Kidney in the net of acute and long-haul coronavirus disease 2019: a potential role for lipid mediators in causing renal injury and fibrosis

Purpose of review Severe COVID-19 disease is often complicated by acute kidney injury (AKI), which may transition to chronic kidney disease (CKD). Better understanding of underlying mechanisms is important in advancing therapeutic approaches. Recent findings SARS-CoV-2-induced endothelial injury initiates platelet activation, platelet–neutrophil partnership and release of neutrophil extracellular traps. The resulting thromboinflammation causes ischemia–reperfusion (I/R) injury to end organs. Severe COVID-19 induces a lipid-mediator storm with massive increases in thromboxane A2 (TxA2) and PGD2, which promote thromboinflammation and apoptosis of renal tubular cells, respectively, and thereby enhance renal fibrosis. COVID-19-associated AKI improves rapidly in the majority. However, 15–30% have protracted renal injury, raising the specter of transition from AKI to CKD. Summary In COVID-19, the lipid-mediator storm promotes thromboinflammation, ischemia–reperfusion injury and cytotoxicity. The thromboxane A2 and PGD2 signaling presents a therapeutic target with potential to mitigate AKI and transition to CKD. Ramatroban, the only dual antagonist of the thromboxane A2/TPr and PGD2/DPr2 signaling could potentially mitigate renal injury in acute and long-haul COVID. Urgent studies targeting the lipid-mediator storm are needed to potentially reduce the heavy burden of kidney disease emerging in the wake of the current pandemic.


INTRODUCTION
Amidst the coronavirus disease 2019 (COVID-19) pandemic, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus has caused immense suffering while placing tremendous burden on healthcare systems worldwide.Although mild cases present with common cold symptoms, severe COVID-19 disease is associated with multiorgan failure and dysfunction of the lung, kidney, heart, liver and brain.Acute kidney injury (AKI) has emerged as a common complication in hospitalized patients with severe COVID-19 with acute tubular injury as the major histopathological finding [1,2].The pathogenesis of COVID-19-induced AKI (CoV2-AKI) is poorly understood.
End organ damage in COVID-19 is fueled by thromboinflammation characterized by platelet and neutrophil activation, and release of neutrophil extracellular traps (NETs) [3 & ] (Fig. 1).It has been proposed that thromboinflammation in COVID-19 is mediated by a cytokine storm.However, cytokine release in COVID-19 is comparable to if not lower than the cytokine levels in influenza patients [4], and yet alveolar capillary microthrombi are nine Although COVID-19 is predominantly a respiratory disorder, the kidney is one of the most common organs affected by COVID-19 second only to the lung [8].AKI is emerging as a common and important sequelae of COVID-19, with rates as high as 33-43% in hospitalized patients [9][10][11][12].CoV2-AKI may present as collapsing glomerulopathy, acute interstitial nephritis, de novo glomerular disease, prerenal azotemia or acute tubular injury [1,13].In kidney biopsy, samples from 17 mild COVID-19 patients, acute tubular injury, collapsing glomerulopathy, endothelial injury or thrombotic microangiopathy were the most common histological findings [14].Interestingly, virus detection was negative in the patient samples [14], and most kidney biopsies performed several weeks after the onset of COVID-19 symptoms failed to show notable SARS-CoV-2 infection [15].Although some studies have reported SARS-CoV-2 particles in urine samples [16][17][18] or in kidney glomerular compartments [19], a direct role of the virus in the development of AKI remains to be demonstrated [15].
Most patients who survive CoV2-AKI regain kidney function but up to 30% may remain on dialysis at discharge [20].Moreover, in a multicenter cohort study of 3099 hospitalized adult COVID-19 patients, one in five patients developed AKI requiring renal replacement therapy (RRT) [11].Of those who survived and were discharged, one in three remained RRT-dependent at discharge, and one in six remained RRT-dependent 60 days after ICU admission [11].This raises the specter of transition from acute to chronic kidney disease.Better understanding of mechanisms underlying COVID-19-induced AKI are critical to designing therapeutic strategies.

