The Role of LPA in Kidney Pathologies Koryun Mirzoyan

The role of LPA in kidney pathologies Koryun Mirzoyan

To cite this version:

Koryun Mirzoyan. The role of LPA in kidney pathologies. Human health and pathology. Université Paul Sabatier - Toulouse III, 2017. English. NNT : 2017TOU30073. tel-01825826

HAL Id: tel-01825826 https://tel.archives-ouvertes.fr/tel-01825826 Submitted on 28 Jun 2018

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. 5)µ4&

&OWVFEFMPCUFOUJPOEV

%0$503"5%&-6/*7&34*5²%&506-064& %ÏMJWSÏQBS Université Toulouse 3 Paul Sabatier (UT3 Paul Sabatier)

1SÏTFOUÏFFUTPVUFOVFQBS Koryun MIRZOYAN le mercredi 20 septembre 2017 5JUSF

The role of LPA in kidney pathologies

²DPMFEPDUPSBMF et discipline ou spécialité ED BSB : Physiopathologie 6OJUÏEFSFDIFSDIF INSERM U1048/I2MC - équipe 12

%JSFDUFVSUSJDF T EFʾÒTF

Dr Jean-Sébastien Saulnier-Blache

Jury:

Pr Isabelle Castan-Laurell (Présidente du Jury) Dr Alberto Ortiz (Rapporteur) Dr Olivier Peyruchaud (Rapporteur) Dr Jean-Sébastien Saulnier-Blache (Directeur de thèse) Aknowledgements

First of all I would like to thank all the members of the jury and reporters for generously accepting to evaluate my work and participating in my thesis defense. Thank you for your valuable advices and comments that highly improved the quality of the thesis.

I would like to thank all organizers of the Erasmus Mundus MEDEA project and especially to Elodie Balonas and Oxana Kunduzova who made big efforts to realize this project and to give me the possibility for this wonderful journey full of with new knowledge, experience and adventures.

Thanks to all members of team 12 who made the 3 years of lab-life filled with lot of work but also with fun and emotions: thanks to Julie, Colette, Bénédicte, Ben, Guylène, Eric, Marion, Dimitri, Stan for sharing the experience and knowledge, for the company, help and advices.

Special thanks to Audrey, who discovered for me all the loveliness of biological experiments and taught me to work in the lab.

Big thanks to Jean-Loup and Joost for accepting me in the team 12 and giving the opportunity to work within this nice team during 3 years.

To PhD students after me, Valerie and Franck, thank you for your company, for belote, pétanque and beer evenings. Look always forward and work hard. The end of the PhD is not so far.

To my mentor Jean-Sébastien: you are the kindest supervisor I have ever met. Thank you for everything, for your supervision, for the advices and the support you gave me. I hope we will meet in the future and continue to collaborate and work together.

Big thanks to my family and all my friends in Armenia, in France and in other pieces all over the world. Without your support and encouragement it wouldn’t be possible to finish this work.

This thesis is dedicated to my parents Samvel Mirzoyan and Janna Sargsyan.

Index

Abstract ...... 6 Résumé ...... 8 Figures and tables legends ...... 10 Abbreviations ...... 11 1. Introduction ...... 13 1.1 The kidneys ...... 13 1.2 Kidney pathologies...... 14 1.2.1 Acute kidney injury ...... 14 1.2.1.1 Definition and classification ...... 14 1.2.1.2 Epidemiology of AKI ...... 17 1.2.1.3 Sepsis-induced AKI ...... 17 1.2.1.3.1 Pathogenesis of sepsis-induced AKI ...... 17 1.2.1.3.1.1 Hemodynamic changes ...... 17 1.2.1.3.1.2 Inflammation in sepsis-induced AKI ...... 19 1.2.1.3.1.3 Adaptive responses in tubular epithelial cells ...... 21 1.2.1.3.2 Treatment strategies ...... 21 1.2.2 Chronic kidney disease ...... 22 1.2.2.1 Definition and classification ...... 22 1.2.2.2 Epidemiology of CKD ...... 23 1.2.2.3 CKD pathogenesis ...... 24 1.2.2.3.1 Glomerulosclerosis ...... 24 1.2.2.3.2 Tubulo-interstitial fibrosis (TIF) ...... 25 1.2.2.3.2.1 Inflammation ...... 25 1.2.2.3.2.2 Appearance of ECM producing cells...... 26 1.2.2.3.2.3 Pro-fibrotic cytokines ...... 27 1.2.2.3.2.4 Accumulation of ECM ...... 27 1.2.2.4 Progression of CKD ...... 29 1.2.2.5 Therapeutic strategies ...... 29 1.2.3 AKI and CKD are interconnected ...... 30 1.2.3.1 AKI is a risk factor for CKD ...... 30 1.2.3.1.1 The mechanisms of AKI progression to CKD ...... 30 1.2.4 Conclusions ...... 31

2. Lysophosphatidic acid ...... 32 2.1-Structure ...... 32 2.2 Synthesis of LPA ...... 32 2.2.1 Intracellular synthesis ...... 33 2.2.2 Extracellular synthesis ...... 33 2.3 Degradation of LPA ...... 34 2.4 Transport of LPA ...... 36 2.5 LPA in biologic fluids ...... 36 2.5.1 LPA in blood ...... 36 2.5.2 LPA in other biological fluids...... 36 2.6 LPA receptors ...... 37 2.6.1 G-protein coupled membrane receptors ...... 37 2.6.1.1. Classification and tissue distribution ...... 37 2.6.1.2 Intracellular signaling pathways ...... 39 2.6.1.3 Relative specificity of LPA receptors towards different LPA species ...... 40 2.6.1.4 Genetic deletions and associated pathologies ...... 41 2.6.2 Intracellular receptors ...... 43 2.6.3 Other potential receptors ...... 44 2.6.4 LPA receptor antagonists ...... 44 2.7 Physiological and pathophysiological effects of LPA ...... 45 2.7.1 Cancer ...... 45 2.7.2 Reproductive system ...... 46 2.7.3 Cardiovascular system ...... 46 2.7.4 Inflammatory diseases ...... 47 2.7.5 Obesity ...... 48 2.7.6 Nervous system ...... 48 2.7.7 Wound healing and fibrotic diseases ...... 49 2.7.7.1 Wound healing ...... 49 2.7.7.2 Pulmonary fibrosis ...... 49 2.7.7.3 Hepatic fibrosis ...... 50 2.7.7.4 Systemic sclerosis ...... 50 2.7.7.5 Kidney fibrosis ...... 50 2.8 LPA in kidney pathology...... 50

2.8.1 LPA and renal obstruction ...... 51 2.8.2 LPA and diabetic nephropathy ...... 52 2.8.3 LPA and other CKDs ...... 54 2.8.4 LPA and acute kidney injury ...... 55 3. Hypothesis ...... 57 4. Results and discussion ...... 58 4.1 LPA and CKD ...... 58 4.1.1 Bibliographic context and objectives ...... 58 4.1.2 Results ...... 59 4.1.3 Conclusions and discussion ...... 77 4.1.3.1 The main results and conclusions ...... 77 4.1.3.2 Discussion ...... 77 4.1.3.2.1 What is the origin of urinary LPA?...... 77 4.1.3.2.2 LPA receptors in subtotal nephrectomy-associated kidney fibrosis ...... 79 4.1.3.2.3 LPC in kidney fibrosis ...... 79 4.2 LPA and AKI ...... 81 4.2.1 Bibliographic context and objectives ...... 81 4.2.2 Results ...... 82 4.2.3 Conclusions and discussion ...... 101 4.2.3.1 The main results and conclusions ...... 101 4.2.3.2 Discussion ...... 101 4.2.3.2.1 How LPA protects against endotoxemia-induced AKI?...... 101 4.2.3.2.2 Urinary LPA decreases in endotoxemia ...... 103 5. Summary ...... 105 6. Annex ...... 107 6.1 Curriculum Vitae...... 107 7. References...... 108

Abstract

Both chronic kidney diseases (CKD) with consecutive development of end stage renal disease (ESRD) and acute kidney injury (AKI) represent worrying problems for healthcare system due to its increased frequency and the lack of efficient treatments. Lysophosphatidic acid (LPA) is a bioactive lysophospholipid that elicits a wide range of cell responses (proliferation, migration, transformation, contraction etc.) through the activation of specific G protein- coupled receptors (LPA1 to LPA6). In this work we were interested in involvement of the LPA and changes in its metabolism in CKD and AKI.

Previous works showed that LPA exerts pro-fibrotic activity and contributes to development of tubulointerstitioal fibrosis (TIF) after ureteral obstruction through activation of LPA1 receptors. In the first part of the thesis we were interested whether LPA signalization is involved in more advanced model of the disease. We found that 5 months after subtotal nephrectomy (SNX) mice develop massive albuminuria, TIF and glomerular hypertrophy compared to control animals. LPA concentration measured by liquid chromatography tandem mass spectrometry was increased in urine but not in plasma of animals. That increase in LPA significantly correlated with albuminuria and TIF. In addition we found a decreased renal expression of lipid phosphate phosphatases (LPP1, 2 and 3) that are responsible for the degradation of LPA by dephosphorylation. Moreover, the expression of LPA1-LPA4 receptors is down-regulated, whereas LPA5 and LPA6 are unchanged. We concluded here that the possible deleterious effect of LPA in the development of CKD in SNX mice was likely related to its increased production rather than an increased sensitivity of the kidney to LPA.

Since LPA was reported previously to protect kidney damage in the course of ischemia/reperfusion injury, and that it was able to mitigate the systemic inflammation and organ damage in sepsis, we were interested in second part of the thesis to determine whether exogenous and/or endogenous LPA might protect against sepsis-associated AKI. C57BL/6 mice were treated with exogenous LPA 18:1 1 hour before being injected with the lipopolysaccharide (LPS) and AKI was analyzed after 24h. LPA pre-treatment significantly mitigated the LPS-induced elevation of plasma urea and creatinine, lessened the up-regulation of inflammatory cytokines (IL-6, TNFα, MCP-1) and completely prevented the down- regulation of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC1α) in kidney. LPA also prevented LPS-mediated alterations of renal mitochondria ultrastructure.

In vitro, pre-treatment with LPA 18:1 (10 μM) significantly attenuated LPS-induced up- regulation of the pro-inflammatory cytokines (TNFα and MCP-1) in RAW264 macrophages. In addition we found that LPS led to the reduction of urinary LPA concentration that was associated with a reduction in LPA anabolic enzymes (autotaxin and acylglycerol kinase), and an elevation in LPA catabolic enzyme (lipid phosphate phosphatase 2) expression in kidney cortex. We concluded hereby that exogenous LPA exerts protection against endotoxemia- induced kidney injury. Moreover, the observation that LPS reduces the renal production of LPA suggests that sepsis-associated AKI could be mediated, at least in part, by alleviation of the protective action of endogenous LPA.

In general our work shows that LPA local metabolism is altered in both forms of kidney diseases. In course of sepsis-induced AKI LPS leads to increased local catabolism of LPA leading to low availability of the phospholipid and alleviating its protective effect whereas in advanced CKD the local catabolism of the phospholipid is decreased with subsequent increase of urine LPA that favors development of the disease. Targeting LPA catabolism can be an interesting approach in treatment of kidney diseases.

Résumé

Les maladies rénales chroniques (MRC) et l'insuffisance rénale aiguë (IRA) sont des problèmes essentiels de santé publique en raison de l’augmentation continue de leur fréquence et du manque de solutions thérapeutiques contre ces maladies. L'acide lysophosphatidique (LPA) est un lysophospholipide bioactif qui induit un large éventail de réponses cellulaires par le biais de récepteurs membranaires spécifiques (LPA1 à LPA6) couplés aux protéines G. Dans ce travail, nous nous sommes intéressés aux effets biologiques et au métabolisme du LPA dans les MRC et l’IRA.

Des travaux antérieurs de l’équipe avaient montré que le LPA contribuait, via le récepteur LPA1, au développement de la fibrose tubulointerstitio (TIF) dans un modèle de MRC chez la souris : l’obstruction urétérale. Dans la première partie de la thèse nous avons étudié l’implication du LPA dans un modèle plus avancé de MRC: la néphrectomie subtotale (SNX) chez la souris. Nos travaux ont montré que 5 mois après chirurgie les souris (SNX) développaient une albuminurie massive associée à une TIF sévère et à une hypertrophie glomérulaire. Chez ces souris la concentration en LPA mesurée par chromatographie liquide en spectrométrie de masse en tandem était augmentée dans l'urine et étroitement corrélée à l'albuminurie et à la TIF. En parallèle, nous avons observé une diminution de l'expression rénale des Lipid-Phosphate Phosphatases (LPP 1, 2 et 3) responsables de l’inactivation du LPA. Nous avons également observé que l'expression rénale des récepteurs LPA1, 2, 3 et 4 était diminuée chez les souris Snx. Nous avons conclu que les effets délétères éventuels du LPA dans le développement de la MRC chez les souris SNX était vraisemblablement lié à une augmentation de sa production rénale plutôt qu’à une sensibilité accrue du rein au LPA.

Des travaux antérieurs avaient montré que l’injection de LPA protégeait contre l’apparition des lésions rénales induites par ischémie/reperfusion chez la souris. Une autre étude avait montré que le LPA permettait d’atténuer l'inflammation systémique et les dommages aux organes induits par un choc septique. Dans la deuxième partie de la thèse, nous avons étudié l’influence du LPA sur l’IRA induite par une endotoxémie au LPS (lipopolysaccharide) chez la souris. Nous avons observé que l’injection de LPA permettait d’atténuer l'élévation d'urée et de créatinine plasmatiques, ainsi que l’augmentation d’expression rénale de cytokines inflammatoires (IL-6, TNFα, MCP-1) induites par le LPS. Le LPA a également empêché la baisse d’expression rénale du facteur PGC1α ainsi que les altérations ultra-structurales des mitochondries rénales induites par le LPS. In vitro, le LPA atténue l’augmentation

8 d’expression des cytokines pro-inflammatoires (TNFα et MCP-1) induite par le LPS dans les macrophages RAW264. Enfin, nos travaux ont montré que l’endotoxémie au LPS chez la souris entrainait une réduction de la concentration urinaire de LPA associée à une réduction des enzymes anaboliques LPA (autotaxine et acylglycérol kinase) et une élévation de l'expression de de LPP2, dans le cortex rénal. Nous en avons conclu que l'IRA associée à l’endotoxémie pourrait être liée, au moins en partie, à une réduction de la production rénale de LPA et, par voix de conséquence, de ses effets anti-inflammatoires protecteurs de la fonction rénale.

En conclusion, notre travail montre que les variations de production rénale de LPA pourraient jouer un rôle important dans le développement des maladies rénales. L’augmentation du LPA dans les MRC favoriserait ses effets délétères (fibrose, inflammation). Sa réduction dans l’IRA réduirait ses effets anti-inflammatoires. Cibler le catabolisme LPA pourrait donc être une approche intéressante dans le traitement des maladies rénales.

Figures and tables legends

Figure 1: Location and structure of kidneys

Figure 2: Etiologies of AKI

Figure 3: Microcirculatory changes and recognition of DAMPs/PAMPs by tubular cells

Figure 4: Development of TIF

Figure 5: Structure of LPA

Figure 6: LPA metabolic pathways

Figure 7: Expression pattern of LPA receptors

Figure 8: LPA receptor signaling pathways

Table 1: Classification of AKI

Table 2: GFR categories of CKD

Table 3: Albuminuria categories of CKD

Table 4: The phenotypic characteristics of LPA receptor KO animals and involvement of LPA receptors in pathologies

Abbreviations

αSMA: alpha smooth muscle actin ADPKD: autosomal dominant polycystic kidney disease AGK: acylglycerolkinase AGPAT: acyl-glycerol-3-phosphate acyltransferase AKI: acute kidney injury Akt: protein kinase B ATN: acute tubular necrosis ATX: autotaxin BAL: broncho-alveolar lavage BUN: blood urea nitrogen CKD: chronic kidney disease CTGF: connective tissue growth factor DAG: diacylglycerol DAGK: diacylglycerol kinase DAMP: disease-associated molecular pattern DN: diabetic nephropathy ECM: extracellular matrix EDG: endothelial differentiation gene EGFR: epidermal growth factor receptor EMT: epithelial mesenchymal transition ESRD: end stage renal disease FGF-2: fibroblast growth factor-2 FSGS: focal-segmental glomerulosclerosis GBM: glomerular basement membrane GFR: glomerular filtration rate GPAT: glycerol phosphate acyltransferase GPCR: G protein coupled receptor HIV: human immunodeficiency virus ICAM-1: intercellular adhesion molecule 1 ICU: intensive care unit iNOs: inducible nitric oxide synthase I/R: ischemia/reperfusion KDIGO: Kidney Disease International Global Outcomes LDL: low-density lipoprotein LPAAT: lysophosphatidic acid acyltransferase LPA: lysophosphatidic acid LPC: lysophosphatidylcholine LPE: lysophosphatidylethanolamine LPL: lysophospholipid LPP: lipid-phosphate phosphatase LPS: lipopolysaccharide

LPS: lysophosphatidylserine MAGK: monoacylglycerol kinase MAPK: mitogen-activated protein kinase MMP: matrix metalloproteinase NO: nitric oxide PA: phosphatidic acid PAF: platelet activating factor PAMP: pathogen-associated molecular pattern PDGF: platelet derived growth factor PI3K: phosphatidylinositol 3-kinase PL: phospholipid PLA: phospholipase A PLD: phospholipase D PPAR: peroxisome proliferator-activated receptor PPRE: peroxisome proliferator response elements PGC1α : peroxisome proliferator-activated receptor gamma coactivator 1α RAAS: renin-angiotensin-aldosterone system RAGE: receptor for advanced glycation endproducts RBF: renal blood flow ROS: reactive oxygen species RRT: renal replacement therapy sEH: soluble epoxide hydrolase TG: triglyceride TGFβ: transforming growth factor β TIF: tubulo-interstitial fibrosis TIMP-1: 1tissue inhibitor of metalloproteinase 1 TLR: toll-like receptor tPA: tissue plasminogen activator TRPV1: transient receptor potential vanilloid 1 uPA: urokinase-plasminogen activator UUO: unilateral ureteral obstruction VCAM-1: vascular cell adhesion molecule 1

1. Introduction

1.1 The kidneys

The kidneys are paired bean-shaped organs located below the rib cage in retroperitoneal space. Their function is to filter plasma and remove waste products, regulate blood and electrolyte balance in the body (Figure 1).

Figure 1.Location and structure of kidney (A) and structure of nephron (B)(adapted from OpenStax College, Anatomy and Physiology. OpenStax CNX).

Histologically kidney consists of the renal cortex and renal medulla. Elementary structural and functional units of the kidney are nephrons, multicellular organizations found along cortex and medulla. The structures responsible for the filtration of the plasma and formation of primary urine are glomerules. Primary urine is further concentrated in kidney tubules, where some more products are reabsorbed or secreted.

In addition to their filtration activities kidneys play an important role in maintaining body homeostasis by producing biologically active compounds such as renin (control of blood

13 pressure), erythropoietin (stimulation of red blood cell formation in bone marrow), calcitriol (active form of vitamin D involved in maintenance of Ca2+ homeostasis).

1.2 Kidney pathologies

Historically diminished kidney function was classified as two distinct syndromes: acute kidney failure and chronic kidney failure.

1.2.1 Acute kidney injury

The concept of acute kidney failure is relatively new. Clinical characterization of patients was described during 2nd World War when occurrence of massive kidney function loss was realized after rhabdomyolysis (crush injury) [1]. Recently Acute Kidney Injury (AKI) has replaced the term acute renal failure to emphasize that a continuum of kidney injury occurs long before sufficient decrease of excretory kidney function can be detected.

1.2.1.1 Definition and classification

AKI is defined as a clinical syndrome characterized by rapid (over the course of hours to days) diminution of kidney excretory function and accumulation of waste products such as creatinine, blood urea nitrogen (BUN) or other toxins not measured routinely. Pathogenetic basis of the accumulation of these waste products is a rapid decline of glomerular filtration rate (GFR) caused by variety of etiologic factors. In addition, electrolyte (sodium, potassium, phosphorus) and fluid balance in body is largely altered in AKI.

The proper staging of AKI is crucial since later stages of AKI are associated with an increased risk of death and the need for renal replacement therapy (RRT) increases. Different classifications of AKI were proposed. Initial criteria was based on the level of creatinine or glomerular filtration rate (GFR) proposed in 2004 (RIFLE) [2]. That classification was later modified and variations in diuresis have been taken into account in the classification proposed by Acute Kidney Injury Network (AKIN) [3]. The actually used classification was proposed in 2012 by Kidney Disease International Global Outcomes group (KDIGO) [4] and uses a combination of previous two scores (RIFLE and AKIN) (Table 1).

Stage Serum creatinine Diuresis 1 1.5-1.9 times baseline < 0.5 ml/kg/h for 6-12 hours OR ≥0.3mg/dl(≥26.5μmol/l) increase 2 2.0-2.9 times baseline < 0.5 ml/kg/h for ≥ 12 hours 3 3.0 times baseline < 0.3 ml/kg/h for ≥ 24 hours OR OR Increase in serum creatinine to anuria for ≥ 12 hours ≥ 4.0mg/dl (≥353.6 μmol/l) OR Initiation of renal replacement therapy OR , In patients < 18 years, decrease in eGFR to < 35 ml/min per 1.73 m2

Table 1. KDIGO classification of AKI.

Clinically AKI can be grouped into three primary types: prerenal, renal and postrenal [5]. (Figure 2)

Prerenal. Maintenance of the normal GFR depends on adequate perfusion of the kidneys. Prerenal AKI is defined as decrease in GFR due to decline in renal perfusion pressure without initial intrinsic damage of the renal parenchyma. Kidneys receive up to 25% of cardiac output and any decline in general or local, intrarenal circulation can alter kidney function. Causes of prerenal AKI are hypovolemia (hemorrhage, diarrhea, vomiting etc.), impaired cardiac output (congestive heart failure, pericardial tamponade etc), decreased vascular resistance and vasodilation (sepsis, vasodilatators etc). Rapid restoration of intravascular volume will improve kidney function. However if decline in kidney perfusion is long enough it can lead to ischemic kidney disease indicating that there could be a continuum from prerenal to renal AKI [5, 6].

Renal. There is a broad list of factors that cause renal type of AKI. Theoretically damage to all different kidney structures can cause AKI. Acute tubular necrosis (ATN) is the histological hallmark of AKI due to damage to kidney tubules and can be caused most frequently by ischemia (caused by severe decline in renal perfusion) or nephrotoxic agents (exogenous such as aminoglycosides, radiocontrast media, etc. or endogenous such as myoglobin in

15 rhabdomyolysis etc). Glomerular damage (in severe cases of glomerulonephritis), interstitial (interstitial nephritis due to an allergic reaction to drugs or to an infection) and vascular alterations (damage to intrarenal vessels due to malignant hypertension, hemolytic-uremic syndrome etc.) are other possible mechanisms for the development of renal AKI [5, 6].

Postrenal. Postrenal AKI is caused by acute obstruction of urine flow that increases intratubular pressure and decreases GFR. Obstruction of the urinary tract at any level may produce AKI. However obstruction must concern the 2 kidneys or a solitary kidney to get AKI development (benign prostatic hyperplasia, prostatic or gynecological cancers, retroperitoneal fibrosis, urolithiasis etc.) [5, 6].

Acute kidney injury

Prerenal AKI Renal AKI Postrenal AKI

Tubular Glomerular Interstitial Vascular damage damage damage damage

Ischemic Nephrotoxic

Figure 2.Etiologies of AKI (adapted from [6]).

1.2.1.2 Epidemiology of AKI

The incidence of AKI is continuously increasing. The number of hospitalizations for dialysis- requiring AKI increased from 63000 to 164000 and the population incidence rate of dialysis- requiring AKI increased from 222 to 533 cases/million person-years between 2000 and 2009 years in US [7]. The total number of inpatient deaths associated with dialysis-requiring AKI increases in parallel to incidence rate from 18000 in 2000 to 39000 in 2009[7]. Incidence of AKI is especially high in intensive care settings and can rich up to 67% [8]. In 47.5% of critically ill patients in intensive care units (ICU) AKI was associated with sepsis representing the most common contributing factor of AKI in ICU patients [9]. Patients with sepsis develop more severe AKI, are more likely to receive RRT, display less frequency of renal function recovery, longer ICU stay and higher ICU mortality [8, 10]. From the economical point of view AKI is a big burden to healthcare systems since reported hospitalized patients increase hospital costs compared to patients who did not develop AKI due to the long ICU stay with greater number of investigations and monitoring as well as RRT use [11].

1.2.1.3 Sepsis-induced AKI

Despite the high incidence and increased mortality, molecular pathways and mechanisms mediating sepsis-induced AKI are poorly defined and no specific targeted therapies exist. The pathogenesis of septic AKI is very complex and involves pathobiological events such as microcirculatory changes with endothelial disturbances and changes in coagulation cascade, inflammation, oxidative stress and adaptive cell response to injury.

Histological changes in septic AKI are sparse and moderate and do not explain the clinical phenotype observed in patients. Tubular cell injury is heterogeneously distributed along kidney and nonspecific changes are more prominent such as apical vacuolization, loss of brush border, tubular cell swelling. Tubular epithelial cell death is rather mediated by the mechanism of apoptosis than necrosis in contrast to ischemic and toxic AKI, where ATN is the prominent histopathological finding [12, 13].

1.2.1.3.1 Pathogenesis of sepsis-induced AKI

1.2.1.3.1.1 Hemodynamic changes Traditionally AKI development in patients with sepsis was attributed to hemodynamic changes secondary to a mismatch between kidney demand and perfusion. Decreased renal blood flow (RBF) due to systemic vasodilatation and decreased systemic vascular resistance

17 were considered as leading factors that cause diminution in kidney perfusion in conditions of increased kidney metabolic demand due to sepsis. There are conflicting results in studies using animal models of sepsis regarding RBF changes in septic AKI with some studies indicating decreased RBF, whereas others declared unchanged or even increased RBF. More recent studies indicated development of AKI in septic animals in parallel with increased RBF and renal vasodilatation [14, 15]. Number of human studies that directly measured RBF by invasive techniques remains low and in all studies RBF was preserved or increased during septic AKI [16]. In addition, Murugan et al showed in their prospective study that AKI is a common finding in patients with community-acquired non severe pneumonia and could develop even in patients who do not present any signs of hemodynamic instability and were not transferred to ICU unit [17]. Today it is generally accepted that GFR decline and AKI development in sepsis are not attributed to decreased RBF but can happen even in case of increased RBF; the term so-called hyperemic AKI was proposed [18].

Further studies showed that the major hemodynamic alterations in kidneys of septic-AKI are in the level of microcirculation. Especially it was shown that there is a decrease in renal capillaries with continuous flow and a parallel increase in capillaries with intermittent or no flow [19, 20]. Studies in porcine model of sepsis showed that in conditions of overall increased RBF, cortical microvascular perfusion was decreased due to shunting of cortical blood flow towards medulla and renal perfusion pressure was significantly decreased due to increased renal venous pressure [21]. In addition, a retrospective observational study showed that elevated central venous pressure was associated with progression to AKI in septic patients suggesting that venous congestion plays a role in septic AKI [22]. The pathogenic basis of microcirculatory changes in septic kidneys could be the endothelial dysfunction leading to vasoconstriction which, in parallel with activation of platelets and coagulation cascade, can cause capillary occlusion. As a result, vascular permeability increases causing interstitial edema, which in turn leads to increased venous pressure and affects GFR.

These data can indicate that microcirculatory alterations and increased renal venous pressure with further decrease in renal perfusion pressure rather than decrease in overall RBF are important in underlying hemodynamic changes in septic AKI.

It is important to mention here that nitric oxide (NO) also plays a role in septic AKI. The inducible NO synthase (iNOs) is constitutively expressed in kidney and its expression is induced during sepsis [23]. Synthesized NO can exert different actions. On one hand it can

18 lead to systemic vasodilatationand hemodynamic instability with further activation of sympathetic system and release of angiotensin. This in turn can lead to constriction of intrarenal vessels with resulting decrease in GFR [24]. On the other hand, reacting with reactive oxygen species (ROS) produced in sepsis, NO can be converted to peroxynitrite (ONOO-) that is a potent oxidant molecule that could be responsible for oxidative stress development and cytotoxic effects of NO. Selective inhibitors of iNOS showed some protective effect in sepsis-induced AKI [25, 26].

1.2.1.3.1.2 Inflammation in sepsis-induced AKI Sepsis is considered as a hyper-inflammatory condition with “cytokine-storm” as a response to the pathogen. In sepsis, pathogen triggers a host response in which inflammatory mechanisms contribute to clearance of infection and tissue recovery on one hand and to organ damage on the other hand. Inflammation is initiated when cells of innate immune system recognize highly conserved pathogen-associated molecular patterns (PAMP) such as lipopolysaccharide (LPS). Pattern-recognition receptors including toll-like receptors (TLR), C-type lectin receptors (CLR), nucleotide-binding oligomerization domain like receptors (NOD-like receptors) and retinoic acid inducible gene 1-like receptors (RIG-1-like receptors) are responsible for the recognition of PAMPs [27]. These receptors also recognize endogenous molecules released from injured cells known as damage-associated molecular patterns (DAMPs) (DNA, RNA, histones, etc.). The engagement of these receptors results in up-regulation of inflammatory gene transcription and initiation of immune response. Different cytokines such as TNFα, IL6, IL10 are found in high concentrations in septic patientsand there is a positive correlation between the level of the cytokines and sepsis-induced AKI [17, 28].

Synthesized cytokines activate different types of cells such as leukocytes, endothelial cells, platelets, epithelial cells. Activated endothelial cells up-regulate expression of adhesion molecules such as E-selectin, P-selectine, intercellular adhesion molecule 1 (ICAM-1) and vascular cell adhesion molecule 1 (VCAM-1) [29-31]. Increased expression of adhesion molecules in parallel with microvascular changes and sluggish blood flow promotes leukocyte adhesion to endothelial cells, leads to increased leukocyte transit through kidney capillaries and increases the general leukocyte recruitment into kidney. These activated leukocytes in turn produce different bioactive molecules (ROS, proteases, elastases etc.) that contribute to further tissue damage. Leukocytes that leave peritubular capillaries are in close proximity with tubular epithelial cells and can activate these cells directly by releasing pro-inflammatory

19 mediators and DAMPs locally. Interestingly epithelial cells can also be activated by the tubular side as PAMPs, DAMPs and pro-inflammatory cytokines can be filtered in glomerulus and reach the tubular epithelial cells. Recognition of DAMPs and PAMPs in epithelial cells can be mediated through TLR receptors that were expressed in murine and human kidney tubules [32]. TLR-4 receptors play important role in tubular epithelium as they are responsible for the recognition and internalization of luminal LPS [33, 34]. Downstream to TLR-4 activation could be the activation of the transcription factor NF-κB leading to subsequent increased production of cytokines and enahanced inflammatory signal. After internalization of LPS, epithelial cells in S1 segment of proximal tubules signal in a paracrine fashion to more distally located epithelial cells and cause oxidative stress in neighboring S2 segment thus giving to the cells of S1 segment properties of sensor for danger signals [34]. Distally located epithelial cells change their metabolism to adapt to a potentially lethal new situation (Figure 3). It remains however not completely understood how TLR-4 activation and induction of downstream inflammatory cascade could lead to decreased GFR and AKI development. One possible mechanism explaining this phenomenon could be the altered electrolyte transport in proximal tubules caused by sepsis with increased delivery of sodium and chloride to macula densa that causes enhanced tubuloglomerular feedback resulting in decreased GFR [35].

Figure 3. Schematic presentation of sepsis-induced microcirculatory changes and recognition of DAMPs/PAMPs by proximal tubular epithelial cells. Taken from [36].

1.2.1.3.1.3 Adaptive responses in tubular epithelial cells The oxidative stress and inflammatory processes trigger responses in tubular epithelial cells that are adaptive in their nature. These adaptive responses lead to conservation of existing energy with expenditure only in processes crucial for cell survival. This “energy conserving” adaptive response of tubular epithelial cells is regulated by mitochondria. It was shown that oxygen consumption by kidney tubular epithelial cells is reduced in septic conditions and was mediated by peroxisome proliferator-activated receptor gamma (PPARγ) coactivator 1α (PGC1α), a major regulator of mitochondrial biogenesis and metabolism. PGC1α was significantly down-regulated in kidneys of septic animals exposed to LPS, whereas increased expression of PGC1α was associated with restored oxygen consumption and improved kidney function [37, 38]. Another component of adaptive cell response in septic kidneys is cell cycle arrest. It was shown that cell cycle arrest in G1-S checkpoint is associated with kidney injury and the recovery of kidney function is associated with cell cycle progression within 48 hours [39]. In agreement with this, two molecules known as markers of cell cycle arrest (tissue inhibitor of metaloproteinases 2 and insulin-like growth factor-binding protein 7) are efficient biomarkers in assessing the risk of development of AKI in critically ill patients [40]. Limiting cell replication conserves energy levels and prevents cells from undergoing an energetically overtaxing behavior that can trigger cell death. One more adaptive response in tubular cells is the process of mitophagy, a process of removal and autodigestion of damaged mitochondria. It was shown that mitophagy is induced very early during sepsis in kidney by different factors including inflammation and oxidative stress and that decreased mitophagy is associated with proximal tubular dysfunction and overall worse outcome in critically ill patients [41, 42]. The biological role of mitophagy is to eliminate the injured mitochondria with subsequent diminution in ROS production and again energy conservation.

It is assumed now that the clinical phenotype of AKI is the early expression of the adaptive response of tubular cells to injurious signal. The inflammatory response and the oxidative stress during sepsis cause an adaptive response in tubular epithelial cells with a down- regulation of cell function aiming to reduce energy consumption and to ensure cell survival. As a result kidney function is reduced [43].

1.2.1.3.2 Treatment strategies

Treatment strategies targeting several pathogenic loops of sepsis were proposed previously. Targeting inflammatory cytokines TNFα [44], IL1β [45] as well as LPS [46] and TLR-4

21 receptors [47] were tested. Neither of these strategies was proven as an efficient treatment option for sepsis or associated complications. In preclinical studies, various molecules displayed protective effects during sepsis-induced AKI (resveratrol [19], 2-methoxyestradiol [48], temsirolimus [49], alkaline phosphatase [50], aminoguanidine and methylguanidine [25, 26], pentoxyfilline [51]). Protective effect of alkaline phosphatase was proven in phase IIa clinical trial and if benefits are proven in large-scale studies it could be a promising treatment approach for sepsis-induced AKI [52]. Despite a better comprehension of the pathogenesis of sepsis-induced AKI and proposed different treatment modalities, pathogenic curative methods are not available so far and the only strategy combating sepsis-induced AKI remains RRT.

1.2.2 Chronic kidney disease

1.2.2.1 Definition and classification

Chronic kidney disease (CKD) is defined as abnormalities of kidney function or structure present for more than 3 months, with implications for health [53]. In contrast to relatively short history of AKI, CKD was identified long time ago (in 19th century) however pathogenetic treatment is not developed so far. In severe failure RRT (peritoneal dialysis or hemodialysis, or kidney transplantation) is the only option. KDIGO classified CKD in 5 stages depending on the severity of the disease, based mainly on measured or estimated GFR [53]. In Stage 1 kidney function is reserved and in Stage 2 it is reduced minimally. RRT is generally initiated in Stage 5.

