Lipid Deposition and Metaflammation in Diabetic Kidney Disease

Abstract

A critical link between metabolic disorders and a form of low-grade systemic and chronic inflammation has been recently established and named “Metaflammation”. Metaflammation has been recognized as a key mediator of both micro- and macrovascular complications of diabetes and as a significant contributor to the development of diabetic kidney disease (DKD). The goal of this review is to summarize the contribution of diabetes-induced inflammation and the related signaling pathways to diabetic complications, with a particular focus on how innate immunity and lipid metabolism influence each other.

Introduction

Obesity and diabetes mellitus (DM) have become a worldwide pandemic, and diabetes-related complications represent a major health problem, with diabetic kidney disease (DKD) remaining the leading cause of kidney disease worldwide. Among several contributors to DKD, inflammation is currently recognized to contribute to the progression of chronic kidney disease (CKD). Moreover, several inflammatory biomarkers, such as adhesion molecules (ICAM-1, VCAM-1), cytokines (IL-6, TNF), c-reactive protein, fibrinogen, serum albumin, white blood cell counts, have been reported to be prognostic to risk-stratify patients for disease progression and all-cause mortality [1].

In obesity and type 2 diabetes, inflammatory biomarkers are traditionally thought to be derived from adipose tissue. As an example, secretion of inflammatory cytokines such as interleukin-1β (IL-1β) [2] and tumor necrosis factor (TNF) [3] has been shown to contribute to the development of insulin resistance and type 2 diabetes (T2D). Secretion of inflammation cytokines may be the consequence of inflammasome activation via caspase-1. Besides the inflammatory cytokine secretion from adipose tissue, de novo recruitment of macrophages into adipose tissue has been described (reviewed in Ref. [4] Once recruited, they become activated by the uptake of saturated fatty acids via Toll-like receptors 2 and 4, which triggers the activation of inflammatory signaling pathways via interferon regulatory factor 3 (IRF3), activator protein 1 (AP1) and nuclear factor-κB (NFκB). Inflammatory mediators can perpetuate lipid dysmetabolism locally in adipose tissue, or affect the concentration of other circulating inflammatory cytokines that in turn will influence the function of peripheral organs including the kidneys. Finally, inflammatory cytokines produced locally in peripheral organs can also influence disease progression, as clearly shown in the pathogenesis of DKD.

In CKD, the concentration of several circulating cytokines correlates with albuminuria. In particular, increased levels of circulating tumor necrosis factor (TNF) and TNF-receptors 1 and 2 are independent predictors of decline in renal function [5]. Several circulating inflammatory cytokines have been linked to progression to end stage renal disease (ESRD) in diabetes and referred to as the Kidney Risk Inflammatory Signature (KRIS). Seventeen of KRIS proteins in the T2D Validation Joslin Cohort and 16 of KRIS proteins in the T2D Replication Pima Cohort were significantly upregulated in patients who developed ESRD compared to those who did not develop ESRD, as discussed in the following section [6]. However, whether systemic or local inflammation contributes to disease progression is still being debated. In fact, increased local production of TNF, independently of circulating TNF levels, has shown to correlate with the loss of eGFR and to contribute to free cholesterol-mediated podocyte apoptosis via nuclear factor of activated T cells (NFAT)-mediated ATP-binding cassette transporter A1 (ABCA1) suppression [7]. In addition, a role of locally produced sphingolipids in kidney diseases has been recently established, and resulted in the discovery of novel key mediators of cell injury such as sphingosine-1-phosphate [8], ceramide-1-phosphate [9] and C16 ceramide [10].

This review will summarize the recent studies on the role of inflammatory mediators in diabetes with focus on the link between innate immunity and lipid metabolism patients, a small number of cross-sectional studies [11,12] and more longitudinal studies [13–17] demonstrated that C-reactive protein (CRP), IL-6 and TNF are associated with an increased risk of developing T2D.

Circulating Inflammatory Mediators and Metaflammation in DKD

Under physiological conditions, inflammation is a process mediated by cytokines, chemokines, and lipid mediators, including prostaglandins and leukotrienes. In healthy patients, a small number of cross-sectional studies [11,12] and more longitudinal studies [13–17] demonstrated that C-reactive protein (CRP), IL-6 and TNF are associated with an increased risk of developing T2D.

Inflammation in diabetes.