MECHANISMS OF CORONAVIRUS DISEASE 2019-INDUCED ACUTE KIDNEY INJURY
The mechanisms underlying the development of CoV2-AKI are likely multifactorial, including viral septicaemia, an enhanced inflammatory response, endothelial damage, hypercoagulability, myocardial dysfunction, drug nephrotoxicity and the effects of general hypoxia and dehydration on renal perfusion [21].Although there is evidence of an association between proinflammatory cytokines and kidney injury [22,23], few patients with COVID-19 exhibit cytokine profiles indicative of cytokine storm syndrome [4].In fact, patients with COVID-19 exhibited lower cytokine levels than patients with influenza [4].Therefore, other underlying mechanisms may play a greater role in fueling the thromboinflammation and AKI in COVID-19 as discussed in this review (Table 1).24,25].Endothelialitis and pyroptosis lead to release of endothelial microvesicles, which activate leukocytes and platelets through surface interaction, receptor activation, cellular fusion and the delivery of intravesicular cargo [25,26].Endothelial cell injury, platelet activation, platelet-leukocyte aggregates and thrombosis in severe COVID-19 disease is evidenced by elevated serum levels of soluble

KEY POINTS
Severe COVID-19-induced lipid mediator storm in the lungs is associated with marked increase in thromboxane A 2 (TxA 2 ), PGD 2 and leukotoxin diols.
PGD 2 /DPr2 signaling promotes a Th2 immune response and apoptosis of renal tubular cells.
The lipid mediator storm presents a therapeutic target in acute and long-haul COVID-19 with potential to mitigate AKI and transition to CKD.FIGURE 1. Potential mechanisms of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) induced acute kidney injury and transition to chronic kidney disease in long-haul coronavirus disease.SARS-CoV-2 virus binds complement MASP-2, thereby activating the lectin pathways and production of membrane attack complex (MAC) in renal vascular beds and tubuli.MAC synergizes with TLR signaling to promote inflammation, and contributes to chemotaxis, leukocyte adhesion and systemic thrombosis.MAC deposition on endothelial cells along with necrosis and pyroptosis of endothelial cells, platelets and monocytes, leads to IL-1a production and release.In turn, IL-1a stimulates the IL-1 receptor expressed on endothelial cells leading to COX-2 and TxAS expression.Subsequent thromboxane A 2 production induces endothelial and platelet activation via the TPr receptor.Activated platelets express P-selectin leading to platelet-neutrophil partnership, neutrophil activation and release of neutrophil extracellular traps (NETs), which fuel thromboinflammation in COVID-19.Renal ischemia-reperfusion injury as a result of renal thromboinflammation leads to NF-kB p65 phosphorylation.SARS-CoV-2 also directly induces NF-kB p65 phosphorylation via binding of S1 protein to ACE2 leading to ER stress and unfolded protein response (UPR) signaling constituting IRE1 and PERK protein activation, and downstream MAP kinase signaling.NF-kB p65 phosphorylation induces COX-2 expression and eicosanoid production, including upregulation of L-PGDS derived PGD 2 .Patients with COVID-19 also exhibit elevated levels of leukotoxin diols.Linoleate is converted into epoxyoctadecenoic acids (EpOMEs) by CYP450 epoxygenase enzymes, and further metabolized into leukotoxin diols by soluble epoxide hydrolase (sEH).PGD 2 /DPr2 signaling and leukotoxin diol-induced mitochondrial dysfunction causes renal proximal tubular epithelial cell apoptosis and Copyright © 2021 Wolters Kluwer Health, Inc.All rights reserved.

Complement activation
Endothelial dysfunction and thrombosis in COVID-19 is also mediated by complement activation (Fig. 1).Nucleocapsid protein of SARS-CoV-2 virus binds to the Mannan-binding lectin-associated serine protease-2 (MASP-2), the lectin pathway's effector enzyme, resulting in complement activation [30].Kidney biopsies in nine COVID-19 patients reveal enhanced renal complement deposition in vascular beds and tubules, along with glomerular MASP-2 and tubular C5b-9 deposition in the majority of cases [8].Complement activation and assembly of the MAC C5b-9 is known to promote endothelial damage, thrombosis and renal injury, such as in atypical hemolytic uremic syndrome [31].
In COVID-19 infection, IL-1a release from necrotic and pyroptotic cells is strongly associated with the lung injury and disease severity [32,33].MAC C5b-9 induces synthesis of IL-1a from porcine endothelial cells, and direct treatment with IL-1a leads to COX-2 but not COX-1 expression [34].Therefore, complement activation and cellular necrosis-led expression of IL-1a may lead to thromboinflammation via COX-2 mediated thromboxane generation as discussed below (Fig. 1).