GFR stage Description GFR (mL/min/1.73m2) 1* Normal or high ≥90 2* Mildly decreased 60-89 3A Mildly to moderately decreased 45-59 3B Moderately to severely decreased 30-44 4 Severely decreased 15-29 5 Kidney failure <15

Table 2. GFR categories of CKD proposed by KDIGO [53]. *In the absence of kidney damage neither stage 1 nor stage 2 fulfill the criteria for CKD

In 2012 it was proposed by KDIGO to include the levels of albuminuria (Table 3) and the cause of the CKD in the classification as they can affect the outcomes and the choice of treatment.

Albumin excretion rate Albuminuria stage Definition (mg/24 hours) 1 Normal to mildly increased <30 2 Moderately increased 30-300 3 Severely increased >300

Table 3. Albuminuria categories in CKD proposed by KDIGO [53].

CKD can develop in case of any damage in any structural compartments of kidney: glomerules, tubulo-interstitium or vasculature. Many factors acting on kidneys can cause CKD such as metabolic, ischemic, immunologic, toxic, hereditary, mechanical etc. CKD can develop as a consequence of systemic diseases (diabetes mellitus, arterial hypertension, autoimmune diseases, systemic vasculitis etc.) or can be the result of a primary kidney disease (glomerulonephritis, focal-segmental glomerulosclerosis, cystic diseases of kidneys etc.). Diabetes is the leading cause of CKD in developed and developing countries where the second most common cause is hypertension [54]. In countries with low income, the major cause is glomerulonephritis or the etiological factors remain unknown [54]. The differences in causative factors of CKD in different areas are due to moving the disease burden from infectious towards chronic lifestyle-related diseases,increased life-expectancy and decreased births in developed countries, whereas infectious diseases continue to be prevalent in countries with low-income due to poor sanitation, inadequate supply of safe water and high concentrations of disease-transmitting vectors [54].

1.2.2.2 Epidemiology of CKD

CKD is one of the major public health problems worldwide and with its final stage (end stage renal disease, ESRD) represents a huge burden to healthcare systems. According to the Global Burden of Diseases study, CKD is among the top three fastest growing main causes of death worldwide, behind HIV/AIDS and diabetes mellitus. The absolute number of deaths from CKD has increased by 82% between 1990 and 2010 [55]. Age-standardized death rates from CKD increased by 15% in this period while rates for most diseases fell, including other non-

23 transmissible diseases such as major vascular diseases, chronic pulmonary disorders, most forms of cancer and liver cirrhosis [55]. It was observed that for most non-transmissible causes of death, the global increase in the number of deaths was due to ageing of the population, while for CKD both ageing of the population and ageing-independent factors were contributing [55]. CKD is associated with an increased risk of dying from cardiovascular causes (including ischemic heart disease, cerebrovascular diseases, heart failure, arrhythmias), but also from non-cardiovascular causes such as infections, neurologic disorders, cancer [56]. Despite these worrying epidemiologic data pathogenetic or regression-inducing treatments are not available and the RRT remains the only option for developed ESRD thus representing a huge economical problem for healthcare systems. About 2-3% of healthcare expenditure in developed countries is used to provide treatment to patients with ESRD [54].

1.2.2.3 CKD pathogenesis

As mentioned above multiple causative factors can underlie the development of CKD. Whatever the nature of the initial injury, the final common manifestation of many CKDs is renal fibrosis. Like in other organs, fibrosis in kidney is characterized by a succession of events which initial final biological aim is to regenerate damaged tissue. Fibrosis is a deregulated form of fibrogenesis characterized by excessive extracellular matrix (ECM) accumulation that leads to the alteration of tissue architecture and ultimately function loss.

Kidney fibrosis comprises the processes of glomerulosclerosis, tubulo-interstitial fibrosis (TIF) or the combination of both.

1.2.2.3.1 Glomerulosclerosis

Two most common causes of CKD (diabetes mellitus and hypertension) as well as a big group of primary glomerulopathies and glomerular alterations due to systemic autoimmune diseases are characterized by initial alterations of glomerules. The initial insult is on glomerular endothelium causing activation and dysfunction of endothelial cells, hemodynamic alterations and loss of glomerular basement membrane (GBM) electrical charge. Later the glomerular inflammatory reaction is initiated leading to interaction between inflammatory and mesangial cells causing activation, proliferation and dysfunction of the latter. Under the influence of inflammatory cytokines (especially TGFβ) mesangial cells undergo phenotypic changes with the formation of cells with embryonic mesenchymal phenotype (mesangioblast). These cells are capable of excessive ECM production leading to mesangial expansion. Another feature of

24 glomerular damage is podocyte injury with decreased podocyte number, foot process effacement and denudation of glomerular basement membrane. Persistent and continuing insult to endothelial-mesangial-epithelial axis will lead to cell death by apoptosis and replacement of glomerular cells by ECM with the final development of glomerulosclerosis [57, 58].

1.2.2.3.2 Tubulo-interstitial fibrosis (TIF)

Even though the majority of CKDs initiates with the glomerular damage and are associated with glomerulosclerosis, the fibrosis develops thereafter towards tubulo-interstitial compartment. Moreover there is a strong association between renal functional impairment and the level of TIF in patients with primary glomerular alterations [59]. Various mechanisms underlie development of the TIF after initial glomerular damage. Abnormally-filtered proteins including albumin, complement and cytokines activate tubular cells and initiate an inflammatory response. In parallel hemodynamic alterations lead to decreased postglomerular blood flow to peritubular capillaries leading to hypoxic damage of tubular epithelium.

The TIF is a complex processes that involves inflammation, appearance of ECM producing cells and eventual ECM accumulation.

1.2.2.3.2.1 Inflammation The striking histological feature of TIF is interstitial inflammatory infiltrate consisting of different inflammatory cells. Increased expression of chemokines (CCL2/MCP1, CCL5/RANTES) and adhesion molecules (E and P selectins, ICAM-1, VCAM-1) by kidney resident cells regulate leukocyte recruitment into kidneys. Moreover, targeting MCP-1 and its receptor CCR2 [60-62] as well as P-selectin [63], ICAM-1 [64, 65] protects against fibrosis in different models of kidney diseases. Although a small population of resident interstitial macrophages may proliferate in response to renal injury, most of the inflammatory cells are recruited from circulation. Recruited monocytes trans-differentiate into macrophages in kidney interstitium. There is a strong correlation between macrophage infiltration and the extent of fibrosis whereas depletion of macrophages ameliorates fibrogenesis in mice [66, 67]. Macrophages can be categorized into classically activated M1 macrophages and alternatively activated M2 macrophages. The M1 macrophages are mostly implicated in formation of ROS and pro-inflammatory cytokines (TNFα, IL1β) which in high concentrations could be detrimental for resident cells [68]. In contrast M2 macrophages produce anti-inflammatory

25 cytokines such as IL10 and proteins of ECM (fibronectine) playing role in tissue repair. These cells produce also pro-fibrotic cytokines transforming growth factor β (TGFβ) and connective tissue growth factor (CTGF) which promote accumulation and activation of myofibroblasts [68].

The other types of recruited inflammatory cells are lymphocytes, mast cells, dendritic cells. Role of lymphocytes in CKD development was shown in different models [69-71], whereas mast cell involvement remains controversial. While some studies indicate that mast cell depletion enhances kidney fibrosis upon ureteral obstruction [72] or puromycine-induced glomerulonephritis [73] indicating possible protective role of these cells, another study showed that mast cells participate in the aggravation of glomerular lesions and kidney function loss in experimental model of crescentic glomerulonephritis [74]. The precise role of dendritic cells in kidney fibrosis is not yet fully understood.

1.2.2.3.2.2 Appearance of ECM producing cells Myofibroblasts in the renal interstitium are considered as the principal source of fibrillar matrix. Schematically myofibroblasts are activated fibroblasts that express alpha smooth muscle actin (αSMA). Sources of myofibroblasts in kidney fibrosis are different:

x The normal interstitium is populated by small number of fibroblasts that are residual embryonic mesenchymal cells and that are relatively quiescent. Under the influence of different molecules released in inflammatory conditions (TGF, fibroblast growth factor-2 (FGF-2), CTGF etc.) resident fibroblasts may proliferate and undergo phenotypic transformation acquiring fibrosis-promoting characteristics [75]. x Another source of myofibroblasts are circulating fibrocytes, that are bone marrow- derived cells expressing leukocytes (CD34 or CD45) and mesenchymal (collagen I or vimentne) markers [76]. While ECM production of fibrocytes is relatively low, they produce large amounts of collagen and fibronectine after transformation into myofibroblasts [76]. x One more important mechanism responsible for the appearance of myofibroblasts is the so called epithelial mesenchymal transition (EMT) during which epithelial cells dedifferentiate into matrix producing fibroblasts and myofibroblasts. It was estimated that during CKD progression about 30% of fibroblasts in kidneys are derived from EMT [77, 78]. In response to inflammatory cytokines tubular cells show down-regulation of their epithelial and up-regulation of a mesenchymal genetic program resulting in a loss of epithelial markers such as E-cadherin, zonula occludens-1, cytokeratine and acquisition of mesenchymal

26 markers such as vimentin, matrix metalloproteinase-2, fibroblast specific protein-1, type I collagen and fibronectine. In addition epithelial cells undergo phenotypic changes such as loss of epithelial cell adhesion and polarity, actin reorganization and αSMA expression, tubular basement membrane disruption and increased cell migration and invasion [79]. x Another source of kidney fibroblasts are endothelial cells that undergo transcriptional reprogramming under certain conditions and regain characteristics of matrix producing fibroblasts (endothelial mesenchymal transition) [80].

1.2.2.3.2.3 Pro-fibrotic cytokines TGFβ: It is a pleiotropic cytokine that controls various cellular processes such as proliferation, differentiation and apoptosis and participates in number of physiological and pathological processes such as angiogenesis, immune processes, tissue reparation, cancer etc [81]. It has strong pro-fibrotic potential and is a key molecule in kidney fibrosis [82]. TGFβ stimulates ECM production and its strong up regulation is described in number of kidney pathologies. Its exogenous administration induces kidney fibrosis, whereas TGFβ signalization blockade attenuates progression of fibrosis [82]. This molecule acts through activation of TGFβ type I or type II receptors and phosphorylation of cytoplasmic Smad proteins which translocate to the nucleus and activate target genes [82].

CTGF (CCN2): It is a secreted growth factor that is implicated in various physiologic processes such as angiogenesis, osteogenesis, tissue repair but also in pathologic conditions such as fibrosis and cancer [83]. It was shown that CTGF induces EMT, promotes production of chemokines and pro-inflammatory cytokines from epithelial cells, recruitment of macrophages, differentiation of fibroblasts to myofibroblasts and ECM production acting alone or in combination with other cytokines like TGFβ [84, 85]. CTGF expression increases in the zones of TIF in kidneys of patients with different forms of nephropathies [86]. Interestingly there is a crosstalk between actions of CTGF and TGFβ: while TGFβ upregulates CTGF expression in different cell types, CTGF itself modulates TGFβ action by activating latent TGFβ and favoring binding of TGFβ to its receptor[87].

1.2.2.3.2.4 Accumulation of ECM Accumulation of ECM proteins within kidney interstitium is a dynamic process that changes over time by intensity and macromolecular composition. Early phases of fibrosis are characterized by accumulation of loose connective tissue composed of fibronectine and collagen type I and III that are sensible to degradation [88]. In advanced stages glycoproteins,

27 proteoglycans and components of basal membrane (laminine, collagen type IV) appear and the proteins undergo irreversible reticulation making the matrix more rigid and more resistant to proteases [88].

The normal kidneys contain proteases that are implicated in matrix degradation and keep the dynamic balance between synthesis and degradation of ECM. Among them, the matrix metalloproteinase 2 and 9 (MMP2 and MMP9) as well as the serine proteases urokinase- plasminogen activator (uPA) and the tissue plasminogen activator (tPA) are known. During fibrosis expression of plasminogen activator inhibitor-1 (PAI-1, a specific inhibitor of uPA and tPA) as well as tissue inhibitor of metalloproteinase 1 (TIMP-1, a specific inhibitor of MMPs) are increased [89, 90]. As a result ECM degradation is inhibited and the balance leans towards ECM accumulation.

Important steps in the development of TIF are schematically presented in Figure 4.

Insults: toxic, immunologic,infectious, metabolic, ischemic

Glomerulus

Proinflammatory Leakage of albumin and Endothelial cytokines complementproteins dysfunction

Inflammation Hypoxia

TubularepithelialcellsT

Secretion Growth factors Transdifferentiation Chemokines, adhesion Cytokines, growth molecules factors

Growth factors

Leukocytes

Extracellular matrix Myofibroblasts

Fibrosis

Figure 4. Development of TIF (adapted from [91]).

1.2.2.4 Progression of CKD

Once fibrosis takes place, an apparently irreversible course towards ESRD occurs. That process is very complex and multiple factors are involved. It is important to notice that the majority of factors responsible for the development of fibrosis in kidneys are also responsible for the progression of the diseases. The major pathogenic contributors of progression are shortly presented:

x Cytokines and growth factors-Sustained tissue aggression and chronic inflammation lead to increased amounts of different chemokines, cytokines and growth factors which in turn aggravate fibrosis. In addition to the action of TGFβ and CTGF, platelet derived growth factor (PDGF) [92], FGF-2 [93], epidermal growth factor receptors (EGFR) are increased in kidney fibrosis and promote progression of the disease. In addition the expression of antifibrotic molecules such as hepatocyte growth factor [94] or bone-morphogenic protein-7 is decreased [95] therefore amplifying fibrosis. x Activation of renin-angiotensin-aldosterone system (RAAS)-The RAAS plays an important role in the control of arterial blood pressure. That system not only plays a critical role in regulation of intraglomerular hemodynamic, but also shows pro-fibrotic activities through multiple mechanisms independent from those involved in their hemodynamic activity [96]. x Proteinuria- The presence of proteins in urine results from damaged glomerular filtration barrier. Proteinuria leads to further tubulo-interstitial damage by different mechanisms: 1) excess reabsorbtion in the proximal tubular cells and lysosomal rupture, 2) direct cytotoxic effect of plasma proteins through induction of ROS, chemokine and adhesion molecule production, 3) co-excretion of complement proteins and different pro-fibrotic molecules (the most important is TGFβ) [97]. x Hypoxia- In biopsy specimens from CKD patients, rarefaction of peritubular capillaries is demonstrated. The progressive loss of microvasculature in parallel with CKD- associated anemia contribute to tissue hypoxia which in turn promotes progression of fibrosis through induction of fibroblast proliferation and differentiation into myofibroblast, increased ECM production and EMT [98].

1.2.2.5 Therapeutic strategies

Clinical studies, trials and preclinical investigations over the last several decades have demonstrated that inhibition of RAAS with angiotensin converting enzyme inhibitors (ACEIs)

29 and angiotensin AT-1 receptor antagonists (ARA) is the most effective single therapy to attenuate progression of CKD [99]. The use of RAAS blockers is the first line treatment of diabetic and non-diabetic kidney diseases. Unfortunately, despite those treatments, CKD continues very often to progress. Alternative molecules or therapies capable to block or even reverse fibrosis are thus necessary. For ESRD the only available treatment option is RRT.

1.2.3 AKI and CKD are interconnected

For many years, AKI and CKD were considered as two distinct syndromes. However recent epidemiologic studies suggest that these two syndromes are closely interconnected in a continuum [100].

1.2.3.1 AKI is a risk factor for CKD

It was considered for long time that patients were completely recovered after AKI. However different studies suggested that the kidney function of patients following AKI is altered increasing the risk of later CKD development. It is now considered that women with a history of preeclampsia (gestational pathology characterized by elevated arterial pressure and vascular kidney injury) have an approximate 5-12 fold increased risk for ESRD development later in life [101]. Another study demonstrated more than 40 fold increased risk of ESRD in patients with pre-existing CKD who suffered AKI episodes compared to healthy individuals. The risk of ESRD is only 8 fold higher in CKD patients without AKI episode [102]. This suggests that AKI triggers signals that are responsible for CKD installation.

Severity and occurrence of multiple AKI episodes is associated with CKD progression whereas duration of the AKI episode seems to be linked to global mortality rather than to CKD progression [103].

1.2.3.1.1 The mechanisms of AKI progression to CKD

Numerous mechanisms are proposed to explain development of fibrosis following an episode of AKI:

x Nephron loss and glomerular hypertrophy- After an episode of AKI, it is possible that a significant permanent loss of nephron mass results in a compensatory hyperfiltration and glomerular hypertrophy in remaining nephrons. Cellular proliferation ensues and ultimately results in TIF and progressive nephron loss [100].

x Interstitial inflammation and fibrosis- Inflammation has been shown to be an important feature of ischemic and septic AKI with the biological aim to remove the causative agent and to induce tissue regeneration. In most models of experimental AKI, renal tubular regeneration occurs. If the alteration is severe enough the recovery is often not complete and the areas of incomplete recovery become fibrotic [100]. x Tubular epithelial repair- In most models of experimental AKI, renal tubular regeneration occurs. Normal repair processes involve multiple growth factors and proliferative and other signaling cascades, which must function in temporally and spatially coordinated manner. These mechanisms are critical for appropriate repair and regeneration. Large number of tubular epithelial cells arrested in G2/M phase of the cell cycle is displayed when kidney injury is accompanied by fibrosis [104]. These arrested cells in opposite to proliferating cells produce greater amounts of TGFβ and CTGF that trigger fibrosis development.

x Endothelial injury and vascular rarefaction- While tubular repair after AKI is robust, the restoration capacity of vasculature remains blunted [105]. Different AKI experimental models demonstrate diminished vascular density (30-50%) after injury [100]. The injured tissue becomes hypoxic due to insufficient vascular supply and this initiates pro-inflammatory and pro-fibrotic changes.

1.2.4 Conclusions

Despite better understanding the pathogenetic mechanisms involved in the development of both forms of kidney pathologies, treatment modalities are sparse. New researches are conducted all over the world with the aim to raise new approaches to treat these diseases.

2. Lysophosphatidic acid

2.1-Structure

Lysophosphatidic acid (LPA) is a bioactive lipid mediator present in intracellular and extracellular compartments that is capable to regulate number of cellular functions and that is associated with increasing numbers of pathologies. LPA is made of a glycerophosphate backbone esterified (acyl or alkyl bond) at its sn1 (Figure 5A) or sn2 (Figure 5B) position with a fatty acid. There exist different molecular species of LPA depending on the length and saturation of the fatty acid chain moiety [106].

ABOH OH HO P O HO P O O 3 O 2 OH 3

1 2 O O

O 1 O OH

Figure 5: Structure of oleoyl-LPA with the C18:1 fatty acid chain esterified (acyl bond) at the sn1 (A) or sn2 (B) position.

2.2 Synthesis of LPA

LPA can be produced either in intracellular or in extracellular compartment.

2.2.1 Intracellular synthesis

LPA is an essential metabolite intermediate of triglyceride and phospholipid anabolism in various cell types and is synthesized through the action of glycerol-3-phosphate acyltransferase (GPAT) on glycerol 3-phosphate found mainly in mitochondria and endoplasmatic reticulum [107, 108]. LPA is then converted to phosphatidic acid (PA) by LPA-acyltransferase (LPAAT) and further to different phospholipids (PL) and triglycerides (TG) [109].

Intracellular LPA can also be synthesized by the action of intracellular phospholipases A1 or A2 (PLA1 and PLA2) that convert PA to LPA by the hydrolytic cleavage of the bond between fatty acid chain and glycerol backbone at sn1 (PLA1) or sn2 (PLA2) positions [110].

One more pathway for intracellular synthesis is the phosphorylation of monoacylglycerol to LPA by the catalytic action of monoacylglycerol kinase (MAGK) [111].

A recent study showed that two members of intracellular glycerophosphodiester phosphodiesterase family, GDE4 and GDE7, possess a lysophospholipase D activity and can produce LPA acting on different substrates such as LPC and lyso-PAF [112].

2.2.2 Extracellular synthesis

Two major pathways of extracellular LPA synthesis have been described so far.

In the first pathway PA is first produced from diacylglycerol through the action of diacylglycerol kinase (DAGK) or from other PLs (ex. phosphatidylcholine (PC) or phosphatidylethanolamine (PE)) through the action of phospholipase D (PLD). PA synthesis occurs mainly in the inner leaflet of the plasma membrane. Synthesized PA is further translocated to the outer leaflet of plasma membrane where it can be deacylated by secreted phospholipase A2 (PLA2) [113] or PA-specific phospholipase A1 (PA-PLA1α and PA-PLA1β also known as LIPH and LIPI respectively) [114, 115] to generate LPA (see Figure 6).

In the second pathway, different lysophospholipids (LPL) such as lysophosphatidylcholine (LPC), lysophosphatidylethanolamine (LPE), lysophosphatidylserine (LPS) are first produced by the hydrolytic cleavage of a fatty acid from their respective phospholipids through the action of secretory phospholipase A2 or phosphatidylserine (PS)-specific PLA1. In plasma, LPC is produced from phosphatidylcholine present in lipoproteins by the lecithin-cholesterol acyltransferase (LCAT). Generated LPLs are subsequently converted to LPA through the

33 activity of a lysophospholipase D named autotaxin (ATX) [116]. This pathway is mainly responsible for the synthesis of LPA in different biological fluids such as plasma or serum.

All above mentioned synthetic pathways are schematically summarized in Figure 6.

Phospholipid Diacylglycerol

PLA PLD Diacylglycerol kinase

Lysophospholipid Phosphatidic acid

PLA LysoPLD (ATX) LPAAT

Glycerol-3 phosphate GPAT LPA MAGK Monoacylglycerol

Lysophospholipase LPPs

: head group : phosphate : glycerol : fatty acid

Figure 6. LPA metabolic pathways, adapted from[117]. PLA-phospholipase A, PLD- phospholipase D, LysoPLD-lysophospholipase D, GPAT-glycerol phosphate acyltransferase, MAGK-monoacylglcerolkinase, LPP-lipid phosphate phosphatase, LPAAT-LPA- acyltransferase

2.3 Degradation of LPA

The half life of circulating LPA is 3 minutes in mice [118]. Such short half-life of the phospholipid indicates that it can be either degraded extracellulary or captured by different cells and further involved in intracellular metabolic pathways or degraded intracellulary. The hydrophobic character of LPA and the lack of specific transmembrane carriers limit the possibility of its transmembrane transport. In contrast, LPA can be degraded extracellularly through dephosphorylation by a group of enzymes called lipid-phosphate phosphatases

(LPPs) that exist under 3 isoforms (LPP1, LPP2, LPP3 and a spliced variant LPP1a) [119] (see Figure 6). These enzymes are integral membrane proteins with six transmembrane α-helices and three catalytic domains located on the extracellular face of the plasma membrane enabling them to dephosphorylate extracellular phospholipids [120]. For example, LPP1 hypomorphic mice have increased plasma LPA concentration with longer half-life (12 minutes versus 3 minutes) compared to control animals indicating that LPP1 is highly involved in the degradation of LPA [118]. LPPs can catalyze the dephosphorylation of a variety of other phospholipids such as PA, sphingosine-1-phosphate, ceramide-1-phosphate [121]. LPP- mediated dephosphorylation of bioactive phospholipids limits their bioavailability to activate their respective GPCRs thereby reducing their biological effects. LPPs are involved in different pathologies, particularly cancer. Indeed, decreased expression of LPP1 and LPP3 was described in several cancer types (cancers of lung, breast, ovary) [122]. On the opposite, overexpression of LPP1 and LPP3 in ovarian cancer cell lines increases LPA hydrolysis therefore inhibiting cell proliferation and colony-forming activity and enhancing apoptosis [123, 124]. LPPs are also localized on the internal membranes of endoplasmic reticulum [125] and Golgi network [126] with the catalytic domains facing luminal sides of the organelles [120] indicating possible regulation of signal transduction processes by dephosphorylation of intracellular lipids.

In intracellular compartment, different enzymes regulate LPA levels. It can be converted to PA by the action of acyl-glycerol-3-phosphate acyltransferase (AGPAT) known also as LPAAT that catalyzes the transfer of an acyl group from acyl-CoA on LPA [127] (Figure 6). LPA can also be deacylated by the action of lysophospholipases [128] (Figure 6).

One more pathway of LPA degradation was described recently. Dephosphorylation of LPA can be catalyzed by the soluble epoxide hydrolase (sEH), an intracellular enzyme with dual enzymatic activities: hydrolase at C-terminus and phosphatase at N-terminus of the protein. The epoxyeicosatrienoic acid was known as an endogenous substrate of the hydrolase activity of sEH and recently it was shown that the phosphatase activity of the enzyme can be responsible for the dephosphorylation of different species of LPA [129].

The presence of LPA synthetic pathways in both intra- and extracellular compartments rises a question: does intracellulary synthesized LPA participate in the formation of extracellular pool? It was shown that over-expression of MAGK (an enzyme responsible for intracellular synthesis of LPA) in PC3 prostate cancer cells increases LPA production in both intra- and

35 extracellular compartments [111]. Similarly, blockade of 2 isozymes of intracellular synthetic enzyme PLA2 (Ca2+-independent PLA2 (iPLA2) and cytosolic PLA2 (cPLA2)) in peritoneal mesothelial cells reduces extracellular LPA [130]. These findings suggest that LPA synthesized in intracellular space can indeed participate in the formation of extracellular pool of LPA. The concept however remains a matter of debates as specific membrane transporters required for the transfer of the hydrophobic LPA from intracellular to extracellular space have not yet been described.

2.4 Transport of LPA

Due to its hydrophobic characteristics LPA has to be carried by transport molecules in order to be soluble in biological fluids. Albumin and gelsolin are described as transporters of LPA in plasma [131, 132]. In addition to their carrier properties these molecules play a role in presenting LPA to its receptors and modulating biological effects of LPA. In intracellular compartments liver fatty acid binding protein (LFABP) is described as a carrier of LPA [133].

2.5 LPA in biologic fluids

2.5.1 LPA in blood

LPA is abundant in serum (up to 10 μM) while its concentration is much lower in plasma (about 0.1μM). Part of the difference is attributed to platelet aggregation that occurs during serum preparation. Indeed, LPA is produced upon in vitro aggregation of platelets, induced by thrombin and/or phospholipase C (PLC) [134] and the action of PLA1 or PLA2 on PA was proposed as the synthetic pathway. However, the amount of LPA produced by platelets is not high enough to fully account for serum LPA concentration [135]. Moreover, it has been proposed that serum LPA would rather originate from hydrolysis, by the circulating ATX, of other lysophospholipids (LPC, LPS, and LPE) produced by platelet activation or brought by lipoproteins [116, 136].

2.5.2 LPA in other biological fluids.

LPA can be found in other biological fluids such as seminal fluid [137], follicular fluid [138], saliva [139], cerebrospinal fluid [140], aqueous humor [141], hen egg white [142], urine

[143]. Variation in the amount of LPA in biological fluids is associated with different pathologies. Increased LPA was observed in ascitic fluid of patients with ovarian [144] and pancreatic cancers [145], synovial fluid of patients with rheumatoid arthritis [146], in vitreous fluid of patients with proliferative diabetic retinopathy [147], in cystic fluid of patients with autosomal dominant polycystic kidney disease (ADPKD) [148], in cerebrospinal fluid after traumatic brain injury [149], in blister fluid [150]. One crucial question is: what are the cellular and metabolic origins of LPA in these biological fluids? In seminal, follicular, cerebrospinal, blister fluids, urine and hen egg white lysoPLD activity was detected suggesting that in these compartments LPA could be produced by ATX. In agreement with this, a recent study showed that choline producing activity of follicular fluid is diminished when it is incubated with anti-ATX antibodies [151]. However ATX involvement in LPA production in various biological fluids requires further confirmation with pharmacological or genetic targeting of ATX. Interestingly enzymatic activity responsible for the degradation of LPA is described in follicular and seminal fluids indicating that the overall amount of LPA in these compartments is the result of a balance between its synthesis and degradation [137, 140, 142, 151]. In addition to possible local production of LPA in various biological fluids, it can theoretically appear in these compartments from circulating plasma pool of the lysophospholipid as well. This last notion requires further verification.

2.6 LPA receptors

LPA is not only an intermediate in lipid/phospholipid metabolism, but is also a bioactive ligand of specific receptors through which it can activate different intracellular signaling pathways.

2.6.1 G-protein coupled membrane receptors

2.6.1.1. Classification and tissue distribution

LPA receptors are 7-transmembrane domains receptors coupled to G proteins (GPCRs). Six receptors (LPA1 to LPA6) coupled to one or more G proteins (Gαi/o, Gαq/11, Gα12/13, Gαs) have been described so far.

Historically LPA1/EDG2 was the first receptor cloned in 1996 and identified as a bona fide receptor for LPA [152]. LPA1 receptor is the most widely expressed among different tissues

37 and organs both in mice and humans [153]. During embryonic development LPA1 receptor expression is restricted to brain, craniofacial region, limb buds, somites and genital tubercle [154]. In postembryonic life highest expression of LPA1 receptors is found in brain and uterus but is also present in heart, lung, stomach, kidney, small intestine, testis and some other organs [154]. In humans, LPA1 receptor is expressed ubiquitously.

Following LPA1, 2 more LPA receptors (LPA2/EDG4 and LPA3/EDG7) were identified [155, 156]. Expression of LPA2 is more restricted compared to LPA1. It is found in embryonic brain but its expression decreases after birth. In adult mouse, high LPA2 expression is found in kidney, uterus and testis, moderate expression in lung and low expression in stomach, spleen and thymus [154]. In human tissues LPA2 expression is more restricted, with high expression in testis and leukocytes and moderate expression in prostate, spleen, thymus and pancreas [154]. LPA3 is found in murine organs such as lung, kidney, uterus, testis, small intestine, heart, stomach, spleen, thymus, whereas it is highly expressed in human heart, testis, prostate, pancreas and lesser expression is found in lung, ovary and brain. LPA1, LPA2 and LPA3 are members of endothelial differentiation gene (EDG) family to which receptors for another bioactive lipid sphingosine-1-phosphate also belong (S1P1/EDG1, S1P2/EDG5, S1P3/EDG3, S1P4/EDG6, S1P5/EDG8) [157]. LPA1, LPA2 and LPA3 share 50-60% amino acid sequence homology.

Three additional LPA receptors were also discovered (LPA4/P2Y9 [158], LPA5/GPR92 [159] and LPA6/P2Y5 [160]). LPA4 has low expression in mouse and is found in heart, ovary, uterus, skin, developing brain and thymus. In humans LPA4 is abundantly expressed in ovary [154]. LPA5 is expressed in number of murine tissues, such as small intestine, lung, heart, stomach, colon, spleen, thymus, skin and liver. In humans, LPA5 highest expression is found in spleen [154]. LPA6 expression is described in hair follicles, kidneys, endothelial cells. LPA4, LPA5 and LPA6 receptors are members of the P2Y purinergic family of receptors and share nearly 35% of amino acid homology. Three more GPCRs are proposed as new receptors for LPA: GPR87 [161], P2Y10 [162] and GPR35 [163]. Nevertheless, further studies are required for the validation of these receptors as bona fide receptors for LPA.

Interestingly for us the most highly expressed receptors in mouse kidneys are LPA2 and LPA6 whereas the expression of LPA1, LPA3 and LPA5 is lower [164, 165]. In human kidneys only moderate LPA1 expression was described [153].

eukocyte brain Embryonic Brain Lung Heart intestine Small Stomach Kidney Skeletal muscle Uterus Ovary Testis Placenta Prostate Colon Spleen Thymus Pancreas Skin Liver L

Lpar1 M NA LPAR1 H NA NA NA NA

Lpar2 M NA NA NA NA NA NA NA LPAR2 H NA NA NA NA

Lpar3 M NA NA NA NA NA LPAR3 H NA NA NA

Lpar4 M NA NA NA NA NA NA LPAR4 H NA NA NA NA

Lpar5 M NA NA NA NA NA NA LPAR5 H NA NA NA NA NA NA NA NA

High Very low Moderate Not expressed Low NA Non available

Figure 7. Gene expression pattern of LPA receptors in human (H) and murine (M) organs and tissues. Adapted from [153].

2.6.1.2 Intracellular signaling pathways

Activation of LPA receptors induces variety of cellular effects by activating several

intracellular pathways. LPA1 and LPA2 receptors couple with Gαi/o, Gαq/11, Gα12/13 and initiate downstream signaling through molecules such as PLC, diacyglycerol (DAG), mitogen-activated protein kinase (MAPK), protein kinase B (Akt), phosphatidylinositol 3- kinase (PI3k) and Rho family of GTPases. Activation of these receptors promotes cell proliferation, survival and migration and cytoskeletal changes [166, 167]. LPA3 couples with 2+ Gαi/o, Gαq/11 and mediates LPA-induced Ca mobilization, adenylyl-cyclase inhibition,

phospholipase C and MAPK activation [167]. LPA4 couples to Gαi/o, Gαq/11, Gα12/13, Gαs and

induces stress fiber formation and neurite retraction through Gα12/13 and downstream

Rho/ROCK pathway [158, 168]. LPA4 is the only LPA receptor that couples to Gαs and leads

39 to intracellular cAMP accumulation upon activation. LPA5 couples to Gα12/13 and induces stress fiber formation and neurite retraction. It couples also to Gαq/11 and increases 2+ intracellular Ca [159]. LPA6 receptor couples to G13-Rho signaling pathway and induces Rho-dependent alteration of cellular morphology [169].

LPA1 LPA2 LPA3 LPA4LPA5 LPA6

Gαi/o Gαs Gα12/13 Gαq/11

Rho PLC RAS PI3K AC

IP3 DAG

SRF ROCK Ca2+ PKC MAPK Akt Rac cAMP

Proliferation, migration, cytoskeletal changes, survival, cell-cell interaction etc.

Figure 8. LPA receptor signaling pathways, adapted from[154]

2.6.1.3 Relative specificity of LPA receptors towards different LPA species

Composition of different LPA species was described in plasma. In healthy individuals the unsaturated 18:2, 18:1 and 20:4 constitute the majority of LPA species [170]. Increased LPA levels were detected in plasma of patients with ovarian or other gynecological cancers, moreover unsaturated fatty acid species of plasma LPA were associated with late stage and recurrent cancers [171, 172]. In parallel, LPA is present in ascitic fluid of patients with ovarian cancer where the majority of LPA species has saturated fatty acid chains [144]. Interestingly for us composition of LPA molecular species was measured in the urine of rats

40 with unilateral ureteral obstruction (UUO). In the urine from non-obstructed kidneys the major constitute is 16:0 followed by 18:1, whereas in urine from obstructed kidneys saturated species (16:0 and 18:0) are predominant [173]. LPA molecular species in plasma of UUO rats doesn’t differ from that of control animals with predominant poly-unsaturated 18:2 and 20:4 species [173]. These discrepancies in LPA molecular composition between plasma and other biological compartments in physiological and pathological states can be due to different metabolic sources in these biological fluids. It can indicate also varying biological roles of different LPA molecular species in pathological settings. This is in agreement with previous studies indicating that LPA species with unsaturated fatty acids (18:1 and 18:2) are more potent than species with saturated fatty acids (16:0 and 18:0) in growth stimulation and induction of intracellular Ca2+ in ovarian cancer cells [144]. Other studies showed the rank order potency for different LPA species in various biological activities. For example, for the stimulation of deoxyglucose transport in Xenopus oocytes the rank order potency of LPAs with different acyl groups in the sn-1 position was LPA18:1 > LPA16:0 > PA = LPA18:0 > LPA12:0 [174]. One more example was the rank order potency of different LPA species to induce Ca2+ mobilization in human A431 carcinoma cell line: LPA18:1 > LPA20:4 > LPA18:3 > LPA18:2 > LPA18:0 = LPA16:0 > LPA12:0 [175]. In addition, different LPA receptors show varying levels of preference towards various LPA species. While LPA1 and LPA2 do not discriminate between saturated and unsaturated LPA species, LPA3 is preferentially activated by unsaturated LPA [156]. LPA4 receptor prefers monounsaturated LPA species with 18 carbon atoms whereas the potency for other species is lower [158]. LPA5 shows 10-fold preference for alkyl-18:1 LPA compared to acyl-18:1 LPA [176]. LPA6 receptor prefers 2-acyl-LPA species and requires much higher LPA concentrations to show an effect compared to LPA1-LPA5 [169]. Despite the varying affinities of the LPA receptors towards different LPA species, and differencies in the potency of various LPA species in inducing a biological effect, LPA18:1 remains the most commonly used molecule in biological research. Therefore it could be interesting to propose the usage of a mix of LPA species, composition of which could simulate the LPA molecular composition in physiological or pathological conditions.