T2D is characterized by peripheral insulin resistance, which is accompanied by chronic low-grade inflammation in peripheral tissues such as muscles, liver and adipose tissue, redefining diabetes as an immune disorder. Classic inflammatory markers in diabetes include lectin, TNF and its receptors 1 and 2, interleukins (IL-6, IL-1β, IL-17, IL-18), monocyte chemoattractant protein 1 (MCP1), adhesion molecules VCAM1 and ICAM1, c-reactive protein (CRP) and nuclear factor-κB (NFκB). Increased production of interleukins has been shown in many animal models of diabetes [18–21] and these findings were confirmed in a cohort of T2D patients, where elevated levels of IL-1α, IL-8 and IL-18 were associated with early stages of DKD and correlated with podocyte damage and proximal tubular dysfunction [22]. Interestingly, IL-37 treatment of C57BL/6 mice exposed to high fat diet was associated with decreased secretion of pro-inflammatory cytokines such as IL-1β, TNF and IL-6 and improved established insulin resistance and glucose intolerance in the liver and the adipose tissue [23], suggesting that IL-37 may be a potential treatment strategy in metabolic diseases. Similar findings were demonstrated in humans as mentioned earlier, in the T2D Validation Joslin Cohort and in the T2D Replication Pima Cohort, where besides the TNFR superfamily proteins, other cytokines including IL15RA, IL17F, CD55, CD300C, CCL14, CCL15, CSF1, HAVCR2, IL1R1 and IL18R1 have been reported to be associated with increased risk of progression to ESRD [6]. In the Tatar ethnic group, Russia (n=440), CCL20-rs6749704 and CCL5-rs2107538 were shown to be associated with T2D [24]. Fractalkine (CX3CL1) was also found to be associated with development of inflammatory conditions in metabolic diseases. In T2D patients (n=47), plasma levels of CX3CL1 has been shown to correlate with inflammatory chemokines/cytokines, such as CCL3, CCL4, CCL11, CXCL1, INF-α2, IL-17A, IL-1β, IL-6 and TNF [25]. Similarly, another study of patients with advanced DKD (n=70) demonstrated strong correlation of C-X-C motif ligand 16 (CXCL16) and endostatin with interstitial fibrosis and tubular atrophy [26].

MCP1 signaling was found significantly increased in patients with T2D [27–29]. A common A/G polymorphism located at the position −2518 in the MCP1 distal regulatory region has been shown to be associated with diabetes. For example, in Turkish T2D patients [30] and Chinese T2D population [31], prevalence of T2D was reported in individuals with the MCP1 G-2518 genotype. A study on Japanese T2D patients revealed that −2518AA carriers have higher serum MCP1 concentration and increased insulin resistance compared to −2518G carriers [32]. The methylation status of CpG sites in the MCP1 promotor was also reported to be associated with diabetes, suggesting that hypomethylation of CpG motifs in T2D patients results in increased serum MCP1 levels [33]. Interestingly, blockade or lack of MCP1 in animal models resulted in reduced renal macrophage/monocyte infiltration and reduction of interstitial fibrosis [34], which will be discussed in details below.

Several adhesion molecules, such as vascular cell adhesion molecule 1 (VCAM1), also known as CD106, and intercellular adhesion molecule 1 (ICAM1), also known as CD54, were also shown to contribute to both types of diabetes. Subsequent studies have demonstrated that elevated levels of plasma ICAM1 in T1D patients (n=1,441) are independently associated with the onset of sustained microalbuminuria [35,36]. Similarly, levels of VCAM1 and ICAM1 were significantly higher in plasma of T2D patients who were enrolled in a cross-sectional study in Saudi Arabia regardless of associated cardiovascular complications [37], as well as in T2D patients with hypertension [38]. In a study on Mexican Americans, a prospective association of serum ICAM1 and VCAM1 with incidence of T2D was determined [39].

Inflammatory mediators and biomarkers of DKD.

Low-grade inflammation is also a common feature of CKD, and evidence suggests that persistent inflammation may be a risk factor for the progression of CKD per se and for the cardiovascular morbidity and mortality associated with CKD. Interestingly, other studies demonstrated that the number of white blood cells (neutrophils, eosinophils, basophils, lymphocytes, and monocytes) is increased in the transition from normalbuminuria to microalbuminuria [40,41], suggesting that white blood cells are also implicated in the development of DKD.

