Effect of ω-3 Polyunsaturated Fatty Acids-Derived Bioactive Lipids on Metabolic Disorders

Arachidonic acid (ARA) is an important ω-6 polyunsaturated fatty acid (PUFA), and docosahexaenoic acid (DHA), eicosapentaenoic acid (EPA) and n-3 docosapentaenoic acid (n-3 DPA) are three well-known ω-3 PUFAs. These fatty acids can be metabolized into a number of bioactive lipids. Eicosanoids derived from ARA have drawn great attention because of their important and complex biofunctions. Although EPA, DHA and n-3 DPA have also shown powerful biofunctions, we have fewer studies of metabolites derived from them than those from ARA. Recently, growing research has focused on the bioaction of ω-3 PUFA-derived metabolites, which indicates their great potential for treating metabolic disorders. Most of the functional studies of these bioactive lipids focused on their anti-inflammatory effects. However, several studies elucidated their direct effects on pancreatic β cells, hepatocytes, adipocytes, skeletal muscle cells, and endothelial cells. These researches revealed the importance of studying the functions of metabolites derived from ω-3 polyunsaturated fatty acids other than themselves. The current review summarizes research into the effects of ω-3 PUFA-derived oxylipins on metabolic disorders, including diabetes, non-alcoholic fatty liver disease, adipose tissue dysfunction, and atherosclerosis.


INTRODUCTION
Polyunsaturated fatty acids (PUFAs) refer to fatty acids with two or more double bonds in their backbone. Arachidonic acid (ARA) is an important ω-6 PUFA, which can be metabolized from linoleic acid (Schmitz and Ecker, 2008). Docosahexaenoic acid (DHA), eicosapentaenoic acid (EPA) and n-3 docosapentaenoic acid (n-3 DPA) are three well-known ω-3 PUFAs and they can be derived from α-linolenic acid (ALA). The estimated conversion rate of ALA to EPA was 8-20% in human, while that to DHA was 0.5-9%, even lesser (Stark et al., 2008). Those PUFAs are precursors of a series of bioactive lipids metabolized by cyclooxygenase (COX), lipoxygenase (LOX), and cytochrome P450s (CYPs) and autoxidized non-enzymatically (Zhang et al., 2015).
Eicosanoids derived from ARA have drawn great attention because of their important and complex biofunctions. Many studies have examined the functions of ARA metabolites, including prostaglandins, thromboxanes, leukotrienes, lipoxins hydroxyeicosatetraenoic acids, and epoxyeicosatrienoic acid. These metabolites play vital roles in many physiological and pathophysiological processes. The effects of dietary supplement of ω-3 PUFAs are mediated not only by the precursor per se and their metabolites but also by competing the enzymes with ARA in the eicosanoid-producing process (Calder, 2020b). The effects of ARA and ARA-derived eicosanoids are well documented by several reviews (Sonnweber et al., 2018;Calder, 2020b). However, although ω-3 PUFAs also showed powerful biofunctions, we have fewer studies of their derived metabolites than those of ARA. Thus, we focused on the ω-3 PUFA derived bioactive lipids in the current review.
Metabolic disorders, such as obesity, diabetes, non-alcoholic fatty liver disease (NAFLD), and cardiovascular disease greatly threaten human health, and the prevalence of the diseases is increasing worldwide (Lavie et al., 2009;Younossi et al., 2016Younossi et al., , 2018Glovaci et al., 2019). In metabolic diseases, the profile of metabolites derived from ω-3 PUFAs is changed because of disturbed PUFA metabolism (Wang et al., 2017;Laguna-Fernandez et al., 2018;Garcia-Jaramillo et al., 2019). In the current review, we summarize the growing research into the effect of ω-3 PUFA-derived bioactive lipids on metabolic disorders, including diabetes, NAFLD, adipose tissue dysfunction and atherosclerosis.
THE METABOLIC PATHWAYS OF ALA, EPA, DHA, AND n-3 DPA The metabolic pathways of ALA, EPA, DHA, and n-3 DPA were profoundly described by several reviews (Gabbs et al., 2015;Kuda, 2017;Drouin et al., 2019) and we briefly summarized as below: ALA can be metabolized into hydroxy fatty acids by the COX and LOX pathway and epoxygenated fatty acids by the CYP pathway (Gabbs et al., 2015). In addition, ALA is the precursor of EPA, n-3 DPA and DHA. The rate limiting step is addition of a fourth double bond by -6 desaturase. Next by elongation and desaturation, EPA is produced (Stark et al., 2008). EPA can be metabolized into 3-series prostaglandins and thromboxanes by the COX pathway; hydroxyeicosapentaenoic acids (HEPEs), E-series resolvins (RvE; RvE1-E3), 5-series leukotrienes and lipoxins by the LOX pathway; and epoxyeicosatetraenoic acids (EEQs) and dihydroxyeicosatetraenoic acids (diHETEs) by the CYP pathway (Zhang et al., 2015). Of note, 18-HEPE is derived from EPA by the CYP pathways or by aspirin-acetylated COX2 and then metabolized into RvEs by the LOX pathway (Figure 1; Gabbs et al., 2015).
In the present review, we focus on the oxylipins enzymatically derived from ω-3 PUFA.