2.6.1.4 Genetic deletions and associated pathologies

The use of gene silencing techniques and pharmacological targeting of different receptors helped to understand the physiological roles of each LPA receptors and their possible involvement in physiopathology (see Table 4).

Receptor Phenotype of KO animal Involvement in pathologies Neuropathic pain, spatial memory 50% perinatal lethality due to defective impairments, schizophrenia [181] olfaction and impaired suckling, Fibrotic diseases [164, 182] decreased survival, reduced body size, LPA1 Cancers [183] frontal cerebral hematomas and Obesity [184] craniofacial dysmorphism [177], defects Rheumatoid arthritis [185] in neurodevelopment [178-180] Male infertility [186] No abnormal phenotype described [187] LPA1 and LPA2 double knockout mice Cancers [188, 189] show no additional phenotypic LPA2 Asthma [190, 191] alterations compared to LPA1 knockout Male infertility [186] animals, except higher rates of perinatal frontal hematomas Delayed embryo implantation and Animals are grossly normal at birth and altered embryo spacing, embryo LPA3 viable but female mice have striking crowding and reduced litter size phenotype in reproductive system [192, 193] Male infertility [186] Decreased gestational survival and altered angiogenesis [194] LPA4 Cell motility[196] Increased tubular bone volume and thickness [195] Neuropathic pain[197] LPA5 No described abnormal phenotype [197]

No information available about KO Hair loss [160, 198] LPA6 animals Development of bladder cancer [199]

Table 4. The phenotypic characteristics of LPA receptor KO animals and involvement of LPA receptors in pathologies.

Some LPA receptors have crucial role during embryonic development. LPA1 receptor signaling is very important for the normal development of nervous system and LPA1 deficiency causes anatomical alterations of the brain as well as neurochemical changes similar to different psychiatric disorders (schizophrenia, depression). LPA1 signaling is implicated in various pathologies such as kidney and lung fibrosis, neoplastic diseases, obesity etc. [164, 182-184]. While LPA2 receptor deficiency doesn’t cause detectable phenotypical changes, its activation plays a role in bronchial asthma and cancers of breast, colon and ovary [188, 189, 191, 200]. The most important phenotypic change observed in LPA3 deficient animals concerns the embryo implantation [192, 193]. Animals lacking LPA4 receptors have high embryonic lethality and demonstrate aberrant angiogenesis with hemorrhages, edema and dilated blood and lymph vessels [194]. Phenotype of LPA4-/- mice resembles that of animals invalidated for ATX gene [201] indicating possible involvement of LPA4 receptors in ATX- induced effects during embryonic development. Surviving LPA4-/- mice show increased osteogenesis [195]. LPA5 deficient animals have normal phenotype and are protected from neuropathic pain [197]. While LPA6 knockout animals have not described so far, a mutation of LPA6 has been reported in genetic forms of hair loss [160, 198]. Interestingly a mutation in a gene encoding PA-PLA1α (LIPH), an enzyme involved in LPA synthesis, also leads to the development of hypotrichosis [202, 203]. It was proposed that PA-PLA1α-LPA-LPA6 axis regulates differentiation and maturation of hair follicles through LPA-induced EGFR transactivation [204].

Interesting for us, LPA receptor deletions showed that loss of basal expression of none of the LPA receptors causes altered kidney structure or function. Nevertheless LPA signaling through certain LPA receptors is involved in the pathogenesis of kidney pathology (discussed later).

2.6.2 Intracellular receptors

PPARs are transcription factors of the nuclear receptor superfamily that heterodimerize with retinoic acid X receptor upon binding with agonists and regulate gene expression by binding to specific response elements in promoter regions that are known as peroxisome proliferator response elements (PPRE). Three isoforms of PPARs, PPAR-alpha (PPARα), PPAR- beta/delta (PPARβ/δ) and PPAR-gamma (PPARγ) have been described so far [205]. PPARγ was identified as intracellular receptor for LPA since it competes with the classical PPARγ agonist rosiglitasone on purified PPARγ protein. It also increases the transcription of cells

43 transfected with PPRE reporter gene. Activation of PPARγ by LPA stimulates also the expression of CD36 and promotes cholesterol uptake by human monocytes forming foam cells, leads to neointima formation and dedifferentiation of vascular smooth muscle cells [206, 207]. These effects are however limited to unsaturated and alkyl-ether forms of the LPA. In addition it was shown that only intracellular LPA synthesized by GPAT is able to activate PPARγ [208]. Moreover, it was shown that extracellular LPA is not able to activate PPARγ in 3T3F442A adipocyte cells [184]. The indirect activation of PPARγ as the consequence of intracellular pathways activated by membrane LPA receptors has also been reported [209]. These contradictory observations could indicate that LPA activation of PPARγ is cell-type specific and depends on chemical composition of the LPA molecule.

2.6.3 Other potential receptors

It was shown that LPA-induced pain-like behavior is at least partially attributed to the direct activation of transient receptor potential vanilloid 1 (TRPV1) ion channel through binding to its C-terminal [210].

Another potential LPA receptor is RAGE (receptor for advanced glycation end products) that is a transmembrane receptor of immunoglobulin superfamily and that was shown to be required for LPA-mediated proliferation and migration of vascular smooth muscle cells as well as for LPA-induced ovarian tumor implantation and metastasis [211].

2.6.4 LPA receptor antagonists

Most antagonists target LPA1 and/or LPA3 receptors with different levels of selectivity. The pharmacological characteristics of some LPA receptor blockers are discussed here.

Ki16425 [3-(4-[4-([1-(2-chlorophenyl)ethoxy]carbonyl amino)-3-methyl-5-isoxazolyl]:This is a non-lipid antagonistthat has been synthesized by Kirin Brewery Co [212] and that is commercially available from Cayman Chemicals. This is the most extensively used LPA receptor antagonist in biological research. It is an LPA1 and LPA3 receptor antagonist without clear subtype preference [212]. Ki16425 is commonly used to demonstrate the receptor-dependence of a cell response (proliferation, migration, calcium mobilization) activated by LPA [213, 214]. Ki16425 is able to inhibit the constitutive activity of LPA receptors reinforcing the blocking activity of this molecule towards LPA [215]. It is noticeable that Ki16425 was reported to exert a partial PPARγ agonist activity indicating that some effects of the molecule could be explained by LPA receptor independent actions [216].

VPC12249 ((S)-phosphoric acid mono-[3-(4-benzyloxyphenyl)-2-octadec-9-enoylamino- propyl] ester): It is a lipid antagonist and is commercially available from Avanti Polar Lipids. VPC12249 behaves as a LPA1 and LPA3 receptor antagonist without clear subtype preference [217]. In vitro experiments showed that it inhibits migration of pancreatic cancer cells induced by malignant ascitic fluid [145], blocks LPA-induced increase of intracellular Ca2+ in skeletal muscle cells of C2C12 cell line and attenuates osteoclast differentiation [218, 219].

Efficiency of these two molecules has been tested in several in vivo experiments as well (see 2.7). The poor water solubility of both of them was the major limiting factor for their use in clinical trials.

AM966 ( (4'-[220]-biphenyl-4-yl)-acetic acid) and AM095 ((sodium, {4'-[3-methyl-4-((R)-1- phenyl-ethoxycarbonylamino)-isoxazol-5-yl]-biphenyl-4-yl}-acetate)): AM966 and AM095 are non-lipid antagonists synthesized by Amira Pharmaceuticals and both exhibit higher LPA1 receptor selectivity when compared to LPA3 subtype as assessed by the ability to induce calcium release or GTPγS binding in CHO cells expressing LPA receptors [220, 221]. These are the only available LPA receptor antagonists that can be administered orally with good pharmacokinetic profile.

2.7 Physiological and pathophysiological effects of LPA

2.7.1 Cancer

The role of LPA in human pathologies has been extensively studied in cancer. LPA has been implicated in initiation and progression of different cancer types, such as ovarian, breast, prostate, colon, thyroid cancers etc. It was shown that ascitic fluid and plasma of patients with ovarian cancer contain elevated amounts of LPA that is responsible for growth and proliferation of ovarian cancer cells in vivo and in vitro [144, 172, 222-224]. Besides promoting proliferation LPA also leads to increased survival of cancer cells through inhibition of apoptosis [225], reduces sensitivity to chemotherapy [226], promotes migration [227], induces angiogenesis [228, 229] and enhances invasiveness and metastatic potency of cancer cells through secretion of uPA and activation of MMPs [230-232]. Expression of enzymes involved in LPA metabolism is changed in different types of cancers: increased ATX expression is found in Hodgkin lymphoma, glioblastoma, hepatocellular carcinoma, renal-cell

45 carcinoma, breast cancer, non-small-cell lung cancer [233], whereas expression of LPPs (LPP1 and LPP3) is decreased in breast, ovarian and lung cancers [122]. In addition LPA receptor expression is also changed in certain cancer types: LPA2 expression is increased in ovarian, colorectal and thyroid cancers [234-236], and LPA1, LPA2, LPA3 in breast cancer cells [237]. Treatment with LPA1/3 receptors antagonist Ki16425 attenuates formation of bone metastasis in xenograft model of breast cancer [183]. Changes in expression of metabolic enzymes leading to increased LPA availability in parallel with increased expression of LPA receptors are the prerequisites for the increased LPA signaling and protumorigenic action. Analyzing all these data it can be concluded that LPA can be a tumor biomarker for certain types of tumors and represents also an interesting target for the development of potential antitumoral chemotherapeutic agents.

2.7.2 Reproductive system

LPA signaling is implicated in different aspects of reproduction such as ovarian function, spermatogenesis, fertilization, embryo implantation and spacing, pregnancy maintenance, parturition etc. LPA is present in follicular fluid and is supposed to be produced by the ATX activity [138]. In vitro experiments showed that LPA promotes oocyte maturation suggesting that LPA in follicular fluid could be responsible for oocyte growth and maturation in vivo [238]. LPA also stimulates mouse ovum transport by contracting smooth muscle cells of oviduct [239] and plays a role in fertilization by inducing acrosomal reaction [240]. LPA- LPA3 signaling in the uterus is important in early pregnancy as LPA3 deficient animals showed uneven embryo spacing and embryo crowding, delayed embryo implantation and prolonged pregnancy. Decreased prostaglandin production is at least partially responsible for the LPA3 deficient female mice phenotype [192, 193]. LPA2 and LPA3 receptors are upregulated in placental tissue of patients with gestational hypertensia and pre-eclampsia [241]. These disorders worsen with the progression of pregnancy and are associated with a parallel increase in LPA levels suggesting the involvement of LPA signaling in these hypertensive pathologies [242]. In addition LPA signaling could be implicated in the initiation of labor since it contracts uterine smooth muscle cells in vitro and increases intrauterine pressure in vivo similar to PGF2α [243].

2.7.3 Cardiovascular system

LPA is involved in different aspects of cardiovascular physiology and pathophysiology. First of all it plays a role in different steps of atherosclerosis. LPA is found in lipid-rich core of

46 atherosclerotic plaques where it is produced through the action of ATX on LPC during oxidation of low-density lipoproteins (LDL) [244]. LPA promotes endothelial chemokine secretion (CXCL1, IL8, MCP-1) and increases expression of adhesion factors (ICAM-1, VCAM-1, E-selectin) leading to recruitment of monocytes into atherosclerotic lesion. These effects are mediated through LPA1 or LPA3 receptors [245-248]. LPA mediates the effects of minimally modified LDL on monocyte activation via LPA1 receptor [249]. It also increases uptake of oxidized LDL in macrophages by increasing scavenger receptor A and induces foam cell formation through LPA1 and LPA3 receptors and PPARγ [206, 250]. LPA induces platelet adhesion and thrombus formation via the alteration of platelet shape, fibronectin deposition and matrix formation, platelet-monocyte interaction [251-254]. LPA is partially responsible of platelet activation induced by mildly oxidized LDL [244, 255, 256] and would thus be involved in thrombus formation upon rupture or erosion of atherosclerotic plaque during myocardial infarction or ischemic stroke. LPA is also implicated in neointima formation by inducing proliferation and migration of vascular smooth muscle cells [257-260] as well as by recruiting smooth muscle progenitor cells through LPA1 and LPA3 mediated activation of CXCL12 [261].

2.7.4 Inflammatory diseases

Different studies showed that LPA can exert anti-inflammatory activities. Treatment with LPA (18:1 or 18:0) increases animal survival upon LPS-induced sepsis, attenuates damage to liver and neuromuscular junction, reduces plasma levels of IL1β and TNFα and decreases myeloperoxidase activity in lungs [262, 263]. Protective effects of LPA18:1 were blocked with both LPA1/3 antagonist (Ki16425) and a PPARγ antagonist (GW9662) whereas the effects of LPA18:0 were blocked with only Ki16425 [263]. These data suggest that the mechanism of protective effect of LPA could be related to acyl chain saturation with LPA18:0 acting via GPCRs and LPA18:1 acting via GPCRs and PPARγ. Interestingly a recent study found increased LPA and decrease LPC plasma levels upon peritoneal sepsis induced by cecal ligation and puncture [264]. In parallel plasma ATX activity was reduced suggesting that increased production of LPA in plasma was not mediated through ATX action. Metabolic sources of increased LPA upon sepsis remain unknown.

On the opposite LPA can exert pro-inflammatory properties. In air pouch model of inflammation, LPA leads to LPA1 and LPA3 receptor dependent and CXCR2 mediated leukocyte recruitment and release of pro-inflammatory cytokines [265]. It was shown that

ATX and LPA levels increase in bronchoalveolar lavage fluid in patients with allergic asthma [190]. In mouse model of allergic asthma, ATX blockade or LPA2 receptor knockout reduces allergic lung inflammation [190, 191]. LPA and ATX are also increased in synovial fluid of patients with rheumatoid arthritis. LPA stimulates cycloxygenase-2 (COX-2) induction and prostaglandin E2 synthesis in synovial fibroblasts via abundantly expressed in synovial tissue LPA1 receptors whereas abrogation of LPA1 signaling reduces cellular infiltration and bone destruction in mouse model of rheumatoid arthritis [146, 185].

The apparent contradictory effects of LPA in inflammatory processes can be explained by different type and activity state of immune cells as well as by the differences of LPA receptors expressed in immune cells in the different pathological contexts [266].

2.7.5 Obesity

Adipose tissue produces LPA that promotes proliferation of preadipocytes [267]. Secreted ATX is mostly responsible for this LPA production and the expression of ATX is higher in differentiated adipocytes compared to preadipocytes [268, 269]. Adipocyte-specific deletion of ATX leads to 40% reduction in plasma LPA demonstrating contribution of adipose tissue in the overall production of LPA in the body [270]. Increased ATX expression is found in adipose tissue of genetically obese diabetic db/db mice and in visceral adipose tissue of obese patients with glucose intolerance [271, 272]. LPA-induced preadipocyte proliferation is LPA1-dependent [273]. At the same time LPA inhibits preadipocyte differentiation and adipogenesis via LPA1-dependant down-regulation of pro-adipogenic factor PPARγ [184]. In addition LPA alters glucose homeostasis in the body by decreasing glucose-induced insulin secretion and possibly by promoting adipose tissue fibrosis [274, 275]. LPA-induced fibrosis of adipose tissue is blocked by the LPA1/3 antagonist Ki16425 [275].

2.7.6 Nervous system

LPA is found in cerebrospinal fluid as well as in different regions of central nervous system and its local production increases in multiple sclerosis and nerve injury [276, 277] indicating its role in normal functioning and pathologies of central nervous system. LPA is highly involved in normal brain development through LPA1 receptors deletion of which causes striking features (see table 4). In addition LPA1 receptor deficient animals demonstrated behavioral changes as well as neurochemical alterations that are characteristic of schizophrenia, depression, anxiety, bipolar psychosis [278]. LPA injection into developing

48 brains leads to hydrocephalus formation that is prevented with genetic invalidation or pharmacologic blockade of LPA1 receptors [279].

ATX expression is increased in frontal cortex of patients with Alzheimer’s disease [280] and LPA increases tau phosphorylation and Aβ production in neurons [281, 282]. Therefore, the LPA/ATX axis appears as a potential interesting target for the development of treatment strategies for Alzheimer’s disease. LPA also initiates neuropathic pain and peripheral nerve demyelination through LPA1 receptor mediated Rho-ROCK activation [283, 284] or through the interaction with TRPV1 channels [210]. Injection of Ki16425 attenuates pain-like behavior in mice [285].

2.7.7 Wound healing and fibrotic diseases

Whereas wound healing after injury is considered as physiologic process with eventual restoration of the tissue function, aberrant wound healing due to chronic injury and sustained production of pro-fibrotic mediators leads to excessive accumulation of ECM and the eventual development of fibrosis. Development of fibrosis leads to altered architecture of the tissue with further loss of function. Depending on tissue context, LPA can either favor wound healing or be a pro-fibrotic factor.

2.7.7.1 Wound healing: Multiple studies involve LPA in wound healing and scar formation of different tissues such as skin, intestine and cornea [286-288]. LPA promotes growth of epithelial cells from tongue, pharynx and esophagus and it thus was proposed that LPA present in saliva can be involved in the healing of wounds of upper gastrointestinal tract [139].

2.7.7.2 Pulmonary fibrosis: LPA level is elevated in bronchoalveolar lavage (BAL) fluid of mice with bleomycine- and radiation-induced pulmonary fibrosis and in patients with idiopathic pulmonary fibrosis [182, 289]. LPA promotes pulmonary fibrosis by inducing vascular leakage and fibroblast recruitment. Genetic deletion of LPA1 receptors as well as pharmacologic blockade with the LPA1 specific antagonists AM095 or AM966 or with the LPA1/LPA3 receptor antagonists VPC12249 or Ki16425 reduces collagen deposition and fibroblast accumulation, decreases expression of TGF-β1, α-SMA, CTGF and protects against lung fibrosis in animal models of the disease [182, 220, 221, 289, 290]. Fibroblasts obtained from patients with idiopathic pulmonary fibrosis show high expression of LPA1 receptors and BAL fluid from same patients induces chemotaxis of human lung fibroblast that was blocked

49 with Ki16425 [182]. ATX level is increased in BAL fluid from bleomycin induced pulmonary fibrosis and specific deletion of ATX from epithelial cells and macrophages as well as pharmacologic inhibition of ATX attenuates fibrotic events in lungs [291]. Thus, LPA1 receptor and ATX are identified as targets for the treatment of pulmonary fibrosis and two molecules, the LPA1 antagonist BMS-986020 (NCT01766817) and the ATX inhibitor GLPG1690 (NCT02738801), entered phase IIa clinical trials to test their efficacy in patients with idiopathic pulmonary fibrosis.

2.7.7.3 Hepatic fibrosis: Patients with chronic hepatitis C and animals with chemically induced liver fibrosis have higher plasma LPA levels and increased serum ATX activity. Both parameters correlate with histological stage of fibrosis and blood markers of liver injury [292, 293]. Involvement of LPA in the pathogenesis of hepatic fibrosis was further reinforced by in vitro studies showing that LPA promotes migration of hepatic myofibroblasts[294], leads to proliferation of stellate cells and increases their contractility [295, 296]. Moreover, a recent study showed that inhibition of ATX activity in mice reduces liver fibrosis [297].

2.7.7.4 Systemic sclerosis: Patients with systemic sclerosis have increased LPA levels in plasma [242]. Activated platelets play an important role during the pathogenesis of this disease and are proposed as the major source of circulating LPA in these patients. Pharmacological blockade of LPA receptors with LPA1/3 antagonist Ki16425 or with the LPA1 selective antagonist AM095, and the genetic deletion of LPA1 receptors attenuated dermal fibrosis in a bleomycin-induced murine model of systemic sclerosis [290, 298]. LPA1 receptor was proposed as an efficient target for the treatment of systemic sclerosis and the LPA1 antagonist SAR100842 (NCT01651143) entered phase IIa clinical trial to test the drug efficacy in patients.

2.7.7.5 Kidney fibrosis.The involvement of LPA in kidney fibrosis is discussed thoroughly in 2.8.

2.8 LPA in kidney pathology

The first report of a change in LPA production associated with kidney disease was made by a japanese group that found increased plasma LPA concentration in 18 patients with renal failure on hemodialysis compared to healthy individualswhile no difference was observed before and after dialysis [299]. Because the cellulose membranes used for hemodialysis were

50 known to activate platelets [300], the authors concluded that platelet aggregation could explain the increase in plasma LPA. Later, metabolomic analysis of serum from patients with CKD showed elevated concentrations of two molecular LPA species (LPA16:0 and LPA18:2) that, in combination with 5 other metabolites, were proposed as useful biomarkers for the discrimination of CKD patients from healthy subjects [301]. In addition it has been demonstrated that serum ATX level was independently associated with proteinuria and kidney lesions in patients with biopsy-proven diabetic nephropathy [302]. Since LPA receptors are expressed in kidneys and since kidney possesses the required metabolic machinery to produce LPA [164, 173], kidney is a likely target of LPA. Whether increased serum LPA in patients with kidney diseases is the consequence and/or the cause in the pathogenesis of these pathologies is an open question that experimental approaches in animal models might help to answer.

2.8.1 LPA and renal obstruction

Unilateral ureteral obstruction (UUO) in rodents is a well-established experimental model that results in rapid development of TIF in the obstructed kidney. Ureter is ligated at the uretero- pelvic junction after abdominal incision of anesthetized animal. Renal blood flow and glomerular filtration rate are reduced within 24 hours and are followed by hydronephrosis, interstitial inflammation and tubular epithelial cell death due to apoptosis or necrosis within the next few days. Subsequent proliferation of interstitial fibroblasts and the phenotypic transformation of resident renal cells into myofibroblasts with further deposition of ECM proteins lead to eventual TIF [303]. In 2007, our group has shown in UUO mice that in vitro release of LPA by explants from obstructed kidneys was nearly 3 fold higher than those prepared from intact kidneys [164]. In UUO rats, it was reported that the concentration of LPA in the urine that accumulates in obstructed kidney was higher (about 3 fold) than in urine collected in bladder [173]. These reports indicate that renal LPA production is increased in obstructed kidneys. Increased ATX and AGK expression in combination with increased choline producing activity in obstructed kidneys was demonstrated as well [173] suggesting that these enzymes are possibly responsible for the increased LPA production from obstructed kidneys. However, the hypothesis remains to be verified through the use of pharmacological tools or genetic ablation of above mentioned enzymes.

What could be the consequences of increased renal production of LPA? UUO is associated with increased renal expression of LPA1 receptor [164, 165]. On the opposite, genetic

51 invalidation of LPA1 or pharmacological blockade of LPA receptors with the LPA1/LPA3 antagonist Ki16425 or the LPA1 specific antagonist AM095 block the development of TIF, decreased the renal expression of CTGF and TGFβ, reduced the amount of proliferating fibroblasts and the accumulation of αSMA positive myofibroblasts associated with UUO [164, 165, 221]. These observations showed that the LPA-LPA1 signaling axis plays an important role in the process of TIF that develops in obstructed kidney.

Epithelial cells of proximal tubules in obstructed kidneys are responsible for the majority of increased CTGF expression after obstruction [165]. In mouse primary tubular epithelial cells all subtypes of LPA receptors are present with LPA2 being the most abundant, followed by LPA1 whereas in human primary tubular epithelial cells high expression of LPA2 and traces of LPA1 are shown [165, 304]. In addition, previous experiments demonstrated that proximal tubular epithelial cells are sensitive to LPA, as it induces Ca2+ mobilization in these cells, inhibits serum deprivation-induced apoptosis and promotes proliferation of these cells through activation of PI3K or ERK pathways [305-307]. Therefore, LPA-induced increased CTGF expression in obstructed kidney could result from the action of increased LPA on tubular cells. This hypothesis was confirmed by in vitro experiments which demonstrated that LPA1 receptor activation induces CTGF and TGFβ expression via a Gα12/13-Rho/ROCK pathway with downstream actin polymerization and induction of myocardin-related transcription factor (MRTF) - serum response factor (SFR) pathway [164, 165]. Another source of CTGF in kidney is resident fibroblasts and LPA was shown to induce CTGF production in cultured renal fibroblasts as well through activation of small GTPase RhoA [308]. It was also shown that LPA2 receptor can mediate LPA-induced CTGF expression in tubular cells. LPA2 receptor knockdown results in diminution of LPA-induced CTGF expression [165]. Moreover, LPA signaling through LPA2 receptors and Gαq proteins in cultured proximal tubular cells transactivates latent TGFβ in Rho/Rho-kinase and αvβ6 integrin dependent manner. Activated TGFβ then induces CTGF and PDGF secretion [309]. Involvement of LPA-LPA2 axis in the development of kidney fibrosis needs however in vivo confirmation. If it is proven, LPA2 receptors in parallel with LPA1 receptors can represent interesting targets for the development of therapeutic strategies against kidney fibrosis.

2.8.2 LPA and diabetic nephropathy

Diabetic nephropathy (DN) was reported to be associated with the accumulation of lysophospholipids in kidney. Increased choline uptake and phosphorylation as well as LPC

52 accumulation were demonstrated in mice with sterptozotocin-induced diabetic nephropathy that was attenuated with insulin therapy [310]. MALDI/MS imaging revealed an increased production of LPC and LPA in glomerular areas of db/db diabetic mouse kidneys expressing clear symptoms of nephropathy. Treatment of animals with an inhibitor of oxidative processes diminishes both symptoms of nephropathy and renal accumulation of the lysophospholipids without impacting blood glucose levels [311]. These observations are in agreement with other reports indicating that DN is associated with increased expression of several phospholipases involved in LPC and LPA synthesis in kidneys of diabetic animals [268, 312, 313] and in mesangial cells exposed to high glucose in vitro [313-315]. These observations have some clinical relevance since it was shown that serum ATX level is independently associated with proteinuria and kidney lesions in patients with DN [302]. Moreover, recent data from our group show that patients with diabetic nephropathy have higher concentrations of LPA and LPC in their urine compared to diabetic patients without nephropathy [316].

Mesangial cells not only produce lysophospholipids [311] but also express LPA receptors [317] and are sensitive to LPA. It acts as a survival factor for cultured rat mesangial cells and promotes their proliferation through activation of MAPK and can be synergistic to the proliferative effect of PDGF [318-320]. In parallel, LPA induces mesangial cell contraction in vitro by mobilization of intracellular Ca2+ through activation of phosphoinositide cascade and by favoring Ca2+ entry across the plasma membrane [321]. In human or rat mesangial cells LPA induces CTGF, a pathway that could play a role in the development of glomerulosclerosis [322, 323]. Since mesangial proliferation, contraction and glomerulosclerosis are hallmarks of DN, LPA could play a role in DN pathogenesis by acting on mesangial cells. The hypothesis is supported by a very recent study showing that albuminuria, ECM deposition in glomerules and TGFβ expression in kidneys of diabetic mice were reduced in animals treated with the LPA receptor antagonist Ki16425 [324]. The same report shows that treatment of cultured mouse mesangial cells with Ki16425 as well as genetic knockdown of LPA1 receptors reduces LPA-induced TGFβ expression [324]. Even though glomerules are the initially affected kidney structures during DN, diabetic microangiopathy causes reduced postglomerular blood flow to peritubular capillaries and, consequently TIF develops in parallel to disease progression. As a result, the whole nephron is damaged in later stages of the disease. Since LPA-LPA1 axis was shown to be responsible for the development of both glomerulosclerois and TIF, it could be an interesting target for the treatment of DN.

RAGE signaling is an important component for the development of DN since overexpression of RAGE in endothelial cells of diabetic mice leads to enhanced glomerular hyperthrophy, measngial expansion and glomerulosclerosis, increased serum creatinine and albuminuria compared to diabetic controls [325]. In contrast, deletion of RAGE in diabetic mice or treatment with soluble RAGE attenuates albuminuria and improves diminished kidney function, mitigates mesangial expansion, glomerulosclerosis and thickening of GBM [326]. Since RAGE is known as a potential receptor for LPA, one could ask whether LPA-induced kidney fibrosis in DN could be mediated by activation of RAGE.

2.8.3 LPA and other CKDs

Autosomal dominant polycystic kidney disease (ADPKD) is the most common form of inherited renal cystic diseases characterized by bilateral renal cysts, various extrarenal manifestations and development of ESRD in 50% of patients in the 4th to 6th decade of life

[327]. The epithelial cell line mpkCCDcl4 exhibits major characteristics of distal tubular epithelial cells that cover the majority of cysts in ADPKD. It was shown that LPA is a bioactive component of cystic fluid of patients with ADPKD able to induce cystic fibrosis - transmembrane conductance regulator (CFTR)-mediated Cl secretion in mpkCCDcl4 cells [148]. LPA-induced Cl- secretion is due to activation of LPA receptors from the basolateral side of cells and could be responsible for the expansion of cysts with future diminution of kidney function in these patients [148].

Human immunodeficiency virus (HIV)-associated nephropathy, is characterized by collapsing focal segmental glomerulosclerosis, microcystic tubular dilatation and interstitial inflammation with progressive renal failure, often accompanied by proteinuria [328]. In a more recent study it was shown that, in human proximal tubular epithelial cells, HIV was able to promote secretion of profibrotic proteins and EMT through activation of different subcellular pathways (FAK/ILK-PI3K-NFκB cascade). LPA plays a critical role in HIV- induced changes that were blocked by either LPA receptor blocker or inhibitors of LPA synthesis [329].

Complementary studies are now required to better understand the role of LPA in both ADPKD an HIV-associated nephropathy.

2.8.4 LPA and acute kidney injury

In contrast to all above-mentioned data where LPA was involved in the development of kidney diseases as a causative factor, it was also found that LPA can exert some protective effects in acute kidney injury (AKI). Especially it was shown that in vivo injection of LPA18:1 to mice reduces ischemia/reperfusion (I/R)-induced increase in plasma creatinine and BUN showing protection against I/R-induced kidney dysfunction [330, 331]. The protection was accompanied by a reduction of caspase dependent apoptosis, inflammation and neutrophil influx into kidney tissue, by an inhibition of the complement activation, reduction in brush border destruction, epithelial edema and number of intraluminal casts [330, 331]. In another study it was shown that low doses of exogenous LPA18:1 perform protection in I/R- induced AKI which was lost with increasing doses of LPA [332]. In opposite to protective effect, AKI was exacerbated when LPA treatment was substituted with LPA3 receptor agonist OMPT. Moreover, LPA1/3 receptor antagonist VPC-12249 by its own reduces leukocyte infiltration and inflammation and protects against I/R-induced AKI. That VPC-12249-induced protection was lost when treatment with VPC-12249 was combined with OMPT [332]. The authors concluded that signaling through LPA3 receptors induces deleterious effects during AKI whereas signaling through other receptors exerts protective effects. Since LPA2 receptors are highly expressed in kidney, the authors conclude that signaling through LPA2 might be responsible for the LPA-induced protection upon I/R-induced AKI. However, relatively low expression of LPA1 receptors couldn’t exclude its involvement in LPA-induced protection. Moreover, at the period of that study other LPA receptors (LPA4-LPA6) were not described and possible protection by signaling through these receptors should also be considered. In addition, TRPV1 activation by specific agonists was reported to protect kidney during I/R [333, 334]. Since TRPV1 is a potential receptor for LPA, kidney protection induced by LPA could be mediated by TRPV1 activation as well. The affinity of LPA3 receptor is lower for LPA18:1 compared with other receptors. At low doses LPA18:1 activates possible "protective receptors", exerting some protection. High doses of LPA18:1 overcome the low affinity of LPA3 receptors resulting in the loss of protection. Interestingly treatment with OMPT, VPC-12249 or combination of both compounds did not modified heart rate or blood pressure indicating that the effects exerted by these molecules could not result from hemodynamic changes [332]. Even though VPC-12249 exerts protection upon AKI, it is not known whether it acts directly on tubular cells or exerts anti-inflammatory effects acting on inflammatory cells.

The discrepancy of results in these experiments shows that there is a need for more experiments to explain the role of LPA during AKI. The unavailability of specific receptor antagonist is a limiting factor for the studies aiming to discover which receptors are involved in LPA signaling.

3. Hypothesis

It was clearly demonstrated previously that LPA1 receptor blockade protects against development of kidney fibrosis in course of UUO and DN [164, 165, 324]. Moreover, LPA production is elevated from obstructed kidneys [164]. These findings have some clinical relevance since increased LPA levels were described in patients with CKD and plasma levels of ATX are associated with kidney lesions in patients with DN [301, 302]. In addition, elevated LPA was found in effluent accumulated in kidneys with obstructed ureters [173] and in urine of patients with DN [316].

In apparent contradiction to the profibrotic action in CKD, LPA was reported to protect kidneys against AKI. Two reports demonstrated that treatment with exogenous LPA preserves kidney function and structure upon I/R-induced AKI [330, 331].

Analysing these information a question could emerge: how the same molecule could perform protection of kidneys on one hand and cause kidney fibrosis on the other hand? It is important to remember here that the two forms of kidney pathologies are different entities. Even though they share the same clinical presentation of lost kidney function and some pathogenetic mechanisms (notably the inflammation), they are however quite different in the velocity of the disease development, in the level of the reversibility and outcomes. If the AKI is an acute condition with the rapid development of clinical presentations that is potentially reversible, CKD develops over years and once the process of fibrosis in the organ is launched no regression could be expected. We hypothesize hereby that LPA could be differently involved in the pathogenesis of AKI and CKD: it protects in course of acutely developed AKI and in opposite, cause disease progression in course of long lasting CKD. To test this hypothesis we studied the role of LPA in both kidney pathologies, CKD and AKI. We studied LPA involvement in subtotal or 5/6 nephrectomy (SNX) model of CKD (Part 1) and the possible protective role of LPA in endotoxemia-induced AKI (Part 2).

