Key Role of 12-Lipoxygenase and Its Metabolite 12-Hydroxyeicosatetraenoic Acid (12-HETE) in Diabetic Retinopathy

Abstract

Purpose

Abnormal lipid metabolism has been proved to be implicated in the complex pathogenesis of diabetic retinopathy (DR). 12-lipoxygenase (12-LOX) is a member of lipoxygenase family responsible for the oxygenation of cellular polyunsaturated fatty acids to produce lipid mediators which modulate cell inflammation. This review explores the role of 12-lipoxygenase and its products in the pathogenesis of DR.

Methods

A comprehensive medical literature search was conducted on PubMed till September 2021.

Results

Emerging evidence has demonstrated that 12-LOX and its main product 12- hydroxyeicosatetraenoic acid (12-HETE) activate retinal cells, especially retinal vascular endothelial cells, through the activation of NADPH oxidase and the subsequent generation of reactive oxygen species (ROS), mediating multiple pathological changes during DR. Genetic deletion or pharmacological inhibition models of 12-LOX in mice show protection from DR.

Conclusion

12-LOX and its product 12-HETE take important part in DR pathogenesis and show their potential as future therapeutic targets for DR. Further studies are needed on the specific mechanism including 12-LOX pathway related molecules, 12-HETE receptors and downstream signaling pathways.

Keywords:

• Diabetic retinopathy;
• 12-lipoxygenase;
• 12-HETE;
• retinal vascular endothelial cells;
• inflammation;
• oxidative stress

Introduction

Diabetic retinopathy (DR), a severe microvascular complication in both type 1 and type 2 diabetes, is the main cause of vision loss in working-age people.¹ Despite the existence of diverse strategies (e.g., intravitreal drugs, laser photocoagulation, and vitreous surgery), the treatment of DR remains a challenge due to its complex pathogenesis. A better understanding of the role of inflammation and the related molecular pathways may help determine early goals for the prevention and intervention of DR.

In recent years, a growing number of studies have emphasized on the close relationship between abnormal lipid metabolism and DR development.^2,3 Polyunsaturated fatty acids (PUFAs) are a class of essential fatty acids that cannot be synthesized endogenously in required amounts and rely on dietary intake.⁴ Among them, ω-3 and ω-6 PUFAs have attracted much attention, and the imbalance of ω-3/ω-6 PUFAs is associated with many chronic conditions, such as cardiovascular disease, diabetes, cancer, obesity, autoimmune diseases, rheumatoid arthritis, asthma and depression.⁵ In the retina, the primary ω-3 and ω-6 PUFA is docosahexaenoic acid (DHA) and arachidonic acid (AA), respectively.⁶ The membrane-bound PUFAs are both released from the second carbon group of glycerol by activated phospholipase A2 (PLA2), and then further converted to functionally active metabolites through 3 major enzymatic pathways – the cyclooxygenases (COXs), lipoxygenases (LOXs) and cytochrome P450 oxidases (CYPs).^7–9 LOX is a family of non-heme iron-containing enzymes widely distributed in the plant and animal kingdoms, which can stereospecifically insert molecular oxygen into certain PUFAs.¹⁰ According to the position of the carbon atom that binds molecular oxygen, LOX is classified into 5-, 8-, 12-, and 15-LOX.¹¹ Among these, 12-LOXs will be the focus of this review, whose products generated from PUFAs take part in anti- and pro-inflammatory activities. Some of the metabolizing pathways are presented in Figure 1.

Figure 1. The 12-LOXs pathways metabolizing ω-3 and ω-6 PUFAs.

Experimental studies have reported a protective role of ω-3 PUFAs against DR or pathological retinal angiogenesis.^12–14 In the PREDIMED (Prevención con Dieta Mediterránea) trial, a 6-year follow-up analysis in middle-aged and older individuals with type 2 diabetes adhering to the Mediterranean diet showed that intake of ≥500 mg/d of dietary ω-3 PUFAs is connected with a reduced incidence of sight-threatening DR.¹⁵ The protective effect of ω-3 PUFAs was abrogated in 5-LOX^−/− mouse models of oxygen-induced ischemic retinopathy (OIR) but not in 12/15-LOX^−/− ones, suggesting that the 5-LOX metabolite 4-hydroxydocosahexaenoic acid (4-HDHA) mediates the major antiangiogenic effect.¹⁶ A recent study found that DHA and its 12-LOX oxylipins (11-HDHA and 14-HDHA in platelets) could inhibit platelet activity and thrombus formation.¹⁷ However, 12-LOX-derived ω-3 oxylipins have not been examined in the retina and additional studies are needed to clarify their particular role in neovascular eye diseases.