THE IDENTIFIED RECEPTORS OF ω-3 PUFA-DERIVED OXYLIPINS
Identifying the receptors of these lipid mediators is vital to investigate their functions and the underlying mechanisms. Several studies have revealed that the effects of metabolites derived from ω-3 PUFA are mediated by G proteincoupled receptors (GPCRs) or nuclear receptors (Table 1). Krishnamoorthy et al. (2010) reported that RvD1 can directly bind to two GPCRs, ALX, and GPR32. ALX was first identified as an LXA4 receptor and GPR32 was considered an orphan receptor. The authors further revealed that RvD1-stimulated phagocytosis in macrophages was mediated by ALX and GPR32 (Krishnamoorthy et al., 2010). GPR18 is identified as a RvD2 receptor (Chiang et al., 2015). The protective effects of PDX on oxidative stress in vascular endothelial cells were mediated by GPR120, thus GPR120 may be a putative receptor of PDX (Hwang et al., 2019). MaR1 derived from DHA specifically binds to and activates human leucine-rich repeat containing G protein-coupled receptor 6 (LGR6) (Chiang et al., 2019). RvE1 binds to leukotriene B4 receptor 1 (BLT-1) and ERV-1 (also known as ChemR23) . 5-HEPE is an agonist of GPR119, a GPCR that regulates insulin secretion in pancreatic β cells (Kogure et al., 2011).