4. Results and discussion

4.1 LPA and CKD

4.1.1 Bibliographic context and objectives

LPA is considered as a profibrotic molecule in various organs (lungs, skin, kidney, etc.). Before the start of this study causative involvement of LPA in kidney fibrosis was initially demonstrated in UUO model [164] and has recently been demonstrated in DN model [324]. The metabolic pathways leading to increased urinary LPA described in UUO-induced kidney fibrosis and DN are not well characterized. Tsutsumi et al described increased renal ATX and AGK expression in parallel with increased choline producing activity of urine in obstructed kidneys of rats with UUO-induced kidney fibrosis, suggesting possible involvement of these enzymes in increased LPA production upon obstruction [173]. However, this hypothesis was not confirmed with experiments specifically targeting ATX or AGK (genetically or pharmacologically).

The UUO model recapitulates all the key features of fibrogenic response in kidneys including activated inflammatory response, increased production of pro-fibrotic cytokines and growth factors, appearance of myofibroblasts with excessive accumulation of ECM proteins followed by progressive tubular loss and peritubular capillary rarefaction [303]. However the very rapid installation of fibrosis in kidney upon UUO differs from progressive appearance of CKD in human. Other limitations of the model are the compensation by the normal contralateral kidney and the absence of urine from the damaged kidney [303]. In contrast to UUO, subtotal or 5/6 nephrectomy model mimics progressive renal failure and is characterized by glomerulosclerosis accompanied by TIF and tubular atrophy that emerge about 2 week after surgery. Functional changes such as decreased GFR and proteinuria and increased systemic arterial pressure are prominent [335].

The objectives of the first part of the thesis were to

1) Investigate the involvement of LPA in this more progressive model of CKD. 2) Study the changes in the renal production of LPA through analysis of both urinary LPA and the renal expression of enzymes involved in LPA metabolism in remnant kidneys 5 months after sub-total nephrectomy.

4.1.2 Results

Study of the LPA involvement in subtotal nephrectomy model of CKD and changes of its local metabolism.

Article 1:

Increased urinary lysophosphatidic acid in mouse with subtotal nephrectomy: potential involvement in chronic kidney disease Mirzoyan K., Baïotto A., Dupuy A., Marsal D., Denis C., Vinel C., Sicard P., Bertrand- Michel J., Bascands JL.,Schanstra JP., Klein J., Saulnier-Blache JS. Journal of Physiology and Biochemistry (2016) Vol. 72, Issue 4, 803-812

J Physiol Biochem DOI 10.1007/s13105-016-0518-0

ORIGINAL PAPER

Increased urinary lysophosphatidic acid in mouse with subtotal nephrectomy: potential involvement in chronic kidney disease

Koryun Mirzoyan & Anna Baïotto & Aude Dupuy & Dimitri Marsal & Colette Denis & Claire Vinel & Pierre Sicard & Justine Bertrand-Michel & Jean-Loup Bascands & Joost P. Schanstra & Julie Klein & Jean-Sébastien Saulnier-Blache

Received: 21 June 2016 / Accepted: 9 September 2016 # University of Navarra 2016

Abstract Increased incidence of chronic kidney disease chronic and progressive model of CKD. Five months after (CKD) with consecutive progression to end-stage renal surgical nephron reduction, SNx mice showed massive disease represents a significant burden to healthcare sys- albuminuria, extensive TIF, and glomerular hypertrophy tems. Renal tubulointerstitial fibrosis (TIF) is a classical when compared to sham-operated animals. Urinary and hallmark of CKD and is well correlated with the loss of plasma levels of LPA were analyzed using liquid chroma- renal function. The bioactive lysophospholipid tography tandem mass spectrometry. LPA was significant- lysophosphatidic acid (LPA), acting through specific G- ly increased in SNx urine, not in plasma, and was signif- protein-coupled receptors, was previously shown to be icantly correlated with albuminuria and TIF. Moreover, involved in TIF development in a mouse model of unilat- SNx mice showed significant downregulation in the renal eral ureteral obstruction. Here, we study the role of LPA in expression of lipid phosphate phosphohydrolases (LPP1, 2, a mouse subjected to subtotal nephrectomy (SNx), a more and 3) that might be involved in reduced LPA bioavailability through dephosphorylation. We concluded that SNx Electronic supplementary material The online version of this increases urinary LPA through a mechanism that could article (doi:10.1007/s13105-016-0518-0) contains supplementary involve co-excretion of plasma LPA with albumin material, which is available to authorized users. associated with a reduction of its catabolism in the kidney. : : : : K. Mirzoyan A. Baïotto A. Dupuy D. Marsal Because of the previously demonstrated profibrotic activ- : : : : C. Denis C. Vinel P. Sicard J. Bertrand-Michel ity of LPA, the association of urinary LPA with TIF : : suggests the potential involvement of LPA in the devel- J. P. Schanstra J. Klein J.

Mirzoyan et al.

(ESRD) that necessitates renal replacement therapy rep- profibrotic action in several organs including the lung, resents a major burden to healthcare systems [7]. CKD is skin, or kidney [6, 16–18]. The causal involvement of encompassing a number of heterogeneous disorders LPA in the development of kidney fibrosis was demon- affecting the structure and function of the kidney. They all strated in a mouse model of accelerated development of lead to a steady decline in nephron numbers that trigger TIF induced by unilateral ureteral obstruction (UUO). multiple adaptive mechanisms, which can be-come We and others demonstrated that UUO-mediated renal dysadaptive and lead to profibrotic changes within the fibrosis can be prevented by transgenic invalidation or kidney. Fibrosis, the process of uncontrolled healing with pharmacological blocking of the LPA1 receptor [16]. excess accumulation of extracellular matrix, is a hallmark Although the UUO model induces renal inflammation, of CKD and is well correlated with future decline of renal fibroblast activation, and accumulation of extracellular function. Development of fibrosis in kidney tissue is a matrix in the tubulointerstitium, it probably does not complex process and comprises pro-gressive development really mirror the progressive appearance of CKD ob- of tubulointerstitial fibrosis (TIF), glomerulosclerosis, and served in humans because of the rapidity of installation loss of glomerular and peritubular capillaries [14]. As most of the disease and the functional compensation by the renal diseases are associated with renal fibrosis, the contralateral kidney [8]. development of treat-ments aiming to slowdown, halt, or Subtotal nephrectomy, or the so-called 5/6 nephrec- even better, reverse fibrosis holds a great potential in the tomy (SNx) or remnant kidney model, recapitulates the fight against ESRD. The multitude of events and factors key features of nephropathy due to nephron reduction involved in the development of renal fibrosis is reflected that characterizes most human CKD progression. The by the large number of experimental reports showing the massive reduction of renal tissue due to SNx triggers potential anti-fibrotic effects of a number of strategies and excessive compensatory mechanisms of remaining com-pounds [11, 12]. However, the only currently nephrons that lead to hypertension, hyperfiltration, clinically available drugs that have shown to slowdown the glomerulosclerosis, TIF, and CKD [22]. pro-gression of CKD are the inhibitors of the renin angio- In the present work, we studied the possible tensin system (RASi). Unfortunately, even when treated involve-ment of LPA in the development of CKD with RASi, CKD continues to progress. Alternative associated with SNx in mouse. molecules or therapies are thus necessary.

Lysophosphatidic acid (LPA) is a lysophospholipid Materials and methods composed of a glycerol backbone with a phosphate head group at sn-3 position and a single ester-linked fatty Animals Female FVB/N mice were used because acid chain (C16:0, C18:0, C18:1, C20:4) at sn-1 or sn-2 they were described as more susceptible to develop position. Three major pathways are responsible for LPA CKD following SNx [13, 15]. FVB/N mice were synthesis: hydrolysis of lysophosphatidylcholine (LPC) purchased from Janvier (Le Genest Saint Isle, by a soluble lysophospholipase D (autotaxin, ATX), France) and housed in pathogen-free, temperature- deacylation of phosphatidic acid by phospholi-pase A1 controlled environment with a 12-h/12-h light/dark or A2 (PLA1 or PLA2), and phosphorylation of cycle. Animals had free access to food and tap water. monoacylglycerol phosphate by an acylglycerol kinase All experiments were conducted in accordance with (AGK) [1]. In contrast, the main LPA catabolic pathway National Institute of Health guidelines for the care of LPA is its dephosphorylation by lipid phosphate and use of laboratory animals and were approved by phosphohydrolase (LPP) that exists as three major sub- a local ethics committee (CEEA#122). types (LPP1, LPP2, and LPP3) [2, 4]. LPA has been shown to be present in various biological fluids includ- Subtotal (5/6) nephrectomy (SNx) Subtotal (5/6) ne- ing plasma and urine. While plasma LPA mainly origi- phrectomy was performed on 8-week-old mice. Animals nates from hydrolysis of LPC by ATX, whether its were anesthetized with 2 % isoflurane. The right kidney presence in urine results from plasma excretion or from was exposed, decapsulated, and removed through flank a local production in kidney remains poorly understood. incision. Next, the left kidney was ex-posed through flank LPA is a bioactive lipid ligand of specific G-protein- incision and was decapsulated to avoid ureter and adrenal coupled receptors (LPA1 to LPA6) that exerts a damage. Upper and lower poles

Increased urinary lysophosphatidic acid were resected and weighed. In parallel, control animals phosphomolybdic acid (2 min), and finally, with aniline underwent sham surgery by incision and suturing the blue dye solution (5 min). Slides were dehydrated, em- two flanks (SHAM). After surgery, animals were re- bedded in microscopic slides, and examined. Image placed into cages and were sacrificed 5 months later. analysis was performed using the ImageJ software Before sacrifice, the mice were placed in metabolic (https://imagej.nih.gov/ij/). Collagen III quantification cages for 12 h to collect urine. Mice were euthanized by was performed by measuring the ratio of the collagen pentobarbital injection after 6 h fasting. Blood was III-specific stained area to the total kidney section area. drawn by intra-cardiac puncture and collected in hepa- The glomerular area was analyzed by measuring the rinized tubes (Microvette Hép-Lithium, Sarstedt, area of 50 glomerules in the sections stained by Marnay, France) to avoid platelet aggregation. Ice-cold Masson’s trichrome. blood was then centrifuged for 5 min at 1500×g and upper phase plasma was collected and immediately LC/MS-MS quantification of LPA and LPC species in frozen at −80 °C. The remnant kidney was removed, plasma and urine LPA and LPC species were extracted and a portion of renal cortex was immediately immersed from urine (150 μL) and plasma (10 μL) using 1- in liquid nitrogen and stored at −80 °C for further RNA butanol containing 50 ng of LPA 17:0 and LPC 17:0 extraction and RT-PCR analysis. The remaining kidney as internal standards. Detection was conducted by LC/ was immediately fixed in Carnoy’s solution (2 % ferric MS-MS using an Agilent 1290 Infinity HPLC system chloride in ethanol/chloroform/acetic acid 48/24/8) and (Agilent Technologies) coupled on-line to an Agilent embedded in paraffin for histological analysis. 6460 triple quadrupole MS (Agilent Technologies) equipped with electrospray ionization source. Measurement of urinary albumin and creatinine Quantification was conducted through comparison with concentration The concentration of creatinine in urine calibration curves of synthetic lipid standards: LPA was measured by the colorimetric method of Jaffe ac- 16:0, LPA 17:0, LPA 18:0, LPA 18:1, LPA 18:2, LPA cording to the protocol Creatinine Assay Kit (Bio Assay 20:0, LPA 20:4, LPC 16:0, LPC 18:0, LPC 18:1, LPC Systems). The concentration of urinary albumin was 20:0, LPC 22:0, and LPC 24:0. More details on the measured by ELISA using the AlbuWell kit (Exocell). method and its validation are provided in supplementary files. Histological analysis Four-micrometer kidney sections were deparaffinized in toluene and rehydrated by a RNA analysis Total RNA was isolated from snap frozen series of ethanol baths containing decreasing concentra- mouse kidneys. Tissue fragments were lysed mechani- tions of ethanol. For anti-collagen III immunohisto- cally by TissuLyser (MM301, Retsch) in RLT buffer chemistry, endogenous peroxidase activity was blocked containing 1/100 of β-mercaptoethanol. Reverse tran- with S2001 (Dako, Trappes, France) to reduce non- scription and PCR amplification were performed as specific binding of the antibodies; the sections were previously described [19]. Primer sequences are listed covered with a saturation buffer (1 % milk). Next, in Table 1. The hypoxanthine phosphoribosyltransferase sections were incubated for 45 min at room temperature (HPRT) mRNA was used as a reference gene. with rabbit primary murine anti-collagen III antibody Normalization of gene expression was calculated ac- (1/250th, reference BP8014, Acris, Herford, Germany) cording to the following formula: 2(Ct HPRT-Ct gene of interest) followed by incubation with the secondary rabbit anti- where Ct corresponds to the number of cycles IgG antibody (DAKO) coupled to horseradish peroxi- needed to generate a fluorescent signal above a dase (HRP). Next, the sections were covered with di- predefined threshold. aminobenzidine (Dako, Trappes, France), the chromo- genic substrate of HRP. Finally, sections were counter- Statistical analysis Continuous variables were given stained by hematoxylin (DakoCytomation, Trappes, as mean ± SE. Comparisons were conducted with France) and examined under light microscope. For M a unpaired Mann-Whitney test. Association between s s o n’s t r i c h r o m e , s t a i n i n g s e c t i o n s w e r variables was analyzed by using the Spearman rank e deparaffinized and rehydrated as described above and correlation test. A P value <0.05 was considered sequentially incubated with hematoxylin-eosine (2 min), statistically significant: *P < 0.05, **P < 0.01, ***P picric acid (30 s), culvert fuchsin (5 min), < 0.001, ****P < 0.0001; ns as nonsignificant.

Mirzoyan et al.

Table 1 PCR primer sequences

Mouse gene Forward Reverse

acylglycerol kinase (AGK) GTGTTTGGCAACCAGCTCATT GCTGCGGGATTGAGAAAGAC autotaxin (ATX) TCCGTGCATCGTACATGAAGA CAGGACCGCAGTTTCTCAATG cell surface glycoprotein F4/80 TGACAACCAGACGGCTTGTG GCAGGCGAGGAAAAGATAGTGT collagen type 1 (COL1) TGTGTGCGATGACGTGCAAT GGGTCCCTCGACTCCTACA collagen type 3 (COL3) AAGGCGAATTCAAGGCTGAA TGTGTTTAGTACAGCCATCC TCTAGAA connective tissue growth factor (CTGF) GGCATCTCCACCCGAGTTAC GATTTTAGGTGTCCGGATGCA hypoxanthine phosphoribosyltransferase TGGCCATCTGCCTAGTAAAGC GGACGCAGCAACTGACATTTC (HRPT) kidney injury molecule-1 (KIM-1) TAGAGACACGGAAGGCAACC TGCTGCTACTGCTCCTTGTG lipid phosphate phosphatase-1 (LPP1) GGGAGACTGGGCAAGACTCTT CACTCGAGAAAGGCCCACAT lipid phosphate phosphatase-2 (LPP2) CGCGATCCAACTTCAACAACT CAGCCCCGAACAGAAAGGT lipid phosphate phosphatase-3 (LPP3) CCATCCTGGCGATCATTACAG AAAGGAAGCATCCCACTTGCT nephrin (npsh1) ATGGGAGCTAAGGAAGCCACA CCACACCACAGCTTAACTGTC neutrophil gelatinase-associated GGACCAGGGCTGTCGCTACT GGTGGCCACTTGCACATTGT lipocalin (NGAL) phospholipase A1 (PLA1) GGCGCTCGGAACAAAGC CGCAGGACAGCGCTGAT phospholipase A2 group 2D (PLA2g2D) TGAAGATCGATGGATGCAAGAG GATAGTGCCCTGGGAGATGCT phospholipase A2 group 2E (PLA2g2E) GGACGAGACGGATTGGTGTT CTCCAGGCGGCCATAGC phospholipase A2 group 4 (PLA2g4) CCAAAGGCACCGAGAATGA TCGCTCACCAAGGCCATTA podocin (npsh2) CCCTGAAGTTCTAGATGATTCCATAA AACAGCGTCGAGCAATTTCC ATA

Results npsh1 (nephrin) and npsh2 (podocin), when compared to SHAM (Table 3). These results are in agreement with Five months after surgery, SNx mice showed significant previous reports [15] showing that SNx leads to a strong higher diuresis (6.1-fold) associated with a significant impairment of kidney function associated with inflam- decrease in creatininuria, when compared to SHAM (Table mation, TIF, and glomerular damage. 2). SNx mice also showed a significant increase in LC/MS-MS analysis of urine lysophospholipid com- albuminuria (9.2-fold when expressed in mg/L and 29.5- position showed increased concentration of five differ- fold when expressed as a ratio to urinary creatinine ent molecular LPA species (LPA16:0, LPA18:0, (uACR)) when compared to SHAM (Table 2). Body LPA18:1, LPA18:2, and LPA20:4) in SNx when com- weight and blood glucose were not significantly differ-ent pared to SHAM mice (Fig. 2a, b). This was associated between SNx and SHAM mice (Table 2). Whereas the with significant increase in urine concentration of six of weight of the remnant kidney from SNx mice was not lysophosphatidylcholine (LPC) molecular species significantly different from the left kidney from SHAM (LPC16:0, LPC18:0, LPC18:1, LPC20:0, LPC22:0, and mice, it showed significant higher glomerular surface area LPC24:0; Fig. 2c, d). Urinary LPA and LPC were associated with significant higher interstitial collagen III expressed as a ratio to urinary creatinine in order to content (Table 2 and Fig. 1). mRNA analy-sis showed that adjust for urine dilution. In contrast to urine, the plasma remnant kidneys from SNx mice also exhibited significant levels of LPA and LPC species were not significantly higher expression of renal injury genes (NGAL, KIM-1 different between SNx and SHAM mice (Fig. 3a, b). COL I, and COLIII) as well as of the inflammatory gene These results indicated that SNx led to a urine-specific F4/80 when compared to SHAM (Table 3). Moreover, SNx alteration in lysophospholipid levels. mice also exhibited signifi-cant lower expression of the Correlation analysis showed that total urinary LPA glomerular specific genes and LPC levels were significantly associated with TIF

Increased urinary lysophosphatidic acid

Table 2 Body weight, organ weight, and urine composition of sham- and SNx-operated mice

Sham SNX

Mean ±SEM n Mean ± SEM n P value

Body weight (g) 23.3±0.2 10 24 ± 1 5 ns Left kidney weight (mg) 154 ± 5 10 167 ± 11 5 ns Blood glucose (mg/dl) 128 ± 15 11 119 ± 13 5 ns Diuresis (ml/18 h) 0.6 ± 0.1 11 4 ± 1 4 * Urinary creatinine (mg/L) 672 ± 69 11 264 ± 45 9 *** Urinary albumin (mg/L) 24 ± 3 11 218 ± 39 9 ** Urinary albumin/creatinine ratio (mg/g) 37 ± 5 11 1082 ± 254 9 ** Mean renal glomerular area (μm2) 2632 ± 91 6 4946 ± 333 4 ** Renal collagen III staining (% total surface) 11 ± 1 9 35 ± 5 10 **

ns not significant *P<0.05, **P<0.01, ***P<0.01 and uACR, but not with the glomerular surface area expression of the three main LPA anabolic enzymes, (Table 4). These results indicated that SNx-mediated PLA1, AGK, and ATX, when compared to kidneys increase of urinary lysophospholipids was correlated from SHAM-operated animals (Fig. 4a–c). SNx rem- with the intensity of kidney damage. nant kidney also exhibited significant lower expression Analysis of kidney mRNA expression showed that of the three main LPA catabolic enzymes, LPP1, LPP2, the SNx remnant kidney exhibited significant lower and LPP3, when compared to kidneys from SHAM-

A B SHAM SNX

C D SHAM SNX

Fig. 1 Histological features of kidney damage. Masson trichrome detection of collagen III (c) and measurement of collagen III- staining (a) and measurement of glomerular area (SHAM: n = 6; positive stained area (SHAM: n = 9; SNx: n = 10; d). Picture SNx: n = 4; b). Arrows localize glomeruli. Immunohistochemical magnification, ×400. Scale 0.05 mm. Values are mean ± SEM

Mirzoyan et al.

Table 3 Renal gene expression in sham- and SNx-operated mice kidneys from SHAM-operated animals (Fig. 4g–i).

Genes Gene expression P value These results indicated that SNx led to altered expression of the enzymes involved in renal lysophospholipid mRNA/HPRT mRNA (×10,000) metabolism. SNx remnant kidney also exhibited signif-

icant lower expression of the LPA receptor subtypes, Sham SNX LPA1, LPA2, LPA3, and LPA4, with no change in Mean ± SEM n Mean ± SEM n LPA5 and LPA6 (Fig. 1S). These results indicated that SNx is not associated with hyper sensitization of the npsh1 2294 ± 218 10 670 ± 202 10 **** kidney to LPA. Therefore, the contribution of LPA in npsh2 12,544 ± 960 10 6268 ± 1523 10 * SNx-mediated renal fibrosis is likely through KIM-1 1641 ± 863 10 5480 ± 1529 10 * modulation of its renal production rather through an NGAL 594 ± 140 10 16,407 ± 6018 10 * increase in its renal sensitivity. COL1 545 ± 72 10 2394 ± 576 10 * COL3 1438 ± 153 10 11,510 ± 2954 10 ** F4/80 125 ± 16 10 238 ± 32 10 ** Discussion operated animals (Fig. 4d–f). Moreover, SNx remnant Our hypothesis was that LPA could be involved in the kidney exhibited no significant in PLA2g4, and signif- development of CKD in the SNx mouse model. As icant higher expression of two isoforms of secreted previously reported [15, 20, 21] and confirmed in the PLA2 (sPLA2g2D and sPLA2g2D) when compared to present study, SNx model is characterized by a strong

Fig. 2 Influence of SNx on urinay LPA and urinary LPC. The urinary LPA (b) and total urinary LPC (d) correspond to the concentration of six LPA (a) and LPC species (c) was quantified sum of all species for each lysophospholipid class. Values are by LC/MS-MS in urine from sham- and SNx-operated mice. Total mean ± SEM of five to nine animals per group

Increased urinary lysophosphatidic acid

Fig. 3 Influence of SNx on plasma LPA and plasma LPC. The Total plasma LPA (b) and total plasma LPC (d) correspond to concentration of six LPA (a) and LPC species (c) was quantified the sum of all species for each lysophospholipid class. Values by LC/MS-MS in plasma from sham- and SNx-operated mice. are mean ± SEM of seven to nine animals per group impairment in kidney function associated with major However, it is particularly useful under conditions of alterations in kidney structure such as TIF and normal kidney function with unchanged creatinine ex- glomer-ular damage. Our results indicate that SNx cretion. When considering the data of Table 2, urinary enhances urinary LPA excretion without changing its creatinine is decreased after SNx; however, the diuresis plasma levels. In addition, the urinary LPA levels are is increased by 6.1-fold. Combining the data from Table signifi-cantly correlated with albuminuria and TIF. 2 allows estimating total creatinuria to 0.403 mg/18 h Quantification of urinary LPA is expressed as a ratio and 1.056 mg/18 h in SHAM and SNx, respectively. to urinary creatinine. This is of current use in clinical This means that the increase in albumin and LPA chemistry and allows correcting for urine dilution. urinary excretion is still higher than described

Table 4 Correlation of urinary lysophospholipids with albuminuria, TIF, and glomerular area

Urinary lysophospholipids

(nmole/g creatinine)

Total LPA Total LPC

P value Spearman r P value Spearman r

Urinary albumin/creatinine ratio (mg/g) 0.00003 0.883 0.00020 0.801 Renal collagen III staining (% total surface) 0.00386 0.733 0.00058 0.820 Mean renal glomerular surface (μm2) 0.13611 0.724 0.11070 0.714

Mirzoyan et al.

Fig. 4 Influence of SNx on renal expression of enzymes involved phospholipase A2 (cPLA2g4A) (g), and in LPA metabolism. mRNAs encoding for phospholipase A1 secretedphospholipases A2 (sPLA2g2D and sPLA2g2E) (h, i) (PLA1) (a), acylglycerol kinase (AGK) (b), autotaxin (ATX) (c), were quantified in the kidney of sham- and SNxoperatedmice. lipid phosphatephosphatases (LPP1–3) (d, e, f), cytosolic Values are mean ± SEM of 10 to 11 animals per group

in the manuscript. Previous reports from our laboratory urinary LPA is closely correlated with albuminuria. SNx- and others showed that LPA participates in TIF devel- mediated increase in urinary LPA could therefore result, at opment in UUO rodent models, through activation of least in part, from its co-filtration with albumin from the LPA1 receptor [16]. Therefore, we propose that the plasma to urine. Nevertheless, our results show that renal profibrotic activity of LPA could participate to the de- expression of several enzymes involved in LPA velopment of SNx-associated CKD. metabolism is altered by SNx. This suggests that changes LPA is present in plasma and urine, and our results in renal LPA production might also contribute to its SNx- show that SNx alters its concentration exclusively in mediated increase in urine. The hypothesis is in agreement urine. Whereas the origin of plasma LPA has clearly with previous reports showing that renal production of been established (i.e., ATX-mediated hydrolysis of LPC LPA is increased following UUO [16] as well as in [3]), the origin of urinary LPA remains unclear. The glomeruli of mice with CKD associated with diabetic presence of a given molecule in urine usually results nephropathy [9]. Tsutsumi et al. previously found that the from combined mechanisms of plasma filtration and LPA concentration increases in urine that accumulates in local production within the kidney. Similarly with other the kidney following UUO in rat, and this was associated lipids present in the plasma, LPA is bound to trans- with increased renal expression of ATX and AGK, two porters such as albumin or lipoproteins. SNx is associ- enzymes involved in LPA synthesis [20]. Our results show ated with severe albuminuria, and our results show that that, conversely to UUO, renal

Increased urinary lysophosphatidic acid

ATX and AGK are downregulated by SNx. We also MUNDUS, MEDEA). All authors participated in the observed a downregulation of PLA1, and no change in conception and design, or analysis and interpretation of the data, contributed to drafting and revising the manuscript, and PLA2g4, two other potential LPA synthetic enzymes. gave final approval of the version to be published. Therefore, these observations suggest that ATX, AGK, PLA1, and PLA2g4 are unlikely responsible for SNx- Compliance with ethical standards All experiments were con- mediated increase in urinary LPA. On the other hand, ducted in accordance with National Institute of Health our results show that SNx led to a clear downregulation guidelines for the care and use of laboratory animals and were approved by a local ethics committee (CEEA#122). of the lipid phosphate phosphatase (LPP1, LPP2, and LPP3) that are catabolic enzymes involved in dephosphosphorylation and inactivation of LPA [2, 5]. References Based on those observations, we propose that SNx- induced increase in urinary LPA could be explained, at 1. Aikawa S, Hashimoto T, Kano K, Aoki J (2015) least in part, by a reduction of its catabolism in kidney Lysophosphatidic acid as a lipid mediator with multiple with the consequence of its increased bioavailability in bio-logical actions. J Biochem 157:81–89 2. Alderton F, Darroch P, Sambi B, McKie A, Ahmed IS, urine. Our results indicate that, in parallel to urinary LPA, Pyne N, Pyne S (2001) G-protein-coupled receptor stimulation of the p42/p44 mitogen-activated protein urinary LPC is also increased by SNx and is correlated kinase pathway is at-tenuated by lipid phosphate with albuminuria and TIF. This suggests that, as for phosphatases 1, 1a, and 2 in human embryonic kidney 293 LPA, LPC could also be involved in the development of cells. J Biol Chem 276: 13452–13460 CKD. LPC, mainly derived from PLA2-dependent 3. Aoki J, Inoue A, Okudaira S (2008) Two pathways for lysophosphatidic acid production. Biochim Biophys Acta deacylation of phosphatidylcholine, is abundant in plas- 1781:513–518 ma as a major component of oxidized low-density lipo- 4. Benesch MG, Tang X, Venkatraman G, Bekele RT, Brindley proteins and is recognized as a carrier of fatty acids, DN (2015) Recent advances in targeting the autotaxin- glycerol, and choline between tissues. Our results show lysophosphatidate-lipid phosphate phosphatase axis in vivo. J Biomed Res 30. doi:10.7555/JBR.29.20150058 that SNx induces a clear upregulation of the renal ex- 5. Brindley DN, Pilquil C (2009) Lipid phosphate phosphatases pression of two potential LPC-producing enzymes, and signaling. J Lipid Res 50(Suppl):S225–S230 PLA2g2D and PLA2g2E. We propose that these en- 6. Castelino FV, Seiders J, Bain G, Brooks SF, King CD, zymes might, among others, play a role in SNx-induced Swaney JS, Lorrain DS, Chun J, Luster AD, Tager AM (2011) urinary LPC. LPC is a modulator of inflamma-tion and Amelioration of dermal fibrosis by genetic deletion or pharmacologic antagonism of lysophosphatidic acid receptor 1 in a immunity with a connection to atherosclerosis. LPC has mouse model of scleroderma. Arthritis Rheum 63:1405–1415 recently been demonstrated to participate to 7. Eckardt KU, Coresh J, Devuyst O, Johnson RJ, Kottgen A, glomerulosclerosis by favoring infiltration of foam cells Levey AS, Levin A (2013) Evolving importance of kidney into glomeruli [10]. LPC was also shown to increase in disease: from subspecialty to global health burden. Lancet 382:158–169 urine that accumulates in kidney following UUO in rat 8. Eddy AA, Lopez-Guisa JM, Okamura DM, Yamaguchi I [20]. Our results on SNx-mice reinforce the possible (2012) Investigating mechanisms of chronic kidney involvement of LPC in CKD through a possible impact disease in mouse models. Pediatr Nephrol 27:1233–1247 on TIF and kidney function. 9. Grove KJ, Voziyan PA, Spraggins JM, Wang S, Paueksakon P, Harris RC, Hudson BG, Caprioli RM (2014) Diabetic ne- In conclusion, since the SNx model is known to phropathy induces alterations in the glomerular and tubule recapitulate the key features of human CKD [22], the lipid profiles. J Lipid Res 55:1375–1385 involvement of LPA and LPC in human 10. Hara S, Kobayashi N, Sakamoto K, Ueno T, Manabe S, nephropathies could therefore be considered. If our Takashima Y, Hamada J, Pastan I, Fukamizu A, Matsusaka T, hypothesis is cor-rect, the enzymes involved in the Nagata M (2015) Podocyte injury-driven lipid peroxidation accelerates the infiltration of glomerular foam cells in focal metabolism of LPA and LPC might represent segmental glomerulosclerosis. Am J Pathol 185:2118–2131 potential targets for the treat-ment of CKD. 11. He J, Xu Y, Koya D, Kanasaki K (2013) Role of the endothelial-to-mesenchymal transition in renal fibrosis of Acknowledgments This work was supported by grants from chronic kidney disease. Clin Exp Nephrol 17:488–497 INSERM and the Société Francophone du Diabète (grant 2015). 12. Iwano M, Neilson EG (2004) Mechanisms of tubulointerstitial We thank Christine Delage and Denis Calise (INSERM US06 fibrosis. Curr Opin Nephrol Hypertens 13:279–284 CREFRE, Toulouse, France) for SNx surgery. Koryun Mirzoyan is 13. Laouari D, Burtin M, Phelep A, Martino C, Pillebout E, supported by a Ph.D. grant from Europe (ERASMUS Montagutelli X, Friedlander G, Terzi F (2011) TGF-alpha

Mirzoyan et al.

mediates genetic susceptibility to chronic kidney disease. J Hart WK, Pardo A, Blackwell TS, Xu Y, Chun J, Luster Am Soc Nephrol 22:327–335 AD (2008) The lysophosphatidic acid receptor LPA1 links 14. Louis K, Hertig A (2015) How tubular epithelial cells dictate pulmo-nary fibrosis to lung injury by mediating fibroblast the rate of renal fibrogenesis? World J Nephrol 4:367–373 recruitment and vascular leak. Nat Med 14:45–54 15. Pillebout E, Burtin M, Yuan HT, Briand P, Woolf AS, 19. Treguer K, Dusaulcy R, Gres S, Wanecq E, Valet P, Friedlander G, Terzi F (2001) Proliferation and Saulnier-Blache JS (2013) Influence of secreted factors remodeling of the peritubular microcirculation after from human adipose tissue on glucose utilization and nephron reduction: association with the progression of proinflammatory reaction. J Physiol Biochem 69:625–632 renal lesions. Am J Pathol 159:547–560 20. Tsutsumi T, Adachi M, Nikawadori M, Morishige J, 16. Pradere JP, Klein J, Gres S, Guigne C, Neau E, Valet P, Tokumura A (2011) Presence of bioactive Calise D, Chun J, Bascands JL, Saulnier-Blache JS, lysophosphatidic acid in renal effluent of rats with Schanstra JP (2007) LPA1 receptor activation promotes unilateral ureteral obstruc-tion. Life Sci 89:195–203 renal interstitial fibrosis. J Am Soc Nephrol 18:3110–3118 21. Verdoorn KS, Lindoso RS, Lowe J, Lara LS, Vieyra A, 17. Rancoule C, Pradere JP, Gonzalez J, Klein J, Valet P, Einicker-Lamas M (2010) Bone marrow mononuclear Bascands JL, Schanstra JP, Saulnier-Blache JS (2011) cells shift bioactive lipid pattern in injured kidney towards Lysophosphatidic acid-1-receptor targeting agents for tissue repair in rats with unilateral ureteral obstruction. fibrosis. Expert Opin Investig Drugs 20:657–667 Nephrol Dial Transplant 25:3867–3874 18. Tager AM, LaCamera P, Shea BS, Campanella GS, Selman 22. Yang HC, Zuo Y, Fogo AB (2010) Models of chronic M, Zhao Z, Polosukhin V, Wain J, Karimi-Shah BA, Kim ND, kidney disease. Drug Discov Today Dis Models 7:13–19

Supplements.

LC/MS-MS quantification of LPA and LPC species in plasma and urine.

Chemicals and reagents.1-Butanol was purchased from Sigma Aldrich. Ammonium Formate (>99%),

Formic acid (FA) and Methanol HPLC-grade (MeOH) were from Sigma Aldrich. Synthetic lipid standards:

LPA 16:0, LPA 17:0, LPA 18:0, LPA 18:1, LPA 18:2, LPA 20:0, LPA 20:4, LPC 16:0, LPC 17:0, LPC 18:0,

LPC 18:1, LPC 20:0, LPC 22:0, LPC 24:0 were purchased from Avanti Polar Lipids. Water used in this study was purified on a milliQ system (Millipore).

Lipid extraction.Urine (150 μL) or plasma (10μl) volume was made up to 500 μL with deionized water and extracted with 500 μL non-acidified 1-Butanol containing 50 ng of LPA 17:0 as internal standards. Non- acidified 1-butanol was preferred to acidified 1-butanol to avoid artificial production of LPA due to acid- mediated hydrolysis from other phospholipids as previously demonstrated[1-3]. The samples were vortexed every 30 minutes for 2 h before centrifugation at 10.000 g for 20 minutes. The upper phase was transferred to a new tube and evaporated to dryness under reduced pressure using a Rotavapor. The extract was dissolved in 160 μL MeOH then filtrated on a 0.45 μmpolyterafluoroethylene membrane filter. The filtrate was evaporated and dissolved in 20 μL MeOH. The extract was then stored at−20 °C before LC-MS/MS analysis.