While the molecular mechanisms that attract monocytes to renal sites are not fully understood, endothelial leukocyte adhesion molecules, including VCAM1and ICAM1, have been shown to play an important role in the initiation of renal inflammation. These molecules are abundantly expressed in kidneys of patients with DKD and VCAM1/ICAM1 circulating levels are independently associated with DKD progression [42]. In a case-control study of a Diabetic Kidney Disease cohort of patients from Saudi Arabia, T2D patients with DKD had higher levels of VCAM1 and ICAM1 compared to those without DKD [43]. Another study, based on a meta-analysis, revealed that ICAM1-rs5498 might be a risk factor of DKD in T1D patients [44]. In T1D patients, ICAM1 K469E polymorphism was also shown to be associated with DKD [45]. In Turkish T2D patients with DKD, elevated levels of plasma VCAM1 and ICAM1 were also demonstrated and what is of particular interest, haemodialysis did not reduce those values [46]. While we do not know the exact role of adhesion molecules in the development of DKD, a recent study using the STZ mouse model of diabetes showed that micro RNA miR-146a expression may regulate levels of VCAM1 and ICAM1 (more miR-146a, less VCAM1 and ICAM1) [47], but no evidence was provided linking the miR-146a/VCAM/ICAM axis to the functional or structural alterations in DKD. Interestingly, ICAM1-deficient mice failed to develop diabetes in a STZ model [48], while the deletion of ICAM1 in db/db mice resulted in decreased renal inflammation [49], suggesting that ICAM1 is critically involved in the pathogenesis of DKD.

Circulating levels of TNF, TNF receptor 1 (TNFR1) and 2 (TNFR2) predict progression to ESRD in patients with type 1 diabetes (T1D) and T2D [5]. Our studies on a small number of patients with T1D enrolled in the FinnDiane Study cohort demonstrated that elevated levels of TNF, TNFR1 and TNFR2 in serum are associated with presence of DKD [7]. In Joslin and Pima cohorts of patients with T2D, elevated plasma concentrations of TNFR1 and TNFR2 were strongly associated with a risk of early renal function decline [6]. Additionally, patients with higher renal TNFR2 values were threefold more likely to experience renal function decline compared to controls [50]. In the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications cohort, increased levels of soluble TNFR1 and TNFR2 was associated with an increase in the odds to develop proteinuria [51]. Table 1 summarizes the clinical studies linking TNF, TNFR1 and TNFR2 to DKD.

Table 1.

Summary of clinical studies linking circulating TNF and TNF receptors and DKD.

	Population	Key Findings	Ref
TNF	T1D Joslin Cohort (n=219)	No association with ESRD risk was found.	[6]
	T2D Joslin Cohort (n=144)	No association with ESRD risk was found.	[6]
	T2D normalbuminuric patients	Association with reduced renal function (eGFR<60mL/min/1.73m2), which was lost after adjusting for TNFR covariates.	[132]
TNFR1/TNFR2	T1D Joslin Cohort (n=275)	Elevated serum levels predict early renal function loss in T1D patients without proteinuria.	[50]
	T1D Joslin Cohort (n=219) and T2D Joslin Cohort (n=144)	Elevated serum levels predict the highest adjusted hazard ratios (HR) for ESRD.	[6]
	Type 1 Diabetes Control and Complications Trial (DCCT)/Epidemiology of Diabetes Interventions and Complications Cohort (n=1,237)	Elevated serum levels predict macroalbuminuria in patients	[51]
	T2D Joslin Cohort (n=267)	Stronger association of elevated serum levels with ESRD in patients without proteinuria.	[133]
	T2D Pima Indians Cohort (n=193)	Elevated serum concentrations are associated with increased risk of ESRD in American Indians with T2D.	[134]
	Indigenous Australians with Diabetes	Association of TNFR1 with ESRD progression independent of albuminuria and eGFR.	[135]
	T2D normalbuminuric patients	Association with reduced renal function (eGFR<60mL/min/1.73m2).	[132]

Higher levels of MCP1 were also found in kidneys of the STZ-induced experimental mouse model of diabetes [52], which were also associated with overexpression of alpha-kinase 1 (ALPK1), an enzyme involved in TNF secretion, NFκB activation and production of pro-inflammatory cytokines (including IL-1β, IL-8 and TGF-β1). Additionally, MCP1 has been shown to be strongly associated with a decline in renal function in patients with diabetes [53,54]. In Malaysian T2D patients, a polymorphism in MCP1 (MCP1-rs3917887) was associated with DKD [55]. The MCP1-rs1024611 polymorphism was also found to be associated with foot ulcers in a Chinese Han population with T2D (n=400) [56]. The role of the MCP1 −2518 polymorphism in DKD remains unclear. While in an Indian population the presence of −2518GG [57] and in Korean patients, the −2518A allele of the MCP1 gene [58] are associated with increased risk of DKD, another study of Korean patients with T2D did not reveal an association of the MCP1 −2518 A/G polymorphism with DKD. However, these differences might be explained by the variations in the experiment design and need further investigation.