Contrary to the protective effect of ω-3 PUFAs on DR, ω-6 PUFAs were found increased in diabetic patients, especially AA, whose pro-inflammatory products are important lipid mediators to regulate DR development.^18,19 12-LOX converts AA into 12-hydroperoxyeicosatetraenoic acid (12-HpETE), which is then reduced by glutathione peroxidase to 12-hydroxyeicosatetraenoic acid (12-HETE). Many studies have shown that 12-LOX and its product, 12-HETE, promote inflammation and oxidative stress, playing a key role in the pathogenesis of diabetes and its complications.^20,21

Here, based on the findings of previous studies, we investigated the biological functions of 12-LOX and 12-HETE, and summarized their role in DR pathogenesis.

An overview of 12-LOX family and its main product 12-HETE

Different types of 12-LOXs

Four subtypes of 12-LOX have been discovered in mice: platelet-type 12-LOX (P-12-LOX), leukocyte-type 12-LOX (L-12-LOX), epidermal-type 12-LOX (E-12-LOX), and 12(R)-LOX.²² Except for 12(R)-LOX, the other three are 12(S)-LOXs and produce the corresponding products in the S-stereo configuration. The isoforms are named after the cells where they were originally identified and subsequently detected in various cell types, such as endothelial cells, smooth muscle cells, glial cells, etc.^23–26 In fact, the phylogenetic correlations among these isomers are quite low.²⁷ Murine L-12-LOX is more closely related to human reticulocyte type 15-LOX-1 than to murine P-12-LOX.²² Human 15-LOX-1 (encoded by ALOX15) and murine L-12-LOX (encoded by Alox15) share 73% sequence homology and are collectively called 12/15-LOX.²⁸ Both of them can produce 12-HETE and 15-HETE; however, the product preference varies as per species, with 9:1 15-HpETE/12-HpETE in humans and 1:4 in mice, considering AA as the main substrate.²⁸ Therefore, even if these two proteins are derived from a homologous gene family, they may not produce the same oxylipins. This must be considered during the development and selection of potential LOX enzyme inhibitors.²⁰ In comparison, human 12(S)-LOX (encoded by ALOX12) and murine P-12-LOX (encoded by Alox12) share 85% sequence homology and both produce only 12-HETE.²⁰ The general expression patterns of different types of 12-LOXs are listed in Table 1.

Table 1. General expression patterns and products of 12-LOXs

Protein	Human gene	Mouse gene	Products	Tissue expression
Platelet type 12-LOX	ALOX12 (12(S)-LOX)	Alox12	12-HETE	Leukocytes, platelets, thrombocytes, macrophages, dendritic cells, skin, and islets (human)
Leukocyte type 12-LOX	ALOX15 (15-LOX-1)	Alox15 (12/15-LOX)	12-HETE, 15-HETE	Macrophages, β-cells, eosinophils, epithelial cells, kidney, smooth muscle, and islets (mouse)
Epidermal-type 12-LOX	Not expressed	Alox12e	12-HETE	Skin and epithelium
12(R)-LOX	ALOX12B	Alox12b	12(R)-HETE	Skin and epithelium

Biological actions of 12-HETE

The impact of 12-LOX is mainly attributed to the production of 12-HETE, which takes part in a wide range of biological processes, such as thrombogenesis, atherosclerosis, and neurodegeneration.^29–32 It also plays an important role in cancer biology as it promotes cell invasion, adhesion, growth, and metastasis.^30,33,34 Moreover, 12-HETE is a pro-inflammatory lipid metabolite that has been implicated in diabetes and vascular complications.^35–37 High glucose treatment increases the 12-HETE production in vascular endothelial and smooth muscle cells, which in turn upregulates the expression of vascular endothelial growth factor (VEGF), a key factor associated with pathogenic neovascularization.^38,39

Pathogenic roles of 12-LOX and 12/15-LOX in DR

Experimental studies have proved the critical role of 12-LOXs and its metabolite 12-HETE in the pathogenesis of DR. For convenience, in the following contents of this review, the simplified term “12-LOX” will refer to the enzyme encoded by human ALOX12 and murine Alox12 genes, while “12/15-LOX” will refer to the one encoded by human ALOX15 and murine Alox15 genes.

The effect of 12/15-LOX in DR

Extensive research has focused on the role of 12/15-LOX in the DR pathogenesis, whose effects are mediated by the occurrence and development of the disease.