Lipid mediator storm promoting thromboinflammation and maladaptive immune response
SARS-CoV-2-induced direct COX-2 expression has been observed in various cell lines and tissues in vitro and in vivo including cardiomyocytes, Calu-3 and A549 lung cancer cell lines, ciliated lung cells, primary human bronchial epithelial cells, human ACE2-expressing mouse lungs and living human lung slices [35][36][37].COX-2 expression is elevated more than four-fold in SARS-CoV-2-infected Calu-3 cells and human lung slices [36] and more than 50fold in infected cardiomyocytes (Professor Srinivasa T. Reddy, UCLA, personal communication following analysis of the supplemental material in reference [36]) [37].
SARS-CoV-2-induced cardiomyocyte COX-2 expression is highly correlated with pro-inflammatory genes and signaling pathways including interferon, IL-1b and NF-kB [37,38].Interestingly, SARS-CoV-2 infection-induced S1-ACE2 receptor interaction leads to NF-kB signaling in various cells including the kidney [39,40].NF-kB activates COX-2 expression and thromboxane A 2 generation [41], leading to renal ischemia-reperfusion injury (Fig. 1).Inflammation, old age and obesity are additional stimulators of COX-2 expression [42,43].Canonical NF-kB signaling pathway regulates early phase inflammation in I/R injury [44,45].In a mouse model of I/R injury induced AKI, widespread NF-kB activation was associated with impaired renal function, tubular apoptosis and neutrophil and macrophage infiltration after I/R injury, all of which were ameliorated by inhibiting expression of NF-kB [45].
Prostaglandin D 2 signaling as a potential mediator of coronavirus disease 2019associated acute kidney injury and chronic kidney disease Prostaglandin D 2 is upregulated during states of ischemia [63,64], and plasma levels of PGD 2 are markedly elevated in severe COVID-19 [7 && ,36].PGD 2 is the most abundant prostanoid in the mammalian brain [65] and exerts its main functions through two receptors, DPr1 and DPr2 (aka CRTh2) [66].In fact, lipocalin PGD 2 synthase and DPr2 expression were elevated in the renal tubules of COVID-19 autopsies [67 && ].SARS-CoV-2 infection of Calu-3 lung cells increased PGD 2 secretion more than two-fold [36].DPr2 expression is upregulated more than 50-fold by TGF-b acting in conjunction with IL-4 [68].In COVID-19, TGF-b is markedly increased and is in fact a determinant of the maladaptive host immune response [69].IL-4 levels are also markedly elevated in severe COVID-19 and post COVID where they correlate with the number of circulating endothelial cells [70].Increased expression of DPr2 receptors in conjunction with local increase in PGD 2 has the potential to massively amplify PGD 2 /DPr2 signaling.

Leukotoxin diols as a potential mediator of coronavirus disease 2019-induced renal inflammation and injury
Leukotoxin diols are a metabolite of linoleic acid (18 : 2n6) that is an essential source for long-chain Copyright © 2021 Wolters Kluwer Health, Inc.All rights reserved.

Leukotrienes as potential mediators of coronavirus disease 2019-induced oxidative stress and acute kidney injury
Leukotrienes, notably LTB 4 , LTE 4 and eoxin E 4 are significantly increased in the BALF of severe COVID-19 patients [7 && ].Leukotrienes have been implicated in renal disease.In rat model of glomerular nephritis, administration of antirat glomerular basement membrane antibodies led to urinary excretion of LTC 4 , LTE 4 and acetylated LTE 4 concomitantly with increased renal LTC 4 synthase activity [86].Moreover, in a mouse model of renal-ischemia reperfusion, inhibition of leukotriene biosynthesis attenuated oxidative stress, histopathological markers of tissue damage, cytokine release and damage to renal function [87].Therefore, leukotrienes likely play a key role in renal ischemia-reperfusion-induced AKI.

