Thèse De Doctorat Présentée En Vue De L’Obtention Du Grade De Docteur En Interactions"Moléculaire"Et"Cellulaire"

École doctorale sciences de la vie et de la santé Unité de recherche : CNRS UMR 7284 INSERM U 1081 UNS

Thèse de doctorat Présentée en vue de l’obtention du grade de docteur en Interactions"moléculaire"et"cellulaire"

de L’UNIVERSITE COTE D’AZUR

par Papa Diogop Ndiaye

RÔLE%ET%REGULATION%DU%VEGF0C%DANS%LES% CANCERS%DU%REIN%A%CELLULES%CLAIRES%

Dirigée par Dr Gilles Pagés

Soutenue le 08 Décembre 2017 Devant le jury composé de :

Barbara%Garmy0Susini%%%%%CR1, Université Toulouse Midi-Pyrénées Rapporteur Fabrice Soncin DR2, Université de Lille 1 Science et technologie Rapporteur Gilles Pagés DR1, Université de Nice Sophia Antipolis Directeur de thèse Marc POCARD PuPH, Université Paris Denis-Diderot Examinateur Patrick Auberger DR1, Université de Nice Sophia Antipolis Président

RESUME&DE&THESE& ! ! Introduction!:!Les!zones!hypoxiques!sont!des!caractéristiques!communes!des!tumeurs!métastatiques.! Le!carcinome!à!cellules!rénales!(RCC)!exprimant!le!facteur!inductible!de!l'hypoxie!(HIF)!en!raison!de! l'inactivation! du! gène! de! von! Hippel! Lindau! (vhl),! représente! un! modèle! d'hypoxie! chronique.! Le! devenir! des! patients! dépend! du! stade! de! dissémination! des! cellules! tumorales.! Par! conséquent,! déchiffrer! les! mécanismes! de! ! métastase! est! une! préoccupation! majeure.! Le! développement! dépendant! du! VEGFMC! (Vascular! Endothelial! Growth! Factor! C)! d'un! réseau! lymphatique! est! en! première!ligne!de!propagation!métastatique.!! ! Objectifs!:!Démontrer!que!le!VEGFMC!est!régulé!par!l’hypoxie!et!étudier!la!régulation!du!VEGFMC!dans! ce!contexte.!Ensuite!étudier!le!rôle!du!VEGFMC!dans!la!tumorigenèse!des!ccRCCs.! ! Résultats&:! Pour! étudier! le! rôle! de! VEGFC! dans! la! dissémination! du! RCC,! nous! avons! étudié! son! expression! dans! l'hypoxie! et! nous! avons! invalidé! son! gène! dans! des! lignées! cellulaires! modèles! de! souris!humaines!et!murines.!L'hypoxie!régule!négativement!l'ARNm!de!VEGFC!par!une!diminution!de! la!transcription!et!de!la!stabilité!de!l'ARNm!mais!l'expression!de!la!protéine!VEGFMC!est!induite!par! l’hypoxie.! Des! capacités! accrues! de! prolifération! et! de! migration,! une! suractivation! de! la! voie! de! signalisation! AKT! et! une! meilleure! expression! des! marqueurs! mésenchymateux! et! des! marqueurs! souches!caractérisent!les!cellules!vegfMc!M/M.!Alors!que!les!cellules!vegfc!M/M!ne!forment!pas!de!tumeurs! chez! les! souris! immunodéficientes,! elles! développent! des! tumeurs! agressives! chez! les! souris! immunocompétentes.! De! plus,! les! cellules! RCC! de! souris! génèrent! des! tumeurs! à! croissance! rapide! chez!les!souris!invalidées!pour!six1!ou!eya2,!deux!régulateurs!majeurs!de!l'expression!de!VEGFC.!La! surexpression!des!marqueurs!lymphangiogéniques,!y!compris!le!VEGFC,!est!liée!à!une!augmentation! de!la!survie!sans!progression!et!globale!chez!les!patients!atteints!de!tumeurs!non!métastatiques!alors! qu'une! diminution! de! la! survie! sans! progression! et! globale! est! observée! chez! les! patients! métastatiques.!! ! Conclusion!:!Nos!expériences!décrivent!une!régulation!subtile!du!VEGFMC!par!hypoxie!et!mettent!en! évidence!son!rôle!bénéfique!ou!péjoratif.!Par!conséquent,!le!ciblage!VEGFMC!pour!la!thérapie!doit!être! considéré!avec!prudence.! REMERCIEMENTS!

Ce"manuscrit"a"pour"but"de"résumer"les"trois"années"que"j’ai"passées"dans"l’équipe"du"Dr" Gilles"Pagès"au"sein"de"L’Institut"de"Recherche"sur"Cancer"et"le"Vieillissement"de"Nice" (IRCAN)."Le"but"du"jeu"est"d’y"mettre"ces"résultats"scientifiques"mais"si"je"devais"retenir" qu’une"chose"de"ces"trois"années"cela"serait"les"rencontres."Il"m’est"donc"impossible"de" finir"cette"thèse"sans"remercier"toutes"les"personnes"rencontrées."" " Mais"tout"d’abord"je"vais"remercier"les"membres"du"jury"qui"ont"accepté"d’évaluer"mon" travail"de"thèse"et"qui"pour"certains"sont"venus"de"très"loin."" " Gilles,"merci"tout"d’abord"de"m’avoir"pris"dans"ton"équipe."Tu"m’as"donné"la"chance"de" pouvoir" effectuer" ma" thèse" malgré" mon" parcours" atypique." L’adaptation" a" été" difficile" pour"moi"en"passant"de"Baharia"à"toi"mais"j’ai"survécu."Avec"toi"j’ai"beaucoup"appris…" de"blagues"sur"les"blondes,"à"ne"pas"manger"avant"midi"et"demi"et"plus"de"choses"qui" resteront" à" jamais" dans" ma" tête." Oui" je" retiens" mieux" les" choses" inutiles." Merci" de" m’avoir" laissé" la" liberté" nécessaire" pour" développer" mon" sujet" de" thèse" et" de" m’avoir" soutenu"durant"tout"ce"temps."Tu"as"été"un"directeur"de"thèse"super"cool"et"aussi"une" belle"personne"qui"prend"soin"des"gens"autour"de"lui."" " Yann,"si"j’ai"fait"cette"thèse"c’est"aussi"grâce"à"toi."Même"si"tu"es"loin,"je"ne"t’oublie"pas." J’ai"beaucoup"apprécié"ta"compagnie."Malgré"le"fait"que"tu"es"une"«"tête"de"con"»,"tu"es"un" vrai"pote."Tu"me"manques."Je"rigole"moins"au"foot"sans"toi."Merci"pour"tout"mec."" " Joff," toi" aussi" tu" as" été" un" acteur" important" dans" mon" recrutement." Je" ne" t’oublie" pas" capitaine."Même"si"tu"es"un"jour"de"niveau"ligue"j’ai"quand"même"joué"avec"toi"et"ça"c’est" cool." Merci" pour" ta" gentillesse" et" tes" conseils" au" labo" comme" au" foot." Dorothée," tu" es" aussi"inclue."" " Julien"Parola,"Sandy,"Maeva"vous"m’avez"intégré"dans"l’équipe"sans"problème."On"a"passé" de"bons"moments"à"la"paillasse"et"aussi"en"dehors"du"laboratoire."Ce"fût"un"plaisir"de" travailler"avec"vous."Les"conditions"de"travail"peuvent"rendre"une"thèse"très"dure"à"vivre" mais"avec"vous,"ces"trois"ans"étaient"aussi"bons"que"du"«"tataki"»."Merci"aussi"pour"votre" aide"dans"le"projet." " Je" remercie" aussi" les" autres" membres" de" l’équipe" Pagès" qui" nous" ont" rejoint":" Sonia" Banana,"Yannick,"Audrey"ma"médecin"préférée,"Jérôme"et"Marilena"qui"n’est"plus"avec" nous"au"labo."Sonia,"tes"chansons"ne"vont"pas"me"manquer"mais"ta"joie"de"vivre"et"ton" sourire"oui."" " Je"n’oublie"pas"toute"l’équipe"Pouysségur"à"commencer"par"Jacques."Je"pense"que"j’ai"de" la"chance"d’avoir"côtoyer"un"scientifique"d’une"telle"qualité."J’ai"beaucoup"appris"à"ses" cotés."Merci"Nathalie"pour"tes"conseils"qui"m’ont"beaucoup"aidé"tout"au"long"de"ma"thèse" et"ton"accessibilité."Philippe"et"Roser,"vous"m’avez"beaucoup"aidé"professionnellement"et" aussi" personnellement." Merci" pour" tout." Cercina," merci" pour" ta" gentillesse" et" tes" conseils";"cela"m’a"beaucoup"aidé."Masa"et"Lucilla,"membre"de"la"famille"Banana"merci" d’avoir"été"là."Même"sans"fenêtre,"j’avais"une"belle"vue."C’est"toujours"bon"pour"le"moral." Je" n’oublie" pas" nos" médecins" Nirvana" et" Julie." Pour" finir," Ibtissam," je" te" remercie" 2" "" beaucoup"pour"ta"compréhension"tes"conseils"ton"soutien"au"labo"et"en"dehors."Tu"es" une"belle"personne."" Je" remercie" aussi" les" stagiaires" qui" sont" passés" par" le" labo" durant" ces" 3" ans" comme" Lorenza," Giulia," Natacha," Alvaro," Monique," Willian" et" une" pensée" spéciale" pour" Janita" avec"qui"j’ai"travaillé"pendant"9"mois"et"qui"m’a"beaucoup"aidé"dans"mon"projet."" " Baharia,"tu"as"joué"un"grand"rôle"dans"mon"parcours"scientifique."Je"te"dois"beaucoup."Tu" as"toujours"été"présente"pour"moi"et"pour"cela"je"ne"pourrais"jamais"te"remercier"assez." Tu"m’as"toujours"soutenu"avant"et"pendant"cette"thèse."Merci"infiniment."" Je"remercie"aussi"toute"l’équipe"Hofman":"Paul"Hofman,"Patrick,"Valerie,"Titou,"Jéremie," Barnabé," Nathalie," Laetitia," Mika," Louis," Simon," Iris," MarieGAngela," Olivia" et" Jonathan" sans" oublier" ceux" qui" sont" partis" " vers" d’autres" aventures" (Amine," Josy" et" Laurie)." Je" remercie" également" les" membres" de" l’équipe" Bulavin" avec" qui" j’ai" partagé" le" labo" pendant"une"partie"de"ma"thèse":"Gwen,"Clément,"Laurent,"Caroline,"Sacha"et"Doria." " Je" remercie" aussi" tous" les" monégasques" du" CSM":" Jérôme" " pour" les" CRISPRs" et" tes" conseils,"Scott,"Vincent,"Valérie,"Milica"et"Renaud."" " Merci"à"tous"mes"partenaires"de"sport"à"commencer"par"Julien"qui"m’a"fait"découvrir"le" Kali" et" qui" est" maintenant" un" ami." Toutes" les" personnes" qui" s’entrainent" avec" nous"maintenant." Les" partenaires" du" Foot" aussi," je" vous" remercie" avec" une" mention" spéciale"pour"Manu"sans"oublier"les"membres"du"botafogo."Le"sport"m’a"beaucoup"aidé" dans"cette"aventure."" " Merci" à" toutes" les" personnes" travaillant" au" CAL" ou" à" la" tour" que" j’ai" découvertes" en" travaillant" avec" Baharia" et" durant" ma" thèse" qui" m’ont" aidé." Les" chercheurs" comme" Pascal"Lopez"et"Didier"les"membres"du"personnel"comme"Wiliam,"Sami,"David"et"Michel."" " Merci" à" la" famille" Gouneaud" pour" leur" soutien" inconditionnel," à" toutes" les" belles" personnes"rencontrées"à"Brignoles,"à"mes"amis"Celia,"Amaury,"Frank,"Audrey,"Amelie"et" Tous"les"membres"de"la"team"Celia"et"la"famille"Huguette." " Je" vais" finir" par" ma" famille" que" je" ne" vais" pas" remercier" car" il" m’est" impossible" de" la" remercier" pour" ce" qu’elle" a" fait" pour" moi" et" pour" que" je" sois" là" où" j’en" suis." J’ai" les" meilleurs"parents"du"monde"qui"m’ont"donné"la"meilleure"éducation"et"je"passerai"ma" vie"à"essayer"de"les"rendre"fier."Ils"m’ont"donné"trois"magnifiques"sœurs"à"qui"j’essaie"de" donner"le"bon"exemple."Je"suis"fier"de"faire"partie"de"cette"famille."" " La"dernière"mais"pas"la"moindre,"Je"remercie"Almut"Brand"qui"s’est"occupée"de"moi"et" sans"cela"tout"aurait"été"beaucoup"plus"difficile."Je"suis"content"que"tu"sois"dans"ma"vie"et" j’espère" pour" longtemps" encore." Merci" pour" ton" soutien," ton" aide," ta" bienveillance" envers"moi,"sans"oublier"les"autres"Brand"pour"leur"soutien"également."" " " " !!

3" "" REMERCIEMENTS!...... !2"

ABBREVIATIONS!...... !7"

INTRODUCTION!...... !12" Le!rein!...... !13" Les"cancers"du"rein"...... "14" Epidémiologie"...... "15" Facteurs"de"risque"...... "15" Diagnostic"...... "16" Pronostic"...... "16" Les"voies"moléculaires"...... "20" L’axe"VHL/HIF/VEGF"...... "20" La"famille"HIFs"...... "21" HIF1"vs"HIF2"...... "22" La"voie"«"mammalian"target"of"rapamycin"(mTOR)"»"...... "23" Les"traitements"...... "25" La"chirurgie"...... "25" L’immunothérapie"de"première"génération"...... "26" Les"thérapies"antiGangiogéniques"...... "27" Les"inhibiteurs"de"mTOR"...... "28" L’immunothérapie"de"nouvelle"génération"...... "28" La"radiothérapie"...... "30" Les"résistances"...... "31" L’activation"d’autres"voies"proGangiogéniques."...... "32" Adaptation"du"microenvironnement"...... "33" Rôle"de"l’hypoxie"...... "33" Résistances"aux"inhibiteurs"de"mTORC1"...... "35" Marqueurs"prédictifs"...... "36" Cytokines"et"facteurs"proGangiogéniques"circulants"...... "37" Les"chimoikines"CXCL"...... "37" Les"chimiokines"CXC"ELR+"...... "37" Les"chimiokines"CXC"ELRG"...... "38" La"vascularisation"des"ccRCCs"...... "39" Le!système!!lymphatique!...... !40" Présentation"...... "40" Origine"...... "41" La"génétique"de"la"lymphangiogenèse"...... "42" Signalisation"du"développement"lymphatique."...... "44" Lymphangiogenèse"et"cancers"...... "45" 4" "" Modification"du"réseau"lymphatique"tumoral"...... "46" Mécanisme"moléculaire"de"la"lymphangiogenèse"tumorale"...... "47" Chimiokines"et"métastases"...... "49" Modification"du"réseau"lymphatique"au"delà"de"la"tumeur"...... "50" Remaniement"lymphatique"et"immunité"tumorale"...... "50" Lymphangiogenèse":"une"cible"anti"tumorale"...... "51" La"régulation"d’expression"du"VEGFGC"...... "52" Les"facteurs"de"transcription"...... "52" Les"cytotkines,"facteurs"de"croissance"et"matrice"extracellulaire"...... "52" Régulation"postGtranscriptionnelle"de"VEGFGC"...... "53" Maturation"de"la"protéine"VEGFGC"...... "53" La"fonction"du"VEGFGC"...... "55" Fonction"canonique"...... "55" Fonctions"non"canoniques"du"VEGFGC"...... "56" EMT!...... !58" La"transition"épithéliumGmésenchyme"...... "58" Les"marqueurs"de"l’EMT"...... "59" Modifications"des"jonctions"...... "59" Modification"du"cytosquelette"et"motilité"...... "60" Les"facteurs"de"transcription"de"l’EMT"...... "60" SNAILS"...... "60" Les"facteurs"de"transcription"basal"helixGloopGhelix"(bHLH)"...... "61" Les"facteurs"de"transcription"ZEB"...... "62" Les"inducteurs"d’EMT"...... "63" Les"protéines"de"la"famille"du"TGFβ"...... "63" Les"récepteurs"tyrosine"kinase"(RTK)"...... "64" Les!cellules!souches!cancéreuses!(CSC)!...... !65" Caractéristiques"des"CSCs"...... "65" Les"voies"de"signalisation"...... "67" La"voie"hedgehog"...... "67" La"voie"du"TGFβ"...... "67" La"voie"Wnt"...... "67" Résistance"aux"traitements,"EMT"et"CSCs"...... "68"

RESULTATS!...... !70"

CONCLUSION!...... !74"

DISCUSSION!...... !80"

PERSPECTIVES!...... !87"

5" "" REFERENCES!...... !90"

ANNEXES!...... !120"

! " !!