LC/MS-MS.High performance liquid chromatography was performed using an Agilent 1290 Infinity (Agilent

Technologies) equipped with an auto sampler, a binary pump and a column oven. The analytical column was an Acquity UPLC BEH-C8 (100 x 2,1 mm, 1,7 廘m) (Waters) maintained at 25 °C. The separation on C8 column allows a short time analysis (8 minutes including column equilibration time). The mobile phases consisted of Water, FA and MeOH (20:0,5:79,5; v/v/v) (A) and MeOH, FA (99,5:0,5, v/v) (B). The two mobile phases contained 5 mM ammonium formate. The gradient was as follows: 0% B at 0 min, 0% B at 1 min, 100% B at 2 min, 100% B at 5 min, 0% B at 6 min and 0% B at 8 min. The flow rate was 0,2 mL/min.

The auto sampler was set at 5 °C and the injection volume was 5 μL. The HPLC system was coupled on-line to an Agilent 6460 triple quadrupole MS (Agilent Technologies) equipped with electrospray ionization source. Electrospray ionization (ESI) was performed in positive (Pos) ion mode for LPC and negative (Neg) ion mode for LPA using two separate acquisitions. Optimization of MS parameters was performed manually for each standard. After optimization, the source parameters used were as follows: source temperature was

70 set at 300°C, nebulizer gas (nitrogen) flow rate was 10 L/min, sheath gas temperature was 300°C, sheath gas

(nitrogen) flow rate was 12 L/min and the spray voltage was adjusted to +4000 V. Collision energy and

Fragmentor voltage were optimized for each LPA and LPC species (Table S1). The most abundant daughter ions generated by tandem mass spectrometry fragmentation of each species were 153 and 184 m/z for LPA and LPC respectively (Table S1). Analyses were performed in Selected Reaction Monitoring detection mode

(SRM) using nitrogen as collision gas. Finally, peak detection, integration and quantitative analysis were done using MassHunterQqQ Quantitative analysis software (Agilent Technologies) and Microsoft Excel software.

Calibration curves.LPA 17:0 and LPC 17:0 was mixed at a concentration of 1000 ng/mL and 200 ng/mL respectively in MeOH and used as internal standard (IS) solution. Stock solution of LPA and LPC species

(except IS) was prepared in MeOH at 1000 ng/mL and 200 ng/mL respectively, then ten calibration points

(dilution by half) were prepared in MeOH. These ten solutions were mixed with a fixed concentration of IS solution to prepare the following LPA standard solutions: 1.95, 3.9, 7.8, 15.6, 31.25, 62.5, 125, 250 and 500 ng/mL containing 100 ng/mL of IS for LPC; and 3.9, 7.8, 15.6, 31.25, 62.5, 125, 250, 500 and 1000 ng/mL containing 500 ng/mL of IS for LPA. Calibration curves were calculated by the ratio between analyte and internal standard peak area. Finally the linearity, the limit of detection (LOD) and the limit of quantification

(LOQ) were determined for the 12 compounds (Table S2). The standard curve of each LPA and LPC species fitted to a linear regression model with correlation factors over 0.99 (Table S2).

Accuracy and precision.To determine the repeatability and precision of the method, 3 samples containing

(3.91, 31.25 and 250 ng/mL for LPC, 7.8, 62.5 and 500 ng/mL for LPA) were analyzed in triplicate on the same day with back-calculated amount using standard curves. This procedure was repeated at day 1, day 7 and day 15. RSD (%) and accuracy (%) within one sample run (intra-day) and between sample runs (inter- day) was determined for each species of LPA and LPC (Table S3 and S4). Intra-day accuracies were ranging from 98.1% to 107.6% for the LPA species (Table S3) and from 96.6 to 119.2% for the LPC species.

Variations for 3 injections was ≤ 5.6% for LPA species and <13.4% for LPC species. Inter-day variations were assessed by re-analyzing the same sample every 7 days (n=3). Accuracies were between 94.6 and

112.2% for LPA species, and between 99.8 and 110.0% for LPC species, indicating suitable reproducibility of the protocol.

Matrix effect and extraction efficiency.Urine or plasma lipid extracts were spiked (post spiked samples) with stock solution of LPA and LPC at two concentrations (500 ng/mL and 62.5 ng/mL for LPA ; 250 ng/mL and 31.25 ng/mL for LPC ) and IS solution (50 μL). Each sample was analyzed in triplicate. In parallel, a separate set of stock solution of LPA, LPC and IS solution was directly analyzed in triplicate. Matrix effect was determined as the percentage difference between each LPA and LPC standard peak area with and without plasma and urine extracts. Urine extract without spiking was analyzed to quantify area of “natural” analyte and subtract this one to post spiked samples. Matrix effect of urine was rather low with a signal recovery of varying from 58.7 to 74.9%, and from 60.9 to 83.4% for LPA and LPC species respectively

(Supplementary Table 5 and 6). Matrix effect of plasma was more important with a signal recovery varying from 23.7 to 68.8 %, and from 21.9 to 42.9% for LPA and LPC species respectively (Supplementary Table

6). To calculate extraction efficiency, plasma and urine (n = 3) were spiked with IS solution (50 μL) and stock solution of LPA and LPC at two concentrations (500 ng/mL and 62.5 ng/mL for LPA ; 250 ng/mL and

31.25 ng/mL for LPC ) before 1-butanol extraction (pre spiked sample). The sample analysis was performed in triplicate and the extraction efficiency was determined as the percent difference between pre spiked and post spiked samples. Urinary and plasma matrix effect and extraction efficiency were calculated for each compound (Tables S5 and S6). Despite the differences in matrix effect, extraction yields of LPA and LPC species were high (65 to 99%) and not different between urine and plasma (Table S5 and S6).

References

[1] Scherer M, Schmitz G and Liebisch G (2009) High-throughput analysis of sphingosine 1-phosphate, sphinganine 1-phosphate, and lysophosphatidic acid in plasma samples by liquid chromatography-tandem mass spectrometry. Clin Chem 55: 1218-22

[2] Zhao Z and Xu Y (2010) An extremely simple method for extraction of lysophospholipids and phospholipids from blood samples. J Lipid Res 51: 652-9

[3] Onorato JM, Shipkova P, Minnich A, Aubry AF, Easter J and Tymiak A (2014) Challenges in accurate quantitation of lysophosphatidic acids in human biofluids. J Lipid Res 55: 1784-96

Table S1. HPLC retention time (RT) and selected reaction monitoring detection mode (SRM) conditions for each LPA and LPC species

Compound RT Precursor ion Product ion Fragmentor Colision (m/z) (m/z) (V) energy (V) LPA (16:0) 3.89 409 153 140 20 LPA (18:0) 4.21 437 153 140 20 LPA (18:1) 3.99 435 153 140 20 LPA (18:2) 3.76 433 153 140 20 LPA (20:0) 4.53 465 153 140 22 LPA (20:4) 3.77 457 153 140 20 LPA (17:0) 4.05 423 153 140 20

LPC (16:0) 3.67 496 184 110 30 LPC (18:0) 4.17 524 184 120 30 LPC (18:2) 4.21 533 184 120 30 LPC (18:1) 3.90 522 184 120 30 LPC (20:0) 4.43 552 184 120 30 LPC (22:0) 4.68 580 184 120 30 LPC (24:0) 4.96 608 184 120 30 LPC (17:0) 3.98 510 184 120 30

Table S2. Calibration curve equation, coefficient of determination (r2), limit of detection (LOD), and limit of quantification (LOQ) for each LPA and LPC species.

LOD LOQ Species Curve equation r2 (ng/mL) (ng/mL) LPA 16:0 y = 8.399681E - 004 * x - 0.009931 0.998 0.98 1.95 LPA 18:0 y = 9.002772E - 004 * x - 0.010106 0.998 0.49 0.98 LPA 18:1 y = 6.852677E - 004 * x - 0.008400 0.998 0.98 1.95 LPA 18:2 y = 0.001747 * x - 0.021235 0.998 0.49 0.98 LPA 20:0 y = 6.721874E - 004 * x - 0.008460 0.998 0.98 1.95 LPA 20:4 y = 0.001408 * x - 0.016777 0.998 0.49 0.98 LPC 16:0 y = 0.004262 * x - 0.028598 0.997 0.06 0.12 LPC 18:0 y = 0.005037 * x - 0.031414 0.998 0.06 0.12 LPC 18:1 y = 0.005391 * x - 0.032945 0.998 0.06 0.12 LPC 20:0 y = 0.004247 * x - 0.020694 0.999 0.06 0.12 LPC 22:0 y = 0.003138 * x - 0.013611 0.999 0.12 0.24 LPC 24:0 y = 0.001920* x - 0.006924 0.999 0.12 0.24

Table S3.Intra- and inter-day precision and accuracy of assay method for each LPA species.

Intra-day Inter-day RSD Accuracy RS Accuracy Species (ng/ml) Measured Measured % % D % % LPA 16:0 7.81 7.56 ± 0.43 5.6 103.2 7.71 ± 0.57 7.4 101.3 62.5 57.75 ± 0.21 0.4 107.6 54.90 ± 0.50 0.9 112.2 500 483.50 ± 7.76 1.6 103.3 461.47 ± 8.54 1.9 107.7 LPA 18:0 7.81 7.78 ± 0.10 1.3 100.3 8.17 ± 0.28 3.4 95.4 62.5 59.61 ± 0.22 0.4 104.6 58.76 ± 1.44 2.5 106.0 500 480.31 ± 2.71 0.6 103.9 481.87 ± 75.79 15.7 103.6 LPA 18:1 7.81 7.96 ± 0.20 2.5 98.0 7.52 ± 0.32 4.3 103.7 62.5 59.61 ± 0.50 0.8 105.5 58.13 ± 0.40 0.7 107.0 500 481.03 ± 1.56 0.3 103.8 476.80 ± 13.11 2.7 104.6 LPA 18:2 7.81 7.77 ± 0.21 2.6 100.4 7.85 ± 0.17 2.2 99.5 62.5 59.24 ± 0.45 0.8 105.2 57.18 ± 0.87 1.5 108.5 500 480.77 ± 1.37 0.3 103.8 467.07 ± 4.37 0.9 106.6 LPA 20 :0 7.81 7.62 ± 0.13 1.7 102.3 8.23 ± 0.44 5.3 94.6 62.5 59.75 ± 0.27 0.5 104.4 56.47 ± 0.73 1.3 109.6 500 478.35 ± 1.54 0.3 104.3 492.97 ± 17.78 3.6 101.4 LPA 20 :4 7.81 7.87 ± 0.18 2.3 99.2 7.76 ± 0.14 1.8 100.6 62.5 59.02 ± 0.23 0.4 105.6 57.51 ± 0.51 0.9 108.0 500 481.16 ± 0.87 0.2 103.8 464.29 ± 13.50 2.9 107.1

Table S4.Intra- and inter-day precision and accuracy of assay method for each LPC species. Intraday Interday Species (ng/ml) Measured RSD Accuracy Measured RSD Accuracy % % % % LPC 16:0 3.91 3.63 ± 0.01 0.3 107.1 3.52 ± 0.05 1.4 110.0 31.25 29.21 ± 0.18 0.5 106.5 28.25 ± 0.15 0.5 109.6 250 239.76 ± 3.23 1.3 104.1 245.41 ± 3.56 1.5 101.8 LPC 18:0 3.91 3.79 ± 0.03 1.6 103.2 3.63 ± 0.01 0.3 107.2 31.25 29.23 ± 0.10 0.9 106.5 29.49 ± 0.10 0.3 105.6 250 243.01 ± 1.31 0.3 102.8 242.92 ± 3.06 1.3 102.8 LPC 18:1 3.91 3.65 ± 0.06 0.8 106.6 3.78 ± 0.01 0.3 103.3 31.25 28.90 ± 0.26 0.3 107.5 29.17 ± 0.11 0.4 106.7 250 243.54 ± 0.75 0.5 102.6 243.88 ± 0.45 0.2 102.4 LPC 20:0 3.91 3.57 ± 0.17 4.8 108.6 3.75 ± 0.04 1.1 104.1 31.25 27.16 ± 1.23 4.5 113.1 29.59 ± 0.15 0.5 105.3 250 249.05 ± 2.17 0.9 100.4 245.36 ± 2.70 1.1 101.9 LPC 22:0 3.91 3.43 ± 0.9 8.5 112.1 3.75 ± 0.05 1.3 104.1 31.25 26.05 ± 1.75 6.7 116.6 29.62 ± 0.18 0.6 105.2 250 255.99 ± 1.93 0.8 97.6 247.12 ± 4.22 1.7 101.2 LPC 24:0 3.91 3.42 ± 0.246 13.4 112.5 3.74 ± 0.16 4.3 104.3 31.25 25.22 ± 2.03 8.0 119.2 29.75 ± 0.23 0.8 104.8 250 258.54 ± 8.02 3.1 96.6 250.57 ± 6.86 2.7 99.8

Table S5. Matrix effect and extraction efficiency for LPAs

Urine Plasma Matrix effect Yield extraction Matrix effect Yield extraction LPA species (ng/ml) (ng/ml) (ng/ml) (ng/ml) 500 62.5 500 62.5 500 62.5 500 62.5 16 :0 58.75 ± 3.9 68.9 ± 4.1 92.3 ± 4.6 83.3 ± 1.9 23.7 ± 1.7 21.4 ± 3.2 72 ± 7.6 86 ± 5.2 18 :0 70.67 ±6.9 83.40 ± 5.9 98.3 ± 2.3 94.1 ± 2.8 36.4 ± 30.9 36.7 ± 1.9 78 ± 4.8 89 ± 3.8 18 :1 64.6 ± 5 73.5 ± 4.2 91.7 ± 3.8 82.4 ± 2.6 27.8 ± 1.6 27.6 ± 3.2 69 ± 6.5 83 ± 4.3 18 :2 60.9 ± 3.5 68.1 ± 4.2 92.1 ± 5.0 80.1 ± 2.4 26 ± 1.3 26.4 ± 3.2 74 ± 3.5 89 ± 5.4 20 :0 85.0 ± 5.0 74.9 ± 6.0 93.6 ± 3.1 82.6 ± 3.8 66.8 ± 2.9 69.3 ± 6.6 65 ± 6.6 74 ± 5 20 :4 59.1 ± 4.2 66.3 ± 4.2 91.4 ± 5.2 78.3 ± 2.4 25.5 ± 2.3 25.6 ± 1.3 72 ± 3.4 86 ± 4.2 17 :0 (IS) 65.6 ± 4.0 58.5 ± 3.2 94.7 ± 4.2 89.2 ± 2.5 31 ± 1.5 27 ± 1.2 74 ± 4.4 91 ± 1.8

Table S6. Matrix effect and extraction efficiency for LPCs

Urine Plasma Matrix effect Yield extraction Matrix effect Yield extraction LPC species (ng/ml) (ng/ml) (ng/ml) (ng/ml) 250 31.2 250 31.2 250 31.2 250 31.2 16 :0 42.3 ± 2.4 36.6 ± 5.2 92.3 ± 3.8 87.0 ± 4.6 25.1 ± 4.1 26.6 ± 3.9 97.4 ± 0.9 80.3 ± 6.5 18 :0 35.0 ± 2.2 37.6 ± 4.6 99.7 ± 2.2 97.3 ± 1.9 28.9 ± 4.5 30.2 ± 2.3 96.9 ± 2.0 67.4 ± 0.8 18 :1 42.3 ± 2.4 36.6 ± 5.2 94.9 ± 1.9 95.1 ± 3.6 31.5 ± 0.8 38.6 ± 5.8 96.4 ± 1.4 70.0 ± 1.0 20 :0 41.5 ± 2.3 56.8 ± 3.1 92.6 ±1.9 87.8 ± 4.7 24.2 ± 2.6 42.9 ± 8.8 76.6 ± 9.3 92.3 ± 6.5 22 :0 43.1 ± 1.8 62.0 ± 3.2 90.7 ± 2.1 86.6 ± 2.5 21.9 ± 1.7 38.3 ± 6.3 88.2 ± 8.3 89.6 ± 5.4 24 :0 55.9 ± 3.4 79.9 ± 6.1 93.5 ± 3.2 90.2 ± 4.8 27.0 ± 2.1 33.8 ± 7.2 92.7 ± 6.5 89.5 ± 6.4 17 :0 (IS) 42.5 ± 2.3 39.1 ± 2.5 93.7 ± 1.3 90.0 ± 2.9 34.6 ± 5.8 39.4 ± 4.3 88.6 ± 6.5 88.4 ± 1.2

LPA receptor subtypes

4000 Sham 3500 SNx 3000 2500 * 2000 1500 500

(x10000) 400 ** 300 ** 200 mRNA / HPRTmRNA 100 *** 0 1 2 4 5 6 A A A A P PA LP3 P P P L L L L L

Figure 1S.Influence of SNx on renal expression of LPA receptor subtypes. mRNAs encoding for LPA1,

LPA2, LPA3, LPA4, LPA5 and LPA6 were quantified in the kidney of SHAM- andSNx-operatedmice.

Values are mean ± SEM of 10 to 11 animals per group.

4.1.3 Conclusions and discussion

4.1.3.1 The main results and conclusions x We showed here that SNX is associated with a strong increase in the level of different species of LPA and LPC in urine that were significantly correlated with kidney fibrosis and albuminuria. These data indirectly indicate that LPA and LPC could be involved in the development of kidney fibrosis after sub-total nephrectomy. x We also observed that the renal expression of the enzymes involved in LPA synthesis (AGK, ATX and PLA1) were significantly reduced by SNX. Expression of the LPA degrading enzymes (LPP1, LPP2 and LPP3) was also decreased. This indicates that in course of kidney fibrosis LPA metabolism is changed in kidneys, and decreased local degradation of LPA could be responsible for its increased availibility.

4.1.3.2 Discussion

4.1.3.2.1 What is the origin of urinary LPA?

LPA is found in plasma where it circulates bound to albumin and other carriers. While plasma LPA is mainly produced by ATX, the origin of urinary LPA remains poorly understood. As for other compounds, the presence of LPA in urine can result either from its excretion or from its local synthesis in kidney. In our study we found that increased urine LPA level in SNX mice is positively correlated with increased albuminuria. This can indicate that increased urine LPA can at least partially result from its co-excretion with albumin. In parallel we found no changes in plasma LPA levels. The increased urinary LPA without changes in its plasma concentration let us to think that increased urinary LPA could also result from changes in kidney LPA metabolism. To test this hypothesis we measured the kidney expression of different enzymes, potentially involved in LPA production. As discussed in the article decreased kidney expression of enzymes involved in LPA synthesis (ATX, MAGK, PLA1 and PLA2g4) indicates that these enzymes are likely not responsible for increased urinary LPA after subtotal nephrectomy. In opposite, decreased expression of three isoforms of LPA catabolic enzymes LPPs suggests that increased urine LPA could be the result of its decreased degradation in remnant kidneys. Interestingly, data from our team showed that LPP1 and LPP3 were also down-regulated in kidneys upon obstruction-induced fibrosis (Jean-Philippe Pradère, unpublished data). Our hypothesis that LPA production is increased from kidneys

77 after sub-total nephrectomy now requires further verification. One approach could be to measure ex vivo production of LPA levels by explants prepared from the remnant kidneys. Decreased expression of LPP1 and LPP3 was described in several types of cancer and overexpression of LPP1 and/or LPP3 in ovarian and breast cancer cells increases LPA hydrolysis, inhibits cell proliferation and colony-forming activity, increases apoptosis, reduces LPA- induced cell migration [123, 124, 179]. Regulation of enzyme expression in biological systems with prescribed compounds seems rather difficult approach however it was shown that gonadotropin-releasing hormone analogs increase LPP expression in ovarian cancer cells [336]. Therefore it could be interesting to test whether the use of these compounds could increase LPPs expression in kidney cells. If it is a case it could be interesting to test whether increase in LPPs expression in vivo will change urinary LPA concentrations and the development of fibrosis in our model of CKD.

In parallel to increased urinary LPA and LPC, their plasma levels were not changed upon subtotal nephrectomy. Plasma levels of lysophospholipids were not changed in UUO-induced kidney fibrosis [173] or in patients with DN as well (accepted manuscript). This finding is in contradiction with a recent report, where metabolomic analysis of plasma samples from rats with two different models of CKD and patients with various forms of CKD revealed increased amounts of two molecular species of LPA (16:0 and 18:2) that were proposed as biomarkers for the discrimination of CKD patients from healthy subjects [301]. How could we explain that discrepancy of results from various studies regarding the association between CKD and plasma lysophospholipids? Plasma samples of the same species (human and rat) were analyzed in experiments with contradictory results regarding plasma levels of lysophospholipids. Moreover the same animal model of CKD (5/6 nephrectomy) was used. This can indicate that species-dependent and model-dependent factors could not account for the observed divergence of results. However, this disagreement of results remains unexplained. The increased urinary lysophospholipids without any changes in plasma level suggests that changes in urinary LPA and LPC levels are likely urine-specific. Therefore it can be proposed to test the different LPA and LPC species in urine as biomarkers able to discriminate CKD patients from healthy subjects. Future large-scale experiments with involvement of patients with different forms of CKDs and various animal models of the disease are required for the validation of urine LPA and LPC species as efficient biomarkers for CKD.

4.1.3.2.2 LPA receptors in subtotal nephrectomy-associated kidney fibrosis

In parallel to positive correlation with the level of albuminuria, SNX-induced increased urine LPA levels are positively correlated with the intensity of TIF. Since LPA was previously shown as a fibrosis-promoting factor in kidney [164, 165, 324] we propose hereby that it could also be responsible for fibrosis development in SNX. Nevertheless, we measured mRNA expression changes of LPA receptors and found a decreased expression of LPA1-4 and no changes in LPA5 and LPA6 expression in kidney from SNX mice. This could indicate that 5 months after subtotal nephrectomy sensitivity of kidney cells to LPA is not increased suggesting that fibrosis likely results from increased LPA levels. Although, knowledge about regulation of LPA receptors is scarce, it was demonstrated that they are prone to agonist- induced phosphorylation followed by receptor internalization and receptor desensitization [337]. Other biological active molecules such as angiotensin II and EGF also could desensitize LPA receptors [338, 339]. Therefore we hypothesize that down-regulation of LPA receptors in remnant kidneys could reflect the desensitization processes induced by increased levels of LPA or other compounds that increase in fibrosis. Future studies are now required to test this hypothesis and the first step could be the study of LPA receptor expression in early stages of kidney fibrosis and its changes during disease progression. Since genetic ablation and pharmacological blockade of LPA1 receptor has been shown previously to attenuate kidney fibrosis [164, 165] it could be of interest to study sub-total nephrectomy-induced kidney fibrosis in LPA1-/- animals and/or in animals treated with LPA1 receptor antagonist. Promotion of kidney fibrosis through LPA2 receptor signaling has also been proposed in vitro [165, 309]. Since specific LPA2 receptor antagonists are not available, the use of LPA2-/- could be an alternative strategy to test in vivo the LPA2 receptor involvement in kidney fibrosis. We found no changes in LPA5 and LPA6 expression whereas they were up-regulated in UUO-induced kidney fibrosis. Whether LPA5 and LPA6 play a role in kidney fibrosis is not known and could be interesting to test in future experiments.

4.1.3.2.3 LPC in kidney fibrosis

We demonstrated in our study that 5 months after subtotal nephrectomy, urinary LPC increases in parallel to LPA and that increased LPC levels are correlated with increased albuminuria. As for LPA, this could suggest that increased urine LPC could at least partially result from its coexcretion with albumin. However, we demonstrated here that 2 enzymes

(sPLA2g2D and sPLA2g2E) potentially involved in LPC production are up-regulated suggesting that LPC production from phosphatidylcholine (PC) by the action of PLA2 might contribute to its increased urine levels. This hypothesis is in agreement with a recent report where decreased PC levels in urine are described in parallel to increased urinary LPC in patients with focal-segmental glomerulosclerosis (FSGS) indirectly indicating that PLA2 dependant degradation of PC indeed could contribute to increase in urinary LPC [340]. The hypothesis could be verified with the use of specific inhibitors of sPLA2 (LY315920) in mice with subtotal nephrectomy or other models of CKD. In addition, increased LPC levels are described in kidney glomerules of animals with DN and FSGS and in urine accumulated in kidneys after obstruction [173, 311, 341], however the biological role of increased LPC in kidneys and urine is poorly understood. We described here that increased urinary LPC correlates positively with TIF as well suggesting that LPC could be involved in development of kidney fibrosis. The underlying mechanisms of possible LPC-induced fibrosis are not well known. In vitro experiments demonstrated that LPC induces expression of adhesion molecules and chemokines in glomerular endothelial and mesangial cells and could lead to macrophage infiltration and glomerulosclerosis [341, 342]. In vivo experiments describing the role of LPC in kidney fibrosis are not available and the lack of specific LPC receptors is a limiting factor in this field of lysophospholipid research.

4.2 LPA and AKI

4.2.1 Bibliographic context and objectives

Previous reports demonstrated that treatment with exogenous LPA could protect against I/R- induced damage of kidneys and heart [330, 331, 343]. In an apparent contradiction another report demonstrated that blockade of LPA receptors with the LPA1/3 antagonist protects against I/R-induced AKI [332]. These differences of the results indicate that more investigations are required to clarify the role of LPA in AKI.

In parallel to these reports LPA was demonstrated to increase animal survival upon challenge with endotoxin and to protect against sepsis-induced damage of lung, liver and neuromuscular junction [262, 263, 344]. Since sepsis is one of the major causes of AKI, we were interested here whether LPA could protect against sepsis-induced AKI. LPS administration is one form of sepsis-induced AKI model. Such models are usually applied in small rodents and are well standardized. LPS triggers a local cell response in proximal tubular epithelial cells through binding to TLR4 receptors, which, together with similar responses elicited in extrarenal cells, promotes the expression of pro-inflammatory cytokines responsible for renal cell damage [345, 346]. Additionally, the production of reactive oxygen and nitrogen species as well as NO release causes hemodynamic alterations that also contribute to AKI [346].

The objectives of the second part of the thesis were to

1) Study whether LPS-induced AKI in mice could be counteracted by exogenous LPA. 2) Test whether LPS treatment could influence the local renal metabolism of LPA.

4.2.2 Results

Study of the LPA involvement in LPS-induced AKI and changes of its local metabolism.

Article 2:

Lysophosphatidic acid protects against endotoxin-induced acute kidney injury

Koryun Mirzoyan, Colette Denis, Audrey Casemayou, Stanislas Faguer, Marion Gilet, Dimitri

Marsal, Jean-Loup Bascands, JoostP. Schanstra, Jean-Sébastien Saulnier-Blache

Accepted to Inflammation (DOI: 10.1007/s10753-017-0612-7)

Lysophosphatidic acid protects against endotoxin-induced acute kidney injury

Koryun Mirzoyan1,2, Colette Denis1,2, Audrey Casemayou1,2, Marion Gilet1,2, Dimitri

Marsal1,2, Dominique Goudounéche 3, Stanislas Faguer1,2, Jean-Loup Bascands4, Joost P.

Schanstra1,2#, Jean-Sébastien Saulnier-Blache1,2##.

1 Institut National de la Santé et de la Recherche Médicale (INSERM), U1048, Institut of

Cardiovascular and Metabolic Disease.

2 Université Toulouse III Paul-Sabatier Toulouse, France.

3Centre de Microscopie Electronique Appliquée à la Biologie (CMEAB), Faculté de

Médecine Rangueil, University of Toulouse, France.

4Institut National de la Santé et de la Recherche Médicale (INSERM), U1188 - Université de La Réunion, France.

## Corresponding author: [email protected]

# Co-corresponding author: [email protected]

Abstract

Septic shock is the most common cause of acute kidney injury (AKI) but the underlying mechanisms remain unclear and no targeted therapies exist. Lysophosphatidic acid (LPA) is a bioactive lipid which in vivo administration was reported to mitigate inflammation and injuries caused by bacterial endotoxemia in liver and lung. The objective of the present study was to determine whether LPA can protect against sepsis-associated AKI. C57BL/6 mice were treated with LPA 18:1 (5mg/kg, i.p.) 1 hour before being injected with the endotoxin lipopolysaccharide (LPS) and AKI was evaluated after 24h. LPA significantly decreased the elevation of plasma urea and creatinine caused by LPS. In the kidney, LPA pre-treatment significantly reduced the up-regulation of inflammatory cytokines (IL-6, TNFα, MCP-1), and completely prevented down-regulation of peroxisome proliferator-activated receptor gamma coactivator 1-alpha and up-regulation of heme oxygenase-1 caused by LPS. LPA also prevented LPS-mediated alterations of the renal mitochondrial ultrastructure. In vitro, pretreatment with LPA 18:1 significantly attenuated LPS-induced up-regulation of the inflammatory cytokines (TNFα and MCP-1) in RAW264 macrophages. Moreover, in vivo LPS-treatment lowered urinary LPA concentration and reduced LPA anabolic enzymes (autotaxin and acylglycerol kinase), and increased the LPA catalytic enzyme (lipid phosphate phosphatase 2) expression in the kidney cortex. In conclusion, exogenous LPA exerted a protective action against renal inflammation and injuries caused by bacterial endotoxemia. Moreover, LPS reduces the renal production of LPA suggesting that sepsis-associated AKI could be mediated, at least in part, by alleviation of the protective action of endogenous LPA.

Key words: endotoxemia, kidney, lysophosphatidic acid, macrophages

Introduction

Acute kidney injury (AKI) describes a multifactorial and complex clinical syndrome characterized by an acute decrease of renal function with accumulation of waste products and electrolyte imbalance [1]. AKI is associated with increased mortality and morbidity during the acute phase as well as in the long term. The incidence of AKI after general surgery is nearly 1% whereas it can reach up to 70% in critically ill patients. So far no targeted therapies exist against AKI and all interventions are not specific. One of the most common causal factors of AKI in hospital intensive care units is septic shock. AKI occurs nearly in half of patients with sepsis and increases mortality six to eight times [2 , 3].

Sepsis-induced AKI is often associated with renal hypoperfusion and ischemia but can also occur in the absence of major hemodynamic alterations [4 , 5 , 6]. The pathogenic mechanisms of sepsis-induced AKI involve renal microcirculatory changes, inflammation, endothelial dysfunction, coagulation disturbances and adaptive cell responses such as downregulating metabolism or cell cycle arrest [7].Sepsis-induced AKI is usually associated with lower rate of tubular apoptosis and necrosis than other forms of AKI linked to a reduction of oxygen consumption resulting from a dedifferentiation process associated with mitochondrial rarefaction that would shift the cells towards an energy-saving state [7]. A central mediator of renal resistance to AKI is PGC1α (peroxisome proliferator activated receptor gamma coactivator-1-alpha) a master regulator of mitochondrial biogenesis and oxidative metabolism that contribute to renal functional recovery following kidney aggression [8 , 9].

Lysophosphatidic acid (LPA) is a lysophospholipid present in various biological fluids that can exert several biological activities through the activation of specific G-protein coupled receptors (LPA1 to LPA6) [10 , 11]. LPA is present in different physiological fluids including plasma and urine. The abundance of urinary LPA increases in kidney disease models including unilateral ureteral obstruction and 5/6 nephrectomy, and is correlated with renal fibrosis [12, 13]. LPA is synthesized by enzymes including lysophospholipase D (also named autotaxin), phospholipase A1 and A2 or acylglycerolkinase [10] and inactivated by dephosphorylation by lipid-phosphate phosphatases (LPP1, LPP2, LPP3) [14].

Previous reports have shown that AKI generated by ischemia-reperfusion of the kidney could be prevented by injecting LPA to the animals before surgery [15 , 16 , 17]. It was also shown that administration of LPA to mice exposed to severe endotoxemia or infection increases their survival and reduces organ damage other than the kidney (liver, neuromuscular junction) [18 ,

19 , 20 , 21]. Moreover, exposure to endotoxin or ischemia was shown to alter LPA production in plasma and bronchoalveolar lavage fluid [21 , 22 , 23 , 24 , 25 , 26 , 27].

The objective of the present study was to assess whether LPA could exert a protective action on sepsis-induced AKI in vivo and in vitro.

Methods

Drugs. LPA18:1, LPA16:0 and LPA18:0 (Avanti Polar, GOGER, Paris, France) and LPS (LPS-EB Ultrapure, Invivogen, San Diego, USA) were solubilized in PBS supplemented with 1% fatty acid free bovine serum albumin (BSA). This solution without the drugs was as control (vehicle, VH).

Animals. Male C57BL6 mice (10 weeks old) were used for experiments that were purchased from Janvier (Le Genest Saint Isle, France) and housed in pathogen-free, temperature- controlled environment with a 12-hour/12-hour light/ dark cycle. Animals had free access to food and tap water. All experiments were conducted in accordance with the Guide for the care and use of laboratory animals from the National Institute of Health and were approved by a local ethics committee (CEEA#122). Mice were injected with LPA 18:1 (5mg/kg, i.p.) 1 hour before i.p. injection of LPS at the dose of 10mg/kg (protocol 1) or 3 mg/kg (protocol 2). Control animals received 100μl of VH (VH group).

In protocol 1, the mice were euthanized by pentobarbital injection 24 hours after LPS injection. Blood was drawn by intra-cardiac puncture and collected in heparinized tubes and stored on ice (Microvette Hép-Lithium, Sarstedt, Marnay, France). Ice-cold blood was then centrifuged for 5 minutes at 1500xg and upper plasma phase was collected and immediately frozen at -80°C. Kidney cortex (upper and lower poles) was dissected out. One part of kidney cortex was immediately immersed in liquid nitrogen and then stored at -80°C for future RNA extraction and RT-PCR analysis. Another part of kidney cortex was dedicated to transmission electron microscopy.

In protocol 2, treated mice were placed into metabolic cages to collect urine after 9 and 24h, urine was centrifuged for 5 minutes at 3000xg and stored at -20°C for LPA quantification. The animals were then euthanized by pentobarbital injection to collect blood samples and kidneys as described in protocol 1.

Transmission electron microscopy. Kidney cortex was fixed with 2% glutaraldehyde in Sorensen phosphate buffer (0.1 M, pH 7.4) for 1 hour, washed with the Sorensen phosphate buffer for 12 hours and post fixed with 1% OsO4 in Sorensen phosphate buffer for 1 hour. Samples were dehydrated with increasing ethanol concentration (30%, 50%, 70%, 95%) until a 100% ethanol solution was reached and then subsequently stored for 30 min in a propylene oxide solution before being embedded in epoxy resin. After 24 h of polymerization at 60°C, ultrathin sections (70 nm) were mounted on 100 mesh collodion-coated copper grids and poststained with 3% uranyl acetate in 50% ethanol and with 8.5% lead citrate before being examined on a HT7700 Hitachi electron microscope at an accelerating voltage of 80 KV. A series of 10 photographs were taken at x7000 magnification.

Measurement of plasma urea and creatinine concentration. Urea and creatinine levels were measured by colorimetric tests on a Pentra 400 analyzer (Horiba Medical, Grabels, France).

LC-MS/MS quantification of LPA species in plasma and urine. LPA species were extracted from urine (150 μl) and plasma (10 μl) using 1-butanol containing 50 ng of LPA 17:0 as internal standard as previously described [13]. Detection was conducted by LC- MS/MS using an Agilent 1290 Infinity HPLC system (Agilent Technologies) coupled on-line to an Agilent 6460 triple quadrupole MS (Agilent Technologies) equipped with electrospray ionization source. Quantification was conducted through comparison with calibration curves of synthetic lipid standards: LPA 16:0, LPA 18:0, LPA 18:1, LPA 18:2, LPA 20:0, LPA 20:4, LPA 22:0, LPA 24:0. More details on the method and its validation are provided in [13].