Other lipid mediators of uncertain significance in causing renal injury in coronavirus disease 2019
Many other lipid mediators have been reported to be increased in the lungs and plasma of COVID-19 patients but their role in causing renal injury in acute and long-haul COVID remains to be defined.These mediators include PGE 2 , specialized proresolving mediators, D-series resolvins and protectin D1 [7  && ,36].

Nitric oxide deficiency
Nitric oxide (NO) has been implicated in many physiological functions in the kidney [88], and is known to ameliorate renal injury in ischemic acute renal failure [89].On the other hand, NO deficiency characterizes transition to chronic kidney disease and adverse kidney events [88,90].SARS-CoV-2 virus-induced endothelial injury and pyroptosis is associated with inhibition of nitric oxide synthesis.Furthermore, TxA 2 directly inhibits endothelial and inducible nitric oxide synthase [91,92].Conversely, NO phosphorylates TxA 2 receptor (TPr) and thereby inhibits platelet activation [93] whereas seratrodast or ramatroban as TPr antagonists enhance NO generation [91].Therefore, TPr antagonism could potentially increase NO production and thereby mitigate AKI in COVID-19.

THERAPIES TARGETING THE LIPID MEDIATOR STORM AND ACUTE KIDNEY INJURY IN CORONAVIRUS DISEASE 2019
Therapies targeting the lipid mediator storm have been proposed in COVID-19.In a prospective study, Hong et al. treated ordinary, severe and critically ill COVID-19 patients with the COX-2 inhibitor, Celecoxib [94].In ordinary COVID-19 cases, patients taking Celebrex did not progress to severe disease compared with 15.7% of ordinary cases under routine treatments.Moreover, none of the severe COVID-19 cases with full dose Celebrex treatment progressed to critical illness.Concern has been raised that blocking upstream mediators including COX-2 and COX-1 may result in more challenges than cures because of their broad inhibition of several essential prostanoids including vasodilatory prostacyclin [95].
Dexamethasone is effective in lowering 28-day mortality and need for renal replacement therapy in hospitalized patients [96].The efficacy of dexamethasone may stem from inhibition of the lipid mediator storm [7  && ].Dexamethasone destabilizes bglobin-Cox-2 reporter mRNAs [97] and upregulates annexin A1 expression [98], thereby inhibiting COX-2 and phospholipase A 2 expression, respectively, and thereby eicosanoid production.
In contrast to dexamethasone, low-dose aspirin failed to limit 28-day mortality, progression to invasive mechanical ventilation or need for renal dialysis in hospitalized COVID-19 patients [99].This may be because of failure of low-dose aspirin in high thromboxane states, as has been reported in the obese and elderly [100,101].NF-kB has also served as a therapeutic target in COVID-19, and NF-kB antagonism may ameliorate I/R by downregulating COX-2 expression and TxA 2 production [102].
It has been proposed that blocking the deleterious effects of PGD 2 and TxA 2 with the dual DPr2/TPr antagonist Ramatroban might be beneficial in COVID-19 [7 && ].Ramatroban as a TPr antagonist is 100 times more potent than aspirin in inhibiting platelet aggregation and P-selectin expression [103,104].In addition to its antiplatelet action, ramatroban also improves vascular responsiveness; while inhibiting endothelial surface expression of ICAM-1 and VCAM-1; inhibiting MCP-1 expression in response to TNF-a or platelet-activating factor; and inhibiting macrophage infiltration [103].In a rat model of endotoxic shock, ramatroban prevented hypotension and reduced mortality by 45%; effects attributable to marked reduction in myeloperoxidase levels in lungs, ileum and heart and more than 90% reduction in plasma TNF-a levels [105].It is notable that infliximab, a TNF inhibitor used in auto-immune conditions, such as Crohn's disease and rheumatoid arthritis is being tested in the WHO-sponsored SOLIDARITY trial [106].
Dual DPr2/TPr antagonism would not only block the activity of both TxA 2 and PGH 2 on platelets and vessels but abrogate the effects of 11-dehydro-thromboxane B 2 , a stable metabolite of TxA 2 and a full agonist of the DPr2 receptor for PGD 2 [107].Ramatroban is a potent inhibitor of IL-4, IL-5 and IL-13 production induced by PGD 2 [108], and therefore, a promising therapy for the maladaptive immune response in severe COVID-19.Ramatroban (Baynas, Bayer Yakuhin Ltd., Japan) has been safely administered for the treatment of allergic rhinitis in Japan since 2000 [103].Ramatroban exhibits surmountable binding to TPr [109], and reversible antiplatelet action [103].This is of advantage in the event of bleeding complications reported in 5.6% of critically ill COVID-19 patients [110].