In contrast to MCP1, epidermal growth factor (EGF) has a protective role in kidney injury. Recent studies have suggested that urinary EGF/MCP1 ratio might be another reliable biomarker of progression to DKD. Thus, in a prospective T2D cohort from Thailand (n=83) a significantly urinary EGF/MCP1 ratio was associated with rapid renal decline, showing significantly lower urine EGF/MCP1 ratio in patients with a glomerular filtration rate (GFR) decline more than 25%/year [54]. The cross-sectional study on Chinese patients (n=1811) showed that urinary EGF- or MCP1-to-creatinine ratios negatively correlated with the occurrence of DKD, while in the longitudinal cohort (n=208) from the same study urinary EFG/MCP1 ratio also correlated with the percentage change of estimated GFR slope [59].

Nuclear factor-κB (NFκB) is another important transcription factor involved in the pathophysiology of diabetes and DKD. Glucose can activate NFκB, resulting in increased inflammatory genes expression. Activation of NFκB leads to activation of MCP1 and macrophage infiltration [60], activation of transforming growth factor β (TGFβ) with extracellular matrix accumulation and fibrosis [61]. A recent study demonstrated that induction of diabetes mellitus in Psammomys obesus using high fat diet results in increased expression of NFκB after three months suggesting its involvement in DKD development [62]. Mesangial cells cultured in high glucose (30 mmol/L), demonstrated activation of TLR4/NFκB p65/NGAL signaling pathway and promoted inflammatory reaction via suppression of Klotho [63].

Taken together, several circulating cytokines have been proposed as biomarkers for risk stratification and may be involved in the pathogenesis of diabetes and DKD (Figure 1), making them a potential drug target for the treatment of patients with DKD. However, as a cluster of cytokines rather than a single cytokine is often involved, targeting of common pathways activated by multiple cytokines rather than targeting single cytokines may prove more efficacious in the treatment of patients with DKD.

Figure 1. The role of lipid deposition and Metaflammation in the pathogenesis of diabetes and its complications.

Hypercaloric diet and genetics affect the function of pancreas, liver, muscles, and adipose tissue, leading to the formation of insulin resistance, hyperglycemia and hyperlipidemia via activation of the innate immune system, the production of pro-inflammatory cytokines and the deposition of lipids. Consequently, this results in diabetic micro- and macro-complications. This image was created using BioRender software (biorender.com).

Abbreviations: TLRs – Toll-like receptors; CLRs – C-type lectin receptors; NLRs – nucleotide-binding domain and leucine-rich repeat-containing receptors; RLRs – retinoic acid-inducible gene I-like receptors; cGAS – cyclic GMP-AMP synthase; STING – stimulator of interferon genes; JNK – c-Jun N-terminal kinase; NF-κB – nuclear factor kappa-light-chain-enchancer of activated B cells; IL – interleukin; TNFα – tumor necrosis factor alpha; ICAM1 – intercellular adhesion molecule 1; VCAM1 – vascular cell adhesion molecule 1; CX3CL1 – C-X3-C motif chemokine ligand 1; CCL5 – C-C motif ligand 5; MCP-1 – monocyte chemoattractant protein-1; TAG – triacylglycerol; FFA – free fatty acids; CD36 – scavenger receptor class B member 3; ABCA1 – ATP-binding cassette subfamily A member 1; ABCG1 – ATP-binding cassette subfamily G member 1; DAG – diacylglycerol; Cer – ceramide; Sph – sphingosine; SphK2 – sphingosine kinase 2; S1P – sphingosine-1-phospahte; C1P – ceramide-1-phosphate; SMPDL3b – sphingomyelin phosphodiesterase acid-like 3b.

Targeting the inflammatory pathways in DKD.

Several approaches have been proposed to treat inflammation in DKD. Besides lifestyle modifications (diet and exercise), some drugs have been reported to be promising in reduction of inflammation in DKD patients.

Bardoxolone is known to target oxidative stress and reduce inflammation via decreasing TGFβ. Clinical trials of Bardoxolone Methyl (BEACON) showed preliminary significant improvement in renal function in CKD patients with T2D, but this trial was stopped due to an increase in heart failure events [64]. The BEACON trial post-hoc analysis (n=2185), however, demonstrated that Bardoxolone Methyl preserves kidney function by preventing decline of estimated GFR, which was sustained 4 weeks post-cessation of treatment [65]. Currently, another trial of Bardoxolone in Japanese T2D patients (AYAME study, NCT03550443) is ongoing with an estimated completion date of March 2022.