(Molsheim, France). Cells were cultured in a humidified atmosphere of 5%CO2-95% O2 air at 37°C. RAW264.7 cells were plated at a density of 1 x 105 cells/cm2 in 12 well plates and grown to 70-80% confluence (about 48h) in DMEM supplemented with 50 U/mL Penicillin, 50 mg/mL Streptomycin and 10% heat-inactivated FBS. The cells were then serum-deprived for 24 h hours before being pretreated with 10 μM of LPA18:1 for 15 minutes before being exposed to LPS (100ng/ml) or VH. After 4 hours, the medium was collected to quantify TNFαconcentration, and the cells were washed once with PBS and total RNA was extracted and analyzed.

Quantification of TNFα. The concentration of TNFα in RAW264.7 cell culture medium (1/10 dilution) was measured using the CisbioBioassays’s mouse TNFα kit (Cisbio International, Bagnols/Ceze, France).

RNA analysis. Total RNA was isolated from frozen kidney or cells using Qiagen RNEasy Plus Mini Kit (Qiagen, Valencia, CA) in accordance with manufacturer protocols. Reverse transcription and PCR amplification were performed on a FlexCycler2 (Analytik Jena Ag, Jena Germany) and StepOne Plus Real-Time PCR systems (Thermofisher,Asnières, France) respectively as previously described [29]. Primer sequences are listed in Table 1. The glyceraldehydes 3-phosphate dehydrogenase (GAPDH) mRNA was used as a reference gene. Normalization of gene expression was calculated according to the following formula: 2(CtGAPDH-Ctgene of interest) where Ct corresponds to the number of cycles needed to generate a fluorescent signal above a predefined threshold.

Mouse gêne Forward Reverse

Acylglcerol kinase (Agk) GTGTTTGGCAACCAGCTCATT GCTGCGGGATTGAGAAAGAC

Autotaxin (Atx) TCCGTGCATCGTACATGAAGA CAGGACCGCAGTTTCTCAATG

Glyceraldehyde 3-phosphate dehydrogenase (Gapdh) TGCACCACCAACTGCTTAGC GGCATGGACTGTGGTCATGAG

Hemeoxygenase 1(Ho-1) AGAAGGGTCAGGTGTCCAGA TGCCCTCTATCTCCTCTTCC

Interleukin 6(Il6) GCCCACCAAGAACGATAGTCA CAAGAAGGCAACTGGATGGAA

Lipid phosphate phosphatase-1(Lpp1) GGGAGACTGGGCAAGACTCTT CACTCGAGAAAGGCCCACAT

Lipid phosphate phosphatase-2(Lpp2) CGCGATCCAACTTCAACAACT CAGCCCCGAACAGAAAGGT

Llipid phosphate phosphatase-3(Lpp3) CCATCCTGGCGATCATTACAG AAAGGAAGCATCCCACTTGCT

Monocyte chemotacticprotein 1 (Mcp1) GCAGTTAACGCCCCACTCA CCAGCCTACTCATTGGGATCA

PPARγcoactivator 1α(Pgc1α) AAAGGATGCGCTCTCGTTCA GGAATATGGTGATCGGGAACA

Tumor necrosis factor α (Tnfα) ATGGCCTCCCTCTCATCAGT CACTTGGTGGTTTGCTACGA

Table 1. Sequences of the primers used in RT-PCR analysis.

Statistical Analysis. Continuous variables were presented as mean ± SEM. Comparisons were made with unpaired Student’s t test where 2 groups of variables were compared and with analysis of variance (ANOVA) where more than 2 groups of variables were compared. A P value <0.05 was considered statistically significant: *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Results

For the sepsis-induced AKI we employed for the major part of the study a classical LPS- induction protocol using 10 mg/kg of LPS (protocol 1) [30]. However, most mice treated with 10 mg/kg of LPS did not survive in metabolic cages necessary for the collection of urine, most likely due to the additional stress imposed by the cages. Hence for the collection of urine a lower dose (3 mg/kg) of LPS was used to induce AKI (protocol 2).

LPA injection counteracts the deleterious effects of LPS on AKI. Using protocol 1 to induce AKI, LPS-treated mice showed a significant increase of plasma urea and plasma creatinine, when compared to control mice (VH) and LPA pretreatment significantly attenuated LPS effects (Figure 1A and B).

Figure 1. Influence of in vivo LPA and LPS treatments on plasma urea and creatinine. Mice were injected with LPA 18:1 (5mg/kg, i.p.) 1 hour before injecting LPS (10 mg/kg, i.p.). Control animals received 100μl of vehicle (VH). Blood was collected after 24h to measure plasma urea (A) and plasma creatinine (B). The number of animal per group was 9, 9, 9 and 10 for VH, LPA, LPS and LPA/LPS respectively. Values are means ± SEM. ANOVAanalysis: *p<0.05, **p<0.01 and ***p<0.001.

LPS treatment significantly elevated the renal expression of the inflammatory cytokine genes IL-6, TNFα and MCP-1 and this was significantly attenuated by LPA (Figure 2A-C).

Figure 2. Influence of in vivo LPA and LPS treatments on renal expression of proinflammatory cytokines. Mice were injected with LPA 18:1 (5mg/kg, i.p.) 1 hour before injecting LPS (10 mg/kg, i.p.). Control animals received 100μl of vehicle (VH). Kidney cortex was collected after 24h to measure the gene expression of IL-6 (A), TNFa (B) and MCP-1 (C). The number of animal per group was 9, 9, 9 and 10 for VH, LPA, LPS and LPA/LPS respectively. Values are means ± SEM. ANOVAanalysis: *p<0.05, **p<0.01 and ***p<0.001.

LPS treatment significantly reduced the renal expression of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC1-α) involved in mitochondriogenesis (Figure 3A). LPS also significantly increased the renal expression of heme oxygenase-1 (HO-1) involved in oxidative stress (Figure 3B). These effects were significantly attenuated by LPA (Figure 3Aand B). Electron microscopy analysis showed that renal mitochondria from LPS-treated mice appeared swollen and vacuolated with a decrease in the number of dense granules when compared to control mice (Figure 3C).Those granules were proposed to correspond a local accumulation of cations inside mitochondria [31]. LPA almost completely restored the alterations of mitochondria ultrastucture generated by LPS (Figure 3C).

A C VH LPA

LPS LPA/LPS B

Figure 3. Influence of in vivo LPA and LPS treatments on renal mitochondria genes and structure. Mice were injected with LPA 18:1 (5mg/kg, i.p.) 1 hour before injecting LPS (10 mg/kg, i.p.). Control animals received 100μl of vehicle (VH). Kidney cortex was collected after 24h to measure the gene expression of PGC1a (A) and HO-1 (B). The number of animal per group was 9, 9, 9 and 10 for VH, LPA, LPS and LPA/LPS respectively. Values are means ± SEM. ANOVA analysis: *p<0.05, **p<0.01 and ***p<0.001. (C) Transmission electron microscopy of kidney slices (magnification x7000). Representative of 5 to 7 images from 1 animal per group (arrows indicate dense granules).

LPA displays direct anti-inflammatory properties on macrophages in vitro. To assess whether the anti-inflammatory effect of LPA could result from its direct action on macrophages, in vitro experiments were performed on the murine macrophage cell line RAW264.7. Exposition of those cells to LPS led to a massive release of TNFα in the culture medium (Figure 4A) and a massive increase in the expression of the inflammatory cytokine genes TNFα(Figure 4B), IL-6 (Figure C), and MCP-1 (Figure 4D). While pre-treatment with LPA 18:1 exerted no effect by itself, it significantly reduced LPS-induced TNFα release (Figure 4A) and gene expression (Figure 4B). LPA also reduced LPS-induced MCP-1 (Figure 4D) but had no significant influence on LPS-induced IL-6 gene expression (Figure 4B). The most abundant LPA receptor subtypes expressed in RAW262.7 cells were LPA2 and LPA5 (data not shown).

Figure 4. Influence of in vitro LPA and LPS treatments on macrophage cytokines. RAW264.7 marcrophages in culture were serum-deprived for 24h before being pretreated with LPA18:1 (10 μM) for 15 minutes and with LPS (100ng/ml) or vehicle (VH). After 4 hours, the release of TNFα in the culture medium was quantified (A), and total RNA was extracted from the cells to measure the expression of TNFα (B), IL-6 (C), and MCP-1 (D). Values are means ± SEM of 3 (A) and 4 (B, C, D) separate experiments. ANOVA analysis: **p<0.01 ,***p<0.001 and ****p<0.0001.

LPS reduces endogenous LPA production. Our next objective was to determine whether LPS could influence endogenous LPA production. For that LPA was quantified in urine and plasma using LC-MS/MS. For this part of the study protocol 2 was employed using 3 mg/kg of LPS to induce AKI [32]. The induction of AKI by this low dose was maintained (increased BUN and plasma creatinine, data not shown) and allowed maintaining the mice alive in metabolic cage. In these conditions, the urine concentration of LPA was significantly reduced (-47%) 9 hours after LPS injection when compared to VH (Figure 5A). LPS-induced reduction of urine LPA was further amplified (-67%) after 24h (Figure 5B). LPS-induced changes in urinary LPA mainly concerned LPA16:0, LPA18:1 and LPA18:2 (Figure 5C).

A B 9h 24h

Figure 5. Influence of LPS treatment on urine LPA content. Mice were injected with LPS (3 mg/kg, i.p.) (n=8). Control animals received 100μl of vehicle (VH) (n=8). The mice were placed into metabolic cages to collect urine after 9 and 24h. LPA was quantified in urine by LC/MS-MS. Concentration of total LPA (sum of LPA species) after 9 h (A) and 24h (B). (C) concentration of the different LPA species from LPS-treated and control (VH) mice. Values are means ± SEM. Unpaired student’s t-test: *p<0.05.

In contrast to urine, plasma LPA concentrations were not significantly modified by LPS injection (Figure 6).

Figure 6. Influence of LPS treatment on plasma LPA content. Mice were injected with LPS (3 mg/kg, i.p.) (n=8). Control animals received 100μl of vehicle (VH) (n=8). Blood was collected after 24h and plasma was prepared. LPA was quantified in plasma by LC/MS-MS. (A) total LPA (sum of LPA species) concentration. (B) concentration of the different LPA species from LPS-treated and control (VH) mice. Values are means ± SEM. Unpaired student’s t-test.

LPS alters the renal expression of renal LPA metabolic enzymes. To assess whether LPS-mediated reduction of urine LPA could result from an alteration of renal LPA metabolism, the expression of the enzymes involved in LPA metabolism was measured in kidney cortex. LPS-treatment significantly reduced the expression of 2 anabolic enzymes, ATX and AGK (-22 and -23% respectively) when compared to control (Table 2). In parallel, LPS significantly increased (272%) the expression of the catabolic enzyme LPP2, while LPP1 and LPP3 subtypes remained unchanged (Table 2).

Gene expression Fold change over mean of VH VH LPS Genes Mean SEM n Mean SEM n P value Atx 1,00 ± 0.05 8 0,78 ± 0.07 8 * Agk 1,00 ± 0,08 8 0,77± 0,06 8 * Lpp1 1,00 ± 0,09 8 1,23± 0,12 8 ns Lpp2 1,00 ± 0,09 8 2,72± 0,28 8 **** Lpp3 1,00 ± 0,14 8 1,43± 0,20 8 ns

Table2. Influence of in vivo LPS treatment on renal expression of LPA metabolic enzymes. Mice were injected with LPS (3 mg/kg, i.p.) (n=8). Control animals received 100μl of vehicle (VH) (n=8). Kidney cortex was collected after 24h to measure the gene expression of LPA metabolic enzymes. Values are means ± SEM. Unpaired student’s t-test: *p<0.05, ****p<0.0001.

Discussion

Previous reports have shown that AKI generated by ischemia-reperfusion could be prevented by administration of exogenous LPA before surgery [15 , 16 , 17]. LPA was also shown to reduce the injury caused by LPS to liver and lung [18 , 19 , 20 , 21]. In the present work, we observed that LPA was able to mitigate several responses characteristic of LPS-induced AKI such as elevation of plasma urea and creatinine, up-regulation of renal cytokine expression, down-regulation of renal PGC1a, up-regulation of renal HO-1 and alteration of renal mitochondrial ultrastructure. From these observations, we can conclude that the protective effect of LPA against AKI is not limited to ischemia-reperfusion and can be extended to sepsis.

Inflammation is a common feature of AKI whether it is induced by ischemia-reperfusion or sepsis, and its consequence is the propagation of renal damage. Indeed, the burst of pro- inflammatory cytokines during the initial phase of sepsis stimulates renal cells (resident leucocytes, endothelial, and epithelial cells) and lead to platelet activation, microvascular dysfunction, hypoxia, and tissue damage [33 , 34 , 35]. Our results showed that LPA counteracts the pro-inflammatory action of LPS both in vivo in kidney and in vitro in cultured macrophages. Those results are in agreement with previous report showing that LPA inhibits bacterial endotoxin-shock through an anti-inflammatory signaling pathway in other organs [18]. Our results are also in agreement with previous reports showing that LPA is an important contributor of the anti-inflammatory action of albumin on renal tubular cells and macrophages [36]. Therefore, the protective activity of LPA against LPS-induced AKI could be mediated, at least in part, by its anti-inflammatory activity on macrophages. Nevertheless, we cannot exclude that LPA could also influence other hallmarks of sepsis-induced AKI such as macro- and micro-vascular dysfunctions, oxidative stress or mitochondrial defects [37]. Indeed, LPA was shown to increase endothelial resistance [38 , 39], a crucial process in vascular dysfunction. LPA was also reported to regulate mitochondrial fragmentation and fusion [40]. Interestingly, our work shows that LPA injection counteracts the swelling and vacuolization of mitochondria, and mitigates the down-regulation of PGC1a and the up- regulation of HO-1 generated by LPS. Therefore, preservation of mitochondrial function and dynamics could also play a role in the protective effect of LPA against LPS-induced AKI.

Our results show that LPS-treatment led to a reduction of LPA concentration in urine. To our knowledge, this has not been reported previously. The presence of a metabolite in urine results either from its excretion from plasma, or from its local production in kidney. In contrast to previous reports [21 , 27], we found no influence of LPS-induced AKI on the plasma LPA concentration. Moreover, we found that LPS-treatment altered the renal expression of several enzymes involved in LPA metabolism, suggesting that kidney could participate to the production of LPA in urine. Moreover, in view of the variation in the expression of these enzymes, the decrease in renal production of LPA result from a decrease in its synthesis by ATX or AGK and an increase in its degradation by LPP2.

Our work shows that administration of exogenous LPA protects against LPS-induced AKI and that LPS-treatment reduces the level of endogenous LPA in urine as a possible reflection of a reduced production of LPA in kidney. We propose that basal renal production of LPA could exert a protective activity on kidney that could be alleviated during sepsis, and thereby

95 favoring AKI development. If this hypothesis is correct therapeutic strategies consisting in increasing renal LPA production could help to prevent sepsis-induced AKI.

Conflict of interest. The authors declare no conflict of interest

Funding & Acknowledgments. This work was supported by grants from INSERM. Koryun Mirzoyan is supported by a PhD grant from Europe (ERASMUS MUNDUS, MEDEA). Justine Bertrand-Michel and Aude Dupuy from the Toulouse’s Lipidomic Core Facility (www.metatoul.fr) are gratefully acknowledged for carrying out LC-MS/MS quantification of the LPA in urine and plasma. We also thank Alexandre Lucas from the Protein Cellular Analysis facility (www.i2mc.inserm.fr/index.php/en/technical-facilities/protein-cellular- analysis-pca) for the quantification of cytokines. All authors participated in the conception and design, or analysis and interpretation of the data, contributed to drafting and revising the manuscript, and gave final approval of the version to be published.

References.

1. Makris, K. and L. Spanou, Acute Kidney Injury: Definition, Pathophysiology and Clinical Phenotypes. Clin Biochem Rev, 2016. 37(2): p. 85-98. 2. Uchino, S., et al., Acute renal failure in critically ill patients: a multinational, multicenter study. JAMA, 2005. 294(7): p. 813-8. 3. Gomez, H. and J.A. Kellum, Sepsis-induced acute kidney injury. Curr Opin Crit Care, 2016. 22(6): p. 546-553. 4. Schrier, R.W. and W. Wang, Acute renal failure and sepsis. N Engl J Med, 2004. 351(2): p. 159-69. 5. Murugan, R., et al., Acute kidney injury in non-severe pneumonia is associated with an increased immune response and lower survival. Kidney Int, 2010. 77(6): p. 527-35. 6. Zarbock, A., H. Gomez, and J.A. Kellum, Sepsis-induced acute kidney injury revisited: pathophysiology, prevention and future therapies. Curr Opin Crit Care, 2014. 20(6): p. 588-95. 7. Dellepiane, S., M. Marengo, and V. Cantaluppi, Detrimental cross-talk between sepsis and acute kidney injury: new pathogenic mechanisms, early biomarkers and targeted therapies. Crit Care, 2016. 20: p. 61. 8. Tran, M.T., et al., PGC1alpha drives NAD biosynthesis linking oxidative metabolism to renal protection. Nature, 2016. 531(7595): p. 528-32. 9. Tran, M. and S.M. Parikh, Mitochondrial biogenesis in the acutely injured kidney. Nephron Clin Pract, 2014. 127(1-4): p. 42-5. 10. Aikawa, S., et al., Lysophosphatidic acid as a lipid mediator with multiple biological actions. J Biochem, 2015. 157(2): p. 81-9. 11. Riaz, A., Y. Huang, and S. Johansson, G-Protein-Coupled Lysophosphatidic Acid Receptors and Their Regulation of AKT Signaling. Int J Mol Sci, 2016. 17(2): p. 215. 12. Tsutsumi, T., et al., Presence of bioactive lysophosphatidic acid in renal effluent of rats with unilateral ureteral obstruction. Life Sci, 2011. 89(5-6): p. 195-203. 13. Mirzoyan, K., et al., Increased urinary lysophosphatidic acid in mouse with subtotal nephrectomy: potential involvement in chronic kidney disease. J Physiol Biochem, 2016. 72(4): p. 803-812. 14. Cholia, R.P., et al., Understanding the Multifaceted Role of Ectonucleotide Pyrophosphatase/Phosphodiesterase 2 (ENPP2) and its Altered Behaviour in Human Diseases. Curr Mol Med, 2015. 15(10): p. 932-43.

15. de Vries, B., et al., Lysophosphatidic acid prevents renal ischemia-reperfusion injury by inhibition of apoptosis and complement activation. Am J Pathol, 2003. 163(1): p. 47-56. 16. Okusa, M.D., et al., Selective blockade of lysophosphatidic acid LPA3 receptors reduces murine renal ischemia-reperfusion injury. Am J Physiol Renal Physiol, 2003. 285(3): p. F565-74. 17. Gao, J., et al., Lysophosphatidic acid and lovastatin might protect kidney in renal I/R injury by downregulating MCP-1 in rat. Ren Fail, 2011. 33(8): p. 805-10. 18. Fan, H., et al., Lysophosphatidic acid inhibits bacterial endotoxin-induced pro- inflammatory response: potential anti-inflammatory signaling pathways. Mol Med, 2008. 14(7-8): p. 422-8. 19. Murch, O., M. Collin, and C. Thiemermann, Lysophosphatidic acid reduces the organ injury caused by endotoxemia-a role for G-protein-coupled receptors and peroxisome proliferator-activated receptor-gamma. Shock, 2007. 27(1): p. 48-54. 20. He, D., et al., Lysophosphatidic acid enhances pulmonary epithelial barrier integrity and protects endotoxin-induced epithelial barrier disruption and lung injury. J Biol Chem, 2009. 284(36): p. 24123-32. 21. Zhao, J., et al., Lysophosphatidic acid receptor 1 modulates lipopolysaccharide- induced inflammation in alveolar epithelial cells and murine lungs. Am J Physiol Lung Cell Mol Physiol, 2011. 301(4): p. L547-56. 22. Li, Z.G., et al., Influence of acetylsalicylate on plasma lysophosphatidic acid level in patients with ischemic cerebral vascular diseases. Neurol Res, 2008. 30(4): p. 366-9. 23. Zhao, J., et al., Autotaxin induces lung epithelial cell migration through lysoPLD activity-dependent and -independent pathways. Biochem J, 2011. 439(1): p. 45-55. 24. Dohi, T., et al., Increased circulating plasma lysophosphatidic acid in patients with acute coronary syndrome. Clin Chim Acta, 2012. 413(1-2): p. 207-12. 25. Dohi, T., et al., Increased lysophosphatidic acid levels in culprit coronary arteries of patients with acute coronary syndrome. Atherosclerosis, 2013. 229(1): p. 192-7. 26. Kurano, M., et al., Possible involvement of minor lysophospholipids in the increase in plasma lysophosphatidic acid in acute coronary syndrome. Arterioscler Thromb Vasc Biol, 2015. 35(2): p. 463-70. 27. Mouratis, M.A., et al., Autotaxin and Endotoxin-Induced Acute Lung Injury. PLoS One, 2015. 10(7): p. e0133619.

28. Raschke, W.C., et al., Functional macrophage cell lines transformed by Abelson leukemia virus. Cell, 1978. 15(1): p. 261-7. 29. Treguer, K., et al., Influence of secreted factors from human adipose tissue on glucose utilization and proinflammatory reaction. J Physiol Biochem, 2013. 30. Wu, L., et al., Peritubular capillary dysfunction and renal tubular epithelial cell stress following lipopolysaccharide administration in mice. Am J Physiol Renal Physiol, 2007. 292(1): p. F261-8. 31. Pasquali-Ronchetti, I., J.W. Greenawalt, and E. Carafoli, On the nature of the dense matrix granules of normal mitochondria. J Cell Biol, 1969. 40(2): p. 565-8. 32. Han, M., et al., Renal neutrophil gelatinase associated lipocalin expression in lipopolysaccharide-induced acute kidney injury in the rat. BMC Nephrol, 2012. 13: p. 25. 33. Jo, S.K., et al., Macrophages contribute to the initiation of ischaemic acute renal failure in rats. Nephrol Dial Transplant, 2006. 21(5): p. 1231-9. 34. Furuichi, K., et al., CCR2 signaling contributes to ischemia-reperfusion injury in kidney. J Am Soc Nephrol, 2003. 14(10): p. 2503-15. 35. Angus, D.C. and T. van der Poll, Severe sepsis and septic shock. N Engl J Med, 2013. 369(9): p. 840-51. 36. Iglesias, J., et al., Albumin is a major serum survival factor for renal tubular cells and macrophages through scavenging of ROS. Am J Physiol, 1999. 277(5 Pt 2): p. F711- 22. 37. Gomez, H., et al., A unified theory of sepsis-induced acute kidney injury: inflammation, microcirculatory dysfunction, bioenergetics, and the tubular cell adaptation to injury. Shock, 2014. 41(1): p. 3-11. 38. Minnear, F.L., et al., Platelet lipid(s) bound to albumin increases endothelial electrical resistance: mimicked by LPA. Am J Physiol Lung Cell Mol Physiol, 2001. 281(6): p. L1337-44. 39. Yin, F. and M.A. Watsky, LPA and S1P increase corneal epithelial and endothelial cell transcellular resistance. Invest Ophthalmol Vis Sci, 2005. 46(6): p. 1927-33. 40. Ohba, Y., et al., Mitochondria-type GPAT is required for mitochondrial fusion. EMBO J, 2013. 32(9): p. 1265-79.

100

4.2.3 Conclusions and discussion

4.2.3.1 The main results and conclusions x We demonstrated here that LPA pretreatment mitigates LPS-induced elevation of plasma urea and creatinine, up-regulation of renal expression of inflammatory cytokines and HO-1, down-regulation of renal PGC1α and ultrastructural changes of mitochondria in kidneys. Therefore LPA-induced protection in course of AKI is not limited to I/R and could be extended to endotoxemia. x LPA pretreatment reduced LPS-induced up-regulation of inflammatory cytokines in cultured macrophages indicating that targeting of these cells by LPA could be responsible for the overall anti-inflammatory action demonstrated in kidneys in course of endotoxemia- induced AKI. x We showed that urinary LPA is decreased in sepsis-induced AKI and this is associated with a down-regulation of the LPA producing enzymes ATX and AGK and an up-regulation of the LPA degrading enzyme LPP2 in kidney. This indicates that in course of endotoxemia, LPA availability in kidney could be decreased due to its’ altered metabolism, especially decreased production and/or increased degradation.

4.2.3.2 Discussion

4.2.3.2.1 How LPA protects against endotoxemia-induced AKI?

We found here that LPA is able to protect against endotoxemia-induced AKI indicating that previously observed LPA-exerted organ protection during sepsis could be extended to kidneys. However that protection requires to be verified in future experiments with the use of more specific functional tests of kidney function (ex: measurement of GFR). We observed that LPA blocks the LPS-induced up-regulation of inflammatory cytokines in kidney. Activated inflammatory reaction is a central event in the establishment and progression of endotoxemia-induced AKI. Therefore we hypothesized here that LPA-induced protection against endotoxemia-induced AKI could at least partially be explained by its anti- inflammatory action. The cells targeted by LPA are not known. Since macrophages play an important role in the induction and progression of the inflammatory reaction we tested whether LPA could directly block the LPS-induced activation of these cells. We found that LPA partially blocks LPS-induced up-regulation of inflammatory cytokines in RAW264.7

101 cell line. This result supports that the blockade of the pro-inflammatory action of LPS by LPA could participate in the protective action of LPA against endotoxemia-induced AKI. It could be interesting next to test whether LPA could also block the activation of primary macrophages from kidneys. For the next step it could be an interesting approach to test whether the LPA-induced blockade of macrophage activation is responsible for the protection exerted by LPA in course of LPS-induced AKI in vivo.

It was previously shown that LPA protects tubular epithelial cells from serum-deprivation- induced apoptosis and induces proliferation of these cells [305-307]. In addition, the tubular cells are activated by LPS and/or by pro-inflammatory cytokines in sepsis-induced AKI. Therefore tubular epithelial cells could be another cell target of LPA in AKI. The histological alterations of tubular epithelial cells upon sepsis-induced AKI are scarce. It was shown previously that vacuolization of proximal tubular epithelial cells upon septic injury reflects mostly the mitochondrial damage [37]. We found here that structural changes of mitochondria in kidneys of animals challenged with endotoxin are reversed by LPA pretreatment, suggesting that LPA could act on tubular cells and reverse LPS-induced changes. This hypothesis requires further verification through in vitro experiments on tubular epithelial cells in culture.

It was shown that sepsis triggers adaptive responses in tubular epithelial cells [36, 43]. One of the consequences of these responses are the metabolic changes required to preserve the source of cell energy. Mitochondria are the central elements of the regulation of this response and the PGC1α is one of the molecular switches that regulate the metabolic activity of tubular cells. We found that LPA pretreatment reverses LPS-induced down-regulation of PGC1α, which in combination with the protection of ultrastructural changes of mitochondria could indicate that LPA-induced protection of kidney function could involve changes in the mitochondrial dynamic.

Other cell types, such as endothelial cells, could also be involved in the LPA-induced kidney protection. This hypothesis is supported by a studies showing that LPA increases endothelial barrier resistance and decreases permeability of endothelial layer [347, 348]. Since endothelial dysfunction and transmigration of leukocytes is a crucial event in inflammatory reaction in sepsis-induced AKI, it would be interesting to test whether LPA increases the resistance of endothelial layer composed of the cells from kidney microvasculature. It could also be of interest to test whether LPA treatment decreases the permeability of capillaries in vivo.

102

4.2.3.2.2 Urinary LPA decreases in endotoxemia

We showed here for the first time that LPA concentration in urine decreases in LPS-induced AKI. In addition the plasma concentration of LPA remains unchanged. This last finding is in apparent contradiction with a recent study, where increased plasma LPA was found in animals with sepsis [264]. However in that study the cecal ligation and puncture was used in ICR mice to induce the sepsis. This difference in animal species and disease models could possibly be responsible for the divergence of results in LPA plasma concentration in these studies and requires further investigations. In addition to decreased LPA levels in urine we also found a down-regulation of the LPA producing ATX and AGK combined with an up-regulation of the LPA degrading LPP2 in kidney. These results suggest that changes in the local metabolism of LPA could at least partially be responsible for its decreased urinary levels. These hypothesis could be verified by measuring the ex vivo production of LPA by explants prepared from septic kidneys.

We found here that exogenous LPA could exert protective effects during endotoxemia- induced-AKI, whereas its endogenous production in kidneys is decreased upon endotoxemia. Based on these data we hypothesize that the basal production of LPA in kidney could exert a constitutive protection of kidney cells which is alleviated by endotoxemia. If this hypothesis is true it could be of particular interest to determine how to maintain renal LPA production. One possibility would be to increase its synthesis. Since ATX is an inducible enzyme and VEGF and retinoic acid could increase its expression [349, 350], it could be interesting to test whether ATX up-regulation in kidney cells could result in increased production of LPA. With the same logic it could be interesting to test whether the blockade of LPP2 could be effective in increasing LPA production from kidney cells. If these strategies are proven in vitro, next step could be to test their efficacy in vivo. There are reports showing that addition of LPC18:1 and LPA18:1 in the chow of animals increases tissue levels of LPA18:1, LPA18:2, LPA20:4 in jejunum and liver and increases their plasma concentration [351]. Moreover, ATX was shown to be responsible for the production of mentioned LPA species from LPC18:1. However the consequence of that diet on kidney and urinary LPA levels has not been investigated in that report and would deserve to be investigated in the future. If addition of lysophospholipids into diet could induce LPA production in kidney it could be an alternative strategy to elevate levels of possibly protective LPA species in kidney. In next step applicability of this approach in human should be verified. Since LPA is known as a

103 profibrotic agent and also is involved in the processes of tumorigenesis and atherosclerosis, the implication of this approach could be limited and the long-term complications of diet induced increase in LPA species in the body should be examined in animal models. Nevertheless, the described diet induced concentration changes take long time to develop and could have only preventive implication in septic patients who are at high risk for the AKI development.

Decreased kidney production of LPA could also have some adaptive meaning. Since LPA promotes proliferation of tubular epithelial cells, its decreased levels in sepsis could act as a signal that triggers cell cycle arrest, which ensures kidney function recovery after the initial injurious factor has alleviated. The hypothesis requires further verification.

The decrease in urinary LPA associated with endotoxemia-induced AKI could be of importance. It is well known that current clinical detection of AKI is based on the measured levels of plasma creatinine as a reflection of GFR reduction. But plasma creatinine is often a late marker of the disease and rather reflects the kidney function decline when the AKI has already been established. Therefore the research of new biomarkers for the early detection of AKI is of particular attention of researchers. We observed here for the first time that 9 hours after the endotoxin injectuion the urinary LPA concentration in mice is decreased. However, urinary LPA changes in earlier stages of the endotoxemia are unknown and it would be interesting to test whether urinary LPA decline could be observed earlier than 9 hours after induction of endotoxemia. The next step could be the detection of urinary LPA changes in patients with sepsis. If the decreased urinary LPA is observed in patients with sepsis as well, it would be interesting in next step to make a cohort of septic patients, follow them and evaluate the AKI development in a prospective study and make multiple measurements of urinary LPA levels in different times after induction of sepsis. This approach could let us to examine whether the decrease in urinary LPA could have a prognostic value in patients with sepsis and would let to discriminate the septic patients who are at higher risk for AKI development.

104

5. Summary

The initial question of the thesis was to check whether LPA is involved in the sub-total nephrectomy model of CKD and whether it could exert some protection in course of AKI.

We found that 5 months after sub-total nephrectomy LPA production increases in remnant kidneys and that increased urinary LPA is positively correlated with kidney fibrosis. This data combined with previous reports let us to suggest that LPA is a profibrotic molecule that uniformly promotes kidney fibrosis in various forms of CKD. In parallel we found that exogenous LPA protects from kidney dysfunction in endotoxemia-induced kidney injury possibly through the reduction of inflammatory reaction in kidney. Moreover, kidney production of LPA decreases during endotoxemia. There is an apparent contradiction in effects of LPA between CKD and AKI. If in CKD it acts as a deleterious factor that promotes kidney fibrosis, in AKI it protects from kidney damage. The inflammation is a key component in the pathogenesis of both AKI and CKD and LPA is differentially involved in acute and chronic inflammatory processes (discussed in 2.7.4). Therefore the contradictory involvement of LPA in AKI and CKD could be explained by the different activity state of immune and inflammatory cells and/ by activation of different LPA receptors and downstream signaling pathways in the processes of acute and chronic inflammation in AKI and CKD respectively.

It is known that in some patients with AKI kidney function do not recover completely and they are in higher risk for the development of CKD later in life. Various mechanisms are responsible for the development of CKD following AKI (discussed in 1.2.3.1.1). As discussed previously, decreased production of LPA could have an adaptive role, since it could possibly trigger the cell cycle arrest that limits energy expenditure. However, it is known also that cells arrested in G2/M phase of cell cycle produce increased amounts of pro-fibrotic cytokines (especially TGFβ and CTGF) that leads to further fibrosis development [104]. Therefore an interesting hypothesis rises: if the decrease in local LPA production upon sepsis-induced AKI is prolonged, the arrested cells could produce pro-fibrotic cytokines that reflects the transition from adaptive cell response to the dysadaptive. As a consequence, kidney fibrosis develops. The first step to test the possible role of LPA in the progression from AKI to CKD could be the thorough follow-up of animals that survive an AKI episode with the measurement of urine LPA changes during the follow-up and analyze of the CKD development later in life. If decreased LPA production could indeed be responsible for the progression of AKI to CKD, it

105 could be interesting next to explore the mechanisms which are responsible for prolonged decrease in LPA production after AKI episode.

We found here that the kidney metabolism of LPA is changed both in AKI and CKD. The cellular sources of LPA in kidney are not described. Theoretically tubular epithelial cells, glomerular cells (mesangial cells, podocytes, endothelial cells), macrophages and myofibroblasts could produce LPA. The separation of different cell types and further in vitro analysis of these cells from control and fibrotic kidneys could be one approach to check the cellular sources of LPA in kidney.