6" "" ABBREVIATIONS*"

" " " " " " " " " " " " " " " " " " " " " " " " " " " " " "

7" "" " 4EGBP"1:"4E"Binding"Protein"1"" AAG:"AntiGangiogéniques"" ABC:"ATP"Binding"Cassette" AKT:"Akt"serine"threonine"kinase="PKB"(Protein"Kinase"B)" ALDH1":"Aldéhyde"Déshydrogénase"" AMM":"Autorisation"de"Mise"sur"le"Marché"" Angpt1":"Angiopoiétine"1"" ARNm":"Acide"RiboGNucléique"messager"

BAP1:"BRCA"(breast"cancer)1"associated"proteinG1" bHLH":"basal"HelixGLoopGHelix"" BMDC:"Bone"marrowGderived"cells"" BMP:"Bone"Morphogenetic"Protein"" CCBE1:"CollagenG"and"calciumGbinding"EGF"domain"1"" CCL:"CCGchemokine"Ligand"" CCR:"CCGchemokine"Receptor" ccRCC":"clear"cell"Renal"Cell"Carcinomas" CECs":"Les"cellules"endothéliales"circulantes"" CEP":"Les"cellules"endothéliales"progénitrices!!

CLEC2":"CGType"Lectin"Domain"Family"2"" CMH":"Complexe"Majeur"d’Histocompatibilité"" CoupGTFII:"Chicken"ovalbumin"upstream"promoter"transcription"factor"="Nr2f2" COX:"Cytochrome"C"assembly"oxidase"factor"" CSC:"Cellules"Souches"Cancéreuses"" CSF1:"Colony"Stimulating""Factor"1"" CTBP:"CGterminal"binding"protein"" CTLA4:"Cytotoxic"T"lymphocyteGassociated"protein"4"" EGF:"Epidermal"Growth"Factor" EMT:"Epithelial–Mesenchymal"Transition" EPO:"érythropoïétine"" ERK:"Extracellular"signalGRegulated"Kinase" FGF1:"Fibroblast"Growth"FactorG1""" FGF2:"Fibroblast"Growth"FactorG2"""

8" "" FH:"Furumate"hydratase" FIH:"Factor"Inhibiting"HIFG1"" FLCN:"Folliculine" FOX:"Forkhead"box" GGCSF:"GranulocyteGColony"Stimulating"Factor"" GAG:"Glycosaminoglycanes" GATA:"Guanine"Adenine"Thymine"Adenine" GTP:"Guanosine"TriGPhosphate" HGF:"Hepatocyte"growth"factor"" HIF:"Hypoxia"Inducible"Factor"" HMGB1:"High"Mobility"Group"Box"1"" HRE:"Hypoxia"Response"Element""" ID:"Inhibitor"of"Differentiation"" IL2":"InterleukineG2"" IL6":"Interleukine"6" IMDC":"International"Metastatic"RCC"Database"Consortium""" INFGα":"InterféronGα"" IRM":"Imagerie"à"Résonance"Magnétique"" IRSG1":"Insulin"Receptor"Substrate"1"" LAMP":"Lysosomal"Associated"Membrane"Protein"" LDH":"Lactate"Deshydrogénase" LEC":"Lymphatic"Endothelial"Cell"" LRP":"LDL"Receptor"Related"Protein"" LT:"Lymphocytes"T"" LyveG1:"Lymphatic"Vessel"Endothelial"Hyaluronan"ReceptorG1"" MAP"Kinases:"Mitogen"Activated"ProteinG"kinases" MET:"Met"protoGoncogène="Hepatocyte"Growth"F"actor"Receptor" MMPG9:"Matrix"MetalloGProtéaseG9"" MSKCC:"Memorial"Sloane"Kettering"Cancer"Center"" mTOR":"mammalian"Target"Of"Rapamycin"" NFκB:"Nuclear"Factor"kappa"B"" NK:"Natural"Killer"" NKX:"NK"homeobox"1"

9" "" NRP:"Neuropiline""" PC:"Proprotéines"Convertases"" PD1:"Programmed""cell"Death"protein"1"" PDGF:"Platelet"Derived"Growth"Factor" PDGFR:"Platelet"Derived"Growth"Factor"Receptor" PDK1:"Phospoinositide"Dependent"Kinase"" PDL1:"Programmed""cell"Death"protein"1"Ligand"1"" Pdpn:"Podoplanine"" PFS:"Progression"Free"Survival" PHD:"Prolyl"Hydroxylase"" PI"(3,4,5)"P3:"phosphatidylinositol"(3,4,5)Gtrisphosphate"" PI"(3,4)"P2:"Phosphatidylinositol"(3,4)Gbisphosphate""" PI"(3)"P:"phosphatidylinositol"3Gphosphate"" PI3K:"PhosphatidylinositolG4,5Gbisphosphate"3Gkinase"" PKB:"Protein"Kinase"B"" PPARγ:"Peroxisome"Proliferator"Activated"Receptor"Gamma"" PROKG"2:"Prokineticin"2" PROX1:"Prospero"Homeodomain"1"" PTEN:"Phosphatase"and"TENsin"homolog" Rheb:"Ras"homolog"enriched"in"the"brain" SDF1:"Stromal"cell"Derived"Factor"1" SDH:"Succinate"deshydrogenase"" SMAD:"Sma"and"Mothers"Against"Decapentaplegic"" SOX18:"Sry"Box"18""

STAT3:"Signal"transducer"and"activator"of"transcription"3" TAFs:"Tumor"Associated"Fibroblast"" TAM:"Tumor"Associated"Macrophages"" TGFβ:"Transforming"Growth"Factor"beta" TKI:"Tyrosine"Kinase"inhibitor" TNFα:"Tumor"Necrosis"Factor"alpha"" TNM:"Tumor"Nodus"Metastasis" TSC:"Tuberous"Sclerosis"" ULK1:"UncG51"like"Autophagy"Activating"KinaseG1""

10" " VEGF:"Vascular"Endothelial"Growth"FactorGA" VEGFGC:"Vascular"Endothelial"GrowthGFactorGC" VEGFR:"Vascular"Endothelial"Growth"Factor"Receptor" VHL:"von"Hippel"Lindau"" WDFY:"D"Repeat"and"FYVE"Domain"Containing"" WNT:"Wingless"plus"Int"(Integration"site)"" " " " " " " " " " " " " " " " " " " " " " " " " " !

11" " INTRODUCTION)!

! " "

12" " Le#rein## Le" rein" est" un" organe" qui" fait" partie" du" l’appareil" urinaire" (Figure" 1)." Il" est" situé" en" dessous"des"côtes,"de"part"et"d’autre"de"la"colonne"vertébrale."Les"reins"sont"facilement" reconnaissables"avec"leur"forme"en"haricot."Un"rein"de"taille""normale"mesure"environ" douze" centimètres" de" longueur," trois" centimètres" de" largeur" et" six" centimètres" de" hauteur.""Il"est"composé"de"trois"parties."De"l’intérieur"vers"l’extérieur,"on"retrouve":"le" calice" et" le" bassinet," le" parenchyme" rénal" et" enfin" la" capsule" qui" est" l’enveloppe" protectrice" du" rein." Le" parenchyme" contient" les" néphrons" qui" " constituent" les" unités" fonctionnelles" structurales" du" rein." Ils" filtrent" le" sang" et" produisent" l’urine." Un" rein" contient"plus"d’un"million"de"néphron."Ce"dernier"est"composé"d’un"glomérule,"d’un"tube" proximal,"d’un"tube"intermédiaire,"d’un"tube"distal"et"d’un"segment"d’union"(Figure"1)." Enfin"l’urine"produite"par"les"néphrons"est"véhiculée"vers"la"vessie"via"respectivement" les"calices,"le"bassinet"et"l’urètre."" La" fonction" principale" rénale" est" le" maintien" de" l’équilibre" hydroGélectrique" dans" l’organisme."Le"rein"permet"d’éliminer"les"molécules"et"substances"en"excès"(eau,"sels)" dont"l’accumulation"peut"engendrer"des"effets"néfastes"(déchets"métaboliques)."Il"a"aussi" une"fonction"endocrine."En"effet,"il"est"responsable"de"la"production"de""l’érythropoïétine" (EPO)"et"de"la"rénine"qui,"jouent"un"rôle"respectivement"dans"la"production"de"globules" rouges"et"la"régulation"de"la"tension"artérielle.""" "

Source:(h*p://www.e/cancer.fr/( " Figure!1!:!Schéma!général!du!rein!

13" " !

Les!cancers!du!rein! Il"existe"plusieurs"types"de"cancer"du"rein."" Le"plus"représenté"est"le!carcinome!à!cellules!claires"(ccRCC)."70%"de"ces"cancers"se" développent"à"partie"des"cellules"du"tube"proximal"distal.""Ils"se"caractérisent"par"une" prolifération"de"cellules"au"cytoplasme"vide"au"microscope""optique"d’où"leur"nom"mais" aussi" par" une" mutation" sur" le" gène" von" Hippel" Lindau"(pvhl)." L’inactivation" de" pVHL" induit" une" expression" constitutive" des" facteurs" de" transcription" "Hypoxia" Inducible" Factor" alpha"(HIFα)"" et" par" conséquent" une" expression" accrue" d’un" de" ses" cibles" le""Vascular"Endothelial"Growth"Factor"A"(VEGF)."Le"VEGFGA"est"un"facteur"de"croissance" des" cellules" endothéliales" vasculaires" ce" qui" explique" l’importante" vascularisation" observée"dans"ces""tumeurs."Le"taux"de"survie"à"5"ans"est"variable"selon"le"stade"de"la" maladie."Il"est"de"91,"74,"67"et"32%"respectivement"pour"les"stades"1,"2,"3"et"4"selon"la" classification" TNM" (Tsui" et" al." 2000)." " C’est" le" type" de" cancer" qui" sera" spécialement" étudié"dans"cette"thèse."" Le!carcinome!rénal!à!cellules!papillaires!(pRCC)"représente"15%"des"cancers"du"rein." Il" est" issu" des" cellules" du" tube" contourné" distal." La" survie" à" 5" ans" des" patients" est" meilleure"par"rapport"au"ccRCC"(85,5%"contre"76,9%)"(Steffens"et"al."2012).""Deux"sousG types"de"pRCC"se"distinguent."Le"sousGtype"1"se"caractérise"par"des"cellules"petites"et" cubiques"à"cytoplasme"basophile"avec"un"noyau"peu"nucléolé"et"peu"atypique."Il"est"de" meilleur" pronostic." Dans" le" sousGtype" 2," l’architecture" est" plus" compacte" avec" des" cellules" plus" cylindriques" à" cytoplasme" éosinophile" avec" stratification" nucléaire." Les" noyaux" sont" plus" atypiques" et" nucléolés" et" il" n’existe" que" très" rarement" des" cellules" spumeuses"et"des"calcifications."Son"pronostic"est"plus"péjoratif"que"le"sousGtype"1."" Le!carcinome!à!cellules!chromophobes!(chRCC)"constitue"le"troisième"type"de"cancer" du"rein."Il"provient"des"cellules"intercalaires"du"tube"collecteur."Il"est"caractérisé"par"des" cellules"avec"des"membranes"cytoplasmiques"nettes"et"non"optiquement""vide."La"survie" à" 5" ans" des" chRCC" est" de" 93%." Ces" tumeurs" ont" donc" un" bon" pronostic" (Volpe" et" al." 2012)."" Enfin,"les!oncocytomes"représentent"un"quatrième"type"de"cancer"du"rein."Ces"tumeurs" proviennent"des"cellules"intercalaires"de"type"A."Elles"prolifèrent"très"lentement"et"sont" très"rarement"sinon"jamais"invasives.""

14" " "

Epidémiologie!! Les" cancers" du" rein" représentent" environ" 3%" de" tous" les" cancers" diagnostiqués" chez" l’adulte."Chaque"année,"295000"nouveaux"cas""et"environ"134"000"décès"sont"recensés." L’incidence"des"cancers"du"rein"est"plus"élevée"dans"les"pays"développés"et"touche"deux" fois"plus"les"hommes"que"les"femmes.""Elle""est"plus"élevée"dans"les"pays"baltiques"pour" des"raisons"que"l’on"ignore"pour"le"moment.""

Facteurs!de!risque! L’obésité," le" tabagisme" et" l’hypertension" sont" des" facteurs" de" risque" établis" pour" les" cancers"du"rein."Le"tabac"est"responsable"de"25"à"30%"des"cas"de"survenue"de"la"maladie." Il"augmente"l’incidence"de"50%"et"de"20%"respectivement"chez"l’homme"et"la"femme"de" risque"de"cancer"par"rapport"à"un"individu"nonGfumeur"(Chow,"Dong,"and"Devesa"2010)." Il"existe"plusieurs"hypothèses"expliquant"le"lien"entre"tabagisme"et"cancer"du"rein"mais" aucun"mécanisme"n’est"décrit"à"ce"jour."" L’obésité""constitue"un"facteur"de"risque."Plusieurs"études"ont"montré"une"corrélation" entre"l’indice"de"masse"corporelle"et"l’apparition"de"cancer"de"rein."En"Europe,"on"estime" que"plus"30%"des"cancers"du"rein"serait"liée"à"un"excès"de"poids"(Pischon"et"al."2006)." Les" perturbations" hormonales" induites" par" le" surpoids" expliqueraient" cette" survenue." Plusieurs" études" ont" montré" un" lien" entre" l’hypertension" et" les" ccRCC." L’hypertension" augmente"le"risque"de"cancer"du"rein"aussi"bien"chez"les"hommes"que"chez"les"femmes" (Weikert" et" al." 2008)." L’incidence" de" l’hypertension" est" croissante" dans" la" population" mondiale" (Kearney" et" al." 2005)." Cependant," il" est" difficile" de" corréler" l’incidence" de" l’hypertension" et" des" RCC." Certaines" études" ont" montré" un" lien" entre" d’autres" paramètres"comme"les"désordres""hormonaux,"l’activité"physique,"le"régime"alimentaire" ou" le" style" de" vie" et" les" RCCs" mais" aucune" étude" clinique" n’a" pu" confirmer" ces" observations" jusqu'à" présent" (Chow," Dong," and" Devesa" 2010)." L’insuffisance" rénale" augmente" l’incidence" de" cancer" du" rein." La" dialyse" à" l’origine" d’une" dysplasie" multiG kystique"(apparition"de"kystes)"sur"les"reins,"augmente"le"risque"de"cancer"du"rein"(Hora" et"al."2008)." Des"facteurs"génétiques""augmentent"le"risque"de"développer"un"cancer"du"rein.""Ainsi," 11"gènes"(BAP1,"FLCN,"FH,"MET,"PTEN,"SDHB,"SDHC,"SDHD,"TSC1,"TSC2&et"VHL)&ont"été" identifiés" pour" leur" rôle" dans" le" développement" des" ccRCC" dont" certains" de" façon"

15" " sporadique" (Haas" and" Nathanson" 2014)." " Le" meilleur" exemple" est" celui" du" gène" pvhl" dont" l’invalidation" est" corrélée" avec" une" forte" incidence" de" ccRCC" (Kaelin" 2007)." Une" mutation"sur"le"gène"pvhl"provoque"la"maladie"de"von"HippelGLindau"qui"est"héréditaire" à" transmission" autosomique" dominante." Il" est" le" principal" facteur" de" risque" de" ccRCC."" Les"mutations"sur"pvhl"sont"responsables"de"1"à"2%"des"ccRCC.""Ce"dernier"se"déclare" vers"l’âge"de"35"ans""et"est"la"principale"cause"de"décès"chez"ces"patients."

Diagnostic! Une"douleur"sur"le"flanc,"de"l’hématurie"et"une"masse"abdominale"palpable"constituent," la"triade"de"symptômes"de"cancer"du"rein."A"ce"stade,"des"signes"de"métastases"doivent" être" recherchés." Les" médecins" vont" alors" tenter" de" déceler" chez" le" patient" ou" sur" son" historique" des" douleurs" osseuses," des" nodules" pulmonaires," de" l’hypercalcémie," de" la" fièvre" ou" une" perte" de" poids" inexpliquée," signes" d’un" syndrome" paranéoplasique." Cependant," de" nos" jours," la" plupart" des" cancers" " du" rein" sont" découverts" de" façon" fortuite." Cette" détection" précoce" est" due" aux" avancées" technologiques" qui" permettent" une" utilisation" large" de" techniques" non" invasives" d’imagerie" réalisées" pour" d’autres" raisons."C’est"donc"l’imagerie"qui"permet"de"diagnostiquer"le"plus"souvent"les"cancers"du" rein"même"si"ces"derniers"peuvent"avoir"des"aspects"très"différents"(Low"et"al."2016)."" Un" scanner" permet" de" compléter" le" diagnostic" et" mettre" en" évidence" l’envahissement" local,"ganglionnaire"ou"métastatique."En"cas"de"contreGindication"pour"le"scanner,"une" imagerie" à" résonance" magnétique" (IRM)""reste" une" alternative." " Afin" de" caractériser" certaines" tumeurs," l’IRM" peut" être" utilisée" avec" le" scanner." Une" biopsie" n’est" pas" recommandée"pour"confirmer"un"diagnostic"sauf"si"un"traitement"ablatif"est"envisagé,"en" cas"de"comorbidité"ou"encore"en"présence"de"métastases"systémiques"pour"lesquelles"la" chirurgie"n’est"pas"recommandée.""" Quand"une"forme"héréditaire"est"suspectée,"un"dépistage"peut"se"faire"chez"les"proches"à" partir"d’un"certain"âge."