EFFECT OF ω-3 PUFA-DERIVED OXYLIPINS ON DIABETES
Type 1 diabetes is described as immune-mediated destruction of pancreatic β cells, and the characteristics of type 2 diabetes are insulin resistance and progressive β-cell failure (Yang et al., 2018). Diabetes is a major metabolic disorder with high prevalence and is a risk factor for relevant public health issues such as cardiovascular disease, retinopathy, microangiopathy, and impaired wound healing (Yang et al., 2018).
EPA and DHA have shown beneficial effects for both type 1 and type 2 diabetes in rodents (Krishna Mohan and Das, 2001;Suresh and Das, 2003;Bi et al., 2017;Lepretti et al., 2018) and there is increasing evidence that the metabolites of EPA and DHA regulate these procedures. However, clinical trials showed conflicting results of dietary supplement of EPA/DHA on metabolic parameters in diabetic patients. A 6-month EPA treatment decreased postprandial glucose level of newly diagnosed impaired glucose metabolism patients (Sawada et al., 2016). Another clinical research also revealed the beneficial effects of ω-3 PUFA supplement on metabolic parameters including glucose and glycosylated hemoglobin in type 2 diabetic patients (Jacobo-Cejudo et al., 2017). However, several clinical studies revealed neutral effects of ω-3 PUFAs on metabolic profiles in type 2 diabetic patients (Wong et al., 2010;Poreba et al., 2017). The disagreement of these studies may be related with different sample sizes, baseline characteristics of patients, different doses and purities of these fatty acid, different time courses of the treatments and different basic medicine of these patients. Moreover, Poreba et al. (2017) also demonstrated highdose ω-3 PUFAs did not increase RvD1 level in patients with atherosclerosis and type 2 diabetes and this is an important clue that the production of bioactive metabolites of ω-3 PUFAs is related to their therapeutic effects (Poreba et al., 2017). Thus, to study the effects and mechanism of ω-3 PUFA-derived FIGURE 2 | Biosynthesis of n-3 DPA-derived bioactive lipids. RvT1: 7,13R, 20-tri hydroxy-n-3 DPA; RvT2: 7,12,13R-tri hydroxy-n-3 DPA; RvT3: 7,8,13R-trihydroxy-n-3 DPA; RvT4: 7,13R-dihydroxy-n-3 DPA. metabolites is important to develop new strategies to confront diabetes.

Effect of DHA-Derived Oxylipins on Diabetes
The levels of RvD1 and 17-HDHA were decreased in adipose tissue of genetic as well as diet-induced obese mice (Neuhofer et al., 2013). 17-HDHA treatment was further found to improve adipose tissue inflammation and insulin sensitivity in highfat-diet (HFD)-fed mice (Neuhofer et al., 2013). Also, RvD1 has beneficial effects on insulin resistance. Hellmann et al. (2011) demonstrated that RvD1 improved glucose tolerance and increased insulin-stimulated pAkt level in liver, adipose tissue and skeletal muscle in db/db mice. The authors further found that RvD1 increased the ratio of M2 and M1 adipose-tissue macrophages (Bathina et al., 2020) and ameliorated adipose tissue inflammation (Hellmann et al., 2011). RvD1 was also reported to improve insulin resistance through the PI3K-Akt-mTOR axis in brain tissue (Bathina et al., 2020). In vitro study also indicated that RvD1 could attenuate interferon γ (IFN-γ)/lipopolysaccharideinduced pro-inflammatory cytokine expression in macrophages (Titos et al., 2011). Collectively, RvD1 improves insulin sensitivity by inhibiting tissue inflammation. Moreover, RvD1 ameliorated streptozotocin induced type1 diabetes in mice (Bathina and Das, 2021). In addition, local RvD1 delivery can accelerate  wound closure in diabetic mice by stimulating macrophage phagocytosis to enhance clearance of apoptotic cells (Tang et al., 2013). In vitro study demonstrated RvD1 reduced the proangiogenic potential of retinal photoreceptors treated by high glucose by increasing anti-angiogenic miRNAs and decreasing VEGF content in exosomes (Maisto et al., 2020). PDX-treated mice showed protection from lipid-induced insulin resistance. Along with this effect, PDX inhibited lipidinduced secretion of C-C motif chemokine ligand (CCL) 2, CCL5, tumor necrosis factor α (TNF-α), IFN-γ, interleukin 1β (IL-1β), IL-2, and IL-17. However, IL-6 level was significantly increased with PDX treatment, which was from skeletal muscle and suppressed gluconeogenic gene expression in liver (White et al., 2014). In addition, PDX can activate AMPK independent of IL-6 (White et al., 2014). Consistent with this finding, PDX improved HFD-induced insulin resistance in mouse skeletal muscle and palmitate-induced insulin resistance in skeletal muscle cells by activating AMPK and increasing PPARα expression (Jung et al., 2017). In hepatocyte, PDX ameliorated palmitate-induced insulin resistance by downregulating the expression of fetuin-A and selenoprotein P . Fetuin-A and selenoprotein P were hepatokines and their levels were increased in the plasma of obesity patients . PDX also improved lipopolysaccharide-induced insulin resistance in adipocytes (Jung et al., 2018a).
MaR1 treatment ameliorated insulin resistance in db/db mice and HFD-fed mice by suppressing inflammation and improving insulin sensitivity in adipose tissue (Martinez-Fernandez et al., 2017). The effects of MaR1 on insulin sensitivity were also confirmed in human adipocytes which were mediated by improving Akt activation (Martinez-Fernandez et al., 2020).
Soluble epoxide hydrolase (sEH) is a member of the epoxide hydrolase family in the CYP pathway (He et al., 2016). It hydrolyses EDPs into DiHDPAs. sEH expression and activity was found increased in retinas of diabetic mice, and the level of its product 19,20-DiHDPA was elevated in eyes. However, levels of other sEH substrates and products were comparable between control and diabetic mice. 19,20-DiHDPA was further found to increase endothelial cell permeability and induce the migration of pericytes into the extravascular space (Hu et al., 2017). Of note, the expression of sEH was increased in retinas of patients with non-proliferative diabetic retinopathy as compared with non-diabetic individuals (Hu et al., 2017), so sEH has potential as a therapeutic target of diabetic retinopathy.