106

6. Annex

6.1 Curriculum Vitae

Koryun Mirzoyan Born on 05.02.1987 in Yerevan, Armenia Education 2003-2009 General Medicine, Yerevan State Medical University

2009-2012 Residency in Urology, Yerevan State Medical University

2014-2015 Master 2 research ”Physiopathologie des maladies circulatoires et métaboliques”, University Toulouse III, Paul Sabatier, Toulouse, France

Since february 2015- PhD student in University Toulouse III Paul Sabatier (Scholarship of European Union-Erasmus Mundus Medea) Working experience 2012-2014 Urologist in “Armenia” medical center, Yerevan, Armenia Publications 1. Increased urinary lysophosphatidic acid in mouse with subtotal nephrectomy: potential involvement in chronic kidney disease. Journal of Physiology and Biochemistry (2016) Vol. 72, Issue 4, 803-812

2. Increased urine acylcarnitines in diabetic ApoE-/- mice: Hydroxytetradecadienoylcarnitine (C14:2- OH) reflects diabetic nephropathy in a context of hyperlipidemia. Biochemical and Biopphysical Research Communications (2017) Vol. 487, Issue 1, 109-115

3. Lysophosphatidic acid protects against endotoxin-induced acute kidney injury. Accepted to Inflammation (DOI:10.1007/s10753-017-0612-7)

107

7. References.

1. Bywaters, E.G. and D. Beall, Crush Injuries with Impairment of Renal Function. Br Med J, 1941. 1(4185): p. 427-32. 2. Bellomo, R., et al., Acute renal failure - definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care, 2004. 8(4): p. R204-12. 3. Mehta, R.L., et al., Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care, 2007. 11(2): p. R31. 4. Section 2: AKI Definition. Kidney Int Suppl (2011), 2012. 2(1): p. 19-36. 5. Thadhani, R., M. Pascual, and J.V. Bonventre, Acute renal failure. N Engl J Med, 1996. 334(22): p. 1448-60. 6. Makris, K. and L. Spanou, Acute Kidney Injury: Definition, Pathophysiology and Clinical Phenotypes. Clin Biochem Rev, 2016. 37(2): p. 85-98. 7. Hsu, R.K., et al., Temporal changes in incidence of dialysis-requiring AKI. J Am Soc Nephrol, 2013. 24(1): p. 37-42. 8. Hoste, E.A., et al., RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care, 2006. 10(3): p. R73. 9. Uchino, S., et al., Acute renal failure in critically ill patients: a multinational, multicenter study. JAMA, 2005. 294(7): p. 813-8. 10. Piccinni, P., et al., Prospective multicenter study on epidemiology of acute kidney injury in the ICU: a critical care nephrology Italian collaborative effort (NEFROINT). Minerva Anestesiol, 2011. 77(11): p. 1072-83. 11. Chertow, G.M., et al., Acute kidney injury, mortality, length of stay, and costs in hospitalized patients. J Am Soc Nephrol, 2005. 16(11): p. 3365-70. 12. Kosaka, J., et al., Histopathology of Septic Acute Kidney Injury: A Systematic Review of Experimental Data. Crit Care Med, 2016. 44(9): p. e897-903. 13. Lerolle, N., et al., Histopathology of septic shock induced acute kidney injury: apoptosis and leukocytic infiltration. Intensive Care Med, 2010. 36(3): p. 471-8. 14. Di Giantomasso, D., C.N. May, and R. Bellomo, Vital organ blood flow during hyperdynamic sepsis. Chest, 2003. 124(3): p. 1053-9. 15. Langenberg, C., et al., Renal blood flow in experimental septic acute renal failure. Kidney Int, 2006. 69(11): p. 1996-2002. 16. Langenberg, C., et al., Renal blood flow in sepsis. Crit Care, 2005. 9(4): p. R363-74.

108

17. Murugan, R., et al., Acute kidney injury in non-severe pneumonia is associated with an increased immune response and lower survival. Kidney Int, 2010. 77(6): p. 527-35. 18. Wan, L., et al., Pathophysiology of septic acute kidney injury: what do we really know? Crit Care Med, 2008. 36(4 Suppl): p. S198-203. 19. Holthoff, J.H., et al., Resveratrol improves renal microcirculation, protects the tubular epithelium, and prolongs survival in a mouse model of sepsis-induced acute kidney injury. Kidney Int, 2012. 81(4): p. 370-8. 20. Tiwari, M.M., et al., Disruption of renal peritubular blood flow in lipopolysaccharide-induced renal failure: role of nitric oxide and caspases. Am J Physiol Renal Physiol, 2005. 289(6): p. F1324-32. 21. Chvojka, J., et al., Renal haemodynamic, microcirculatory, metabolic and histopathological responses to peritonitis-induced septic shock in pigs. Crit Care, 2008. 12(6): p. R164. 22. Legrand, M., et al., Association between systemic hemodynamics and septic acute kidney injury in critically ill patients: a retrospective observational study. Crit Care, 2013. 17(6): p. R278. 23. Ahn, K.Y., et al., In situ hybridization localization of mRNA encoding inducible nitric oxide synthase in rat kidney. Am J Physiol, 1994. 267(5 Pt 2): p. F748-57. 24. Schrier, R.W. and W. Wang, Acute renal failure and sepsis. N Engl J Med, 2004. 351(2): p. 159- 69. 25. Wang, L., X.M. Fan, and H.X. Tang, [Effects of aminoguanidine in different dosages on renal function in endotoxin induced rabbits shock model]. Zhonghua Er Ke Za Zhi, 2004. 42(3): p. 206-9. 26. Marzocco, S., et al., Effect of methylguanidine in a model of septic shock induced by LPS. Free Radic Res, 2004. 38(11): p. 1143-53. 27. Takeuchi, O. and S. Akira, Pattern recognition receptors and inflammation. Cell, 2010. 140(6): p. 805-20. 28. Payen, D., et al., A multicentre study of acute kidney injury in severe sepsis and septic shock: association with inflammatory phenotype and HLA genotype. PLoS One, 2012. 7(6): p. e35838. 29. Wu, X., et al., The role of ICAM-1 in endotoxin-induced acute renal failure. Am J Physiol Renal Physiol, 2007. 293(4): p. F1262-71. 30. Herter, J.M., et al., Adhesion molecules involved in neutrophil recruitment during sepsis- induced acute kidney injury. J Innate Immun, 2014. 6(5): p. 597-606.

109

31. Wu, L., et al., Peritubular capillary dysfunction and renal tubular epithelial cell stress following lipopolysaccharide administration in mice. Am J Physiol Renal Physiol, 2007. 292(1): p. F261-8. 32. Wolfs, T.G., et al., In vivo expression of Toll-like receptor 2 and 4 by renal epithelial cells: IFN- gamma and TNF-alpha mediated up-regulation during inflammation. J Immunol, 2002. 168(3): p. 1286-93. 33. El-Achkar, T.M., et al., Sepsis induces changes in the expression and distribution of Toll-like receptor 4 in the rat kidney. Am J Physiol Renal Physiol, 2006. 290(5): p. F1034-43. 34. Kalakeche, R., et al., Endotoxin uptake by S1 proximal tubular segment causes oxidative stress in the downstream S2 segment. J Am Soc Nephrol, 2011. 22(8): p. 1505-16. 35. Morrell, E.D., et al., Epithelial transport during septic acute kidney injury. Nephrol Dial Transplant, 2014. 29(7): p. 1312-9. 36. Gomez, H., et al., A unified theory of sepsis-induced acute kidney injury: inflammation, microcirculatory dysfunction, bioenergetics, and the tubular cell adaptation to injury. Shock, 2014. 41(1): p. 3-11. 37. Tran, M., et al., PGC-1alpha promotes recovery after acute kidney injury during systemic inflammation in mice. J Clin Invest, 2011. 121(10): p. 4003-14. 38. Yang, R.L., et al., Energy and oxygen metabolism disorder during septic acute kidney injury. Kidney Blood Press Res, 2014. 39(4): p. 240-51. 39. Yang, Q.H., et al., Acute renal failure during sepsis: potential role of cell cycle regulation. J Infect, 2009. 58(6): p. 459-64. 40. Kashani, K., et al., Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury. Crit Care, 2013. 17(1): p. R25. 41. Gunst, J., et al., Insufficient autophagy contributes to mitochondrial dysfunction, organ failure, and adverse outcome in an animal model of critical illness. Crit Care Med, 2013. 41(1): p. 182-94. 42. Hsiao, H.W., et al., The decline of autophagy contributes to proximal tubular dysfunction during sepsis. Shock, 2012. 37(3): p. 289-96. 43. Zarbock, A., H. Gomez, and J.A. Kellum, Sepsis-induced acute kidney injury revisited: pathophysiology, prevention and future therapies. Curr Opin Crit Care, 2014. 20(6): p. 588-95. 44. Abraham, E., et al., Double-blind randomised controlled trial of monoclonal antibody to human tumour necrosis factor in treatment of septic shock. NORASEPT II Study Group. Lancet, 1998. 351(9107): p. 929-33.

110

45. Opal, S.M., et al., Confirmatory interleukin-1 receptor antagonist trial in severe sepsis: a phase III, randomized, double-blind, placebo-controlled, multicenter trial. The Interleukin-1 Receptor Antagonist Sepsis Investigator Group. Crit Care Med, 1997. 25(7): p. 1115-24. 46. Willatts, S.M., S. Radford, and M. Leitermann, Effect of the antiendotoxic agent, taurolidine, in the treatment of sepsis syndrome: a placebo-controlled, double-blind trial. Crit Care Med, 1995. 23(6): p. 1033-9. 47. Opal, S.M., et al., Effect of eritoran, an antagonist of MD2-TLR4, on mortality in patients with severe sepsis: the ACCESS randomized trial. JAMA, 2013. 309(11): p. 1154-62. 48. Yeh, C.H., et al., Anticancer agent 2-methoxyestradiol improves survival in septic mice by reducing the production of cytokines and nitric oxide. Shock, 2011. 36(5): p. 510-6. 49. Howell, G.M., et al., Augmenting autophagy to treat acute kidney injury during endotoxemia in mice. PLoS One, 2013. 8(7): p. e69520. 50. Peters, E., et al., Alkaline phosphatase: a possible treatment for sepsis-associated acute kidney injury in critically ill patients. Am J Kidney Dis, 2014. 63(6): p. 1038-48. 51. Wang, W., et al., Pentoxifylline protects against endotoxin-induced acute renal failure in mice. Am J Physiol Renal Physiol, 2006. 291(5): p. F1090-5. 52. Pickkers, P., et al., Alkaline phosphatase for treatment of sepsis-induced acute kidney injury: a prospective randomized double-blind placebo-controlled trial. Crit Care, 2012. 16(1): p. R14. 53. Chapter 1: Definition and classification of CKD. Kidney Int Suppl (2011), 2013. 3(1): p. 19-62. 54. Jha, V., et al., Chronic kidney disease: global dimension and perspectives. Lancet, 2013. 382(9888): p. 260-72. 55. Lozano, R., et al., Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet, 2012. 380(9859): p. 2095-128. 56. Thompson, S., et al., Cause of Death in Patients with Reduced Kidney Function. J Am Soc Nephrol, 2015. 26(10): p. 2504-11. 57. Cybulsky, A.V., Growth factor pathways in proliferative glomerulonephritis. Curr Opin Nephrol Hypertens, 2000. 9(3): p. 217-23. 58. Wolf, G. and F.N. Ziyadeh, Cellular and molecular mechanisms of proteinuria in diabetic nephropathy. Nephron Physiol, 2007. 106(2): p. p26-31. 59. Risdon, R.A., J.C. Sloper, and H.E. De Wardener, Relationship between renal function and histological changes found in renal-biopsy specimens from patients with persistent glomerular nephritis. Lancet, 1968. 2(7564): p. 363-6. 60. Kitagawa, K., et al., Blockade of CCR2 ameliorates progressive fibrosis in kidney. Am J Pathol, 2004. 165(1): p. 237-46.

111

61. Shimizu, S., et al., Anti-monocyte chemoattractant protein-1 gene therapy attenuates nephritis in MRL/lpr mice. Rheumatology (Oxford), 2004. 43(9): p. 1121-8. 62. Tesch, G.H., et al., Monocyte chemoattractant protein-1 promotes macrophage-mediated tubular injury, but not glomerular injury, in nephrotoxic serum nephritis. J Clin Invest, 1999. 103(1): p. 73-80. 63. Naruse, T., et al., P-selectin-dependent macrophage migration into the tubulointerstitium in unilateral ureteral obstruction. Kidney Int, 2002. 62(1): p. 94-105. 64. Cheng, Q.L., et al., Effects of ICAM-1 antisense oligonucleotide on the tubulointerstitium in mice with unilateral ureteral obstruction. Kidney Int, 2000. 57(1): p. 183-90. 65. Janssen, U., et al., Improved survival and amelioration of nephrotoxic nephritis in intercellular adhesion molecule-1 knockout mice. J Am Soc Nephrol, 1998. 9(10): p. 1805-14. 66. Ko, G.J., et al., Macrophages contribute to the development of renal fibrosis following ischaemia/reperfusion-induced acute kidney injury. Nephrol Dial Transplant, 2008. 23(3): p. 842-52. 67. Nishida, M. and K. Hamaoka, Macrophage phenotype and renal fibrosis in obstructive nephropathy. Nephron Exp Nephrol, 2008. 110(1): p. e31-6. 68. Ricardo, S.D., H. van Goor, and A.A. Eddy, Macrophage diversity in renal injury and repair. J Clin Invest, 2008. 118(11): p. 3522-30. 69. Burne-Taney, M.J., et al., Transfer of lymphocytes from mice with renal ischemia can induce albuminuria in naive mice: a possible mechanism linking early injury and progressive renal disease? Am J Physiol Renal Physiol, 2006. 291(5): p. F981-6. 70. Lebleu, V.S., et al., Lymphocytes are dispensable for glomerulonephritis but required for renal interstitial fibrosis in matrix defect-induced Alport renal disease. Lab Invest, 2008. 88(3): p. 284-92. 71. Tapmeier, T.T., et al., Pivotal role of CD4+ T cells in renal fibrosis following ureteric obstruction. Kidney Int, 2010. 78(4): p. 351-62. 72. Kim, D.H., et al., Mast cells decrease renal fibrosis in unilateral ureteral obstruction. Kidney Int, 2009. 75(10): p. 1031-8. 73. Miyazawa, S., et al., Role of mast cells in the development of renal fibrosis: use of mast cell- deficient rats. Kidney Int, 2004. 65(6): p. 2228-37. 74. Timoshanko, J.R., et al., A pathogenetic role for mast cells in experimental crescentic glomerulonephritis. J Am Soc Nephrol, 2006. 17(1): p. 150-9. 75. Picard, N., et al., Origin of renal myofibroblasts in the model of unilateral ureter obstruction in the rat. Histochem Cell Biol, 2008. 130(1): p. 141-55.

112

76. Bellini, A. and S. Mattoli, The role of the fibrocyte, a bone marrow-derived mesenchymal progenitor, in reactive and reparative fibroses. Lab Invest, 2007. 87(9): p. 858-70. 77. Iwano, M., et al., Evidence that fibroblasts derive from epithelium during tissue fibrosis. J Clin Invest, 2002. 110(3): p. 341-50. 78. Kalluri, R. and R.A. Weinberg, The basics of epithelial-mesenchymal transition. J Clin Invest, 2009. 119(6): p. 1420-8. 79. Liu, Y., Epithelial to mesenchymal transition in renal fibrogenesis: pathologic significance, molecular mechanism, and therapeutic intervention. J Am Soc Nephrol, 2004. 15(1): p. 1-12. 80. Zeisberg, E.M., et al., Fibroblasts in kidney fibrosis emerge via endothelial-to-mesenchymal transition. J Am Soc Nephrol, 2008. 19(12): p. 2282-7. 81. Govinden, R. and K.D. Bhoola, Genealogy, expression, and cellular function of transforming growth factor-beta. Pharmacol Ther, 2003. 98(2): p. 257-65. 82. Meng, X.M., A.C. Chung, and H.Y. Lan, Role of the TGF-beta/BMP-7/Smad pathways in renal diseases. Clin Sci (Lond), 2013. 124(4): p. 243-54. 83. Shi-Wen, X., A. Leask, and D. Abraham, Regulation and function of connective tissue growth factor/CCN2 in tissue repair, scarring and fibrosis. Cytokine Growth Factor Rev, 2008. 19(2): p. 133-44. 84. Sanchez-Lopez, E., et al., CTGF promotes inflammatory cell infiltration of the renal interstitium by activating NF-kappaB. J Am Soc Nephrol, 2009. 20(7): p. 1513-26. 85. Zhang, C., et al., Connective tissue growth factor regulates the key events in tubular epithelial to myofibroblast transition in vitro. Cell Biol Int, 2004. 28(12): p. 863-73. 86. Ito, Y., et al., Expression of connective tissue growth factor in human renal fibrosis. Kidney Int, 1998. 53(4): p. 853-61. 87. Abdel Wahab, N. and R.M. Mason, Connective tissue growth factor and renal diseases: some answers, more questions. Curr Opin Nephrol Hypertens, 2004. 13(1): p. 53-8. 88. Eddy, A.A., Progression in chronic kidney disease. Adv Chronic Kidney Dis, 2005. 12(4): p. 353- 65. 89. Horstrup, J.H., et al., Elevation of serum and urine levels of TIMP-1 and tenascin in patients with renal disease. Nephrol Dial Transplant, 2002. 17(6): p. 1005-13. 90. Eddy, A.A. and A.B. Fogo, Plasminogen activator inhibitor-1 in chronic kidney disease: evidence and mechanisms of action. J Am Soc Nephrol, 2006. 17(11): p. 2999-3012. 91. Meng, X.M., D.J. Nikolic-Paterson, and H.Y. Lan, Inflammatory processes in renal fibrosis. Nat Rev Nephrol, 2014. 10(9): p. 493-503. 92. Eitner, F., et al., PDGF-C is a proinflammatory cytokine that mediates renal interstitial fibrosis. J Am Soc Nephrol, 2008. 19(2): p. 281-9.

113

93. Strutz, F., et al., Basic fibroblast growth factor expression is increased in human renal fibrogenesis and may mediate autocrine fibroblast proliferation. Kidney Int, 2000. 57(4): p. 1521-38. 94. Mizuno, S., et al., Reciprocal balance of hepatocyte growth factor and transforming growth factor-beta 1 in renal fibrosis in mice. Kidney Int, 2000. 57(3): p. 937-48. 95. Wang, S.N., J. Lapage, and R. Hirschberg, Loss of tubular bone morphogenetic protein-7 in diabetic nephropathy. J Am Soc Nephrol, 2001. 12(11): p. 2392-9. 96. Ruster, C. and G. Wolf, Renin-angiotensin-aldosterone system and progression of renal disease. J Am Soc Nephrol, 2006. 17(11): p. 2985-91. 97. Strutz, F.M., EMT and proteinuria as progression factors. Kidney Int, 2009. 75(5): p. 475-81. 98. Fine, L.G. and J.T. Norman, Chronic hypoxia as a mechanism of progression of chronic kidney diseases: from hypothesis to novel therapeutics. Kidney Int, 2008. 74(7): p. 867-72. 99. Brewster, U.C. and M.A. Perazella, The renin-angiotensin-aldosterone system and the kidney: effects on kidney disease. Am J Med, 2004. 116(4): p. 263-72. 100. Chawla, L.S. and P.L. Kimmel, Acute kidney injury and chronic kidney disease: an integrated clinical syndrome. Kidney Int, 2012. 82(5): p. 516-24. 101. Paauw, N.D., et al., Long-term renal and cardiovascular risk after preeclampsia: towards screening and prevention. Clin Sci (Lond), 2016. 130(4): p. 239-46. 102. Ishani, A., et al., Acute kidney injury increases risk of ESRD among elderly. J Am Soc Nephrol, 2009. 20(1): p. 223-8. 103. Thakar, C.V., et al., Acute kidney injury episodes and chronic kidney disease risk in diabetes mellitus. Clin J Am Soc Nephrol, 2011. 6(11): p. 2567-72. 104. Yang, L., et al., Epithelial cell cycle arrest in G2/M mediates kidney fibrosis after injury. Nat Med, 2010. 16(5): p. 535-43, 1p following 143. 105. Basile, D.P., et al., Impaired endothelial proliferation and mesenchymal transition contribute to vascular rarefaction following acute kidney injury. Am J Physiol Renal Physiol, 2011. 300(3): p. F721-33. 106. Aoki, J., A. Inoue, and S. Okudaira, Two pathways for lysophosphatidic acid production. Biochim Biophys Acta, 2008. 1781(9): p. 513-8. 107. Daae, L.N. and J. Bremer, The acylation of glycerophosphate in rat liver. A new assay procedure for glycerophosphate acylation, studies on its subcellular and submitochondrial localization and determination of the reaction products. Biochim Biophys Acta, 1970. 210(1): p. 92-104. 108. Haldar, D., W.W. Tso, and M.E. Pullman, The acylation of sn-glycerol 3-phosphate in mammalian organs and Ehrlich ascites tumor cells. J Biol Chem, 1979. 254(11): p. 4502-9.

114

109. Dircks, L. and H.S. Sul, Acyltransferases of de novo glycerophospholipid biosynthesis. Prog Lipid Res, 1999. 38(5-6): p. 461-79. 110. Nakajima, K., et al., A novel phospholipase A1 with sequence homology to a mammalian Sec23p-interacting protein, p125. J Biol Chem, 2002. 277(13): p. 11329-35. 111. Bektas, M., et al., A novel acylglycerol kinase that produces lysophosphatidic acid modulates cross talk with EGFR in prostate cancer cells. J Cell Biol, 2005. 169(5): p. 801-11. 112. Ohshima, N., et al., New members of the mammalian glycerophosphodiester phosphodiesterase family: GDE4 and GDE7 produce lysophosphatidic acid by lysophospholipase D activity. J Biol Chem, 2015. 290(7): p. 4260-71. 113. Fourcade, O., et al., Secretory phospholipase A2 generates the novel lipid mediator lysophosphatidic acid in membrane microvesicles shed from activated cells. Cell, 1995. 80(6): p. 919-27. 114. Sonoda, H., et al., A novel phosphatidic acid-selective phospholipase A1 that produces lysophosphatidic acid. J Biol Chem, 2002. 277(37): p. 34254-63. 115. Hiramatsu, T., et al., Biochemical and molecular characterization of two phosphatidic acid- selective phospholipase A1s, mPA-PLA1alpha and mPA-PLA1beta. J Biol Chem, 2003. 278(49): p. 49438-47. 116. Aoki, J., et al., Serum lysophosphatidic acid is produced through diverse phospholipase pathways. J Biol Chem, 2002. 277(50): p. 48737-44. 117. Okudaira, S., H. Yukiura, and J. Aoki, Biological roles of lysophosphatidic acid signaling through its production by autotaxin. Biochimie, 2010. 92(6): p. 698-706. 118. Tomsig, J.L., et al., Lipid phosphate phosphohydrolase type 1 (LPP1) degrades extracellular lysophosphatidic acid in vivo. Biochem J, 2009. 419(3): p. 611-8. 119. Pyne, S., K.C. Kong, and P.I. Darroch, Lysophosphatidic acid and sphingosine 1-phosphate biology: the role of lipid phosphate phosphatases. Semin Cell Dev Biol, 2004. 15(5): p. 491- 501. 120. Zhang, Q.X., et al., Identification of structurally important domains of lipid phosphate phosphatase-1: implications for its sites of action. Biochem J, 2000. 345 Pt 2: p. 181-4. 121. Waggoner, D.W., et al., Phosphatidate phosphohydrolase catalyzes the hydrolysis of ceramide 1-phosphate, lysophosphatidate, and sphingosine 1-phosphate. J Biol Chem, 1996. 271(28): p. 16506-9. 122. Tang, X., M.G. Benesch, and D.N. Brindley, Lipid phosphate phosphatases and their roles in mammalian physiology and pathology. J Lipid Res, 2015. 56(11): p. 2048-60. 123. Tanyi, J.L., et al., Role of decreased levels of lipid phosphate phosphatase-1 in accumulation of lysophosphatidic acid in ovarian cancer. Clin Cancer Res, 2003. 9(10 Pt 1): p. 3534-45.

115

124. Tanyi, J.L., et al., The human lipid phosphate phosphatase-3 decreases the growth, survival, and tumorigenesis of ovarian cancer cells: validation of the lysophosphatidic acid signaling cascade as a target for therapy in ovarian cancer. Cancer Res, 2003. 63(5): p. 1073-82. 125. Barila, D., et al., The Dri 42 gene, whose expression is up-regulated during epithelial differentiation, encodes a novel endoplasmic reticulum resident transmembrane protein. J Biol Chem, 1996. 271(47): p. 29928-36. 126. Sciorra, V.A. and A.J. Morris, Sequential actions of phospholipase D and phosphatidic acid phosphohydrolase 2b generate diglyceride in mammalian cells. Mol Biol Cell, 1999. 10(11): p. 3863-76. 127. Leung, D.W., The structure and functions of human lysophosphatidic acid acyltransferases. Front Biosci, 2001. 6: p. D944-53. 128. Thompson, F.J. and M.A. Clark, Purification of a lysophosphatidic acid-hydrolysing lysophospholipase from rat brain. Biochem J, 1994. 300 ( Pt 2): p. 457-61. 129. Oguro, A. and S. Imaoka, Lysophosphatidic acids are new substrates for the phosphatase domain of soluble epoxide hydrolase. J Lipid Res, 2012. 53(3): p. 505-12. 130. Ren, J., et al., Lysophosphatidic acid is constitutively produced by human peritoneal mesothelial cells and enhances adhesion, migration, and invasion of ovarian cancer cells. Cancer Res, 2006. 66(6): p. 3006-14. 131. Goetzl, E.J., et al., Gelsolin binding and cellular presentation of lysophosphatidic acid. J Biol Chem, 2000. 275(19): p. 14573-8. 132. Tigyi, G. and R. Miledi, Lysophosphatidates bound to serum albumin activate membrane currents in Xenopus oocytes and neurite retraction in PC12 pheochromocytoma cells. J Biol Chem, 1992. 267(30): p. 21360-7. 133. Vancura, A. and D. Haldar, Regulation of mitochondrial and microsomal phospholipid synthesis by liver fatty acid-binding protein. J Biol Chem, 1992. 267(20): p. 14353-9. 134. Mauco, G., et al., Phosphatidic and lysophosphatidic acid production in phospholipase C-and thrombin-treated platelets. Possible involvement of a platelet lipase. Biochimie, 1978. 60(6- 7): p. 653-61. 135. Gaits, F., et al., Lysophosphatidic acid as a phospholipid mediator: pathways of synthesis. FEBS Lett, 1997. 410(1): p. 54-8. 136. Tokumura, A., et al., Involvement of lysophospholipase D in the production of lysophosphatidic acid in rat plasma. Biochim Biophys Acta, 1986. 875(1): p. 31-8. 137. Tanaka, M., et al., Prostatic acid phosphatase degrades lysophosphatidic acid in seminal plasma. FEBS Lett, 2004. 571(1-3): p. 197-204.

116

138. Tokumura, A., et al., Production of lysophosphatidic acids by lysophospholipase D in human follicular fluids of In vitro fertilization patients. Biol Reprod, 1999. 61(1): p. 195-9. 139. Sugiura, T., et al., Lysophosphatidic acid, a growth factor-like lipid, in the saliva. J Lipid Res, 2002. 43(12): p. 2049-55. 140. Sato, K., et al., Identification of autotaxin as a neurite retraction-inducing factor of PC12 cells in cerebrospinal fluid and its possible sources. J Neurochem, 2005. 92(4): p. 904-14. 141. Liliom, K., et al., Growth factor-like phospholipids generated after corneal injury. Am J Physiol, 1998. 274(4 Pt 1): p. C1065-74. 142. Nakane, S., et al., Hen egg yolk and white contain high amounts of lysophosphatidic acids, growth factor-like lipids: distinct molecular species compositions. Lipids, 2001. 36(4): p. 413- 9. 143. Ohkawa, R., et al., Lysophospholipase D activity exists in the urine to catalyse the formation of lysophosphatidic acid. Nephrol Dial Transplant, 2006. 21(12): p. 3612-3. 144. Xu, Y., et al., Characterization of an ovarian cancer activating factor in ascites from ovarian cancer patients. Clin Cancer Res, 1995. 1(10): p. 1223-32. 145. Yamada, T., et al., Lysophosphatidic acid (LPA) in malignant ascites stimulates motility of human pancreatic cancer cells through LPA1. J Biol Chem, 2004. 279(8): p. 6595-605. 146. Nochi, H., et al., Stimulatory role of lysophosphatidic acid in cyclooxygenase-2 induction by synovial fluid of patients with rheumatoid arthritis in fibroblast-like synovial cells. J Immunol, 2008. 181(7): p. 5111-9. 147. Abu El-Asrar, A.M., et al., Expression of lysophosphatidic acid, autotaxin and acylglycerol kinase as biomarkers in diabetic retinopathy. Acta Diabetol, 2013. 50(3): p. 363-71. 148. Blazer-Yost, B.L., et al., Lysophosphatidic acid is a modulator of cyst growth in autosomal dominant polycystic kidney disease. Cell Physiol Biochem, 2011. 28(6): p. 1255-64. 149. Crack, P.J., et al., Anti-lysophosphatidic acid antibodies improve traumatic brain injury outcomes. J Neuroinflammation, 2014. 11: p. 37. 150. Mazereeuw-Hautier, J., et al., Production of lysophosphatidic acid in blister fluid: involvement of a lysophospholipase D activity. J Invest Dermatol, 2005. 125(3): p. 421-7. 151. Yamamoto, J., et al., Preferable existence of polyunsaturated lysophosphatidic acids in human follicular fluid from patients programmed with in vitro fertilization. Prostaglandins Other Lipid Mediat, 2016. 126: p. 16-23. 152. Hecht, J.H., et al., Ventricular zone gene-1 (vzg-1) encodes a lysophosphatidic acid receptor expressed in neurogenic regions of the developing cerebral cortex. J Cell Biol, 1996. 135(4): p. 1071-83.

117

153. Choi, J.W., et al., LPA receptors: subtypes and biological actions. Annu Rev Pharmacol Toxicol, 2010. 50: p. 157-86. 154. Yung, Y.C., N.C. Stoddard, and J. Chun, LPA receptor signaling: pharmacology, physiology, and pathophysiology. J Lipid Res, 2014. 55(7): p. 1192-214. 155. An, S., et al., Characterization of a novel subtype of human G protein-coupled receptor for lysophosphatidic acid. J Biol Chem, 1998. 273(14): p. 7906-10. 156. Bandoh, K., et al., Molecular cloning and characterization of a novel human G-protein- coupled receptor, EDG7, for lysophosphatidic acid. J Biol Chem, 1999. 274(39): p. 27776-85. 157. Chun, J., et al., International Union of Basic and Clinical Pharmacology. LXXVIII. Lysophospholipid receptor nomenclature. Pharmacol Rev, 2010. 62(4): p. 579-87. 158. Noguchi, K., S. Ishii, and T. Shimizu, Identification of p2y9/GPR23 as a novel G protein-coupled receptor for lysophosphatidic acid, structurally distant from the Edg family. J Biol Chem, 2003. 278(28): p. 25600-6. 159. Lee, C.W., et al., GPR92 as a new G12/13- and Gq-coupled lysophosphatidic acid receptor that increases cAMP, LPA5. J Biol Chem, 2006. 281(33): p. 23589-97. 160. Pasternack, S.M., et al., G protein-coupled receptor P2Y5 and its ligand LPA are involved in maintenance of human hair growth. Nat Genet, 2008. 40(3): p. 329-34. 161. Tabata, K., et al., The orphan GPCR GPR87 was deorphanized and shown to be a lysophosphatidic acid receptor. Biochem Biophys Res Commun, 2007. 363(3): p. 861-6. 162. Murakami, M., et al., Identification of the orphan GPCR, P2Y(10) receptor as the sphingosine- 1-phosphate and lysophosphatidic acid receptor. Biochem Biophys Res Commun, 2008. 371(4): p. 707-12. 163. Oka, S., et al., GPR35 is a novel lysophosphatidic acid receptor. Biochem Biophys Res Commun, 2010. 395(2): p. 232-7. 164. Pradere, J.P., et al., LPA1 receptor activation promotes renal interstitial fibrosis. J Am Soc Nephrol, 2007. 18(12): p. 3110-8. 165. Sakai, N., et al., Lysophosphatidic acid signaling through its receptor initiates profibrotic epithelial cell fibroblast communication mediated by epithelial cell derived connective tissue growth factor. Kidney Int, 2016. 166. Fukushima, N., Y. Kimura, and J. Chun, A single receptor encoded by vzg-1/lpA1/edg-2 couples to G proteins and mediates multiple cellular responses to lysophosphatidic acid. Proc Natl Acad Sci U S A, 1998. 95(11): p. 6151-6. 167. Ishii, I., et al., Functional comparisons of the lysophosphatidic acid receptors, LP(A1)/VZG- 1/EDG-2, LP(A2)/EDG-4, and LP(A3)/EDG-7 in neuronal cell lines using a retrovirus expression system. Mol Pharmacol, 2000. 58(5): p. 895-902.

118

168. Lee, C.W., et al., LPA(4)/GPR23 is a lysophosphatidic acid (LPA) receptor utilizing G(s)-, G(q)/G(i)-mediated calcium signaling and G(12/13)-mediated Rho activation. J Biol Chem, 2007. 282(7): p. 4310-7. 169. Yanagida, K., et al., Identification and characterization of a novel lysophosphatidic acid receptor, p2y5/LPA6. J Biol Chem, 2009. 284(26): p. 17731-41. 170. Baker, D.L., et al., Direct quantitative analysis of lysophosphatidic acid molecular species by stable isotope dilution electrospray ionization liquid chromatography-mass spectrometry. Anal Biochem, 2001. 292(2): p. 287-95. 171. Shen, Z., et al., Fatty acid composition of lysophosphatidic acid and lysophosphatidylinositol in plasma from patients with ovarian cancer and other gynecological diseases. Gynecol Oncol, 2001. 83(1): p. 25-30. 172. Xu, Y., et al., Lysophosphatidic acid as a potential biomarker for ovarian and other gynecologic cancers. JAMA, 1998. 280(8): p. 719-23. 173. Tsutsumi, T., et al., Presence of bioactive lysophosphatidic acid in renal effluent of rats with unilateral ureteral obstruction. Life Sci, 2011. 89(5-6): p. 195-203. 174. Thomson, F.J., et al., Lysophosphatidic acid stimulates glucose transport in Xenopus oocytes via a phosphatidylinositol 3'-kinase with distinct properties. Biochem J, 1996. 316 ( Pt 1): p. 161-6. 175. Jalink, K., et al., Lysophosphatidic acid-induced Ca2+ mobilization in human A431 cells: structure-activity analysis. Biochem J, 1995. 307 ( Pt 2): p. 609-16. 176. Jongsma, M., et al., LPA is a chemorepellent for B16 melanoma cells: action through the cAMP-elevating LPA5 receptor. PLoS One, 2011. 6(12): p. e29260. 177. Contos, J.J., et al., Requirement for the lpA1 lysophosphatidic acid receptor gene in normal suckling behavior. Proc Natl Acad Sci U S A, 2000. 97(24): p. 13384-9. 178. Estivill-Torrus, G., et al., Absence of LPA1 signaling results in defective cortical development. Cereb Cortex, 2008. 18(4): p. 938-50. 179. Matas-Rico, E., et al., Deletion of lysophosphatidic acid receptor LPA1 reduces neurogenesis in the mouse dentate gyrus. Mol Cell Neurosci, 2008. 39(3): p. 342-55. 180. Castilla-Ortega, E., et al., Aggravation of chronic stress effects on hippocampal neurogenesis and spatial memory in LPA(1) receptor knockout mice. PLoS One, 2011. 6(9): p. e25522. 181. Frisca, F., et al., Biological effects of lysophosphatidic acid in the nervous system. Int Rev Cell Mol Biol, 2012. 296: p. 273-322. 182. Tager, A.M., et al., The lysophosphatidic acid receptor LPA1 links pulmonary fibrosis to lung injury by mediating fibroblast recruitment and vascular leak. Nat Med, 2008. 14(1): p. 45-54.