Pronostic!! Il"est"très"difficile"de"faire"un"pronostic"sur"l’évolution"d’un"cancer"du"rein."Cependant," quelques" caractéristiques" qui" peuvent" donner" des" indications" comme" le" stade" de" la" maladie,"la"taille"de"la"tumeur"et"le"grade"histologique.""La"classification"TNM"est"la"plus" utilisée"en"clinique"de"nos"jours"avec"la"gradation"de"Fuhrman."Elle"permet"de"classer"les" tumeurs"selon"leur"taille"T,"si"elle"a"atteint"les"ganglions"N"et"la"présence"de"métastases"

16" " M."la"classification"TNM"sert"aussi"à"distinguer"les"formes"localisées"(stade"1"et"2)"des" formes"localement"avancées"(stade"3)"et"des"formes"métastatiques"(Sobin"and"Compton" 2010)(stade" 4)"(Tableau" 1)." Par" contre" pour" les" ccRCCs" métastatiques," les" modèles" utilisées" sont" principalement" les" scores" du" Memorial" Sloane" Kettering" Cancer" Center" (MSKCC)"ou"de"Motzer"et"de"l’international"Metastatic"RCC"Database"Consortium"(IMDC)" ou" de" Heng" qui" permettent" de" classer" les" patients" en" 3" groupes":" favorable," intermédiaire"et"péjoratif"selon"la"médiane"de"survie"associée"(Tableau"2)"" " Tableau"1:"CLASSIFICATION!UICC!TNM!DES!TUMEURS!MALIGNES,!7E!ÉDITION,!2009" TNM!(2009)! Nomenclature! Statut!

Tumeur"(T)! Tx! Le"statut"tumoral"ne"peut"être"défini! " T1a! Tumeur"≤"à"4"cm"localisée"au"rein!

" T1b! Tumeur" >" à" 4"cm" et" ≤" 7"cm" localisée" au"rein! " T2a! Tumeur">"7"à"≤"10"cm"localisée"au"rein!

" T2b! Tumeur">"10"cm"localisée"au"rein!

Envahissement"du"tissu"adipeux"périG rénal" et/ou" le" tissu" adipeux" hilaire" mais" pas" le" fascia" de" Gerota" et/ou" " T3a! thrombus" macroscopique" dans" la" veine" rénale" ou" dans" l’une" de" ses" branches! " T3b! Thrombus" dans" la" veine" cave" sous" le" diaphragme! " T3c! Tumeur"s’étendant"dans"la"veine"cave" auGdessus" du" diaphragme" ou" envahissant"la"paroi"musculaire"de"la" veine"cave! " T4! Tumeur"infiltrant"auGdelà"du"fascia"de"

17" " Gerota" et/ou" envahissement" par" contiguïté"de"la"surrénale! Métastase"ganglionnaire! Nx! Pas"d’évaluation"du"statut"GG! " " " "(N)" " " " N0! Pas"de"métastase"GG!

" N1! Métastase"régionale"GG"dans"1"seul"GG!

" N2! Métastase"régionale"GG"dans"plus"de"1" GG! Métastase"à"distance"(M)! Mx! Pas" d’évaluation" du" statut" métastatique! " M0! Pas"de"métastase!

" M1! Métastase"tissulaire"à"distance!

Tableau!2!:!Score!MSKCC!pour!les!RCCs!métastatiques!

Score"MSKCC" " GIndice"de"Karnofsky"(péjoratif"si"<"80%)" ! Favorable"(0critére)":30"mois" GHémoglobinémie" (péjoratif" si" <13g/dL" ! Intermédiaire" (1" à" 2" critères)":14" chez" l’homme" et" <" 11,5" g/dL" chez" la" mois" femme" ! Mauvais"(>3"critères)" GDurée"de"l’intervalle"entre"le"diagnostic"et" le"début"systémique"(péjoratif"si"<1"an)" GCalcémie" corrigée" (péjoratif" si" >10mg/dL)" GValeur" de" LDH" (péjoratif" si" >1,5" X" la" normale)" Le" grade" de" Fuhrman" permet" de" d’évaluer" la" malignité" d’une" tumeur." Elle" permet" de" classer"les"tumeurs"en"grade"de"1"à"4"selon"leur"agressivité""en"se"basant"sur"l’aspect"

18" " (morphologie" et" taille)" des" noyaux" et" nucléoles" ainsi" que" la" présence" de" cellules" «"monstrueuses"»"(Tableau"3)""

Tableau!3:!Classement!grade!Fuhrman/nucléolaire!!

Grade!! Aspect!!noyau! Contours! Nucléoles! Cellules! noyau! monstrueuses!!

I! Rond," petite" réguliers! Absent" ou" Néant"! taille" environ" imperceptibles"! 10"microns!

II! Plus" Discrètes"" Visible" à" Néant! volumineux:" irrégularités! grossissement" environ" 15" 400X! microns"!

III! Volumineux":" Nettement" Visible" à" Néant! environs" 20" irréguliers! grossissement" microns"! 100X!

IV! Volumineux! Irrégulier" et" Visible" à" Présente"! multilobés! grossissement" 100X!

Cette" classification" compare" l’aspect" des" cellules" tumorales" à" celle" des" cellules" saines." Ainsi"plus"les"cellules"malines"sont"différenciées,"comparables"aux"cellules"saines,"plus"la" maladie"est"précoce"et"donc"de"bon"pronostic.""Le"comité"de"cancérologie"de"l’association" française" " d’urologie" dans" ses" dernières" recommandations" 2016G2018" indique" que" l’utilisation"du"grade"de"Fuhrman"devrait"être"remplacée"par"celle"du"grade"nucléolaire" de"l’International"Society"of"Urological"Pathology"ISUP"sauf"dans"le"cas"des"ccRCC"et"des"

19" " pRCC" (Bensalah" et" al." 2016)." Ces" classifications" permettent" de" définir" des" groupes" de" bon"ou"mauvais"pronostic"et"ainsi"orienter"le"suivi"et"le"traitement"des"patients."" "

Les!voies!moléculaires!! Les" efforts" de" recherche" de" ces" dernières" années" ont" montré" que" certaines" voies" de" signalisation"sont"dérégulées"lors"de"la"tumorigenèse"des"cancers"du"rein."Ces"résultats" ont"permis"de"mieux"connaître"cette"maladie"et"de"mieux"la"combattre."Notamment"dans" le"cancer"du"rein,"des"disfonctionnements"au"niveau"de"voies"de"signalisation"majeures" ont"été"mises"en"évidence.""

L’axe!VHL/HIF/VEGF! pvhl" est" le" gène" le" plus" fréquemment" muté" dans" les" ccRCCs" (Gnarra" et" al." 1994)." Une" altération" du" gène" pvhl" est" l’événement" le" plus" précoce" " qui" amorce" l’apparition" des" ccRCC." " Cependant" la" mutation" de" pvhl" est" insuffisante" pour" induire" un" ccRCC" (Kaelin" 2007)."La"fonction"de"VHL"la"plus"connue"est"celle"de"dégradation"des"protéines"HIF1α"et" HIF2α" dans" les" cellules" saines" en" condition" de" normoxie" (état" dans" lequel" l’environnement" contient" assez" l’oxygène" pour" permettre" une" activité" normale" de" la" cellule).""En"hypoxie"HIFα"n’est"plus"dégradé"et"peut"donc"transloquer"dans"le"noyau"afin" d’induire"l’expression"de"ses"gènes"cibles"comme"le"VEGFGA"(figure"2)"

20" " "

Figure!2!:!Régulation!des!protéines!HIFα!par!VHL!

La!famille!HIFs! Elle"est"responsable"de"la"plupart"des"changements"induits"par"la"cellule"pour"s’adapter" à"l’hypoxie"(manque"d’oxygène"dans"l’environnement)."Il"existe"quatre"membres"dans"la" famille"des"HIFs"(figure"3)".HIF1α"été"le"premier"découvert"il"y"a"25"ans"lors"d’un"étude" sur" l’EPO" (Semenza" and" Wang" 1992)." HIF1" est" un" facteur" de" transcription" hétérodimérique" formé" de" HIF1β" exprimé" de" manière" constitutive" et" HIF1α" dont" l‘expression" dépend" majoritairement" de" l’oxygène.""HIF1α" et" HIF2α" sont" très" proches" (figure"3)."Ils"partagent"certaines"cibles"mais"ont"chacun"des"cibles"spécifiques"(Mole"et" al."2009).""HIF3,"le"plus"récemment"découvert,"a"une"régulation"complexe"et"encore"mal" comprise." Quant" à" HIF1α" et" HIF2α," leur" régulation" est" connue" et" est" principalement" postGtranscriptionnelle."Dans"les"cellules"en"normoxie,"HIF1α"et"HIF2α"sont"hydroxylés" sur" des" prolines" situées" au" niveau" de" leur" domaine" «"Oxygen" Dpendent" Domain" ou" ODD"»." Cette" hydroxylation" est" l’action" d’enzymes" de" la" classe" des" Proxyl" hydroxylase" (PHD)."Il"existe"trois"PHD"nommées"de"1"à"3."Les"trois"sont"capables"de"se"lier"à"HIF1α" mais"PHD2"a"une"affinité"plus"forte"avec"HIF1α"et"PHD1"et"3"avec"HIF2α"(Berra"et"al." 2003)." Toutes" les" PHDs" ont" besoin" de" Fe2+" et" de" αGketoglutarate" (αGKG)" pour" être"

21" " actives"et"l’oxygène"est"un"cofacteur"essentiel"à"leur"activité."C’est"la"raison"pour"laquelle" leur"activité"est"fortement"réduite"en"hypoxie"(Epstein"et"al."2001).""L’hydroxylation"de" HIF1α" induit" son" interaction" avec" pVHL" " qui" recrute" le" complexe" ElonginBGElonginCG Cullin2GE3" ubiquitin" ligase," entraînant" l’ubiquitination" sur" la" lysine" 48" de" HIF1α" et" sa" dégradation"par"le"protéasome"(Ivan"et"al."2001,"Yu"et"al."2001)."En"hypoxie,"les"PHDs" ont" une" activité" fortement" réduite" voire" inexistante" à" cause" de" leur" dépendance" à" l’oxygène." Ainsi" HIF1α" n’est" pas" complexé" avec" pVHL" et" se" lie" à" HIF1β" qui" est" constitutivement" exprimé" dans" le" noyau." Le" complexe" HIF1αGHIF1β" peut" alors" se" lier" aux"promoteurs"des"gènes"contenant"un"«"Hypoxia"Response"Element"(HRE";"RCGTG)"»" afin"de"déclencher"leur"transcription."Ces"gènes"permettent"l’adaptation"de"la"cellule"au" manque"d’oxygène(Kaelin"and"Ratcliffe"2008)."" "

Figure!3!:!Schéma!!représentant!les!différents!domaines!des!protéines!HIFs.!

HIF1!vs!HIF2! Bien"qu’ils"aient"des"structures"similaires"et"se"fixent"avec"la"même"efficacité"sur"les"sites" HRE,"il"subsiste"des"différences"entre"HIF1"et"HIF2."Beaucoup"d’articles"étudiant"le"rôle" de"chacun"ont"été"publiés"ces"dernières"années."Cependant"ce"qui"ressort"de"l’ensemble"

22" " de"ces"études"reste"flou"mais"peut"être"résumé"comme"suit."Le"rôle"de"HIF1"ou"HIF2"lors" de"la"tumorigenèse"varie"selon"le"type"cellulaire.""Dans"le"cancer"du"côlon,"l’inhibition" spécifique"de"HIF1α"ou"HIF2α"donne"des"résultats"différents"selon"une"étude"publiée"en" 2009."L’inhibition"de"HIF1α"induit"une"diminution"de"la"croissance"tumorale"ainsi"que"de" la"prolifération"la"migration"et"l’invasion"cellulaire"in&vitro."Quand"l’expression"de"HIF2α" est"réduite"les"cellules"de"cancer"du"côlon"prolifèrent"autant"que"les"cellules"sauvages" mais"forment"des"colonies"plus"grosses."Les"tumeurs"expérimentales"générées"avec"des" cellules" invalidées" pour" HIF2α" se" développent" presque" deux" fois" plus" vite" que" les" contrôles"(Imamura"et"al."2009)."HIF2α"serait"donc"un"suppresseur"de"tumeur"dans"ce" cas."Des"résultats"similaires"ont"été"observés"dans"le"glioblastome"(Acker"et"al."2005)."" Cependant" dans" le" cancer" du" rein," la" surexpression" HIF2α" mais" pas" HIF1α" accélère" la" croissance" tumorale" (Raval" et" al." 2005)." Ces" résultats" montrent" que" malgré" leur" similitude,"les"HIFα"peuvent"induire"des"réponses"biologiques"complètement"différentes" voire"opposées"selon"le"fond"génétique"et"le"type"cellulaire"étudié."""" Dans"les"ccRCCs,"les"HIFs"ont"un"statut"spécial."A"cause"de"l’inactivation"de"pvhl"dans"ces" tumeurs"par"mutation,"délétion"ou"modification"épigénétique"de"son"promoteur,"les"HIFs" sont"constitutivement"exprimés"même"en"normoxie."Ainsi"les"gènes"cibles"des"HIFs"sont" surexprimés."VEGFGA"est"l’une"des"cibles"les"plus"décrites"de"HIF1α."La"production"de" VEGFGA"par"ces"cellules"tumorales"induit"un"phénotype"de"tumeurs"hypervascularisées." Cette" vascularisation" exacerbée" supporte" la" prolifération" cellulaire" au" niveau" de" la" tumeur"en"apportant"les"nutriments"nécessaires"à"ce"processus.""

La!voie!«!mammalian!target!of!rapamycin!(mTOR)!»! La"voie"mTOR"est"importante"dans"la"biologie"du"cancer"du"rein."Cette"voie"est"en"aval" des" récepteurs" des" facteurs" de" croissance." Elle" débute" avec" la" liaison" d’un" facteur" de" croissance"avec"son"récepteur,"en"général"un"récepteur"tyrosine"kinase"et"aboutie"à"des" effets" multiples" sur" la" transcription" de" gènes" impliqués" dans" la" prolifération," la" différenciation" et" la" survie" cellulaire." La" liaison" d’un" facteur" de" croissance" avec" son" récepteur"stimule"la!PhosphatidylinositolG4,5Gbisphosphate"3Gkinase"(PI3K)."La"classe"I" des"PI3K"est"responsable"de"la"production"de"phosphatidylinositol"3Gphosphate"(PI(3)P)," phosphatidylinositol"(3,4)Gbisphosphate"(PI(3,4)P2)"et"de"phosphatidylinositol"(3,4,5)G trisphosphate" (PI(3,4,5)P3)" (Okkenhaug" 2013)." Les" phosphoinositides" formés" à" la" membrane" permettent" le" recrutement" " de" la" protéine" kinase" B" (PKB/AKT)" et" la"

23" " «"phospoinositide" dependent" kinase" (PDK1)"»." Cette" voie" est" inhibée" par" Phosphatase" and"Tensin"homolog"(PTEN)"qui"bloque"la"formation"de"PIP3"(Sansal"and"Sellers"2004)" et"régule"négativement"l’activation"d’AKT."Parmi"les"multiples"cibles"d’AKT,"mTOR"est" activée" de" manière" indirecte." AKT" phosphoryle" TSC2" (tuberous" sclerosis" complex" 2)" induisant" la" dissociation" du" complexe" TSC1/TSC2." La" libération" de" TSC2" bloque" son" action" inhibitrice" sur" " Rheb" (Ras" homologue" enriched" in" the" brain)" qui" sous" sa" forme" RhebGGTP"(active)"active"mTOR"(Manning"and"Cantley"2007)."mTOR"activée"stimule"la" synthèse"protéique"en"phosphorylant"la"S6"kinase"(p70S6K)"(Jastrzebski"et"al."2007)"et" le"facteur"de"transcription"4EGBP"1"(4E"binding"protéine"1)"(Hay"and"Sonenberg"2004)."" L’activation"de"mTOR"induit"un"blocage"de"l’autophagie"via"la"phosphorylation"d’UncG51" like" Auntophagy" Activating" KinaseG1" (ULK1)" et" déclenche" la" lipogenèse" via" le" (Peroxisome" Proliferator" Activated" Receptor" Gamma)"PPARγ." (Saran," Foti," and" Dufour" 2015)." " La" voie" mTOR" interagit" avec" la" voie" des" MAP" Kinases" (Mitogen" Activated" ProteinG" kinases)." mTOR" peut" être" activée" directement" par" Extracellular" signalG Regulated" Kinase"(ERK)(Sabatini" 2006)." " La" voie" AKT/mTOR" est" décrite" comme" étant" plus"activée"dans"les"ccRCCs"par"rapport"aux"cellules"rénales""normales."Dans"une"étude" sur"128"tumeurs"primaires"ccRCC,"22"tumeurs"métastatiques,"24"reins"sains,"le"niveau" de"protéine"70S6K,"mTOR,"et"AKT"phosphorylées"est"décrit"supérieur"dans"les"cellules" de"ccRCC"par"rapport"aux"cellules"saines"(Lin"et"al."2006)."" Dans"les"ccRCCs,"la"voie"mTOR"agit"à"plusieurs"niveaux."mTOR"stimule"la"traduction"de" l’ARNm" HIF1α." Ainsi" mTOR" agit" sur" la" néoGangiogenèse" de" manière" indépendante" de" pVHL."D’autre"part,"la"stimulation"de"traduction"induite"par"mTOR"étant"peu"spécifique," elle" permet" l’augmentation" des" protéines" favorisant" la" croissance" tumorale." Ainsi," elle" promeut"la"prolifération"et"la"survie"cellulaire"et"bloque"l’autophagie"dans"les"cellules" tumorales"(Figure"4)."Elle"agit"également"au"niveau"de"la"cellule"endothéliale,"en"aval"de" la"signalisation"par"le"VEGFR,"favorisant"donc"son"action"proGangiogénique.""La"mise"en" évidence"de"ces"voies"de"signalisation"a"permis""de"mieux"comprendre"la"physiologie"des" ccRCCs." Elle" a" ainsi" permis" de" développer" plusieurs" molécules" thérapeutiques" pour" lutter"de"façon"efficace"contre"ces"cancers.""