Immunolocalization has shown that 12/15-LOX is expressed in retinal cells, including retinal microvascular endothelial and retinal pigment epithelial cells.⁴⁰ The earliest study has explored the role of 12/15-LOX in retinal angiogenesis, which is one of the most destructive pathological events occurring during advanced DR.⁴⁰ In a mouse model of OIR, the retinal levels of 12/15-LOX, 12-HETE, and 15-HETE were found to be increased significantly. In addition, the 12-HETE and 15-HETE levels were significantly elevated in the vitreous of diabetic patients with proliferative DR (PDR) compared with those without PDR. The blockade of 12/15-LOX with baicalein, a non-specific LOX inhibitor, or deletion of the 12/15-LOX gene could attenuate the expression of VEGF and ameliorate the retinal neovascularization. In addition, 12-HETE induced and inhibited the expression of VEGF and pigment epithelium-derived factor (PEDF) in rat Müller cells and murine primary astrocytes, respectively.⁴⁰ These results indicate that 12-HETE and 15-HETE produced by 12/15-LOX could promote retinal angiogenesis by disrupting the balance between VEGF and PEDF.

It is well known that inflammatory damage to the retina is involved in the early development of DR. Therefore, subsequent research was concentrated on studying the effect of 12/15-LOX on the barrier function of retinal vascular endothelial cells (RECs). Othman et al. found that the 12/15-LOX metabolites promote high permeability of the retinal vascular barrier, which is connected with the enhanced NADPH oxidase (NOX2) expression and reactive oxygen species (ROS) generation and decline in the level of zonula occludens-1 (ZO-1), while NADPH oxidase (NOX) inhibitors could prevent the 12/15-LOX pro-permeability effect.⁴¹ Treatment of Ins2^Akita mice with baicalein significantly diminished the levels of 12-HETE, inflammatory molecules (ICAM-1, VCAM-1, and IL-6), ROS production, NOX2, and pVEGFR2 expression in the retina, while restoring the protein tyrosine phosphatase pSHP1 expression.⁴¹ This suggests that 12/15-LOX leads to vascular hyperpermeability during DR through a NOX-dependent mechanism, involving the inhibition of protein tyrosine phosphatase and activation of the VEGFR2 signaling pathway. In another study, LC/MS lipidomics analysis of human RECs incubated with high glucose (30 mM) found that 12/15-LOX products were significantly upregulated. 12/15-LOX activates RECs with the help of NOX system, resulting in leukocyte adhesion, high permeability, and ultimately increased neovascularization.⁴² To further understand the pathogenesis of the LOX-NOX pathway in DR, Elmasry et al. investigated the role of endoplasmic reticulum (ER) stress in mediating the effect of 12/15-LOX metabolites on REC dysfunction.⁴³ They found that knocking out the 12/15-LOX gene could significantly mitigate the ER stress in the retina caused by diabetes, whereas 15-HETE could induce ER stress in human RECs. Inhibition of NOX or deletion of NOX2 did not attenuate 12/15-LOX-induced ER stress, while inhibition of ER stress downregulated the NOX activity and the expression of NOX2 and P47phox, its catalytic and regulatory subunits, respectively. It also lowered 15-HETE-induced leukocyte adhesion and VEGFR2 phosphorylation. In addition, 15-HETE enhances the intracellular calcium levels in human RECs.⁴³ These findings support the hypothesis that ER stress is upstream of the NOX pathway, promoting 12/15-LOX-induced retinal inflammation in DR by activating NOX and VEGFR2. The destruction of Ca²⁺ homeostasis may be an indispensable step in DR to initiate this signaling pathway.

Thus, hyperglycemia activates PLA2 to release AA from the retinal cell membrane, which is then converted into 12-HETE or 15-HETE. Then, ROS is generated through NOX, thereby forming a state of oxidative stress. Oxidative stress activates RECs through various inflammatory signaling pathways, leading to pathological changes in DR at different stages (Figure 2).

Figure 2. 12/15-LOX pathway in RECs in DR pathogenesis.

In the aforementioned studies, the authors emphasized that considering the species differences in 12/15-LOX, they used 12-HETE in all murine experiments and 15-HETE in all human cell experiments.⁴⁴ This action ensured the consistency of 12/15-LOX, the research object, in both species. However, a key issue may have been overlooked, that is, it is the product of the enzyme that exerts the biological effect, not the enzyme itself. There is no guarantee that the role of 12-HETE in mice is the same as that of 15-HETE in humans. As reported, 12-HETE is an effective vasodilator, whereas 15-HETE can cause vasoconstriction.^19,45 In addition, the studies used baicalein to block the function of 12/15-LOX, which is a nonspecific LOX inhibitor and may have off-target effects.¹⁹ On one hand, such an inhibitor with antioxidant properties could affect cell redox homeostasis, making it difficult to determine whether LOX inhibition or redox homeostasis is the main reason for the observed biological results.⁴⁶ On the other hand, not all the effects that cause DR microvascular disease are attributed to 12/15-LOX, because other studies have demonstrated the role of other LOXs, such as 5-LOX, and their products in DR, which can also be blocked by baicalein.^20,47