119

183. Boucharaba, A., et al., The type 1 lysophosphatidic acid receptor is a target for therapy in bone metastases. Proc Natl Acad Sci U S A, 2006. 103(25): p. 9643-8. 184. Simon, M.F., et al., Lysophosphatidic acid inhibits adipocyte differentiation via lysophosphatidic acid 1 receptor-dependent down-regulation of peroxisome proliferator- activated receptor gamma2. J Biol Chem, 2005. 280(15): p. 14656-62. 185. Miyabe, Y., et al., Necessity of lysophosphatidic acid receptor 1 for development of arthritis. Arthritis Rheum, 2013. 65(8): p. 2037-47. 186. Ye, X., et al., Age-dependent loss of sperm production in mice via impaired lysophosphatidic acid signaling. Biol Reprod, 2008. 79(2): p. 328-36. 187. Contos, J.J., et al., Characterization of lpa(2) (Edg4) and lpa(1)/lpa(2) (Edg2/Edg4) lysophosphatidic acid receptor knockout mice: signaling deficits without obvious phenotypic abnormality attributable to lpa(2). Mol Cell Biol, 2002. 22(19): p. 6921-9. 188. Lin, S., et al., The absence of LPA2 attenuates tumor formation in an experimental model of colitis-associated cancer. Gastroenterology, 2009. 136(5): p. 1711-20. 189. Kitayama, J., et al., Over-expression of lysophosphatidic acid receptor-2 in human invasive ductal carcinoma. Breast Cancer Res, 2004. 6(6): p. R640-6. 190. Park, G.Y., et al., Autotaxin production of lysophosphatidic acid mediates allergic asthmatic inflammation. Am J Respir Crit Care Med, 2013. 188(8): p. 928-40. 191. Zhao, Y., et al., Role of lysophosphatidic acid receptor LPA2 in the development of allergic airway inflammation in a murine model of asthma. Respir Res, 2009. 10: p. 114. 192. Ye, X., et al., LPA3-mediated lysophosphatidic acid signalling in embryo implantation and spacing. Nature, 2005. 435(7038): p. 104-8. 193. Hama, K., et al., Embryo spacing and implantation timing are differentially regulated by LPA3- mediated lysophosphatidic acid signaling in mice. Biol Reprod, 2007. 77(6): p. 954-9. 194. Sumida, H., et al., LPA4 regulates blood and lymphatic vessel formation during mouse embryogenesis. Blood, 2010. 116(23): p. 5060-70. 195. Liu, Y.B., et al., LPA induces osteoblast differentiation through interplay of two receptors: LPA1 and LPA4. J Cell Biochem, 2010. 109(4): p. 794-800. 196. Lee, Z., et al., Role of LPA4/p2y9/GPR23 in negative regulation of cell motility. Mol Biol Cell, 2008. 19(12): p. 5435-45. 197. Lin, M.E., R.R. Rivera, and J. Chun, Targeted deletion of LPA5 identifies novel roles for lysophosphatidic acid signaling in development of neuropathic pain. J Biol Chem, 2012. 287(21): p. 17608-17. 198. Nahum, S., et al., A novel mutation in LPAR6 causes autosomal recessive hypotrichosis of the scalp. Clin Exp Dermatol, 2011. 36(2): p. 188-94.

120

199. Lee, S., et al., Forerunner genes contiguous to RB1 contribute to the development of in situ neoplasia. Proc Natl Acad Sci U S A, 2007. 104(34): p. 13732-7. 200. Yu, S., et al., Lysophosphatidic acid receptors determine tumorigenicity and aggressiveness of ovarian cancer cells. J Natl Cancer Inst, 2008. 100(22): p. 1630-42. 201. van Meeteren, L.A., et al., Autotaxin, a secreted lysophospholipase D, is essential for blood vessel formation during development. Mol Cell Biol, 2006. 26(13): p. 5015-22. 202. Kazantseva, A., et al., Human hair growth deficiency is linked to a genetic defect in the phospholipase gene LIPH. Science, 2006. 314(5801): p. 982-5. 203. Tariq, M., et al., A novel mutation in the Lipase H gene underlies autosomal recessive hypotrichosis and woolly hair. Sci Rep, 2012. 2: p. 730. 204. Inoue, A., et al., LPA-producing enzyme PA-PLA(1)alpha regulates hair follicle development by modulating EGFR signalling. EMBO J, 2011. 30(20): p. 4248-60. 205. Berger, J. and D.E. Moller, The mechanisms of action of PPARs. Annu Rev Med, 2002. 53: p. 409-35. 206. McIntyre, T.M., et al., Identification of an intracellular receptor for lysophosphatidic acid (LPA): LPA is a transcellular PPARgamma agonist. Proc Natl Acad Sci U S A, 2003. 100(1): p. 131-6. 207. Zhang, C., et al., Lysophosphatidic acid induces neointima formation through PPARgamma activation. J Exp Med, 2004. 199(6): p. 763-74. 208. Stapleton, C.M., et al., Lysophosphatidic acid activates peroxisome proliferator activated receptor-gamma in CHO cells that over-express glycerol 3-phosphate acyltransferase-1. PLoS One, 2011. 6(4): p. e18932. 209. Rodway, H.A., et al., Lysophosphatidic acid attenuates the cytotoxic effects and degree of peroxisome proliferator-activated receptor gamma activation induced by 15- deoxyDelta12,14-prostaglandin J2 in neuroblastoma cells. Biochem J, 2004. 382(Pt 1): p. 83- 91. 210. Nieto-Posadas, A., et al., Lysophosphatidic acid directly activates TRPV1 through a C-terminal binding site. Nat Chem Biol, 2011. 8(1): p. 78-85. 211. Rai, V., et al., Lysophosphatidic acid targets vascular and oncogenic pathways via RAGE signaling. J Exp Med, 2012. 209(13): p. 2339-50. 212. Ohta, H., et al., Ki16425, a subtype-selective antagonist for EDG-family lysophosphatidic acid receptors. Mol Pharmacol, 2003. 64(4): p. 994-1005. 213. Hama, K., et al., Lysophosphatidic acid and autotaxin stimulate cell motility of neoplastic and non-neoplastic cells through LPA1. J Biol Chem, 2004. 279(17): p. 17634-9.

121

214. Yamada, T., et al., Lysophosphatidic acid stimulates the proliferation and motility of malignant pleural mesothelioma cells through lysophosphatidic acid receptors, LPA1 and LPA2. Cancer Sci, 2008. 99(8): p. 1603-10. 215. Moughal, N.A., et al., Protean agonism of the lysophosphatidic acid receptor-1 with Ki16425 reduces nerve growth factor-induced neurite outgrowth in pheochromocytoma 12 cells. J Neurochem, 2006. 98(6): p. 1920-9. 216. Durgam, G.G., et al., Synthesis, structure-activity relationships, and biological evaluation of fatty alcohol phosphates as lysophosphatidic acid receptor ligands, activators of PPARgamma, and inhibitors of autotaxin. J Med Chem, 2005. 48(15): p. 4919-30. 217. Heise, C.E., et al., Activity of 2-substituted lysophosphatidic acid (LPA) analogs at LPA receptors: discovery of a LPA1/LPA3 receptor antagonist. Mol Pharmacol, 2001. 60(6): p. 1173-80. 218. David, M., et al., Lysophosphatidic acid receptor type 1 (LPA1) plays a functional role in osteoclast differentiation and bone resorption activity. J Biol Chem, 2014. 289(10): p. 6551- 64. 219. Xu, Y.J., et al., Mechanisms of the lysophosphatidic acid-induced increase in [Ca(2+)](i) in skeletal muscle cells. J Cell Mol Med, 2008. 12(3): p. 942-54. 220. Swaney, J.S., et al., A novel, orally active LPA(1) receptor antagonist inhibits lung fibrosis in the mouse bleomycin model. Br J Pharmacol, 2010. 160(7): p. 1699-713. 221. Swaney, J.S., et al., Pharmacokinetic and pharmacodynamic characterization of an oral lysophosphatidic acid type 1 receptor-selective antagonist. J Pharmacol Exp Ther, 2011. 336(3): p. 693-700. 222. Westermann, A.M., et al., Malignant effusions contain lysophosphatidic acid (LPA)-like activity. Ann Oncol, 1998. 9(4): p. 437-42. 223. Mills, G.B., et al., Ascitic fluid from human ovarian cancer patients contains growth factors necessary for intraperitoneal growth of human ovarian adenocarcinoma cells. J Clin Invest, 1990. 86(3): p. 851-5. 224. Mills, G.B., et al., A putative new growth factor in ascitic fluid from ovarian cancer patients: identification, characterization, and mechanism of action. Cancer Res, 1988. 48(5): p. 1066- 71. 225. Meng, Y., S. Kang, and D.A. Fishman, Lysophosphatidic acid inhibits anti-Fas-mediated apoptosis enhanced by actin depolymerization in epithelial ovarian cancer. FEBS Lett, 2005. 579(5): p. 1311-9.

122

226. Frankel, A. and G.B. Mills, Peptide and lipid growth factors decrease cis- diamminedichloroplatinum-induced cell death in human ovarian cancer cells. Clin Cancer Res, 1996. 2(8): p. 1307-13. 227. Bian, D., et al., Lysophosphatidic Acid Stimulates Ovarian Cancer Cell Migration via a Ras-MEK Kinase 1 Pathway. Cancer Res, 2004. 64(12): p. 4209-17. 228. Hu, Y.L., et al., Lysophosphatidic acid induction of vascular endothelial growth factor expression in human ovarian cancer cells. J Natl Cancer Inst, 2001. 93(10): p. 762-8. 229. Shida, D., et al., Lysophosphatidic acid (LPA) enhances the metastatic potential of human colon carcinoma DLD1 cells through LPA1. Cancer Res, 2003. 63(7): p. 1706-11. 230. Fishman, D.A., et al., Lysophosphatidic acid promotes matrix metalloproteinase (MMP) activation and MMP-dependent invasion in ovarian cancer cells. Cancer Res, 2001. 61(7): p. 3194-9. 231. Pustilnik, T.B., et al., Lysophosphatidic acid induces urokinase secretion by ovarian cancer cells. Clin Cancer Res, 1999. 5(11): p. 3704-10. 232. Park, S.Y., et al., Lysophosphatidic acid augments human hepatocellular carcinoma cell invasion through LPA1 receptor and MMP-9 expression. Oncogene, 2011. 30(11): p. 1351-9. 233. Nakanaga, K., K. Hama, and J. Aoki, Autotaxin--an LPA producing enzyme with diverse functions. J Biochem, 2010. 148(1): p. 13-24. 234. Shida, D., et al., Aberrant expression of lysophosphatidic acid (LPA) receptors in human colorectal cancer. Lab Invest, 2004. 84(10): p. 1352-62. 235. Schulte, K.M., et al., Lysophosphatidic acid, a novel lipid growth factor for human thyroid cells: over-expression of the high-affinity receptor edg4 in differentiated thyroid cancer. Int J Cancer, 2001. 92(2): p. 249-56. 236. Goetzl, E.J., et al., Distinctive expression and functions of the type 4 endothelial differentiation gene-encoded G protein-coupled receptor for lysophosphatidic acid in ovarian cancer. Cancer Res, 1999. 59(20): p. 5370-5. 237. Liu, S., et al., Expression of autotaxin and lysophosphatidic acid receptors increases mammary tumorigenesis, invasion, and metastases. Cancer Cell, 2009. 15(6): p. 539-50. 238. Komatsu, J., et al., The signaling pathways linking to lysophosphatidic acid-promoted meiotic maturation in mice. Life Sci, 2006. 79(5): p. 506-11. 239. Kunikata, K., et al., Effect of lysophosphatidic acid on the ovum transport in mouse oviducts. Life Sci, 1999. 65(8): p. 833-40. 240. Garbi, M., et al., Activation of protein kinase calpha in the lysophosphatidic acid-induced bovine sperm acrosome reaction and phospholipase D1 regulation. Biol Reprod, 2000. 63(5): p. 1271-7.

123

241. Li, L.X., et al., [Expression and significance of Edg4 and Edg7 in the placentas of patients with hypertensive disorder complicating pregnancy]. Zhonghua Fu Chan Ke Za Zhi, 2007. 42(6): p. 386-9. 242. Tokumura, A., et al., Altered activity of lysophospholipase D, which produces bioactive lysophosphatidic acid and choline, in serum from women with pathological pregnancy. Mol Hum Reprod, 2009. 15(5): p. 301-10. 243. Tokumura, A., et al., Stimulatory effect of lysophosphatidic acids on uterine smooth muscles of non-pregant rats. Arch Int Pharmacodyn Ther, 1980. 245(1): p. 74-83. 244. Rother, E., et al., Subtype-selective antagonists of lysophosphatidic Acid receptors inhibit platelet activation triggered by the lipid core of atherosclerotic plaques. Circulation, 2003. 108(6): p. 741-7. 245. Rizza, C., et al., Lysophosphatidic acid as a regulator of endothelial/leukocyte interaction. Lab Invest, 1999. 79(10): p. 1227-35. 246. Zhou, Z., et al., Lipoprotein-derived lysophosphatidic acid promotes atherosclerosis by releasing CXCL1 from the endothelium. Cell Metab, 2011. 13(5): p. 592-600. 247. Lee, H., et al., Lysophospholipids increase ICAM-1 expression in HUVEC through a Gi- and NF- kappaB-dependent mechanism. Am J Physiol Cell Physiol, 2004. 287(6): p. C1657-66. 248. Gustin, C., M. Van Steenbrugge, and M. Raes, LPA modulates monocyte migration directly and via LPA-stimulated endothelial cells. Am J Physiol Cell Physiol, 2008. 295(4): p. C905-14. 249. Fueller, M., et al., Activation of human monocytic cells by lysophosphatidic acid and sphingosine-1-phosphate. Cell Signal, 2003. 15(4): p. 367-75. 250. Chang, C.L., et al., Lysophosphatidic acid-induced oxidized low-density lipoprotein uptake is class A scavenger receptor-dependent in macrophages. Prostaglandins Other Lipid Mediat, 2008. 87(1-4): p. 20-5. 251. Olorundare, O.E., et al., Assembly of a fibronectin matrix by adherent platelets stimulated by lysophosphatidic acid and other agonists. Blood, 2001. 98(1): p. 117-24. 252. Retzer, M. and M. Essler, Lysophosphatidic acid-induced platelet shape change proceeds via Rho/Rho kinase-mediated myosin light-chain and moesin phosphorylation. Cell Signal, 2000. 12(9-10): p. 645-8. 253. Haseruck, N., et al., The plaque lipid lysophosphatidic acid stimulates platelet activation and platelet-monocyte aggregate formation in whole blood: involvement of P2Y1 and P2Y12 receptors. Blood, 2004. 103(7): p. 2585-92. 254. Gerrard, J.M., et al., Lysophosphatidic acids. Influence on platelet aggregation and intracellular calcium flux. Am J Pathol, 1979. 96(2): p. 423-38.

124

255. Siess, W., et al., Lysophosphatidic acid mediates the rapid activation of platelets and endothelial cells by mildly oxidized low density lipoprotein and accumulates in human atherosclerotic lesions. Proc Natl Acad Sci U S A, 1999. 96(12): p. 6931-6. 256. Maschberger, P., et al., Mildly oxidized low density lipoprotein rapidly stimulates via activation of the lysophosphatidic acid receptor Src family and Syk tyrosine kinases and Ca2+ influx in human platelets. J Biol Chem, 2000. 275(25): p. 19159-66. 257. Damirin, A., et al., Role of lipoprotein-associated lysophospholipids in migratory activity of coronary artery smooth muscle cells. Am J Physiol Heart Circ Physiol, 2007. 292(5): p. H2513- 22. 258. Zhou, Z.B., J.P. Niu, and Z.J. Zhang, Receptor-mediated vascular smooth muscle migration induced by LPA involves p38 mitogen-activated protein kinase pathway activation. Int J Mol Sci, 2009. 10(7): p. 3194-208. 259. Kim, J., J.R. Keys, and A.D. Eckhart, Vascular smooth muscle migration and proliferation in response to lysophosphatidic acid (LPA) is mediated by LPA receptors coupling to Gq. Cell Signal, 2006. 18(10): p. 1695-701. 260. Komachi, M., et al., Signaling pathways involved in DNA synthesis and migration in response to lysophosphatidic acid and low-density lipoprotein in coronary artery smooth muscle cells. Vascul Pharmacol, 2009. 50(5-6): p. 178-84. 261. Subramanian, P., et al., Lysophosphatidic acid receptors LPA1 and LPA3 promote CXCL12- mediated smooth muscle progenitor cell recruitment in neointima formation. Circ Res, 2010. 107(1): p. 96-105. 262. Fan, H., et al., Lysophosphatidic acid inhibits bacterial endotoxin-induced pro-inflammatory response: potential anti-inflammatory signaling pathways. Mol Med, 2008. 14(7-8): p. 422-8. 263. Murch, O., M. Collin, and C. Thiemermann, Lysophosphatidic acid reduces the organ injury caused by endotoxemia-a role for G-protein-coupled receptors and peroxisome proliferator- activated receptor-gamma. Shock, 2007. 27(1): p. 48-54. 264. Ahn, W.G., et al., Alteration of Lysophosphatidylcholine-Related Metabolic Parameters in the Plasma of Mice with Experimental Sepsis. Inflammation, 2016. 265. Zhao, C., et al., TNF-alpha promotes LPA1- and LPA3-mediated recruitment of leukocytes in vivo through CXCR2 ligand chemokines. J Lipid Res, 2011. 52(7): p. 1307-18. 266. Goetzl, E.J., Y. Kong, and J.K. Voice, Cutting edge: differential constitutive expression of functional receptors for lysophosphatidic acid by human blood lymphocytes. J Immunol, 2000. 164(10): p. 4996-9. 267. Valet, P., et al., Alpha2-adrenergic receptor-mediated release of lysophosphatidic acid by adipocytes. A paracrine signal for preadipocyte growth. J Clin Invest, 1998. 101(7): p. 1431-8.

125

268. Ferry, G., et al., Autotaxin is released from adipocytes, catalyzes lysophosphatidic acid synthesis, and activates preadipocyte proliferation. Up-regulated expression with adipocyte differentiation and obesity. J Biol Chem, 2003. 278(20): p. 18162-9. 269. Gesta, S., et al., Secretion of a lysophospholipase D activity by adipocytes: involvement in lysophosphatidic acid synthesis. J Lipid Res, 2002. 43(6): p. 904-10. 270. Dusaulcy, R., et al., Adipose-specific disruption of autotaxin enhances nutritional fattening and reduces plasma lysophosphatidic acid. J Lipid Res, 2011. 52(6): p. 1247-55. 271. Boucher, J., et al., Potential involvement of adipocyte insulin resistance in obesity-associated up-regulation of adipocyte lysophospholipase D/autotaxin expression. Diabetologia, 2005. 48(3): p. 569-77. 272. Rancoule, C., et al., Depot-specific regulation of autotaxin with obesity in human adipose tissue. J Physiol Biochem, 2012. 68(4): p. 635-44. 273. Pages, C., et al., Endothelial differentiation gene-2 receptor is involved in lysophosphatidic acid-dependent control of 3T3F442A preadipocyte proliferation and spreading. J Biol Chem, 2001. 276(15): p. 11599-605. 274. Rancoule, C., et al., Lysophosphatidic acid impairs glucose homeostasis and inhibits insulin secretion in high-fat diet obese mice. Diabetologia, 2013. 56(6): p. 1394-402. 275. Rancoule, C., et al., Pro-fibrotic activity of lysophosphatidic acid in adipose tissue: in vivo and in vitro evidence. Biochim Biophys Acta, 2014. 1841(1): p. 88-96. 276. Hammack, B.N., et al., Proteomic analysis of multiple sclerosis cerebrospinal fluid. Mult Scler, 2004. 10(3): p. 245-60. 277. Ma, L., et al., Evidence for de novo synthesis of lysophosphatidic acid in the spinal cord through phospholipase A2 and autotaxin in nerve injury-induced neuropathic pain. J Pharmacol Exp Ther, 2010. 333(2): p. 540-6. 278. Yung, Y.C., et al., Lysophosphatidic Acid signaling in the nervous system. Neuron, 2015. 85(4): p. 669-82. 279. Yung, Y.C., et al., Lysophosphatidic acid signaling may initiate fetal hydrocephalus. Sci Transl Med, 2011. 3(99): p. 99ra87. 280. Umemura, K., et al., Autotaxin expression is enhanced in frontal cortex of Alzheimer-type dementia patients. Neurosci Lett, 2006. 400(1-2): p. 97-100. 281. Sayas, C.L., et al., The neurite retraction induced by lysophosphatidic acid increases Alzheimer's disease-like Tau phosphorylation. J Biol Chem, 1999. 274(52): p. 37046-52. 282. Shi, J., et al., Lysophosphatidic acid induces increased BACE1 expression and Abeta formation. Biochim Biophys Acta, 2013. 1832(1): p. 29-38.

126

283. Inoue, M., et al., Autotaxin, a synthetic enzyme of lysophosphatidic acid (LPA), mediates the induction of nerve-injured neuropathic pain. Mol Pain, 2008. 4: p. 6. 284. Inoue, M., et al., Initiation of neuropathic pain requires lysophosphatidic acid receptor signaling. Nat Med, 2004. 10(7): p. 712-8. 285. Ma, L., et al., Evidence for lysophosphatidic acid 1 receptor signaling in the early phase of neuropathic pain mechanisms in experiments using Ki-16425, a lysophosphatidic acid 1 receptor antagonist. J Neurochem, 2009. 109(2): p. 603-10. 286. Balazs, L., et al., Topical application of the phospholipid growth factor lysophosphatidic acid promotes wound healing in vivo. Am J Physiol Regul Integr Comp Physiol, 2001. 280(2): p. R466-72. 287. Sturm, A., et al., Modulation of intestinal epithelial wound healing in vitro and in vivo by lysophosphatidic acid. Gastroenterology, 1999. 117(2): p. 368-77. 288. Xu, K.P., J. Yin, and F.S. Yu, Lysophosphatidic acid promoting corneal epithelial wound healing by transactivation of epidermal growth factor receptor. Invest Ophthalmol Vis Sci, 2007. 48(2): p. 636-43. 289. Gan, L., et al., Blockade of lysophosphatidic acid receptors LPAR1/3 ameliorates lung fibrosis induced by irradiation. Biochem Biophys Res Commun, 2011. 409(1): p. 7-13. 290. Ohashi, T. and T. Yamamoto, Antifibrotic effect of lysophosphatidic acid receptors LPA1 and LPA3 antagonist on experimental murine scleroderma induced by bleomycin. Exp Dermatol, 2015. 24(9): p. 698-702. 291. Oikonomou, N., et al., Pulmonary autotaxin expression contributes to the pathogenesis of pulmonary fibrosis. Am J Respir Cell Mol Biol, 2012. 47(5): p. 566-74. 292. Watanabe, N., et al., Both plasma lysophosphatidic acid and serum autotaxin levels are increased in chronic hepatitis C. J Clin Gastroenterol, 2007. 41(6): p. 616-23. 293. Watanabe, N., et al., Plasma lysophosphatidic acid level and serum autotaxin activity are increased in liver injury in rats in relation to its severity. Life Sci, 2007. 81(12): p. 1009-15. 294. Tangkijvanich, P., et al., Platelet-derived growth factor-BB and lysophosphatidic acid distinctly regulate hepatic myofibroblast migration through focal adhesion kinase. Exp Cell Res, 2002. 281(1): p. 140-7. 295. Ikeda, H., et al., Effects of lysophosphatidic acid on proliferation of stellate cells and hepatocytes in culture. Biochem Biophys Res Commun, 1998. 248(2): p. 436-40. 296. Yanase, M., et al., Lysophosphatidic acid enhances collagen gel contraction by hepatic stellate cells: association with rho-kinase. Biochem Biophys Res Commun, 2000. 277(1): p. 72-8. 297. Bain, G., et al., Selective Inhibition of Autotaxin Is Efficacious in Mouse Models of Liver Fibrosis. J Pharmacol Exp Ther, 2017. 360(1): p. 1-13.

127

298. Castelino, F.V., et al., Amelioration of dermal fibrosis by genetic deletion or pharmacologic antagonism of lysophosphatidic acid receptor 1 in a mouse model of scleroderma. Arthritis Rheum, 2011. 63(5): p. 1405-15. 299. Sasagawa, T., et al., The significance of plasma lysophospholipids in patients with renal failure on hemodialysis. J Nutr Sci Vitaminol (Tokyo), 1998. 44(6): p. 809-18. 300. Cases, A., et al., In vivo evaluation of platelet activation by different cellulosic membranes. Artif Organs, 1997. 21(4): p. 330-4. 301. Zhang, Z.H., et al., Metabolomic Signatures of Chronic Kidney Disease of Diverse Etiologies in the Rats and Humans. J Proteome Res, 2016. 15(10): p. 3802-3812. 302. Shimizu, M., et al., Serum Autotaxin Levels Are Associated with Proteinuria and Kidney Lesions in Japanese Type 2 Diabetic Patients with Biopsy-proven Diabetic Nephropathy. Intern Med, 2016. 55(3): p. 215-21. 303. Eddy, A.A., et al., Investigating mechanisms of chronic kidney disease in mouse models. Pediatr Nephrol, 2012. 27(8): p. 1233-47. 304. Kumagai, N., et al., Mitogenic action of lysophosphatidic acid in proximal tubular epithelial cells obtained from voided human urine. Clin Sci (Lond), 2000. 99(6): p. 561-7. 305. Dixon, R.J. and N.J. Brunskill, Lysophosphatidic acid-induced proliferation in opossum kidney proximal tubular cells: role of PI 3-kinase and ERK. Kidney Int, 1999. 56(6): p. 2064-75. 306. Levine, J.S., et al., Lysophosphatidic acid: a novel growth and survival factor for renal proximal tubular cells. Am J Physiol, 1997. 273(4 Pt 2): p. F575-85. 307. Dixon, R.J., K. Young, and N.J. Brunskill, Lysophosphatidic acid-induced calcium mobilization and proliferation in kidney proximal tubular cells. Am J Physiol, 1999. 276(2 Pt 2): p. F191-8. 308. Heusinger-Ribeiro, J., et al., Expression of connective tissue growth factor in human renal fibroblasts: regulatory roles of RhoA and cAMP. J Am Soc Nephrol, 2001. 12(9): p. 1853-61. 309. Geng, H., et al., Lysophosphatidic acid increases proximal tubule cell secretion of profibrotic cytokines PDGF-B and CTGF through LPA2- and Galphaq-mediated Rho and alphavbeta6 integrin-dependent activation of TGF-beta. Am J Pathol, 2012. 181(4): p. 1236-49. 310. Suzuki, Y., et al., Stimulation of phosphatidylcholine biosynthesis in diabetic hypertrophic kidneys. Endocrinology, 1987. 120(2): p. 595-601. 311. Grove, K.J., et al., Diabetic nephropathy induces alterations in the glomerular and tubule lipid profiles. J Lipid Res, 2014. 55(7): p. 1375-85. 312. Furuya, Y., et al., Increased glomerular cytosolic phospholipase A2 activity of OLETF rats with early diabetes. Exp Clin Endocrinol Diabetes, 1999. 107(5): p. 299-305.

128

313. Vlachojannis, G.J., et al., Enhanced expression of group IIA secreted phospholipase A2 by elevated glucose levels in cytokine-stimulated rat mesangial cells and in kidneys of diabetic rats. Clin Nephrol, 2005. 63(5): p. 356-67. 314. Haneda, M., et al., Mitogen-activated protein kinase cascade is activated in glomeruli of diabetic rats and glomerular mesangial cells cultured under high glucose conditions. Diabetes, 1997. 46(5): p. 847-53. 315. Hayama, M., et al., High glucose-induced cytosolic phospholipase A2 activation responsible for eicosanoid production in rat mesangial cells. J Biochem, 1997. 122(6): p. 1196-201. 316. Saulnier-Blache, J.S., et al., Urinary lysophopholipids are increased in diabetic patients with nephropathy. J Diabetes Complications, 2017. 31(7): p. 1103-1108. 317. Xing, Y., et al., Cell density-dependent expression of EDG family receptors and mesangial cell proliferation: role in lysophosphatidic acid-mediated cell growth. Am J Physiol Renal Physiol, 2004. 287(6): p. F1250-7. 318. Gaits, F., J.P. Salles, and H. Chap, Dual effect of lysophosphatidic acid on proliferation of glomerular mesangial cells. Kidney Int, 1997. 51(4): p. 1022-7. 319. Inoue, C.N., et al., Lysophosphatidic acid and platelet-derived growth factor synergistically stimulate growth of cultured rat mesangial cells. Proc Soc Exp Biol Med, 1997. 216(3): p. 370- 9. 320. Inoue, C.N., et al., Bimodal effects of platelet-derived growth factor on rat mesangial cell proliferation and death, and the role of lysophosphatidic acid in cell survival. Clin Sci (Lond), 2001. 101(1): p. 11-9. 321. Inoue, C.N., H.G. Forster, and M. Epstein, Effects of lysophosphatidic acid, a novel lipid mediator, on cytosolic Ca2+ and contractility in cultured rat mesangial cells. Circ Res, 1995. 77(5): p. 888-96. 322. Goppelt-Struebe, M., et al., Regulation of connective tissue growth factor (ccn2; ctgf) gene expression in human mesangial cells: modulation by HMG CoA reductase inhibitors (statins). Mol Pathol, 2001. 54(3): p. 176-9. 323. Hahn, A., et al., Induction of connective tissue growth factor by activation of heptahelical receptors. Modulation by Rho proteins and the actin cytoskeleton. J Biol Chem, 2000. 275(48): p. 37429-35. 324. Li, H.Y., et al., Blocking lysophosphatidic acid receptor 1 signaling inhibits diabetic nephropathy in db/db mice. Kidney Int, 2017. 325. Yamamoto, Y., et al., Development and prevention of advanced diabetic nephropathy in RAGE-overexpressing mice. J Clin Invest, 2001. 108(2): p. 261-8.

129

326. Wendt, T.M., et al., RAGE drives the development of glomerulosclerosis and implicates podocyte activation in the pathogenesis of diabetic nephropathy. Am J Pathol, 2003. 162(4): p. 1123-37. 327. Grantham, J.J., Clinical practice. Autosomal dominant polycystic kidney disease. N Engl J Med, 2008. 359(14): p. 1477-85. 328. Wyatt, C.M., P.E. Klotman, and V.D. D'Agati, HIV-associated nephropathy: clinical presentation, pathology, and epidemiology in the era of antiretroviral therapy. Semin Nephrol, 2008. 28(6): p. 513-22. 329. Ayasolla, K.R., et al., Tubular cell phenotype in HIV-associated nephropathy: role of phospholipid lysophosphatidic acid. Exp Mol Pathol, 2015. 99(1): p. 109-15. 330. de Vries, B., et al., Lysophosphatidic acid prevents renal ischemia-reperfusion injury by inhibition of apoptosis and complement activation. Am J Pathol, 2003. 163(1): p. 47-56. 331. Gao, J., et al., Lysophosphatidic acid and lovastatin might protect kidney in renal I/R injury by downregulating MCP-1 in rat. Ren Fail, 2011. 33(8): p. 805-10. 332. Okusa, M.D., et al., Selective blockade of lysophosphatidic acid LPA3 receptors reduces murine renal ischemia-reperfusion injury. Am J Physiol Renal Physiol, 2003. 285(3): p. F565- 74. 333. Chen, L., et al., Role of TRPV1 channels in ischemia/reperfusion-induced acute kidney injury. PLoS One, 2014. 9(10): p. e109842. 334. Ueda, K., et al., Preventive effect of SA13353 [1-[2-(1-adamantyl)ethyl]-1-pentyl-3-[3-(4- pyridyl)propyl]urea], a novel transient receptor potential vanilloid 1 agonist, on ischemia/reperfusion-induced renal injury in rats. J Pharmacol Exp Ther, 2009. 329(1): p. 202- 9. 335. Zhang, Y. and A.R. Kompa, A practical guide to subtotal nephrectomy in the rat with subsequent methodology for assessing renal and cardiac function. Nephrology (Carlton), 2014. 19(9): p. 552-61. 336. Imai, A., et al., A gonadotropin-releasing hormone-responsive phosphatase hydrolyses lysophosphatidic acid within the plasma membrane of ovarian cancer cells. J Clin Endocrinol Metab, 2000. 85(9): p. 3370-5. 337. Alcantara-Hernandez, R., et al., Phosphorylation and Internalization of Lysophosphatidic Acid Receptors LPA1, LPA2, and LPA3. PLoS One, 2015. 10(10): p. e0140583. 338. Avendano-Vazquez, S.E., A. Garcia-Caballero, and J.A. Garcia-Sainz, Phosphorylation and desensitization of the lysophosphatidic acid receptor LPA1. Biochem J, 2005. 385(Pt 3): p. 677-84.

130

339. Colin-Santana, C.C., et al., EGF and angiotensin II modulate lysophosphatidic acid LPA(1) receptor function and phosphorylation state. Biochim Biophys Acta, 2011. 1810(12): p. 1170- 7. 340. Erkan, E., et al., Distinct urinary lipid profile in children with focal segmental glomerulosclerosis. Pediatr Nephrol, 2016. 31(4): p. 581-8. 341. Hara, S., et al., Podocyte injury-driven lipid peroxidation accelerates the infiltration of glomerular foam cells in focal segmental glomerulosclerosis. Am J Pathol, 2015. 185(8): p. 2118-31. 342. Mizobata, R., [Effects of lysophosphatidylcholine on expression of monocyte chemoattractant protein-1 in glomerular endothelial cells]. Nihon Jinzo Gakkai Shi, 2003. 45(2): p. 76-83. 343. Chen, H., et al., Lysophosphatidic Acid Pretreatment Attenuates Myocardial Ischemia/Reperfusion Injury in the Immature Hearts of Rats. Front Physiol, 2017. 8: p. 153. 344. He, D., et al., Lysophosphatidic acid enhances pulmonary epithelial barrier integrity and protects endotoxin-induced epithelial barrier disruption and lung injury. J Biol Chem, 2009. 284(36): p. 24123-32. 345. Heyman, S.N., S. Rosen, and C. Rosenberger, Animal models of renal dysfunction: acute kidney injury. Expert Opin Drug Discov, 2009. 4(6): p. 629-41. 346. Ortiz, A., et al., Translational value of animal models of kidney failure. Eur J Pharmacol, 2015. 759: p. 205-20. 347. Minnear, F.L., et al., Platelet lipid(s) bound to albumin increases endothelial electrical resistance: mimicked by LPA. Am J Physiol Lung Cell Mol Physiol, 2001. 281(6): p. L1337-44. 348. Yin, F. and M.A. Watsky, LPA and S1P increase corneal epithelial and endothelial cell transcellular resistance. Invest Ophthalmol Vis Sci, 2005. 46(6): p. 1927-33. 349. Dufner-Beattie, J., R.S. Lemons, and A. Thorburn, Retinoic acid-induced expression of autotaxin in N-myc-amplified neuroblastoma cells. Mol Carcinog, 2001. 30(4): p. 181-9. 350. Ptaszynska, M.M., et al., Positive feedback between vascular endothelial growth factor-A and autotaxin in ovarian cancer cells. Mol Cancer Res, 2008. 6(3): p. 352-63. 351. Navab, M., et al., Source and role of intestinally derived lysophosphatidic acid in dyslipidemia and atherosclerosis. J Lipid Res, 2015. 56(4): p. 871-87.

131