24" " "

Figure!4!:!Voie!de!signalisation!Akt/mTOR!et!mécanismes!physiologiques!associés!

Les!traitements!!

La!chirurgie! La"chirurgie"constitue"le"traitement"de"référence"des"cancers"du"rein"localisés."Elle"peut" aussi"être"utilisée"en"combinaison"avec"un"traitement"médical"pour"les"patients"avec"une" tumeur"métastatique."Elle"peut"être"partielle"ou"totale"c’est"à"dire"soit"un"partie"du"rein" soit" la" totalité" du" rein" peut" être" enlevée." La" radiofréquence" et" la" cryothérapie" sont" d’autres"techniques"qui"sont"utilisées"mais"elles"sont"recommandées"que"pour"les"petites" tumeurs," chez" des" patients" avec" une" morbidité" importante" due" à" l’âge," donc" mauvais" candidats" pour" la" chirurgie" (Ljungberg" et" al." 2015)." Une" néphrectomie" partielle" est" recommandée" chez" les" patients" avec" une" tumeur" T1" selon" la" classification" TNM," les" patients"avec"un"seul"rein,"ou"avec"une"tumeur"bilatérale"synchronisée"ou"encore"avec" syndrome" de" pvhl." Le" but" est" d’enlever" la" totalité" de" la" tumeur" tout" en" préservant" la" fonction" rénale." Bien" sûr" une" néphrectomie" partielle" provoque" une" diminution" de" la" fonction" rénale" mais" comparée" à" une" néphrectomie" totale," des" résultats" similaires" en"

25" " terme"de"survie"chez"les"patients"avec"une"tumeur"T1"ont"été"obtenus"(Van"Poppel"et"al." 2011)."Dans"le"cas"de"tumeurs"plus"avancées,"la"néphrectomie"totale"est"recommandée." Elle"consiste"à"enlever"le"rein"entièrement"avec"le"tissu"adipeux"autour"du"rein,"la"glande" surrénale" ainsi" que" les" ganglions" lymphatiques" les" plus" proches." Une" néphrectomie" totale"est"recommandée"chez"les"patients"avec"des"tumeurs"multiples"sur"un"rein"ou"une" tumeur"localement"avancée"mais"ni"intraGcapsulaire,"ni"métastatique)."Lorsqu’un"patient" a" une" tumeur" métastatique," une" néphrectomie" cytoréductrice" peut" être" réalisée" avant" un"traitement"médical"s’il"est"de"bon"pronostic"ou"de"pronostic"intermédiaire"(Zini"et"al." 2008).""Une"métastasectomie"totale"peut"aussi"être"réalisée"chez"les"patients"avec"des" métastases" pulmonaires," hépatiques" et" pancréatiques." Cela" améliore" la" survie" globale" (Alt"et"al."2011)."

L’immunothérapie!de!première!génération! L’immunothérapie" est" la" première" stratégie" qui" a" été" utilisée" contre" les" ccRCCs" métastatiques."Jusqu’en"2005,"les"cytokines":"InterleukineG2"(IL2)"et"interféronGα"(INFGα)" étaient"les"molécules"données"aux"patients."La"survie"moyenne"était"alors"de"10"mois." Cela" correspond" à" la" période" sombre" " des" thérapies" contre" le" cancer" du" rein." Ces" cytokines" étaient" les" seules" molécules" à" montrer" une" efficacité" modeste" mais" réelle."" L’INFGα" est" en" effet" capable" d’induire" une" action" antiGangiogénique" via" l’inhibition" de" l’expression" de" CXCLG8" et" de" VEGFGA." En" plus," il" active" aussi" des" facteurs" antiG angiogéniques" comme" CXCL9," CXCL10" et" CXCL11" (Indraccolo" 2010)." Même" des" rémissions"spectaculaires"ont"été"induites"après"traitement"avec"l’IL2"mais"que"dans"une" très"faible"proportion"des"patients."Surtout,"les"mécanismes"expliquant"ces"rémissions" sont"inconnus."""

" Figure!5!:!Historique!de!l’évolution!des!thérapies!anti!RCC!

26" " Les!thérapies!antiJangiogéniques! A"partir"de"2005,"une"révolution"est"arrivée"dans"le"traitement"des"ccRCCs."" La"recherche"biologique"sur"les"mécanismes"et"voies"de"signalisation"impliquées"dans" leur" développement" a" permis" de" mieux" connaître" ces" cancers." Ainsi" l’exploitation" des" données"issues"de"cette"recherche"a"contribué"au"développement"de"nouvelles"thérapies." Dans" le" cas" des" ccRCCs," le" résultat" est" l’apparition" des" thérapies" antiGangiogéniques" (AAG)" (Figure" 5)." L’inactivation" de" pVHL" dans" ces" tumeurs" aboutissant" à" une" néovascularisation"tumorale"importante"est"la"raison"de"l’utilisation"des"AAGs."Les"AAGs" peuvent"être"divisés"en"deux"sousGfamilles.""

La&sousOfamille&des&anticorps&antiOVEGFOA"est"représentée"par"le"bevacizumab"(Avastin)" un"anticorps"monoclonal"humanisé"dirigé"contre"le"VEGFGA."L’Avastin"est"commercialisé" par" ROCHE" et" a" obtenu" son" autorisation" de" mise" sur" le" marché" (AMM)" en" 2007." Son" mécanisme"d’action"est"le"suivant":"il"va"se"fixer""sur"les"molécules"de"VEGFGA"circulant" avant" qu’elles" ne" puissent" se" lier" à" leurs" récepteurs" (Figure" 5)." Son" efficacité" thérapeutique"contre"les"ccRCC"métastatiques"a"été"prouvée"en"association"avec"l’INFα." Dans"ces"études,"l’association"Avastin"+"INFα"augmente"la"survie"sans"progression"par" rapport" au" traitement" contrôle" mais" n’a" pas" d’effet" significatif" sur" la" survie" globale" (Escudier,"Pluzanska,"et"al."2007,"Escudier"2010b)."Par"manque"d’efficacité"sur"la"survie" l’AMM"a"été"retirée"pour"les"ccRCC"en"Juillet"2016.""

La& sous& famille& des& inhibiteurs& de& récepteurs& tyrosineOkinase." Ces" molécules" ciblent" les" récepteurs" de" facteurs" de" croissance" comme" le" VEGFGA." Elles" inhibent" également" d’autres"facteurs"qui"contribuent"au"développement"du"cancer."Parmi"ces"inhibiteurs,"on" retrouve" le" sunitinib" (sutent)," développé" par" Pfizer," qui" a" obtenu" son" AMM" en" 2006" (Figure" 5)." Le" sunitinib" bloque" l’activité" des" récepteurs" du" VEGF""(VEGFRG1," 2," 3)," du" Platelet"Derived"Growth"Factor"(PDGFR),"du"«"FMSGlike"Tyrosine"kinaseG3"»"du"Colony" Stimulatinf" Factor" 1" (CSF1)" et" de" cGKIT." Une" étude" clinique" en" 2007" a" montré" que" le" sunitinib" apportait" un" avantage" thérapeutique" au" niveau" de" la" survie" globale" des" patients" traités" en" seconde" ligne" (Motzer," Michaelson," et" al." 2007)." Dans" une" étude" comparative" avec" l’INFα," la" survie" globale" des" patients" traités" avec" sunitinib" est" meilleure"(11"mois"versus"5"mois)"(Motzer,"Hutson,"et"al."2007)." Le" sorafenib"(Nexavar)" commercialisé" par" Bayer" est" un" autre" inhibiteur" de" récepteur" tyrosine"kinase"a"pour"cible"les"mêmes"que"la"sunitinib"mais"également"la"kinase"RAF."Il"

27" " a"été"approuvé"en"2006."Les"patients"avec"ccRCCs"métastatiques"traités"avec"le"sorafenib" ont"gagné"quelques"mois"en"survie"sans"progression"et"en"survie"globale"par"rapport"aux" patients"traité"avec"du"placebo"lors"d’une"étude"en"2005"(Escudier,"Eisen,"et"al."2007)"."" D’autres"inhibiteurs"viennent"compléter"cet"arsenal"de"TKI":"le"pazopanib,"l’axitinib,"le" cabozantinib" (Choueiri" et" al." 2016)" et" le" lenvatinib" (Motzer" et" al." 2016)" qui" ont" été" approuvés" respectivement" entre" 2010" et" 2015." Toutes" ces" molécules" thérapeutiques" ciblent"la"voie"VEGF/VEGFGR.""

Les!inhibiteurs!de!mTOR! Il"s’agit"de"l’everolimus"(Afinitor)""et"du"temsirolimus"(Torisel)."Ces"deux"inhibiteurs"ont" été"approuvés"pour"le"traitement"des"ccRCCs"métastatiques"respectivement"en"2007"et" 2009." En" bloquant" mTOR" ces" inhibiteurs" peuvent" ainsi" agir" au" niveau" de" plusieurs" processus" " cellulaires" comme" l’angiogenèse" via" " HIFα." " Un" bénéfice" sur" la" survie" sans" progression"des"patients"traités"avec"temsirolimus"en"deuxième"ligne"a"été"reporté"par" rapport"aux"patients"traités"avec"de"l’INFα."En"effet"cette"étude"regroupait"626"patients" de" très" mauvais" pronostic," traités" soit" avec" INFα" seul," soit" avec" le" temsirolimus" ou" encore"une"combinaison"des"deux"molécules"(Hudes"et"al."2007)."La"survie"globale"des" patients"traités"avec"le"temsirolimus"était"statistiquement"plus"longue"que"les"patients" traités"avec"l’INFα."Et"la"combinaison"des"deux"n’apportait"pas"de"bénéfice"par"rapport"à" l’INF"seul"(Battelli"and"Cho"2011)."Ces"résultats"ont"permis"d’intégrer"le"temsirolimus" comme" traitement" en" première" ligne" pour" contre" les" RCCs" avancés." " Par" ailleurs" l’" évérolimus"a"été"testé"avec"des"patients"sur"lesquels"le"sunitinib"et"sorafenib"n’ont"pas" marché." Ces" patients" ont" donc" reçu" de" l’everolimus" ou" du" placebo." La" survie" sans" progression" des" patients" a" été" meilleure" chez" le" groupe" everolimus" que" le" groupe" placebo" (Motzer" et" al." 2010)." Ainsi" l’everolimus" a" été" approuvé" en" seconde" ligne" thérapeutique"pour"les"patients"réfractaires"aux"AAGs."" Ces"antiGVEGFGA"et"antiGmTOR"ont"occupé"la"scène"des"thérapies"ciblées"contre"le"ccRCC" entre" 2005" et" 2015" (Figure" 6)." Cette" décennie" correspond" à" la" période" moderne" de" traitements"contre"les"cancers"du"rein."Depuis"2015,"on"est"rentré"dans"une"nouvelle"ère," celle"des"inhibiteurs"de"checkGpoint"immunitaire.""

L’immunothérapie!de!nouvelle!génération! Les"thérapies"antiGangiogéniques"à"base"de"cytokines"ne"fonctionnent"que"sur"certains" patients"avec"des"rémissions"de"durée"importante."Leur"utilisation"entraine"une"certaine"

28" " toxicité"à"forte"dose"et"ceci"est"particulièrement"vrai"pour"IL2."De"nos"jours"plusieurs" études" sont" en" cours" pour" tester" l’efficacité" de" la" combinaison" des" AAGs" avec" une" nouvelle" génération" d’immunothérapie" qui" sont" des" inhibiteurs" de" «"checkGpoint"»" immunitaires"des"lymphocytes"T"(LT)."Cette"stratégie"utilise"des"anticorps"dirigés"contre" le"«"Programmed""cell"Death"protein"1"Ligand"1"(PDL1)"»"comme"avelumab"(Bavencio)"et" atezolizumab"(Tecentriq)"ou"contre"le"«"Programmed"cell"death"protein1"PD1"»"comme" le" nivolumab" (Opdivo)" et" pembrolizumab" (Keytruda)." L’activation" de" l’axe" PDL1/PD1" induit" une" tolérance" immune.""PD1" est" exprimé" au" niveau" des" lymphocytes" et" inhibe" l’activité"de"ces"derniers"quand"il"est"activé"par"son"ligand"PDLG1."PDLG1"est"fortement" exprimé" au" niveau" des" cellules" tumorales." Ainsi" le" blocage" de" cet" axe" permettrait" la" réactivation"des"lymphocytes"T"et"la"destruction"des"cellules"cancéreuses"par"le"système" immunitaire." Par" ailleurs" l’ipilimumab" (Yervoi)," un" autre" inhibiteur" de" checkGpoint" immunitaire" est" aussi" en" cours" d’étude." Il" a" pour" cible" le" cytotoxic" T" lymphocyteGassociated" protein" 4" (CTLA4)." CTLAG4" inhibe" l’activation" des" LTs."" L’ipilimumab"pourrait"lever"cette"inhibition"afin"de"réactiver"les"LTs."Le"nivolumab"a"été" approuvé" après" une" étude" montrant" une" augmentation" en" survie" globale" en" comparaison" avec" l’everolimus" sur" des" patients" devenus" résistants" au" sunitinib" et" au" pazopanib"(Motzer"et"al."2015)."Cependant"le"taux"de"réponse"au"nivolumab"était"que"de" 25%" contre" 5%" pour" l’everolimus." La" plupart" des" patients" traités" avec" le" nivolumab" n’ont" pas" eu" de" régression" tumorale." Néanmoins," ces" inhibiteurs" restent" très" prometteurs." Il" est" très" difficile" de" prédire" une" réponse" à" ce" type" de" traitement":" l’expression" de" PDL1" n’est" pas" corrélée" à" une" réponse" au" traitement" visant" l’axe" PD1/PDL1"comme"reporté"dans"l’étude"clinique"CheckMate"025"(Escudier"et"al."2017)."" En" ce" moment," différentes" combinaisons" entre" des" AAGs" et" agents" d’immunothérapie" sont"en"cours"de"test."Les"résultats"préliminaires"sont"assez"encourageants"et"suggèrent" l’entrée"dans"la"«"période"dorée"»"des"traitements"contre"le"cancer"du"rein""(Hsieh"et"al." 2017)."Cependant"malgré"toutes"ces"avancées"dans"la"lutte"contre"le"cancer"du"rein,"le" principal" problème" reste" les" récidives." Les" cellules" tumorales" ont" une" incroyable" capacité" à" trouver" des" mécanismes" afin" de" contrecarrer" l’action" des" traitements." Ces" mécanismes"sont"divers"et"variés"et"leur"compréhension"est"primordiale"pour"améliorer" l’efficacité"des"traitements"actuels.""""