Post-hoc analysis of using of Selonsertib, a highly selective small-molecular inhibitor of apoptosis-regulating kinase 1 (which in turn regulates activity of p38 MAPK and c-JNK), suggest that it may slow DKD progression [66]. Currently, Selonsertib is in a clinical trial Phase 3 (MOSAIC; NCT04026165).

Another potential medication to treat inflammation in DKD is Pentoxifylline, a methylxantine derivative with nonspecific inhibition of phosphodiesterases. In animal models, Pentoxifylline was shown to have a marked antiproteinuric effect with attenuating interstitial inflammation and renal injury progression [67,68] and decreasing levels of TNF and IL-6 [69]. In the PREDIAN trial, using of 2 years of Pentoxifylline therapy in combination with RAAS blockade resulted in decreased proteinuria and urinary concentration of TNF and slowed the decline in estimated GFR [70].

Metaflammation Activates Pathways of the Innate Immune System Involved in Insulin Resistance and DKD

Even if DKD is not considered as a primary “immune-mediated” renal disease, more and more evidence supports the involvement of many components of the immune system in DKD initiation and progression. As an evolutionarily conserved system, the innate immune system is the first line of host defense against pathogens and internal danger signals, which relies on pattern recognition receptors, such as toll-like receptors (TLRs), C-type lectin receptors (CLRs), RIG-I-like receptors (RLRs), intracellular Nod-like receptors (NLRs), HIN-200 receptors and the recently discovered cyclic GMP-AMP synthase-stimulator of interferon genes pathway (cGAS-STING).

More than 90% of scientific reports investigating role of TLRs in diabetes revealed that TLR2 is positively associated with T2D and its complications, while T1D is also associated with TLR1, TLR3, TLR4, TLR7 and TLR9 [71–75]. In NOD mice, a model of T1D, TLR4 mediates cardiac lipid accumulation and diabetic heart disease [76], while deletion of TLR4 results in acceleration of diabetes onset [77]. In human monocytes, it has been demonstrated that high glucose induces increased expression of TLR2 and TLR4 via protein kinase C (PKC-α and PKC-δ, respectively) [78].

NLRs signal via the inflammasome, a cytosolic multiprotein complex of the innate immune system that detects pathogens and activates pro-inflammatory response via IL-1β and IL-18. Within the NLR family of proteins, pyrin domain containing 1 (NLRP1) and 3 (NLRP3) have attracted much attention in recent studies as potential targets to modulate inflammation in T1D and T2D. NLRP1 was the first protein shown to form inflammasome [79]. A recent study demonstrated that alleles of the Nlrp1b gene in C57BL/6 mice play an important role in the glucose metabolism when fed with a high fat diet, where the Nlrp1b1 allele downregulates inflammation and attenuates glucose intolerance and insulin resistance in adipose tissue in obese mice [80]. Polymorphisms of NLRP1 were found to be associated with lupus erythematosus and T1D [81,82]. Interestingly, in the blood mononuclear cells and granulocytes of patients with T1D expression of NLRP1 and NLRP3 was markedly decreased, suggesting that NLRP1/NLRP3 inflammasomes signaling pathways may be a protective factor in the early stage of T1D [83]. In a recent study the functional specificity of the NLRP1 inflammasome for the activation of IL-18 in adipose tissue of mice on a high fat diet was demonstrated [84], which opens novel avenues for the targeting therapy.

Activation of NLRP3 is promoted by TLRs activation, which senses dangerous signals such as lipopolysaccharides and high glucose levels, thereby triggering NFκB pathway and production of IL-1β, IL-18 and IL-33 proinflammatory cytokines [85,86]. Another study also demonstrated that high glucose promotes expression of NLRP3, IL-1β, IL-18 in rat glomerular mesangial cells [87]. A decline in renal function is associated with accumulation of inflammatory cells in the kidney, which suggests the NLRP3 inflammasome as a potential therapeutic target in patients with CKD [88]. In glomerular mesangial cells of the STZ-treated rats, high glucose and lipopolysaccharides activate reactive oxygen species (ROS) and the NLRP3 inflammasome, contributing to the development of DKD [89]. In addition, in rats with STZ-induced DKD renal NLRP3 inflammasome activation is shown to be associated with lipid accumulation (based on the Oil Red O staining), which was ameliorated using quercertin (100 mg/kg) and allopurinol (10 mg/kg) [90]. In T2D patients with DKD (n=45), increased mRNA expression of NLRP3 and caspase-1 compared to T2D patients without DKD (n=45) was also reported, while mRNA expression of caspase-1 only was significantly associated with decreased estimated glomerular filtration rate, representing the severity of DKD [91]. Thus, the NLRP3-IL-1β pathway seems to be activated by food intake, obesity, and insulin resistance, while the NLRP1-IL-18 pathway reduces energy intake, lipid accumulation and improves insulin sensitivity.