The effect of 12-LOX in DR

It has been confirmed that 12-LOX participates in diabetes and its vascular complications such as diabetic nephropathy and diabetic heart disease.^21,48,49 A previous study reported the existence of 12-LOX in the retina, and its expression was significantly increased in the OIR mouse model compared with the control.⁴⁰ Moreover, an obvious increase in 12-LOX protein levels was observed in the retinas of diabetic subjects compared with non-diabetic subjects.⁴⁰ However, no follow-up experimental investigations were conducted on the relationship between 12-LOX and DR. Recently, a multiplatform metabolomics study revealed a significantly augmented level of serum 12-HETE in DR subjects compared to those with diabetes without DR, and predicted it to be a promising diagnostic biomarker of DR.⁵⁰ Unlike in mice, the main source of 12-HETE in humans is 12-LOX, not 12/15-LOX. Therefore, the significantly elevated level of 12-HETE in DR patients suggests a strong association between 12-LOX and DR. On some occasions, there is even a statement that the murine 12/15-LOX is functionally closest to the human 12-LOX, which produces 12-HETE almost exclusively.³⁷

Pharmacological inhibition of 12-LOX and 12/15-LOX

Nonspecific LOX inhibitors are easy to discover, many of which are utilized widely and effectively in animal models of various diseases including DR, like baicalein and cinnamyl-3,4-dihydroxy-alpha-cyanocinnamate (CDC).^51,52 Nevertheless, using nonspecific inhibitors could easily confuse the interpretation of experimental results due to the too many targets (as discussed in Section 3.1). Thus, great efforts have been made to identify selective inhibitors of certain LOXs.

The specific human 12/15-LOX inhibitor ML351 which was discovered by high-throughput screening (HTS) showed its effectiveness in inhibiting murine 12/15-LOX.⁵³ This attribute is important because the distinct product preference between the two enzymes implies their different active sites, while ML351 inhibits both. That means the effects of ML351on human versus mouse are not exactly the same, which should be paid attention to when utilizing ML351 in preclinical studies.

ML355, the specific inhibitor of 12-LOX, efficiently improves the functions of cytokine-treated human islets and type 2 diabetic islets in vitro ⁵⁵. ML355 is a benzenesulfonamide also identified by HTS with high potency (IC50 = 0.3 μM) and selectivity.^54,55 It has been employed in research in multiple fields, including diabetes, thrombosis, LPS-induced inflammation, and ischemia-reperfusion-induced hepatic injury, displaying good pharmacokinetic characteristics in mice without observable toxic effects.^56–58 It has advantages over other LOX inhibitors in terms of explaining 12-LOX-related molecular mechanisms and drug development.

Conclusion and prospect

In this review, we have briefly introduced the 12-LOX family and its primary metabolite 12-HETE, elaborated the functions of 12/15-LOX in the onset and progression of DR, and indicated the necessity for further research on 12-LOX in DR.

Existing studies have shown (as described in Section 3.1) that 12-HETE activates the NOX system to generate ROS, causing oxidative stress and triggering inflammation in the retina. Nevertheless, the specific mechanism by which 12-HETE activates NOX remains unclear. An unresolved issue regarding 12-HETE is whether its cellular effects are mediated by its binding to a receptor or acting as a second messenger. G protein-coupled receptor 31 (GPR31) is a specific receptor with high affinity for 12-HETE and renamed as 12-HETER.³³ 12-HETE–GPR31 mediates a variety of signaling pathways, such as MAPK, MEK, and NF-κB.³⁴ It is involved in thrombosis, hepatic ischemia-reperfusion injury, and regulation of cancer proliferation, migration, and invasion, implicating its relevance in inflammation and angiogenesis.^58–62 As the development of 12-LOX inhibitors has been challenged by the structural and functional differences of enzymes among species, it can be worth concentrating on 12-HETE and its receptors.

To summarize, an in-depth exploration of the role of lipid metabolism such as 12-LOX and 12-HETE in DR would help elucidate the pathogenesis of this disease, thereby, providing a potential new strategy for treatment of DR.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was funded by Chinese National Nature Science Foundation [Project number 82071012], The Project of Shanghai Shen Kang Hospital Development Centre SHDC2018110,SHDC2020CR30538 [Grant No. SHDC2020CR30538 and SHDC2018110], Shanghai engineering research center of precise diagnosis and treatment of eye diseases, Shanghai, China [Project No. 19DZ2250100], The Science and Technology Commission of Shanghai Municipality [Project No. 20DZ1100200], Shanghai public health system three-year plan-Key Subjects [Project No. GWV10.1-XK7], and Shanghai General Hospital, Clinical Research [CTCCR-2018Z01].