29" " La!radiothérapie!! Elle" est" rarement" utilisée" dans" le" traitement" du" cancer" du" rein." Cependant," elle" est" parfois"proposée"pour"traiter"les"métastases"qui"se"sont"formées"dans"le"cerveau"ou"les" os." Elle" a" pour" objectif" de" réduire" la" taille" de" ces" métastases" afin" de" soulager" les" symptômes"qu’elles"provoquent."La"radiothérapie"utilise"des"rayonnements"ionisants,"de" très" forte" énergie" pour" détruire" les" cellules" cancéreuses." Elle" consiste" à" diriger" précisément" ces" rayonnements" (appelés" aussi" rayons" ou" radiations)" sur" les" cellules" cancéreuses" tout" en" préservant" le" mieux" possible" les" tissus" et" les" organes" sains" avoisinants."Ces"rayonnements"sont"produits"par"un"accélérateur"de"particules."Ils"sont" dirigés" en" faisceau," à" travers" la" peau," sur" les" zones" où" se" trouvent" les" métastases." La" radiothérapie" a" longtemps" été" considérée" comme" inefficace" contre" les" RCCs" (Deschavanne" and" Fertil" 1996)" mais" cette" notion" est" peut" être" fausse." En" effet" le" développement" des" nouvelles" technologies" a" remis" à" jour" la" radiothérapie" dans" le" traitement" des" cancers" du" rein." Ainsi" la" «"stereotactic" ablative" radiotherapy" " (SABR)"»" qui"permet"de"donner"des"doses"plus"importantes"de"radiation,"plus"ciblée"c’est"à"dire" avec" moins" de" dommages" sue" les" cellules" saines," montre" des" résultats" encourageants" dans" le" traitement" des" RCC" (Alongi" et" al." 2017," Siva" et" al." 2017)." En" plus" cette" radiothérapie" a" pour" effet" de" stimuler" le" système" immunitaire" anti" tumoral." On" peut" donc"envisager"sa"combinaison"avec"les"thérapies"ciblées"ou"l’immunothérapie"(Siva"et" al."2017)" "

30" " " Figure! 6!:! Schéma! du! mode! d’action! des! AAGs! sur! les! différentes! voies! de! signalisation!dans!les!RCCs!(Hsieh!et!al.!2017)! "

Les!résistances!! Malgré"la"forte"amélioration"apportée"par"les"AAGs"aux"traitements"de"cancer"du"rein," l’impact" de" ces" traitements" sur" la" survie" globale" des" patients" reste" décevant." Ainsi" beaucoup" d’efforts" ont" été" réalisés" dans" le" but" le" comprendre" les" mécanismes" de" résistance"à"ces"traitements."En"général,"les"résistances"aux"thérapies"dans"les"cancers" du"poumon"ou"les"leucémies"sont"dues"à"des"mutations"sur"leurs"cibles"(Pao"et"al."2005)." Cependant," cela" semble" ne" pas" être" le" cas" pour" les" AAGs" qui" ciblent" entre" autres" les" récepteurs"du"VEGF"qui"se"trouvent"sur"les"cellules"endothéliales."Ces"cellules"sont"plus" stables" génétiquement" que" les" cellules" tumorales." La" majorité" des" changements" au" niveau"des"xénogreffes"résistantes"au"sorafenib"sont"réversibles"quand"les"xénogreffes" sont" réimplantées" dans" des" souris" non" traitées" (Zhang" et" al." 2011)." " Il" est" donc" fort" probable" que" les" changements" conduisant" aux" résistances" ne" soient" pas" d’ordre" génétique" (permanente)." Des" résultats" non" publiés" de" mon" équipe" montrent" des"

31" " résultats" identiques" sur" des" cellules" résistantes" au" sunitinib." Cependant," beaucoup" de" cellules" tumorales" expriment" les" récepteurs" ciblés" par" les" AAGs" (Rizkalla" et" al." 2003)."" Ces" derniers" exercent" donc" une" pression" de" sélection" aboutissant" à" une" adaptation" génétique"des"cellules"tumorales."

L’activation!d’autres!voies!proJangiogéniques.!! Les"traitements"bloquant"le"l’axe"VEGFGA/VEGFR"induisent"généralement"une"régression" tumorale" chez" la" souris" ou" dans" les" essais" cliniques." Cependant" cette" réponse" est" souvent" transitoire" et" est" suivie" d’une" phase" de" progression" tumorale." Cette" seconde" phase"peut"être"la"conséquence"d’une"revascularisation."En"effet,"une"étude"sur"le"cancer" du" pancréas" a" montré" que" cette" revascularisation" est" associée" à" l’expression" d’autres" facteurs" proGangiogéniques." L’expression" d’autres" facteurs" d’angiogenèse" a" été" également" montrée" par" notre" équipe" sur" les" tumeurs" du" rein" (Grepin" et" al." 2012)." Le" mode" d’action" de" ces" facteurs" est" indépendant" de" la" voie" VEGFGA/VEGFR." Ainsi" dans" cette" étude," les" facteurs" comme" le" «"Fibroblast" Growth" FactorG1" (FGF1)," Fibroblast" Growth" FactorG2" (FGF2)," et" angiopoiétine" 1" (Angpt1)" et" cytokines" de" la" famille" ELR+CXCL" sont" surexprimés" lors" de" cette" phase" de" progression" après" un" traitement" bloquant"VEGFRG2"par"rapport"aux"tumeurs"non"traitées"(Casanovas"et"al."2005,"Grepin" et"al."2012)."Lorsque"les"animaux"porteurs"de"tumeurs"expérimentales"sont"traités"avec" un" FGFGtrap" (molécule" " bloquant" l’action" du" FGF)," la" revascularisation" tumorale" et" la" progression" tumorale" sont" réduites" comparées" aux" tumeurs" traitées" seulement" avec" l’inhibiteur" de" VEGFRG2." D’autres" facteurs" comme" l’interleukine" 8," induisent" de" l’angiogenèse" dans" un" contexte" où" la" voie" VEGFGA/VEGFR" est" bloquée." " Ces" résultats" montrent" que" le" blocage" de" la" voie" VEGFGA" provoque" l’activation" d’autres" voies" proG angiogéniques" restaurant" l’angiogenèse" tumorale." Les" cellules" tumorales" peuvent" utiliser" des" facteurs" d’angiogenèse" redondants" du" VEGF" pour" résister" aux" AAGs" standard." Certaines"tumeurs"sont"même"insensibles"aux"AGGs."L’existence"de"divers"signaux"proG angiogéniques"redondants"au"préalable"pourrait"être"une"explication"à"ce"phénomène." Le" FGFG2" est" préférentiellement" exprimé" dans" les" stades" tardifs" de" cancer" du" sein." L’expression" du" VEGFGA" est" plutôt" précoce" (Relf" et" al." 1997)." L’inefficacité" d’un" traitement" anti" VEGFGA" d’un" cancer" du" sein" avancé" devient" logique" dans" ce" contexte." D’autre"part,"la"présence"de"cellules"myéloïdes"au"sein"d’une"tumeur"jamais"traitée"peut" conférer" une" résistance" préexistante" aux" AAGs." Ces" cellules" sécrètent" beaucoup" de" 32" " facteurs" proGangiogéniques." Des" tumeurs" expérimentales" ne" répondant" pas" à" un" AAG" présentent" une" infiltration" de" cellules" CD11b+" GR1+" plus" importante" que" celles" répondant"à"un"AGG."Inhiber"le"recrutement"de"ces"cellules"myéloïdes"reGsensibilise"les" tumeurs"résistantes"(Shojaei,"Wu,"Malik,"et"al."2007).""Finalement,"les"cellules"tumorales" détiennent"un"arsenal"de"stratégies"pour"survivre"aux"AAGs."Ces"stratégies"peuvent"être" des" modifications" au" niveau" des" cellules" tumorales" ou" des" cellules" du" microenvironnement."Cependant"les"AAGs"ne"sont"pas"les"seuls"traitements"des"ccRCCs.""""

Adaptation!du!microenvironnement! La"résistance"aux"AAGs"peut"aussi"provenir"du"microenvironnement"tumoral."Ce"dernier" est" composé" de" cellules" endothéliales," de" péricytes," de" fibroblastes" et" de" cellules" hématopoïétiques." Les" péricytes" sont" des" cellules" en" contact" direct" avec" les" cellules" endothéliales" qui" forment" les" vaisseaux." Ils" entourent" les" vaisseaux" par" de" longs" prolongements"cytoplasmiques."Les"péricytes"constituent"alors"un"élément"important"de" la" vascularisation" tumorale." Ainsi" la" tumeur" peut" les" utiliser" afin" de" maintenir" une" intégrité"vasculaire"en"cas"d’inhibition"de"la"voie"du"VEGFGA.""Effectivement,"cibler"les" péricytes" en" inhibant" le" récepteur" du" PDGF" (PDGFR)," un" élément" essentiel" de" leur" physiologie"permet"de"potentialiser"l’action"des"inhibiteurs"de"l’angiogenèse"(Bergers"et" al." 2003)." Ainsi," une" régression" maintenue" dans" le" temps" de" tumeurs" du" pancréas" expérimentales" a" été" obtenue" en" combinant" AAG" et" inhibiteur" du" PDGFR." La" perturbation" du" dialogue" entre" les" péricytes" et" les" cellules" endothéliales" semble" prometteuse" pour" la" lutte" contre" les" résistances" aux" AAGs." Cependant" cette" stratégie" peut" avoir" des" effets" pervers" en" induisant" une" perméabilisation" des" vaisseaux" et" en" facilitant" l’entrée" des" cellules" tumorales" dans" la" circulation" et" donc" la" propagation" métastatique"(Xian"et"al."2006).""

Rôle!de!l’hypoxie! L’hypoxie"induite"par"les"AAGs"entraine"le"recrutement"au"sein"de"la"tumeur"de"«"Bone" marrowGderived" cells" (BMDC)"»." Les" BMDCs" comme" les" «"Tumor" Associated" Macrophages" (TAM)"»," les" monocytes" immatures" (monocytes" TIE2+," hémiangiocytes" VEGFRG1+"et"les"cellules"myéloïdes"(CD11b+)"exercent"une"action"proGangiogénique"en" exprimant"des"facteurs"de"croissance,"des"cytokines"et"des"protéases"mais"ne"font"pas" partie"intégrante"du"réseau"vasculaire"(Du"et"al."2008)."Les"cellules"CD11b+""produisent" une"grande"quantité"de"«"Matrix"MetalloGProtéaseG9"(MMPG9)"»"qui"rend"bioGdisponible"

33" " le"VEGF"piégé"dans"la"matrice"extracellulaire."Elles"peuvent"également"se"différencier"en" cellules"endothéliales."Les"cellules"myéloïdes"peuvent"supporter"l’angiogenèse"tumorale" via" ces" deux" processus" (Yang" et" al." 2004)." Le" nombre" de" cellules" " CD11b+" " est" effectivement" plus" élevé" dans" les" tumeurs" résistantes" aux" AAGs." Le" fait" d’ajouter" un" anticorps" antiGCD11b" permet" améliore" l’efficacité" des" AAGs" (Shojaei," Wu," Malik," et" al." 2007)."Les"cellules"CD11b+"présentes"dans"les"tumeurs"résistantes"aux"AAGs"produisent" plus"de"Prokineticin"2"(PROKG2),"un"facteur"proGangiogénique"qui"stimule"l’angiogenèse" indépendamment"de"VEGFGA."L’expresion"de"PROKG2"est"stimulée"par"le"«"GranulocyteG Colony" Stimulating" Factor" (GGCSF)"»," le" «"Stromal" cell" Derived" Factor" 1" (SDF1)"»" et" l’interleukine" 6," des" cytokines" exprimées" par" les" cellules" tumorales" et" stromales." Le" blocage" de" PROKG2" diminue" le" recrutement" " de" cellules" CD11b+" après" traitement" par" AAG"et"par"conséquent"une"diminution"de"la"croissance"tumorale"(Shojaei,"Wu,"Zhong,"et" al."2007)."Des"résultats"similaires"ont"été"obtenus"lors"d’une"étude"sur"les"ccRCCs"(Panka" et"al."2013)."Toutes"ces"études"montrent"l’importance"du"microenvironnement"tumoral" dans"la"tumorigenèse"mais"aussi"dans"la"résistance"aux"traitements."D’autres"cellules"du" microenvironnement" comme" les" «"Tumor" Associated" Fibroblast" (TAFs)"»" sont" décrites" pour" leur" rôle" dans" l’angiogenèse" et" la" progression" tumorale." L’inhibition" efficace" de" l’angiogenèse"tumorale"nécessite"le"ciblage"de"plusieurs""voies"pour"ne"donner"aucune" échappatoire"à"la"tumeur."" L’inhibition"de"la"vascularisation"induit"de"l’hypoxie"dans"la"tumeur,"et"les"conséquences" de"cette"hypoxie"peuvent"être"catastrophiques"avec"l’apparition"de"résistances." Les"thérapies"antiGVEGF,"en"détruisant"la"vascularisation"tumorale"entraine"l’apparition" de" zones" hypoxiques" dans" la" tumeur." L’hypoxie" induit" la" sélection" de" cellules" plus" agressives"et"invasives"et"moins"sensibles"aux"AAGs"(Loges"et"al."2009).""L’hypoxie"induit" notamment" l’expression" de" récepteurs" cGMET" à" la" surface" des" cellules" tumorales" (Pennacchietti" et" al." 2003)." L’activation" de" ces" récepteurs" par" «"l’Hepatocyte" growth" fator" (HGF)"»" qui" provient" essentiellement" des" cellules" du" stroma" entraine" une" augmentation"des"capacités"invasives"et"proliférative"et"la"survie"des"cellules"tumorales" via"plusieurs"voies"de"signalisation"(PI3K/AKT,"STAT3"et"MAP"Kinase)."La"combinaison" d’un" inhibiteur" de" VEGFGA" et" de" cGMET" réduit" la" croissance" tumorale" (Shojaei" et" al." 2010)."Ainsi,"l’axe"HGF/cGMET"est"à"l’origine"de"certaines"résistances"au"sunitinib"dans" les" ccRCCs." "L’inhibition" concomitante" des" voies" HGF/cGMET" et" VEGFGA" entraine" une" suppression" de" la" croissance" tumorale," de" l’angiogenèse," et" de" la" dissémination"

34" " métastatique"(Sennino"et"al."2012,"Shojaei"et"al."2010)."L’hypoxie"induite"par"les""AAGs" peut" donc" avoir" des" conséquences" dramatiques" sur" la" progression" tumorale" et" est" certainement"un"des"facteurs"à"l’origine"des"résistances"aux"AAGs."Par"ailleurs,""parmi" tous"les"patients"traités,"il"existe"un"petit"groupe"de"patients"qui"ne"répondent"pas"du" tout"au"traitement"par"sunitinib,""bevacizumab"ou"sorafenib."Ces"résultats"sont"dus"soit"à" une"adaptation"extrêmement"rapide"et"peu"probable"de"la"tumeur,"soit"à"une"résistance" intrinsèque."Les"cellules"de"ces"patients"auraient"un"fond"génétique"ou"épigénétique,"mis" en"place"tout"au"long"de"leur"vie,"aboutissant"à"l’activation"préalable"des"mécanismes"de" résistance" aux" AAGs." Cette" mise" en" place" serait" due" à" une" sélection" imposée" par" le" microenvironnement"bien"avant"l’apparition"et"le"développement"de"la"tumeur"(Bergers" and" Hanahan" 2008)." Les" mécanismes" pouvant" expliquer" cette" résistance" préexistante" ont"été"discutés"par"Bergers"et"Hanahan.""