The recently discovered cGAS-STING pathway was originally described as a signaling cascade that can be activated by double-stranded DNA during a pathogen infection. Studies suggest that host DNA damage induces cytokines through the cGAS-STING signaling pathway thus contributing to increased sterile inflammation. More evidences appeared connecting the cGAS-STING signaling pathway with a state of chronic inflammation in kidneys. A recent study demonstrated that STING activation in tubular epithelial cells contributes to renal inflammation and fibrosis and that pharmacological inhibition of STING has a beneficial effect in a mouse model of chronic kidney disease [92], thus providing further evidence that connects the cGAS-STING signaling pathway with a state of chronic inflammation in kidneys. However, less is known about the role of the cGAS-STING pathway in DKD. A recent study demonstrated that in eNOS^db/db mice and T2D rats the cGAS-STING pathway contributes to the development of DKD particularly in [93]. Thus, targeting of the cGAS-STING signaling pathway may present another novel treatment approach in DKD. However, further investigations are needed.

Overall, there is an increasing number of studies linking Metaflammation to diabetic complications, including DKD (Figure 1). While it is important to understand the roles of innate immunity and adaptive immunity in obesity/diabetes, which could reveal novel immunotherapeutic approaches to modulate metabolic inflammation and insulin resistance, the contribution of lipid metabolism to inflammation remains largely unexplored.

Role of Metaflammation and Lipid/Sphingolipid Dysmetabolism in Kidney Injury

Basal fat cell lipolysis, a process of triacylglycerol breakdown into fatty acids and glycerol, is elevated in obesity conditions and is closely associated with insulin resistance. Both circulating lipids and lipid tissue deposition have been shown to contribute to cellular dysfunction and injury. Lipotoxicity is a process of lipid accumulation in non-adipose tissue such as kidney, liver, heart and muscles. Lipid droplets, which are specialized cellular organelles that play a dynamic role in energy homeostasis, were shown to accumulate in islets as well as in podocytes in diabetes and DKD [94–96]. While accumulation of lipid droplets was initially considered a defense mechanism, it is now well recognized that intracellular lipid deposition may play an essential causative role in the progression of diabetic complications.

Both fatty acid and cholesterol metabolism have been shown to contribute to lipotoxic damage in metabolic tissues in patients with diabetes. High glucose has been shown to exacerbate fatty acid uptake and deposition by increasing the class B scavenger receptor CD36 expression via AKT-PPARγ signaling pathway. Plasma CD36 is associated with microparticles in patients with T2D and suggested to be a biomarker of T2D [97]. Interestingly, fatty acids, which are often increased in conditions of insulin resistance, stimulate the production of pro-inflammatory cytokines such as IL-1β directly via activation of TLR2 and TLR4 or indirectly via ceramide production [98]. Another study demonstrated an important role of fatty acid synthase (FAS) in membrane lipid composition and inflammation in diabetes [99], where deletion of Fasn in macrophages of HFD-fed mice prevented insulin resistance, recruitment of macrophages into adipose tissue and chronic inflammation by impairing the retention of plasma membrane cholesterol and disturbing Rho GTPase trafficking. However, lipid deposition may also be glucose-independent and driven by altered cholesterol metabolism rather than altered fatty acid metabolism. In fact, we have demonstrated that human podocytes exposed to the sera of patients with DKD accumulate lipid droplets when compared to the euglycemic sera of patients with diabetes without DKD [100]. Studies investigating the role of ABCA1, a protein involved in cholesterol and phospholipid efflux to apolipoproteins, demonstrated reduced expression of ABCA1 in podocytes treated with sera obtained from patients with both T1D and T2D [7,95] as well as in the kidney cortexes of db/db and ob/ob DKD mouse models. While reduced ABCA1 expression is not sufficient to cause DKD, genetic or pharmacological ABCA1 overexpression were sufficient to completely rescue mice from DKD or DKD-like glomerulosclerosis [95], suggesting that ABCA1 may be a potential therapeutic target. Moreover, glomerular ABCA1 expression in early disease stages inversely correlates with markers of DKD progression both clinically and experimentally [7], suggesting that ABCA1 may be a susceptibility factor contributing to podocyte injury in DKD.