Effect of EPA-Derived Oxylipins on Diabetes
BLT-1 and ERV-1 are two receptors for RvE1 . In type 2 diabetic patients' neutrophils, ERV-1 expression was significantly upregulated and BLT-1 expression was decreased. In addition, the serum level of RvE1 was decreased in type 2 diabetic patients versus healthy controls. RvE1 was further found to facilitate neutrophil phagocytosis from healthy individuals, and a higher dose was needed to achieve a similar response in neutrophils of diabetic patients . These data indicate that repressed RvE1 signaling is involved in neutrophil phagocytosis dysfunction in type 2 diabetes. In addition, overexpression of the RvE1 receptor ERV-1 in myeloid cells attenuated diet-induced obesity, hepatic steatosis and glucose intolerance in mice. A mechanism study revealed that ERV-1 overexpression maintained peripheral blood monocyte and adipose-tissue macrophage skewing to an M2 phenotype in mice with an HFD (Sima et al., 2017). Besides, RvE1 was reported to improve hyperinsulinemia and hyperglycemia in HFD fed mice by activating ERV-1. The authors further demonstrated genetic diversity and variability defined the therapeutic effects of RvE1 by using the diversity outbred mice. This research highlights the genetic variants in the RvE1 response need to be considered when exploring the therapeutic effects of EPA clinically (Pal et al., 2020).
Eicosapentaenoic acid could increase glucose-stimulated insulin secretion from ob/ob mice (Neuman et al., 2017) 5-HEPE derived from EPA could increase glucose-stimulated insulin secretion in MIN6 cells by activating the GPR119/cAMP pathway (Kogure et al., 2011). Thus, the effect of EPA on insulin secretion may be mediated by its metabolites, which needs further investigation.

Effect of ω-3 PUFA-Derived Oxylipins on Non-alcoholic Fatty Liver Disease
Non-alcoholic fatty liver disease is defined as the accumulation of excess fat in the liver in the absence of excessive alcohol drinking and any secondary cause and thus a hepatic manifestation of metabolic syndrome (Ahmed, 2015). In NAFLD, simple steatosis can progress into non-alcoholic steatohepatitis (NASH), estimated to be the major reason for liver transplantation in the United States by 2020 (Diehl and Day, 2017). EPA and DHA showed protective effects on NAFLD (Scorletti and Byrne, 2018;Yan et al., 2018;Jordao Candido et al., 2019). Moreover, to better understand the underlying mechanisms, increasing studies have focused on the functions of their derived metabolites in NAFLD.

