD-induced steatohepatitis liver, suggesting that PGI₂ signaling inhibits inflammation and influences the antioxidant reaction in NASH (166). Thus, PG appears to play a protective role against hepatic steatohepatitis, most likely under disturbed metabolism in NAFLD and T2DM progression.

The key mechanisms of hepatic fibrosis include a disbalance between fibrogenesis and fibrinolysis and the activation of hepatic stellate cells (HSCs) and Kupffer cells in response to various stimuli (167,168). Numerous studies have revealed that PGs facilitate the development of hepatic steatosis, steatohepatitis and fibrosis (169-171). In a prospective cohort research of 361 patients with NAFLD, daily aspirin use induced less severity of histologic characteristics of NAFLD and significantly decreased the risk of fibrosis initiation and progression in a duration-dependent manner, when compared with the non-regular use of aspirin (172). It was further suggested that the antifibrotic effect of long-term aspirin treatment is attributed to its involvement in inhibiting NF-κB and IKK-β signaling (173). Furthermore, plasma bioactive lipids, such as PGE₂ and PGI₂, have been regarded as useful markers for prognosis in liver cirrhosis (174). In accordance with clinical evidence, the upregulation of COX-2 was positively correlated with fibrosis formation in liver from a carbon tetrachloride (CCl₄)-induced fibrotic mouse model (175). Conversely, it was revealed that COX-2-derived PGE₂ could suppress fibrogenesis and NASH progression (176,177). In hCOX-2 Tg mice with diet-induced NASH, PGE₂ attenuated CCl₄-induced liver fibrosis by decreasing the activation and proliferation of HSCs and increasing apoptosis by suppressing microRNA (miR)-23a and miR-28a expression (164,178). In addition, COX-2-derived PGE₂ was revealed to suppress collagen synthesis through the downregulation of collagen type I α1, α smooth muscle actin and collagen binding protein-1 in HSCs under TGF-β1-induced conditions (178,179). These results demonstrated that the COX-2/PGE₂ pathway prevents the development of liver fibrosis through growth-suppressive and pro-apoptotic effects on HSCs.

To sum up, during the progression of T2DM and NAFLD, PGs may primarily act by interacting with other inflammatory factors, as well as mediating the NF-κB signaling pathway, which plays an important role in the chronic inflammation caused by glucotoxicity and lipotoxicity in a variety of organs. PGs can serve as pro-inflammatory mediators in the impairment of insulin sensitivity, glycemia and hepatic lipid metabolism. However, PGD₂, PGE₂ and PGI₂ also exert anti-inflammatory effects and improve peripheral insulin resistance, NASH and related fibrosis. Due to the complex action of PGs in the inflammatory process, the use of COX inhibitors in T2DM and NAFLD treatment should be given more consideration, and further explorations are highly warranted.

The comorbidity of T2DM and NAFLD is well recognized and has become an area of increased investigation over past decades. Nowadays, considerable evidence has highlighted the roles of 2-series PGs in the pathogenesis of T2DM and NAFLD. 2-Series PGs are important lipid molecules that are widely distributed in various organs. These exert multifunctional effects on the four highly intertwined pathogeneses of T2DM and NAFLD, including insulin resistance, hyperglycemia, hepatic lipid accumulation and chronic inflammation. PGs potently mediate insulin resistance, which subsequently induces pathological alterations including hyperinsulinemia, hyperglycemia, dyslipidemia and ectopic lipid accumulation. In addition, PGs can directly impact hyperglycemia by decreasing insulin secretion, pancreatic β-cell proliferation and increasing gluconeogenesis. In addition, PGs contribute to hepatic lipid accumulation by enhancing hepatic lipogenesis and decreasing TG output. Moreover, PGs distinctly establish a close interaction with inflammatory processes in the progression of T2DM and NAFLD.

Most 2-series PGs exert negative effects on the progression of T2DM and NAFLD. Therefore, the application of COX inhibitors such as aspirin and celecoxib beyond their conventional use on vascular diseases, rheumatoid arthritis and pain is emerging as a promising option for T2DM and NAFLD treatment. However, certain aspects of the application of PG pathways should be considered. First, some PGs are beneficial to the prevention of T2DM and NAFLD development to a certain extent, suggesting that the clinical use of COX inhibitors requires careful consideration and highlighting the potential therapeutic use of PGs and their derivates in the prevention and control of T2DM and NAFLD. Secondly, since the existing NSAIDs and COXIBs are associated with several side effects, it is meaningful to perform molecular modification of these drugs and develop new treatment strategies, to aim to accurately modulate the PG pathway in related organs such as the pancreas, liver and adipose tissues. Overall, due to the important role of 2-series PGs in T2DM and NAFLD, additional studies associated with the molecular mechanisms of PGs in the pathogenesis of T2DM and NAFLD are highly warranted. These studies will provide new and more precise therapeutic strategies based on targeting PG pathways in the treatment of these two diseases.