Résistances!aux!inhibiteurs!de!mTORC1! Les" inhibiteurs" de" mTOR" sont" aussi" utilisés" pour" le" traitement" des" ccRCCs." Ces" inhibiteurs" utilisés" en" deuxième" ligne" thérapeutique" induisent" aussi" des" résistances" à" long"terme"et"les"patients"rechutent"inexorablement"après"une"période"de"réponse"plus" ou"moins"importante."" Afin" d’expliquer" les" résistances" aux" inhibiteur" de" mTORC1," plusieurs" hypothèses" sont" plausibles."La"recherche"sur"les"voies"de"signalisation"a"permis"de"bien"connaître"la"voie" mTOR""et"surtout""aussi"de"mettre"en"évidence"un"certain"nombre"de"rétrocontrôles"de" cette"voie."Ainsi"la"boucle"impliquant"la"p70S6K"qui"après"son"activation"bloque"l’action" de"«"l’Insulin"Receptor"Substrat"1"(IRSG1)"»"dont"l’activité"est"essentielle"à"l’activation"de" la"PI3K"et"donc"de"la"voie"AKT/mTOR."Même"si"les"inhibiteurs"de"mTORC1"sont"efficaces," leur"efficacité"pourrait"être"augmentée"par"le"blocage"de"ce"rétrocontrôle"qui"active"AKT" (Sun" et" al." 2005)." " De" plus," l’inhibition" de" mTORC1" active" le" complexe" mTORC2" qui" stimule"AKT"en"induisant"sa"phosphorylation"sur"Ser473"(Sarbassov"et"al."2005).""Enfin," la" voie" des" MAP" Kinases" est" décrite" comme" inhibé" par" mTOR" via" la" boucle" S6K/PI3K/Ras"(Carracedo"et"al."2008)."Des"études"précliniques"et"cliniques"ont"montré" que"le"traitement"par"everolimus"induit"une"activation"des"MAP"Kinases."L’inhibition"de" mTORC1"entraine"une"répression"des"boucles"de"rétrocontrôle"négatives"aboutissant"à" l’activation" d’AKT." AKT" est" impliqué" dans" plusieurs" processus" antiGapoptotiques" et" de" prolifération"cellulaire."Ces"mécanismes"peuvent"donc"être"à"l’origine"des"résistances"à" l’everolimus" et" au" temsirolimus." Des" doses" d’inhibiteurs" plus" importantes" sont" 35" " nécessaires" pour" bloquer" mTORC2" par" rapport" à" mTORC1." Celles" utilisées" pour" les" ccRCCs" ciblent" plus" le" complexe" 1." Sachant" qu’HIF2α" est" plus" oncogénique" dans" les" ccRCCs" (Kondo" et" al." 2003)" et" que" son" expression" est" plus" dépendant" de" mTORC2" (Toschi"et"al."2008),"l’incapacité"des"inhibiteurs"de"mTORC1"à"réprimer"l’expression"de" HIF2α"peut"être"un"autre"mécanisme"de"résistance."Cette"observation"est"cohérente"avec" le"fait"que"les"ccRCCs"les"plus"agressifs"n’expriment"qu’HIF2α"(Gordan"et"al."2008)."Cette" liste"des"mécanismes"de"résistance"aux"traitements"contre"les"RCCs"met"en"évidence"la" nécessité" de" découvrir" des" marqueurs" prédictifs" de" réponse" aux" AAGs" afin" de" mieux" sélectionner"les"groupes"de"patients"et"les"traitements"correspondants.""

Marqueurs!prédictifs! Il"n’existe"aucun"marqueur"validé"en"clinique"pour"anticiper"les"possibles"réponses"aux" traitements"pour"une"utilisation"optimale."Cependant,"certaines"pistes"sont"en"cours"de" validation" " Les!cellules!endothéliales!circulantes!(CECs)& Il"en"existe"deux"groupes":" Les! cellules! endothéliales! matures":" elles" sont" issues" de" la" paroi" des" vaisseaux" sanguins"et"décrochées"quand"ceuxGci"sont"endommagés."Lors"de"l’angiogenèse"normale," elles" constituent" le" groupe" le" plus" important." La" présence" de" ces" cellules" indiquerait" l’efficacité"des"AAGs." Les!cellules!endothéliales!progénitrices!(CEP)"proviennent"de"la"moelle"osseuse"et"se" différencient"en"cellules"endothéliales"matures"et"participent"à"la"néoGvascularisation." Le" nombre" de" cellules" endothéliales" circulantes" est" plus" élevé" dans" les" patients" cancéreux"par"rapport"aux"personnes"saines"(Beerepoot"et"al."2004)."Dans"les"ccRCCs,"le" nombre"de"CEPs"est"aussi"plus"élevé"et"cette"augmentation"dépend"du"HIF1α"(Bhatt"et"al." 2011)."Ces"résultats"suggèrent"fortement"un"rôle"des"CEPs"dans"le"développement"des" ccRCCs."Elles"présentent"à"leur"surface"du"VEGFRG2,"ce"qui"en"fait"des"cibles"potentielles" d’AAG"comme"le"sunitinib."Ainsi,"le"sunitinib"provoque"une"réduction"des"CECs"associées" à" une" meilleure" survie" sans" progression" (PFS)" chez" les" patients" avec" un" ccRCC" métastatique"(Gruenwald"et"al."2010)."Par"ailleurs,"un"nombre"élevé"de"CEPs"est"associé" à"un"plus"haut"risque"de"progression"et"un"mauvais"pronostic"(Farace"et"al."2011)."Les" CECs"serviraient"de"marqueur"d’efficacité"du"sunitinib.""

36" " Cytokines!et!facteurs!proJangiogéniques!circulants! Les" taux" sanguins" de" «"Tumor" Necrosis" Factor" alpha" (TNFα)" et" de" MMPG9" sont" plus" élevés"chez"les"patients"qui"progressent"rapidement"sous"traitement"(PerezGGracia"et"al." 2009)."Les"ligands"circulants"des"VEGFRs"(VEGFGA"et""PIGF)"sont"augmentés"au"cours"du" traitement"avec"du"sunitinib."Une"forte"augmentation"de"ces"deux""facteurs"ainsi"qu’une" baisse" de" VEGFGC" sont" associées" à" une" meilleur" réponse" au" sunitinib" et" une" PFS" plus" longue"(Rini"et"al."2008)."Les"facteurs"comme"l’interleukine"8,"VEGFRG2"et"VEGFRG3"sont" décrits"comme"potentiels"marqueurs"prédictifs"de"la"réponse"au"sunitinib"dans"les"ccRCC" métastatiques" (Rini" et" al." 2008," Harmon" et" al." 2014)." La" recherche" de" marqueurs" prédictifs""de"réponse"aux"traitements"est"active"mais"pour"le"moment"aucun"marqueur" n’a"été"validé"cliniquement."Cependant"durant"les"prochaines"années"la"validation"d’un" marqueur" pourrait" être" un" grand" pas" dans" la" prise" en" charges" des" patients" avec" un" ccRCC.""

Les!chimiokines!CXCL! La" famille" des" chimiokines" est" subdivisée" en" 4" groupes" selon" le" positionnement" de" la" cystéine"conservée"sur"la"partie"NGterminale"de"la"protéine."Elle"groupe"ainsi"les"CXC,"CC," CX3C" et" finalement" les" XC" (Zlotnik" and" Yoshie" 2000)." " Le" groupe" des" CXCs" comprend" deux" sousGgroupes":" ELR+" et" ELRG"correspondant" à" la" présence" ou" non" respectivement" des"acides"aminés"(acide"glutamique,"leucine"et"arginine"avant"le"domaine"CXC."Les"ELR+" ont"une"fonction"angiogénique"alors"que"les"ELRG"n’ont"pas"d’activité"angiogénique"sauf" le" «"Stromal" cell" Derived" Factor" (SDFG1)"»" qui" est" ERLG" mais" avec" une" fonction" angiogénique" (Vandercappellen," Van" Damme," and" Struyf" 2008)." La" dimérisation," l’oligomérisation"et"l’interaction"avec"des"glycosaminoglycanes"(GAG)"est"essentielle"à"la" fonction"des"chimiokines"in&vivo&(Proudfoot"et"al."2003)."L’interaction"des"chimiokines"et" des" GAGs" entraîne" une" accumulation" des" chimiokines" facilitant" l’établissement" d’un" gradient" indispensable" à" l’attraction" des" cellules" comme" les" leucocytes." Les" GAGs" participent" aussi" à" d’autres" activités" cellulaires" comme" la" réplication" cellulaire," l’invasion"et"la"vascularisation"(Folkman"and"Shing"1992).""

Les"chimiokines"CXC"ELR+"" Les"CXC"ELR+"se"lient"à"leur"récepteur"le"CXCRG1"et/ou"CXCRG2"(Tableau"4)."L’interaction" de"ces"chimiokines"avec"le"CXCRG2"induit"un"chimiotaxisme"des"neutrophiles"ainsi"que" l’angiogenèse" et" le" chimiotaxisme" des" cellules" endothéliales" (Addison" et" al." 2000)." Le"

37" " recrutement" des" neutrophiles" joue" un" rôle" important" dans" l’angiogenèse." En" effet" les" neutrophiles" peuvent" synthétiser" et" stocker" les" facteurs" proGangiogéniques" comme" le" VEGFGA"(Scapini"et"al."2004)."Le"CXCL5,"membre"de"ce"groupe,"est"associé"à"l’agressivité" tumorale"dans"les"cancers"gastriques"(Park"et"al."2007)."De"plus,"le"traitement"de"souris" avec" un" carcinome" du" poumon" non" à" petites" cellules" avec" un" anticorps" antiGCXCL5" bloquant" l’activité" de" ce" dernier" entraine" une" réduction" de" la" croissance," la" vascularisation"et"des"métastases"tumorales"(Arenberg"et"al."1998)"

Les"chimiokines"CXC"ELRG" Ces"chimiokines"sont"des"inhibiteurs"de"l’angiogenèse"et"du"chimiotaxisme"des"cellules" endothéliales" (Strieter" et" al." 1995)." Parmis" " les" CXC" ELRG" ," l’expression" des" CXCL9," CXCL10" et" CXCL11" est" induite" par" les" interferons" INF/" et" " respectivement" (Vandercappellen," Van" Damme," and" Struyf" 2008)." Ces" chimiokines" interagissent" avec" CXCRG3" et" inhibent" l’activité" angiogénique" induite" par" les" CXC" ELR+"." CXCRG3A," un" des" trois"variants"de"CXCRG3"est"impliqué"dans"le"chimiotaxisme"des"CXCLs"induit"par"l’INF " des" lymphocytes" T" et" des" Natural" Killers" (NK)." Le" recrutement" de" ces" cellules" est" associé"à"une"activité"immunitaire"antiGtumorale"et"une"régression"tumorale"(Luster"and" Leder"1993,"Wenzel"et"al."2005)."" " " " " " " " " " " " " " " "

38" " Tableau!4!:!liste!et!quelques!caractéristiques!des!CXCL!humains!! Chimiokine!! Récepteurs! Chimiotaxisme!! Angiogenèse!! SousBgroupe!! CXCL1" CXCRG2" Neutro,"C.E" +" ELR"+" CXCL2" CXCRG2" Neutro,"C.E" +" " CXCL3" CXCRG2" Neutro,"C.E" +" CXCL5" CXCRG2" Neutro,"C.E" +" CXCL6" CXCRG2,"CXCRG1" Neutro,"C.E" +" CXCL7" CXCRG2" Neutro,"C.E" +" CXCL8" CXCRG2,"CXCRG1" Neutro,"C.E,"Ba,"Mo" +" CXCL4" CXCRG3" Fibro,"CE" G" ELRG" CXCL9" CXCRG3" LTa,"NK" G" CXCL10" CXCRG3" LTa,"NK" G" CXCL11" CXCRG3" LTa,"NK" G" CXCL12" CXCRG4" Prog"CD34+,"C.H" G" CXCL13" CXCRG5" LB"naîve,"CD4+" G" CXCL14" Inconnu" Mo,"CDi" G" CXCL15" Inconnu" Neutro" G" CXCL16" CXCRG6" CD8naîve,"CD4+,"NK" G" Neutro":" Neutrophiles," C.E":" cellule" endothéliale," Ba":" basophile," Mo":" monocyte," Fibro":" fibroblaste,"LTa":"lymphocyte"T"actif,"NK":"natural"killer,"Prog":"cellule"progénitrice,"C.H" cellules"hématopoïétiques.""

La!vascularisation!des!ccRCCs!! La"vascularisation"est"un"processus"très"important"au"cours"du"développement."Tous"les" tissus" de" notre" organisme" ont" besoin" d’un" apport" en" oxygène" et" doivent" évacuer" " les" déchets"du"métabolisme"cellulaire."Les"cellules"tumorales"n’échappent"pas"à"cette"règle." Afin"de"maintenir"une"prolifération"soutenue,"les"cellules"tumorales"induisent"une"néoG vascularisation"au"sein"de"la"tumeur"pour"assurer"l’apport"de"nutriments"et"d’oxygène" en" plus" de" l’évacuation" des" déchets" métaboliques." Plusieurs" types" de" vaisseaux" parcourent"notre"corps"mais"je"vais"essentiellement"me"concentrer"sur"la"formation"des"" vaisseaux" sanguins" (angiogenèse)" et" des" vaisseaux" lymphatiques" (lymphangiogenèse)." L’angiogenèse" est" la" formation" de" nouveaux" vaisseaux" sanguins" à" partir" de" 39" " l’endothélium" vasculaire" préexistant." C’est" un" mécanisme" très" complexe" dépendant" d’une"balance"entre"facteurs"proG"et"antiGangiogéniques."Il"fait"partie"des"«"Hallmarks"of" Cancers"»"(Hanahan"and"Weinberg"2000)."Cependant"au"cours"de"ma"thèse,"je"me"suis" concentré" sur" le" VEGFGC" (Vascular" Endothelial" GrowthGFactorGC)" dont" l’expression" est" associée"au"développement"des"vaisseaux"lymphatiques.""

Le!système!!lymphatique!

Présentation!! Le"système"lymphatique"est"formé"de"vaisseaux"qui"drainent"les"liquides"interstitiels"de" tout" l’organisme" (Figure" 7)." Il" est" composé" d’un" réseau" de" canules" qui" drainent" la" lymphe."Ces"canules"présentent"des"valvules"tronconiques"qui"font"que"la"circulation"de" la"lymphe"est"unidirectionnelle,"des"tissus"vers"les"ganglions"lymphatiques"et"ensuite"les" veines."Les"ganglions"lymphatiques"sont"des"filtres"mécaniques"mais"aussi"des"barrières" immunitaires" pouvant" bloquer" la" progression" de" certains" pathogènes" et" de" cellules" cancéreuses.""

" Figure! 7!:! Présentation! du! système! lymphatique! humain! A! et! schéma! d’un! vaisseau!lymphatique!B!(Leclers!et!al.!2005)! !

40" " Origine! Deux"théories"existent"pour"expliquer"l’origine"embryonnaire"du"réseau"lymphatique."La" théorie" centrifuge" a" été" émise" par" l’anatomiste" Florence" Sabin" dès" 1902." Avec" des" injections"d’encre"de"chine"dans"des"embryons"de"porc,"Sabin"a"mis"en"évidence"que"les" vaisseaux" lymphatiques" bourgeonnaient" à" partir" de" la" veine" cardinale" (Sabin" et" al.," 1902)."Par"ailleurs,"Georges"Huntington"et"Charles"F."W."McClure,"en"étudiant"des"coupes" d’embryons" de" chat," montrent" que" les" vaisseaux" lymphatiques" proviennent" de" la" différenciation" des" cellules" mésenchymateuses" (Huntington" 1912)." Ils" utilisèrent" la" méthode" Wax" qui" consiste" en" une" reconstruction" en" trois" dimensions" à" partir" des" coupes" de" tissus." Les" théories" de" Sabin" et" Huntington" sont" respectivement" appelées" centrifuge" et" centripète." Cette" dualité" a" duré" à" peu" près" 100" ans." Les" nouvelles" technologies" ont" permis" d’élucider" ce" mystère." Les" progrès" de" la" génétique," de" la" biologie" moléculaire" et" l’imagerie" ont" réconcilié" les" deux" théories." L’identification" de" Prospero" Homeodomain" 1" (prox1)," le" gène" qui" est" exprimé" dans" certaines" cellules" endothéliales" situées" au" niveau" de" la" veine" cardinale" (Wigle" et" al." 2002)," a" permis" de" valider"la"théorie"de"Sabin."Ces"cellules"exprimant"Prox1"forment"le"bourgeon"qui"donne" naissance"au"sac"lymphatique"primitif"et"plus"tard"au"réseau"lymphatique."Prox1"code" pour"un"facteur"de"transcription"et"son"expression"est"essentielle"au"développement"du" système" lymphatique." Une" invalidation" du" gène" prox1" entrainant" plusieurs" problèmes" lors"du"développement"embryonnaire,"est"létale"chez"la"souris"(Wigle"and"Oliver"1999)." Ces" résultats" ont" été" confirmés" par" une" méthode" de" «"lineage" tracing"»" qui" permet" de" marquer" les" cellules," de" les" suivre" dans" l’organisme" et" d‘indiquer" leur" origine" embryonnaire"(Yaniv"et"al."2006)."Le"traçage"de"cellules"endothéliales"veineuses,"suivi" d’un"marquage"avec"un"anticorps"reconnaissant"Prox1"a"mis"en"évidence"l’existence"d’un" petit" nombre" de" cellules" positives" pour" Prox1" mais" pas" marquées" (non" endothéliale)." Une" expression" insuffisante" de" la" «"Cre" recombinase"»" permettant" d’activer" le" traceur" peut" expliquer" cette" observation." Cependant" ce" résultat" peut" aussi" être" la" preuve" de" l’existence" de" cellules" endothéliales" lymphatiques" ne" provenant" pas" de" la" veine" cardinale." La" théorie" de" Sabin" sur" l’origine" embryonnaire" des" vaisseaux" lymphatiques" est"maintenant"bien"établie"mais"la"participation"de"cellules"d’une"autre"origine"n’est"pas" à"exclure.""