Effect of ω-3 PUFA-Derived Derived Oxylipins on Non-alcoholic Steatohepatitis
Non-alcoholic steatohepatitis is characterized by liver steatosis, inflammation, hepatocellular injury and different degrees of fibrosis and is the progressive form of NAFLD (Schuster et al., 2018). A recent study found RvD1 treatment mitigated lipid accumulation, inflammation and hepatic fibrosis in MCD-diet induced NASH mice. For the mechanism, RvD1 suppressed oxidative stress by activating nuclear factor E2-related factor 2 and ameliorated inflammation by inhibiting NF-κB and MAPK signaling pathways (Li et al., 2020). In addition, RvD1 had additional protective effects on calorie restrictive-improved NASH, as evidenced by decreased macrophage infiltration with decreased expression of M1 macrophage markers and increased expression of M2 macrophage markers (Rius et al., 2014). Also, Han et al. (2019) demonstrated that MaR1 derived from DHA increased the M2 polarity of liver macrophages and then ameliorated NASH by activating RORα. RORα, as a nuclear receptor, in turn increased MaR1 production by transcriptional induction of 12-lipoxygenase expression (Han et al., 2019). These studies suggest that these specialized pro-resolving lipid mediators derived from ω-3 PUFAs have therapeutic potential for NASH by promoting M2 polarization of liver macrophages.

EFFECT OF ω-3 PUFA-DERIVED OXYLIPINS ON ADIPOSE TISSUE FUNCTION
Depending on the adipocyte, adipose tissue can be divided into white and brown adipose tissue. Also, inducible cells within white adipose tissue, called "beige" adipocytes, can generate heat under cold exposure (Rosen and Spiegelman, 2014;Ye et al., 2020). Adipose tissue functions, including adipose tissue inflammation, lipolysis, adipogenesis, endocrine function, and browning, are closely related to obesity-related diseases. The studies of the effects of ω-3 PUFA derivatives on adipose tissue function mainly focused on the immune response of adipose tissue. Their influence on macrophage function contributing to adipose tissue inflammation was discussed in the previous section ( Table 2). In addition, Onodera et al. (2017) demonstrated that EPA increased the number and proportion of T regulatory cells in epididymal adipose tissue of db/db mice. This result was mediated by 5-HEPE, which is derived from EPA by 5-LOX (Onodera et al., 2017).
In addition to the immune response, other adipose tissue functions are regulated by ω-3 PUFA-derived bioactive metabolites.

Effect of ω-3 PUFA-Derived Oxylipins on Lipogenesis and Lipolysis
The imbalance of lipogenesis and lipolysis of adipose tissue can increase the risk of obesity-induced disease (Lafontan, 2014). MaR1 inhibited TNF-α-induced lipolysis in 3T3-L1 adipocytes (Laiglesia et al., 2018a). Increased adipocyte lipolysis may increase plasma free fatty acid level and lead to insulin resistance and fatty liver disease (Matsuzaka and Shimano, 2011).
Nevertheless, PDX treatment inhibited lipid accumulation in 3T3-L1 cells during differentiation (Jung et al., 2018a). GPR120, also called free fatty acid receptor 4, is a free fatty acid receptor. Recently, DHA is found to promote adipogenesis by activating GPR120 in the cilia of preadipocytes. For the mechanism, GPR120 activation induced a rapid increase in ciliary cyclic AMP (cAMP) level, which in turn promoted adipogenesis by activating exchange factor directly activated by cAMP (EPAC) (Hilgendorf et al., 2019). Because GPR120 can be activated by PDX, this research implies the complicated effects of ω-3 PUFA-derived bioactive lipids on adipogenesis. In addition, more studies are needed to demonstrate whether ω-3 derived bioactive lipids can affect the lipid storage and release function of adipose tissue in vivo.