There is increasing evidence of ongoing lung inflammation in patients with long-haul COVID, which may culminate in lung fibrosis [111,112].There is no animal model of long-haul COVID to date.However, an animal model of silicosis exhibits massive increases in lung PGD 2 and TxA 2 associated with lung inflammation and fibrosis [113 && ].In this model, ramatroban reduced macrophage, lymphocyte and neutrophil infiltration of the lungs while inhibiting TNFa, IL-6, IL-1b, IL-18 and NLRP3 activation, thereby reducing inflammation, fibrosis and cardiopulmonary dysfunction [113 && ].Therefore, ramatroban is a promising candidate for chemoprophylaxis and treatment of lung and cardiac injury in COVID-19 thereby mitigating secondary kidney injury.
Soluble epoxide hydrolase inhibitors (sEHIs) have two activities that could be beneficial in COVID-19-induced AKI.Inhibition of sEH can lower linoleic-derived leukotoxin diols and increase arachidonic acid-derived epoxyeicosatrienoic acids (EETs) [81].The therapeutic benefits for sEHIs have been demonstrated in several animal models of kidney disease including AKI [81,114].Cisplatin-induced AKI was opposed by sEH inhibition with decreased NF-kB activation and TNF-a inflammation [115,116].The conversion of AKI to chronic kidney disease can also be opposed by sEHIs.Renal fibrosis induced by unilateral ureter obstruction is prevented by sEH inhibitor administration, which was attributed to increased EET levels and decreased inflammation associated with decreased neutrophil influx [117].Antifibrotic actions for sEHIs can be attributed to suppression of epithelial to mesenchymal transition by increasing E-cadherin, decreasing a-SMA and preventing the phenotype transition [117,118].These kidney animal model studies have demonstrated that sEHIs provide renal protection through anti-inflammatory and antifibrotic mechanisms.
Development of sEHIs have been tested in clinical trials and SEHIs did not demonstrate major side effects or toxicity [119,120].Interestingly, a major finding for the human clinical trial with GSK2256294 found improved endothelial function in obese smokers assessed by forearm blood flow response to the vasodilator bradykinin [121].These actions on endothelial function would be beneficial in COVID-19-induced AKI.Taken together, sEHIs ability to lower leukotoxin diols and increase EETs have vascular, inflammatory, and epithelial actions that could treat COVID-19-induced AKI and prevent the progression to CKD.
In view of the massive increase in leukotrienes in patients with severe COVID-19, Montelukast, a cysteinyl leukotriene receptor 1 antagonist has recently been shown to reduce clinical deterioration in hospitalized patients with COVID-19 in a small retrospective study [122].In a rat model of chronic renal failure, montelukast attenuated oxidative damage, myeloperoxidase, LTB 4 , cytokine levels, reduced glutathione, and finally, damage to glomerular structure [123].Moreover, in a rat model of lipopolysaccharide-induced sepsis, Montelukast reduced renal injury, and levels of inflammatory and oxidative stress markers [124].Montelukast also reduced AKI induced by rhabdomyolysis or intestinal ischemia in rats [125,126].

CONCLUSION
SARS-CoV-2 infection induces a lipid mediator storm with massive increases in pulmonary and systemic thromboxane A 2 , PGD 2 and leukotoxin diols.Thromboxane A 2 and PGD 2 appear to play a key role in COVID-19-associated thromboinflammation and maladaptive immune response, respectively while leukotoxin diols induce mitochondrial dysfunction and cytotoxicity.The lipid mediators may promote AKI, renal fibrosis and development of CKD.Approaches targeting the lipid mediator storm merit further exploration in order to reduce the heavy burden of kidney disease emerging in the wake of the current pandemic.This was the first study to describe splitting angiogenesis as a result of endothelial injury in the lungs of severe COVID-19 patients. 6.

Table 1 .
Renal involvement in coronavirus disease 2019: lipid mediators, pre and inflammatory cytokine, main clinical consequences and therapeutic options