41" " La!génétique!de!la!lymphangiogenèse!!! Chez"la"souris,"les"premières"cellules"précurseurs"de"LEC"sont"détectées"à"E9,5"au"niveau" de" la" veine" cardinale" (Srinivasan" et" al." 2007)." A" ce" stade," ces" cellules" " expriment" spécifiquement" Prox1" (Wigle" and" Oliver" 1999)." Elles" sont" alors" entrées" en" différenciation" pour" donner" des" cellules" endothéliales" lymphatiques" (LECs)." Elles" prolifèrent"et"bourgeonnent"à"partir"de"la"veine"pour"donner"le"sac"lymphatique"primitif" (Figure"8)."Cependant,"quelques"LECs"restent"au"niveau"de"la"veine"afin"de"constituer"des" valves" qui" vont" empêcher" le" reflux" vers" le" sang" (Srinivasan" and" Oliver" 2011)." L’expression" de" Prox1" est" régulée" par" Sry" Box" 18" (Sox18)" qui" est" un" facteur" de" transcription" exprimé" par" une" sous" population" de" cellules" endothéliales" dans" la" veine" cardinale"de"l’embryon"de"souris."Sox18"est"détectable"dans"l’embryon"de"souris"avant" l’apparition" de" Prox1" et" est" régulé" par" la" voie" des" MAP" Kinase/ERK" (François" et" al." 2008)."L’invalidation"de"sox18"provoque"des"œdèmes"sousGcutanés"et"est"létale"à"E14,5." La"mutation"du"gène"sox18"est"connue"chez"l’homme"pour"provoquer"des"lymphœdèmes" (Irrthum" et" al." 2003)." Sox18," en" se" fixant" sur" son" promoteur," induit" l’expression" de" Prox1" mais" aussi" d’autres" marqueurs" lymphatiques." Il" est" nécessaire" à" l’induction" de" Prox1."Cependant,"les"cellules"endothéliales"artérielles"expriment"sox18"mais"pas"prox1." Sox18"est"donc"nécessaire"mais"pas"suffisant"pour"induire"Prox1."L’expression"de"Sox18" est" perdue" dans" les" stades" tardifs" du" développement" embryonnaire" du" système" lymphatique" (Francois," Harvey," and" Hogan" 2011)." En" plus" de" Sox18" et" Prox1," le" «"Chicken" ovalbumin" upstream" promoter" transcription" factor" (CoupGTFII)"»" ou" encore" connu"sous"le"nom"de"Nr2f2,"est"un"gène"important"pour"le"développement"du"système" lymphatique." CoupGTFII" est" exprimé" dans" les" cellules" endothéliales" veineuses" " à" E8,5" (You"et"al."2005)"et"dans"les"LECs"durant"l’embryogenèse"et"à"l’âge"adulte"(Lin,"Chen,"et" al."2010)."CoupTFII"interagit"directement"avec"Prox1,"et"a"un"rôle"de"coGrégulateur"de" Prox1"dans"le"maintien"des"LECs"(Lee"et"al."2009)."Son"invalidation"provoque"un"défaut" de"formation"des"vaisseaux"lymphatiques"durant"le"développement"embryonnaire,""(Lin," Chen," et" al." 2010)." L’expression" du" VEGFRG3" est" régulée" par" COUPTFII" dans" les" LECs" (Yamazaki"et"al."2009)."COUPTFII"a"plusieurs"rôles"durant"le"développement"du"réseau" lymphatique" au" cours" du" temps." Son" action" sur" Prox1" est" nécessaire" pour" la" différenciation"et"le"maintien"des"LECs"jusqu'à"E13,5"chez"la"souris."Ensuite"l’expression" de" Prox1" devient" indépendante" de" CoupTFII" (Srinivasan" et" al." 2010)." Le" VEGFRG3" est" aussi" très" important" pour" le" développement" lymphatique." Il" joue" un" rôle" dans"

42" " l’angiogenèse" et" dans" la" lymphangiogenèse." Il" est" exprimé" d’abord" dans" les" cellules" endothéliales"des"vaisseaux"sanguins"(E10,"5)"(Dumont"et"al."1998)."Son"expression"se" restreint"ensuite"aux"cellules"endothéliales"lymphatiques"et"elle"est"réprimée"dans"les" cellules" endothéliales" vasculaires" (Wigle" et" al." 2002)." L’expression" de" VEGFRG3" est" induite" par" Prox1." Cependant" l’expression" de" Prox1" dépend" de" VEGFRG3" en" rétrocontrôle""afin"de"maintenir"l’identité"et"le"nombre"de"cellules"progénitrices"de"LEC" (Srinivasan"et"al."2014)."Le"ligand"le"mieux"caractérisé"de"VEGFRG3"est"le"VEGFGC."Il"est" exprimé" à" E10,5" par" les" cellules" mésenchymateuses" situées" à" proximité" des" LECs" bourgeonnantes" et" exprimant" VEGFRG3" (Karkkainen" et" al." 2004," Kukk" et" al." 1996)." L’invalidation"de"VEGFGC"n’a"pas"d’effet"sur"Prox1"et"donc"sur"la"différenciation"des"LECs." Par" contre," sans" VEGFGC" les" LECs" ne" bourgeonnent" pas" et" ne" forment" pas" de" sac" lymphatique"primitif"(Karkkainen"et"al."2004,"Srinivasan"et"al."2014)."Le""CollagenG"and" calciumGbinding" EGF" domain" 1" (CCBE1)" (Figure" 8)"" est" requis" pour" la" maturation" et" l’activation"de"VEGFGC"(Koltowska"et"al."2015)."Son"invalidation"induit"une"diminution" des"LEC"exprimant"Prox1"et"le"«"Lymphatic"Vessel"Endothelial"Hyaluronan"ReceptorG1" (LyveG1)"»" au" niveau" de" la" veine" cardinale." Cela" a" pour" conséquence" une" mauvaise" vascularisation""lymphatique,"de"sévères"œdèmes"et"une"mortalité"in&utéro"chez"la"souris" (Bos"et"al."2011)."LyveG1"est"un"marqueur""très"utilisé"pour"les"vaisseaux"lymphatiques."Il" est" effectivement" exprimé" très" tôt" au" niveau" de" l’endothélium" lymphatique" veineux" (Banerji"et"al."1999)."Il"est"exprimé"à"E"9.5–10.5"au"niveau"des"cellules"endothéliales"de" la"veine"cardinale"avant"l’expression"de"Prox1"(Wigle"et"al."2002)."L’invalidation"de"son" gène"n’empêche"aucunement"le"développement"normal"du"réseau"lymphatique"(Gale"et" al."2007)."Néanmoins,"il"constitue"avec"Prox1,"l’un"des"marqueurs"les"plus"précoces"de"la" lymphangiogenèse."Son"expression"n’est"pas"restreinte"aux"LECs."Il"est"présent"dans"une" sous"population"de"macrophages"et"de"cellules"de"la"rate."" La" podoplanine" (pdpn)" intervient" également" dans" la" différenciation" des" LECs." Elle" est" exprimée" par" les" cellules" fibroblastiques" réticulaires" des" ganglions" lymphatiques" " et" dans" les" LECs" (Pan" and" Xia" 2015)." Son" expression" dans" les" LECs" débute" à" E" 10,5" seulement" dans" les" cellules" exprimant" Prox1" et" qui" vont" migrer" hors" de" la" veine" cardinale"(Yang,"GarcíaGVerdugo,"et"al."2012)."Prox1"régule"directement""l’expression"de" pdpn."Des"sites"de"fixation"de"Prox1"ont"été"identifiés"sur"le"promoteur"du"gène"pdpn"et" Prox1"régule"l’expression"transcriptionnelle"de"pdpn."Pdpn"est"alors"en"aval"de"Prox1" dans"la"signalisation"induisant"le"développement"du"système"lymphatique"(Pan,"Wang,"

43" " and" Yago" 2014)." Pdpn" est" nécessaire" au" processus" d’initiation" et" de" maintien" de" la" séparation" des" vaisseaux" sanguins" et" lymphatiques." Ce" processus" fait" aussi" appel" à" l’expression"de"pdpn"à"la"surface"des"LECs"et"à"l’expression"de"«"CGType"Lectin"Domain"

Family" 2"(CLEC2)"»," un" récepteur" présent" à" la" surface" des" plaquettes" dont" l’activation" induit" leur" agrégation" afin" de" séparer" le" sang" de" la" lymphe" (Bertozzi" et" al." 2010)." L’invalidation"de"CLEC2"provoque"un"mélange"du"sang"et"de"la"lymphe"dans"l’embryon" de"souris"et"chez"la"souris"adulte"(Hess"et"al."2014)."" "

"" Figure!8!:!Mécanisme!de!différenciation!des!LECs!

Signalisation!du!développement!lymphatique.!! Le"développement"du"système"lymphatique"fait"intervenir"plusieurs"voies"moléculaires" qui" interagissent" entre" elle" pour" aboutir" à" la" formation" du" système" lymphatique." Les" membres"de"la"famille"des"«"Bone"Morphogenetic"Protein"(BMP)"»""sont"des"répresseurs" de"la"différenciation"des"LECs."Bmp2b"induit"l’expression"de"mir31"et"mir"181a"via""Sma" and" Mothers" Against" Decapentaplegic" (smad)"," ce" qui" entraine" une" inhibition" de" l’expression"de"Prox1"chez"le"poisson"zèbre"(Dunworth"et"al."2014)."Mir"31"et"mir"181a" induisent"plus"une"spéciation"veineuse."Par"un"mécanisme"inconnu"à"ce"jour,"ces"microG

44" " ARNs" vont" être" réprimés" durant" l’embryogenèse" provoquant" l’initiation" de" la" différenciation"des"LECs."Pour"les"mammifères,"les"études"chez"la"souris"ont"démontré" un"rôle"de"Bmp9"semblable"à"celui"de"Bmp2b"chez"le"poisson"zèbre."Une"invalidation"du" gène"Bmp9"provoque"une"dilatation"des"vaisseaux"lymphatiques"à"E15.5"ce"qui"suggère" une"hyperGprolifération"des"LECs"(Yoshimatsu"et"al."2013)."" La" voie" de" signalisation" Wnt" («"Wingless" plus" Int"»)" semble" aussi" intervenir" dans" la" formation"du"système"lymphatique."Wnt5b"est"nécessaire"et"suffisant"pour"induire"une" différenciation"lymphatique"chez"le"poisson"zèbre"(Nicenboim"et"al."2015)."Wnt5b"agit" via" la" voie" βGcatenin/TFC" afin" d’induire" l’expression" de" Prox1" dans" les" cellules" progénitrices" de" LECs." Cette" voie" est" conservée" au" cours" de" l’évolution" car" l’ajout" de" Wnt5b" in& vitro" sur" des" angioblastes" dérivés" de" cellules" souches" embryonnaires" est" suffisant"pour"induire"l’expression"de"Prox1"dans"ces"cellules"(Nicenboim"et"al."2015)." Ces"résultats"montrent"l’action"proGlymphatique"de"Wnt5b."" Cependant""la"voie"«"Jagged/Notch"»"a"plutôt""une"action"antiGlymphatique."Elle"inhibe" l’axe" CoupTFII/Prox1" entraînant" l’inhibition" de" la" lymphangiogenèse." Ainsi" cette" voie" permet" le" maintien" d’une" identité" veineuse" des" cellules" (Murtomaki" et" al." 2013)." La" surexpression"de"Notch"in&vitro"inhibe"l’expression"de"marqueurs"lymphatiques"via"ses" effecteurs"(Kang"et"al."2010)."Cependant,"l’invalidation"du"gène"«"Ritinol"Binding"Protein" Rbpj&»"un"des"principaux"effecteurs"de"Notch"n’a"aucun"effet"sur"l’expression"de"Prox1"et" de"LyveG1"dans"les"cellules"endothéliales"(Srinivasan"et"al."2010)."L’effet"de"Notch"sur"la" formation"des"vaisseaux"lymphatiques"pourrait"dépendre"d’une"voie"non"canonique.""

Lymphangiogenèse!et!cancers! Le"réseau"lymphatique"draine"la"lymphe"vers"les"organes"lymphatiques"et"le"cœur."Les" vaisseaux"lymphatiques"parcourent""en"certaines"régions"les"ganglions"lymphatiques"qui" constituent"une"barrière"mécanique"mais"aussi"immunitaire."Ainsi"certains"pathogènes" sont"détruits"au"niveau"des"ganglions"lymphatiques."C’est"le"cas"des"cellules"tumorales." Cependant"les"cellules"tumorales"peuvent"aussi"se"développer"dans"les"ganglions"avant" de"se"propager"dans"d’autres"organes."Ainsi"le"système"lymphatique"peut"avoir"un"rôle" pro"ou"antiGtumoral."L’étude"de"son"organisation"et"des"mécanismes"de"sa"mise"en"place" est" donc" primordiale" avant" d’envisager" le" développement" des" traitements" visant" ce" système.""

45" " "

Figure! 9!:! Vascularisation! tumorale! et! dissémination! métastatique/Description! des!évènements!séquentiels! "

Modification!du!réseau!lymphatique!tumoral! Le"système"lymphatique"intervient"lors"des"échanges"entre"les"cellules"tumorales"et"le" sang." En" effet" du" plasma" s’infiltre" constamment" dans" le" milieu" interstitiel" constituant" ainsi"la"lymphe"et"assurant"le"transport"des"nutriments"jusqu’aux"cellules."Ces"échanges" sont" bidirectionnels" et" permettent" l’évacuation" des" déchets" métaboliques." Ceci" est" indispensable"à"la"croissance"des"cellules"tumorales"qui"produisent"beaucoup"de"déchets" à" cause" de" leur" forte" prolifération." Ces" déchets" sont" donc" canalisés" par" le" réseau" lymphatique"avant"leur"élimination"via"la"circulation"sanguine"(Saharinen"et"al."2004)." Cependant,"plus"particulièrement,"les"cellules"endothéliales"lymphatiques"(LECs)"jouent" un"rôle"important"dans"le"développement"des"tumeurs"(Farnsworth,"Achen,"and"Stacker" 2006)." La" prolifération" et" la" migration" des" LECs" contribuent" au" développement" du" réseau"lymphatique."En"plus,"l’élargissement"des"vaisseaux"lymphatiques"observé"dans"

46" " plusieurs"cancers"(Dadras"et"al."2003)"est"associé"à"la"prolifération"des"LECs"(Figure"9)." Cependant,"durant"les"dernières"années,"la"recherche"sur"le"système"lymphatique"s’est" surtout"concentrée"sur"son""rôle"dans"la"dissémination"métastatique"(Padera"et"al."2002," Leu" et" al." 2000)." Les" facteurs" de" croissance" VEGFGC" et" VEGFGD" sont" les" principaux" inducteurs" de" la" lymphangiogenèse" tumorale." Jadis," le" dogme" stipulait" l’absence" de" vaisseaux" lymphatiques" au" sein" de" la" tumeur" (Leu" et" al." 2000)." " VEGFGC" et" VEGFGD" induisent"le"développement"de"vaisseaux"lymphatiques"dans"la"tumeur."Cependant,"ces" vaisseaux"semblent"non"fonctionnels"car"ils"sont"collapsés"à"cause"de"la"pression"intraG tumorale" (Shayan" et" al." 2013)." En" plus," la" présence" des" vaisseaux" lymphatiques" n’a" aucun" impact" sur" le" développement" de" métastases" lymphatiques" ou" vers" d’autres" organes"(Padera"et"al."2002)."Par"contre,"le"réseau"lymphatique"périGtumoral"peut"subir" certaines"modifications"au"cours"de"la"tumorigenèse,"notamment,"un"élargissement"des" vaisseaux"induit"par"la"sécrétion"de"VEGFGC"et"VEGFGD."Cet"élargissement"augmenterait" la"surface"de"contact"entre"les"cellules"tumorales"et"les"vaisseaux"lymphatiques"facilitant" l’entrée" des" cellules" tumorales" dans" la" circulation" lymphatique" (Leu" et" al." 2000)." Cela" pourrait"être"un"point"essentiel"pour"la"dissémination"métastatique."