Effect of ω-3 PUFA-Derived Oxylipins on Brown and Beige Adipose Tissue
Brown and beige adipocytes, as heat-producing cells, are considered to counteract metabolic diseases, including obesity and type 2 diabetes. Leiria et al. (2019) found that the 12-LOX biosynthetic pathway was activated in brown adipose tissue under cold exposure, which promoted the generation and release of 12-HEPE. Then, 12-HEPE exerted a glucose-shuttling effect on tissues to support thermogenesis (Leiria et al., 2019).
GPR120 is highly expressed in brown adipose tissue and significantly upregulated in beige adipose tissue induced by cold exposure. It was further found to mediate ω-3 PUFA-induced thermogenic gene expression in beige adipocytes by upregulating fibroblast growth factor 21 expression (Quesada-Lopez et al., 2016). However, the role of ω-3 PUFA metabolites in white adipose tissue browning remains unknown. GPR120 can be activated by PDX, but whether these ω-3 PUFA-derived bioactive lipids could regulate this process is worth studying.
Besides the direct effects on adipose tissue, ω-3 PUFA metabolites are reported to indirectly regulate adipose tissue function. GPR18, the receptor for RvD2, is widely expressed in hypothalamus and was decreased in level by HFD feeding in mice. In addition, the production of hypothalamic RvD2 was decreased in HFD-fed mice. When obese mice were treated with intra-cerebroventricular injection of RvD2, visceral fat was reduced, and hypothalamic leptin resistance was reversed (Pascoal et al., 2017).

EFFECT OF ω-3 PUFA-DERIVED OXYLIPINS ON ATHEROSCLEROSIS
Atherosclerosis causes ischemic heart disease, strokes, and peripheral vascular disease (Kobiyama and Ley, 2018). Metabolic syndrome is responsible for the initial disease and disease progression (Varghese et al., 2018). Endothelial-cell dysfunction is the initial step of atherosclerosis. Plaque is chronically built up with the assistance of macrophages differentiated from monocytes, smooth muscle cells and multiple chemokines and growth factors (Gimbrone and Garcia-Cardena, 2016). The metabolites derived from EPA or DHA,including RvE1,RvD2,MaR1,and 17,have shown positive effects on anti-atherosclerosis ( Table 2).
In the ApoE-/-mouse aorta, RvD2 and MaR1 levels are correlated negatively with vulnerability plaque index, which is decreased by HFD treatment. In addition, RvD2 and MaR1 administration suppressed atheroprogression. The protective effects of RvD2 and MaR1 on atherosclerosis were mediated by preventing the macrophage inflammatory response (Viola et al., 2016). RvD1 was decreased in vulnerable regions as compared with stable regions in human carotid atherosclerotic plaques. Additionally, its level was decreased in advanced versus early atherosclerotic lesion in western diet-fed mice deficient in low-density-lipoprotein receptor (Fredman et al., 2016). These studies suggest that several metabolites of EPA and DHA are beneficial for atherosclerosis. However, more in vivo and mechanistic studies are needed to better understand their effects on atherosclerosis.