A tight connection between metabolic and inflammatory pathways exists. Persistent inflammation or the failure of tissue to return to the normal physiological state favors tissue destruction and resolution of inflammation is a critical step in the process of regaining tissue homeostasis. Over recent years, pro-resolution lipid mediators were brought into focus in the pathogenesis of diabetes.

Prostaglandins are important lipid mediators of arachidonic acid produced via the sequential catalyzation of cyclooxygenases (COXs) and downstream prostaglandin synthases. One of the best studied lipid mediators is prostaglandin E2 (PGE2), which has been shown to be associated with hypertension and diabetes [101,102]. Urinary PGE2 production is elevated in parallel with activation of COX2 in mice with STZ-induced diabetes [103] and B6-Ins(Akita) mice [104], while using of PGE2 receptor inhibitors [105] and COX2 inhibition [106] may protect against DKD progression. Deletion of PTGER1, a G-protein coupled receptor of PGE2 action, in diabetic OVE26 mice reduced diabetes-induced expression of fibrotic markers, prevented downregulation of nephrin mRNA expression down-regulation and ameliorated glomerular basement membrane thickening and foot processes effacement [107]. Interestingly, in humans with uncomplicated T1D, inhibition of COX2 using 200 mg celecoxib resulted in a significant decrease of glomerular filtration rate (GFR) in the hyperfiltration group and increased in GFR in the normofiltration group [108].

Leukotrienes are a family of eicosanoid derived from arachidonic acid and the essential fatty acid eicosapentaenoic acid by the enzyme arachidonate-5-lipoxygenase (5-LO). Increased levels of leukotrienes synthesis was also reported in diabetic retinopathy (reviewed in Ref. [109]). Like PGE2, urinary excretion levels of leukotriene E4 (LTE4) was increased in T1D patients [110], suggesting a distinguished role of LTE4 in renal inflammation. In addition, in patients with T2D (n=30) lower urinary LTE4 levels were associated with decreased renal function and correlated with serum creatinine and estimated glomerular filtration rate [111].

A Role for eicosanoids, including protectins, resolvins and lipoxins was reported in T1D and T2D. Many studies report low concentrations of eicosanoids in diabetes milieus [112–117], suggesting affected resolution of inflammation. However, not too many data exist on the role of eicosanoids in DKD. Thus, in human renal mesangial cells in DKD, LXA4 is a potent modulator of matrix accumulation, where it affects the expression of many genes, including transforming growth factor beta 1, fibronectin, matrix metalloproteinase 1, and several collagens [118]. The role of prostaglandins, leukotrienes and eicosanoids in diabetes and its complications is also summarized in Figure 2.

Figure 2. Prostaglandins, leukotrienes and eicosanoids in the pathogenesis of diabetes and its complications.

Hyperglycemia and release of cytokines lead to the malfunction in the polyunsaturated fatty acids metabolism. In turn it causes overexpression of prostaglandins (mainly PGE2) and leukotrienes (mainly LTB4 and LTE4) and inhibition of eicosanoids (mainly RvE1, LXA4, LXB4 and RvD1), resulting in inability to inhibit inflammation and promotion of diabetic complications.

The image is created using BioRender software (biorender.com).

Abbreviations: ω3-PUFA – omega 3 polyunsaturated fatty acids; ω6-PUFA – omega 6 polyunsaturated fatty acids; EPA – eicosapentaenoic acid; DHA – docosahexaenoic acid; COX2 – cyclooxygenase 2; HPETE – hydroperoxyeicosatetraenoic acid; HPDHA – hydroperoxydocahexaenoic acid; LOX – lipoxygenase; HDHA – hydroxy-docosahexaenoic acid; RvD1 – resolvin D 1; RvE1, RvE2, RvE3 – resolvin E 1, 2 and 3, respectively; LXA4 – lipoxin A4; LXB4 – lipoxin B4; AA – arachidonic acid; PGG2 – prostaglandin G2; PGH2 – prostaglandin H2; PGE2 – prostaglandin E2; mPGES-1 – microsomal prostaglandin E synthase-1; 5-LO – 5-lipoxygenase; LTA4 – leukotriene A4; LTB4 – leukotriene B4; LTC4 – leukotriene C4; LTD4 – leukotriene D4; LTE4 – leukotriene E4; MyD88 – innate immune signal transduction adaptor.

Involvement of the sphingolipid metabolic pathway in renal injury in diabetes [119] and in inflammation [120] has been also reported. Major bioactive sphingolipids include ceramide, sphingosine, sphingosine-1-phosphate (S1P), and ceramide-1-phosphate (C1P). A role of ceramides has been demonstrated in inflammation of adipose tissue as well, suggesting that regulation of C16-ceramide synthesis may be beneficial in the treatment and even prevention of diabetes [10]. Elevation of muscle C20- and C22-ceramide was shown in patients with general or abdominal obesity in association with muscle insulin resistance, suggesting that tissue ceramide levels may predict insulin resistance [121].

S1P is a member of the sphingolipid family, which is involved in lymphocyte development, B- and T-cell recirculation and mediations of the effects of numerous of biological stimuli, including cytokines. S1P is positively associated with obesity. Circulating levels of S1P have been shown to be significantly increased in diabetic db/db mice and obese humans where a positive correlation of human plasma S1P with homeostatic model assessment of insulin resistance (HOMA-IR), hemoglobin A1C, low density lipoproteins and total cholesterol was detected [122]. A cross-sectional study demonstrated a significant association between serum S1P and TNF in healthy overweight adolescents [123], suggesting increased levels of S1P as a risk factor for metabolic syndrome. Additionally, a previous study demonstrated elevated levels of S1P in obese diabetic BTBR ob/ob mice [124]. The same study showed that S1P and ceramide, another important centerpiece of sphingolipid metabolism, induced expression of IL-6, TNF, and MCP1. S1P can also induce adipocyte dysfunction by promoting chronic inflammation and blocking of insulin signaling [125]. Finally, studies in MIN6 pancreatic β-cells showed that decreased levels of S1P via treatment with sphingosine kinase inhibitor results in reduced glucose-stimulated insulin secretion [126]. However, the role of S1P in DKD remains largely unknown. Some studies demonstrated that excessive S1P/S1P2 receptor pathway activation in tubular cells in DKD may induce activation of Rho kinase and facilitate renal fibrosis in DKD [127], while the activation of S1P/S1P1 receptor pathway may attenuate early-stage DKD [128].

Sphingomyelin phosphodiesterase acid-like 3b (SMPDL3b), an enzyme localized in lipid raft domains and upregulated in DKD [129], was previously shown to act as a major regulator of TLR4 signaling in macrophages [130]. We demonstrated that overexpression of SMPDL3b in human podocytes results in the activation of multiple inflammatory pathways [9], suggesting an important role of sphingolipids in the immune system homeostasis. In the same study, we showed that in vitro treatment of SMPDL3b overexpression human podocytes with C1P or exogenous administration of C1P in db/db mice results in restored AKT phosphorylation and improved renal outcome [9], suggesting that SMPDL3b excess interferes with ceramide kinase function in podocytes [131].

Concluding Remarks

The existence of ample cumulative data that support a role of inflammation in the pathogenesis of diabetes and its complications such as DKD thus warranting their consideration as therapeutic strategies. Many clinical trials have been initiated to study anti-inflammatory approaches to treat T2D or pre-diabetes. Based on the reviewed literature, we can conclude that abnormalities in lipid metabolism per se are not sufficient to induce renal injury, but that lipotoxicity represents an important component of the disease mechanism that may involve multiple hits. Patterns of cytokine production differ due to various inflammatory conditions in patients with diabetes, making a personalized approach urgently important when planning a treatment for a patient. Many of the inflammatory and lipid mediators are measurable in blood and urine and may therefore serve as candidate DKD biomarkers, however, therapeutic developments specifically targeting these mediators for the treatment of patients with DKD still need to be established. Moreover, the crosstalk between lipids and the immune system is still elusive. Further studies are needed to investigate whether anti-inflammatory treatment alone or together with drugs preventing peripheral lipid dysmetabolism can prevent the progression of diabetes and its complications.

Highlights.

• Lipotoxicity and glucotoxicity are linked to the activation of pro-inflammatory cytokines in diabetic kidney disease.

• The innate immune system, including the cGAS-STING signaling pathway, is highly involved in the pathogenesis of diabetic complications.

• Impaired cholesterol efflux (primarily via ATP-binding cassette transporter 1) and an imbalance in biologically active sphingolipids (such as ceramide-1-phosphate and sphingosine-1-phosphate) contribute to the development of diabetic kidney disease.