EFFECT OF n-3 DPA AND ITS DERIVATIVES ON METABOLIC DISORDERS
n-3 DPA, an important ω-3 PUFA, is also a precursor of various docosanoids. Besides, it is an important intermediate in the conversion pathway of EPA and DHA (Figure 2; Drouin et al., 2019). n-3 DPA supplement significantly improved homeostasis model assessment of insulin resistance (HOMA-IR) in HFD fed mice, while DHA and EPA showed a minor effect (Guo et al., 2018). In human, n-3 DPA and its proresolving mediators have beneficial effects on cardiometabolic disease . Moreover, It has been proved to be more potent than EPA in inducing the differentiation process in preadipocytes, and inhibits the pro-inflammatory signaling pathways (Murali et al., 2014). Although, it showed more beneficial effects on those metabolic disorders mentioned above than EPA and DHA, the functions of its metabolites are poorly studied. n-3 DPA can be metabolized into PD n-3DPA , RvD n-3DPA , MaR n-3DPA , hydroxylated derivatives from n-3 DPA, 13-serie Rvs etc. Several functional studies about n-3 DPA derivatives indicates their anti-inflammation function. A recent study found significant decreases in plasma RvD n-3DPA concentrations in CVD patients and RvD n-3DPA reduce leukocyte and platelet activation in peripheral blood from healthy volunteers as well as CVD patients. In addition, RvD5 n-3DPA reduced aortic lesions in western diet-fed ApoE-/-mice (Colas et al., 2018). PD n-3DPA also found to play an important role in regulating macrophage resolution responses (Pistorius et al., 2018). PD1 n-3DPA and RvD5 n-3DPA were reported to decrease leukocyte-endothelial interaction and attenuate intestinal inflammation (Gobbetti et al., 2017). Although these n-3 DPA derivatives are identified as novel specialized proresolving lipid mediators, their effects on metabolic disorders, such as diabetes, NAFLD, obesity and atherosclerosis are still largely unknown.

CONCLUSION
Although we have fewer studies of the biofunctions of ω-3 PUFAderived bioactive lipids than ARA metabolites, the former have been increasingly emphasized recently, especially for metabolic disorders ( Table 2). Most of the functional studies focused on their anti-inflammatory effects. These metabolites can be more effective against inflammation than the precursors per se. Because PUFAs are vulnerable to lipid peroxidation, ω-3 PUFA supplement can lead to increased lipid peroxidation products, which may limits their clinical applications (Zaloga, 2021). It is important to increase their anti-inflammatory efficiency and decrease the dosage. Therefore, studying the function ω-3 PUFA metabolites may help us to find novel lipid mediators to treat metabolic disorders better than dietary supplement of EPA and DHA. Moreover, the anti-inflammatory efficiency of these metabolites should be further compared to provide more information for the future clinical applications.
In addition, several studies revealed the direct effects of ω-3 PUFA-derived oxylipins on pancreatic β cells, hepatocytes, adipocytes, skeletal muscle cells and endothelial cells. These bioactive lipids may have potential effects other than antiinflammatory effects, which needs more exploration. Metabolites derived from ω-3 PUFAs are numerous, with attention to RvEs, RvDs, and PDs. Other metabolites such as EEQs, EDPs, HEPEs, and n-3 DPA derivatives need more mechanistic studies. In addition, the explorations of the biofunctions of ω-3 PUFAderived bioactive metabolites, including their effects on cellular function, tissue micro-environment and interactions among metabolic tissues, are important for understanding their roles in energy metabolic disorders and related diseases. We also need more studies to identify their receptors and elucidate the downstream signaling pathway, which may provide potential therapeutic strategies for metabolic disorders.
In animal studies, the age, sex, and background of animals are well controlled. However, plasma and tissue levels of EPA and DHA and their metabolites in human can be altered by age, sex and disease status (Calder, 2020a), which indicates the complexity of clinical application of EPA and DHA. The genetic variants in the specialized pro-resolving mediator response also need to be considered when exploring the therapeutic effects of EPA and DHA clinically. Thus, the individualized treatment regimens of clinical applications of ω-3 PUFAs may achieve better effects on metabolic disorders. Moreover, according to a recent clinical trial, high-dose ω-3 PUFA supplement failed to increased RvD1 levels in diabetic patients, indicating the importance to study the disturbance of ω-3 PUFA metabolism in some disease status.

AUTHOR CONTRIBUTIONS
JD and YS contributed to the drafting, figure composition, subsequent edits, and final composition of the manuscript. XZ provided the comments and corrections. CW contributed to the concept and design, drafting of the manuscript, and guarantor of the manuscript. All authors have read and approved the final manuscript.

FUNDING
This work was supported by the National Natural Science Foundation of China (81822006 and 81770836) and the Natural Science Foundation of Tianjin (20JCYBJC01120).