Mécanisme!moléculaire!de!la!lymphangiogenèse!tumorale! L’axe"VEGFGC/VEGFGD/VEGFRG3"est"la"voie"principale"promotrice"du"développement"du" réseau""lymphatique"au"sein"de"la"tumeur"(Figure"10)."Dans"la"tumeur,"ces"facteurs"sont" sécrétés"par"les"cellules"tumorales"ellesGmêmes,"les"cellules"immunitaires"ainsi"que"les" fibroblastes" (Schoppmann" et" al." 2002)" (Alitalo" 2011)." Ils" sont" activés" après" plusieurs" étapes"de"modifications"postGtraductionnelles"et"induisent"des"métastases"lymphatiques" chez"les"modèles"animaux"(Skobe"et"al."2001,"Stacker"et"al."2001)."Le"blocage"de"cet"axe" inhibe"les"métastases"lymphatiques"ainsi"que"la"lymphangiogenèse"tumorale"(He"et"al." 2005)." Par" contre," le" blocage" de" Neuropiline" 2" (NRP2)," le" corécepteur" de" VEGFRG3" n’inhibe"que"la"migration"mais"pas"la"prolifération"des"LECs"in"vitro"et"in"vivo"et"réduit" l’incidence" de" métastases" lymphatiques" (Caunt" et" al." 2008)." Ainsi" toutes" les" voies" de" signalisation" qui" peuvent" influencer" cet" axe" vont" avoir" une" action" sur" la" lymphangiogenèse"tumorale"ainsi"que"l’incidence"des"métastases"lymphatiques."C’est"le" cas" de" WNT1" qui" inhibe" l’expression" de" VEGFGC" dans" le" mélanome" et" par" conséquent" réduit" la" lymphangiogenèse" et" les" métastases" (Niederleithner" et" al." 2012)." La" prostaglandine"et"ses"récepteurs"EP2,"EP3"et"EP4"exprimés"par"les"cellules"tumorales"et"

47" " les" cellules" immunitaires" forment" un" axe" stimulant" l’expression" de" VEGFGC" et"" promouvant" la" lymphangiogenèse" ainsi" que" les" métastases" tumorales" (Su" et" al." 2004," Zhang," Huang," et" al." 2008)." Par" ailleurs" l’inhibition" de" mTOR" inhibe" aussi" la" lymphangiogenèse"et"les"métastases"(Patel"et"al."2011)"dans"les"cancers"de"la"tête"et"du" cou."Beaucoup"d’autres"molécules"comme"le"VEGFGA,"le"FGF2,"l’EGF"etc..,"exercent"une" action" positive" sur" le" développement" du" système" lymphatique" dans" le" microenvironnement" tumoral" (Hirakawa" et" al." 2005," Cao" et" al." 2012," Bracher" et" al." 2013)."Cependant,"le"TGF"a"plutôt"une"action"répressive"sur"la"lymphangiogenèse."En" effet" l’inhibition" de" TGFβ" dans" un" model" murin" de" xénogreffe" d’adénocarcinome" pancréatique" induit" le" développement" du" système" lymphatique" via" l’expression" de" VEGFGC" (Oka" et" al." 2008)." Par" ailleurs," la" matrice" extra" cellulaire" joue" aussi" un" rôle" important"dans"le"développement"du"système"lymphatique"tumoral."Elle"sert"de"support" aux" LECs" lors" de" leur" migration" après" stimulation" par" des" facteurs" proG lymphangiogéniques.""L’intégrine"α4β1"permet"l’interaction"entre"les"LECs"et"la"matrice" extra"cellulaire."Ainsi"l’invalidation"du"gène"de"l’intégrine"41"bloque"l’induction"de" la"lymphangiogenèse"tumorale"et"les"métastases"au"niveau"des"ganglions"lymphatiques" (GarmyGSusini" et" al." 2010)." La" tétraspanine" CD9" est" nécessaire" à" la" transmission" des" signaux"entre"les"VEGFRs"et"les"intégrines."Elle"est"fortement"exprimée"sur"les"cellules" endothéliales" lymphatiques." L’invalidation" du" gène" CD9" entraine" une" baisse" de" la" lymphangiogenèse" et" des" métastases" lymphatiques" (Iwasaki" et" al." 2013)." Plusieurs" molécules" sont" donc" capables" de" moduler" le" développement" lymphatique" dans" l’environnement"tumoral."Les"nouvelles"technologies""de"criblage"génétique"permettront" certainement" de" trouver" d’autres" voies" impliquées" dans" ce" processus" car" les" mécanismes"moléculaires"provoquant"les"modifications"génétiques"dans"les"LECs"lors"de" croissance"tumorale"sont"encore"mal"connus.""

48" " "

Figure!10!:!Interactions!cellulaires!dans!le!microenvironnement!tumoral!

Chimiokines!et!métastases!! "Le"réseau"lymphatique"est"utilisé"par"les"cellules"du"système"immunitaire"pour"transiter" des" tissus" périphériques" aux" ganglions" lymphatiques" et" aux" autres" sites" lymphoïdes" secondaires." Les" chimiokines" " facilitent" ce" type" de" mouvement" cellulaire." En" effet," les" LECs"expriment"par"exemple"à"leur"surface"le"CCGchemokine"Ligand"21"(CCL21)"qui"se" lie" à" son" récepteur" le" CCR7" exprimé" par" les" cellules" dendritiques." Ainsi" les" cellules" dendritiques" peuvent" rentrer" plus" facilement" dans" le" réseau" lymphatique" (Förster," DavalosGMisslitz," and" Rot" 2008)." D’autre" part," les" cellules" cancéreuses" sont" aussi" capables" d’entrer" ou" de" se" lier" au" réseau" lymphatique" grâce" à" l’expression" de" chimiokines."Ce"processus"est"impliqué"dans"le"développement"des"métastases"(Shields," Emmett,"et"al."2007)."Par"exemple"l’expression"de"CCR7"par"les"cellules"tumorales"est" associée"à"plus"de"métastases"dans"les"cancers"gastriques,"colorectaux"et"du"sein"(Müller" et"al."2001,"Mashino"et"al."2002)."Le"chimiotactisme"autologue"est"un"autre"mécanisme" faisant"intervenir"les"chimiokines"afin"que"les"cellules"cancéreuses"puissent"entrer"dans"

49" " le"réseau"lymphatique."Ce"mécanisme"est"basé"sur"le"flux"du"fluide"interstitiel"généré"par" le" drainage" lymphatique." Par" exemple," plusieurs" types" de" cellules" cancéreuses" expriment"CCL19"et"CCL21."En"même"temps"elles"expriment"aussi"les"récepteurs"de"ces" chimiokines"le"CCR7."Le"flux"interstitiel"génère"un"gradient"de"molécules"sécrétées"pas" les" cellules" cancéreuses" qui" migrent" en" fonction" ce" gradient" vers" les" vaisseaux" lymphatiques"constitués"de"cellules"qui"expriment"leur"récepteur"(Shields,"Fleury,"et"al." 2007)" ." L’invasion" par" les" gliomes" du" tissu" nerveux" a" aussi" " été" décrite" comme" dépendante" de" ce" mécanisme" (Munson," Bellamkonda," and" Swartz" 2013)." De" plus," la" sécrétion"de"VEGFGC"par"les"cellules"tumorales"entraîne"la"surexpression"de"VEGFRG3"à" la"surface"des"cellules"endothéliales"lymphatiques"résultant"en"la"sécrétion"de"CCL21"par" ces" mêmes" LECs." Ainsi," les" cellules" tumorales" exprimant" CCR7" migrent" vers" les" LECs" (Issa"et"al."2009)."Les"chimiokines"jouent"un"rôle"dans"la"mise"en"place"des"métastases"en" régulant" l’entrée" des" cellules" tumorales" dans" les" vaisseaux" lymphatiques" (Das" et" al." 2013)." Les" chimiokines" accélèrent" le" développement" de" métastases" par" différents" mécanismes.""

Modification!du!réseau!lymphatique!au!delà!de!la!tumeur! Les"mécanismes"qui"gouvernent"les"modifications"du"réseau"lymphatique"dans"et"au"delà" de"la"tumeur"sont"différents."Pendant"que"l’élargissement"des"vaisseaux"lymphatiques"au" niveau"de"la"tumeur"dépend"du"VEGFGC"et"de"la"prolifération"des"LECs,"l’élargissement" des"vaisseaux"auGdelà"de"la"tumeur"dépend"de"la"voie"des""prostaglandines"(Karnezis"et" al."2012)."Le"VEGFGC"est"impliqué"dans"le"remaniement"du"réseau"lymphatique"auGdelà" du"ganglion"lymphatique"sentinelle"qui""est"le"premier"ganglion"drainant"la"tumeur."Ces" modifications" font" intervenir" les" cellules" musculaires" vasculaires" lisses" et" nécessitent" l’expression"de"NRP2"(Gogineni"et"al."2013).""

Remaniement!lymphatique!et!immunité!tumorale! La" plupart" des" études" sur" le" VEGFGC" et" le" VEGFGD" régulant" la" lymphangiogenèse" tumorale"ont"été"réalisées"sur"des"souris"immunoGdéficientes."Ainsi,"elles"ne"prennent" pas"en"compte"les"effets"de"la"lymphangiogenèse"sur"l’immunité"tumorale."Les"LECs"ont" certaines" caractéristiques" des" cellules" présentatrices" d’antigènes." Elles" expriment" les" Complexe"Majeur"d’Histocompatibilité"(CMH)"I"et"II."Les"LECs"peuvent"donc"stimuler"la" prolifération"et"l’activation"des"cellules"lymphocytes"T"CD8"(LT8)"dans"les"mélanomes."" En" absence" de" coGstimulation" par" l’antigène," les" lymphocytes" expriment" PDG1."

50" " L’expression"de"PDG1"inhibe"l’expression"d’interleukine"2"(IL2)."L’IL2"étant"nécessaire"à" la" survie" des" LT8," ces" derniers" vont" mourir" par" apoptose" entrainant" ainsi" une" immunotolérance" (Tewalt" et" al." 2012)." Dans" ce" contexte," le" VEGFGC" qui" induit" la" lymphangiogenèse"à"la"périphérie"de"la"tumeur"et"au"niveau"des"ganglions"lymphatiques" drainant"permet"aux"cellules"cancéreuses"d’échapper"au"système"immunitaire"(Lund"et" al." 2012)." Ces" études" ont" démontré" un" rôle" du" système" lymphatique" sur" l’immunité" tumorale"qui"mériterait"d’être"investiguée"de"manière"plus"approfondie.""

Lymphangiogenèse!:!une!cible!anti!tumorale! La"lymphangiogenèse"tumorale"est"essentiellement"régulée"par"l’axe"VEGFGC/VEGFRG3" et" VEGFGD/VEGFRG3." Elle" est" aussi" corrélée" avec" une" incidence" plus" élevée" de" métastases." Ainsi" cibler" l’axe" VEGFGC/VEGFRG3" et" VEGFGD/VEGFRG3" pourrait" être" considéré"comme"thérapie"anti"dissémination"métastatique"qui"est"la"cause"principale" des"décès"par"cancer."Ce"ciblage"restreindrait"la"tumeur"à"son"site"primaire."L’inhibition" de"cet"axe"dans"plusieurs"modèles"murins"réduit""significativement"la"lymphangiogenèse" tumorale," l’élargissement" des" vaisseaux" lymphatiques" tumoraux" et" les" métastases" au" niveau"des"ganglions"lymphatiques"(Achen,"McColl,"and"Stacker"2005,"Lin"et"al."2005,"He" et" al." 2002)." De" plus," le" VEGFRG3" est" impliqué" dans" le" développement" des" vaisseaux" sanguins" et" est" exprimé" au" niveau" des" bourgeons" vasculaires" (Siekmann" and" Lawson" 2007," Tammela" et" al." 2008)." L’inhibition" de" VEGFRG3" dans" ce" cas" pourrait" avoir" une" double" action" sur" la" croissance" tumorale" et" sur" l’apparition" de" métastases" respectivement"grâce"à"son"effet"antiGangiogénique"et"son"effet"antiGlymphangiogénique." Plusieurs"inhibiteurs"antiGlymphangiogenèse"ciblant"VEGFGC,"VEGFGD,"VEGFRG3"et"NRPG2" ont" été" développées" et" pourraient" être" bénéfiques" pour" les" patients." Cibler" d’autres" voies" impliquées" dans" le" développement" des" vaisseaux" lymphatiques" comme" l’axe" Angiopoietin" (ANGPT)" et" son" récepteur" ANGTP1" ou" encore" l’enzyme" «"Cytochrome" C" assembly"oxidase"factor"(COX)"»"important"pour"l’élargissement"des"vaisseaux"drainant" la" tumeur" pourrait" représenter" une" approche" thérapeutique" pertinente." Cependant," aucun" traitement" antiGlymphangiogenèse" n’a" été" approuvé" contre" le" cancer" bien" que" deux"brevets"existent"sur"des"anticorps"dirigés"contre"le"VEGFGC"(WO2011071577A1"et" WO201127519A1)." Evidemment" ce" domaine" de" recherche" doit" être" approfondi." Il" possède" un" fort" potentiel" dans" le" développement" de" traitements" innovants" et" de" nouveaux"outils"de"diagnostic"et"de"pronostic"pour"les"patients.""

51" " La!régulation!d’expression!du!VEGFBC! Le""Vascular"Endothelial"Growth"Factor"(VEFGGC)"fait"partie"de"la"famille"du"VEGF/PDGF." Sa" principale" fonction" connue" est" d’induire" la" lymphangiogenèse" en" se" fixant" sur" le" VEGFRG3"exprimé"à"la"surface"des"LECs"(Joukov"et"al."1996)."Il"joue"aussi"un"rôle"dans"la" formation" des" vaisseaux" sanguins" et" la" perméabilité" des" vaisseaux" via" VEGFRG3" (Tammela" et" al." 2008)""ou" encore" son" second" récepteur" le" VEGFRG2." Durant" le" développement" embryonnaire," le" VEGFGC" intervient" aussi" dans" le" développement" neuronal"(Le"Bras"et"al."2006)"

Les!facteurs!de!transcription! Le"VEGFGC"est"le"principal"facteur"de"croissance"induisant"le"développement"d’un"réseau" lymphatique"normal"ou"tumoral.""La"régulation"de"son"expression"est"assez"méconnue" mais" les" études" récentes" notamment" dans" le" cadre" du" cancer" ont" permis" de" mieux" comprendre" ce" mécanisme." Par" exemple," le" facteur" de" transcription" «"Sine" oculis" homoebox" homolog" 1" (Six1)"»" qui" est" surexprimé" dans" les" cancers" du" sein," stimule" la" transcription" de" vegfOc" ce" qui" est" corrélé" avec" une" augmentation" du" nombre" de" métastases"(Wang"et"al."2012)."Dans"le"cancer"de"la"prostate,"le"facteur"de"transcription" NKX3.1"qui"s’exprime"de"manière"spécifique"dans"les"cellules"de"la"prostate"contrôle"la" transcription"de"vegfOc."NKX3.1"qui"est"réprimé"dans"les"cancers"de"la"prostate,"inhibe" l’expression"du"VEGFGC"en"coopérant"avec"une"histone"désacétylase"(Zhang,"Muders,"et" al."2008)."Plus"intéressant,"un"site"de"fixation"du"«"nuclear"factor"kappa"B"(NFκB)"»"a"été" identifié" sur" le" promoteur" de" vegfOc." Ce" résultat" suggère" un" rôle" de" VEGFGC" dans" l’inflammation." De" plus," NκB" stimule" la" transcription" de" vegfOc" dans" le" cancer" de" la" vésicule" biliaire" en" se" fixant" directement" sur" son" promoteur" (Du" et" al." 2014)" afin" de" promouvoir" la" lymphangiogenèse." Plusieurs" autres" facteurs" de" transcription" comme" COUPTFII," «"high" mobility" group" box" 1" (HMGB1)"»" régulent" l’expression" de" VEGFGC" probablement"de"manière"indirecte"puisque"l’interaction"directe"entre"le"promoteur"de" vegfOc"et"ces"facteurs"de"transcription"n’a"pas"été"démontrée"(Lin,"Chen,"et"al."2010,"Chen," Tang,"and"Yu"2012).""

Les!cytotkines,!facteurs!de!croissance!et!matrice!extracellulaire!! La" présence" d’un" site" de" fixation" de" NFκB" sur" le" promoteur" de" vegfOc" suggère" que" l’expression"de"VEGFGC"pourrait"être"régulée"par"des"stimuli"inflammatoires."En"effet,"les" cytokines" inflammatoires" comme" l’interleukine" 1β (IL1β)" et" l’interleukine" 6" (IL6)"

52" " induisent" une" augmentation" de" la" quantité" d’ARNm" de" VEGFGC" (Shinriki" et" al." 2011)." D’autre"part,"plusieurs"facteurs"de"croissances"dont"le"TGF,"le"PDGF"et"l’EGF"activent" l’expression"de"VEGFGC"(Enholm"et"al."1997,"Ristimäki"et"al."1998)."L’enzyme"COXG2"est" coGexprimé"avec"le"VEGFGC"dans"plusieurs"cancers"notamment"dans"ceux"de"l’œsophage" dont" la" composante" inflammatoire" est" importante" (von" Rahden" et" al." 2005)." De" plus," l’endoglycosidase" héparanase" responsable" du" clivage" de" l’héparine" sulfate" stimule" l’expression"de"VEGFGC"et"la"lymphangiogenèse"(CohenGKaplan"et"al."2008)."Cet"enzyme" joue" un" rôle" dans" le" remodelage" de" la" matrice" extracellulaire." Enfin," un" produit" d’épissage" alternatif" de" la" fibronectine" active" le" VEGFGC" dans" les" cellules" de" cancer" colorectal"(Xiang"et"al."2012)."Il"existe"plusieurs"voies"de"signalisation"impliquées"dans" l’expression" de" VEGFGC";" p38" MAP" Kinase/NFκB," PI3K/Akt" et" MAPK/ERK" selon" le" contexte"cellulaire"(Shinriki"et"al."2011,"Luangdilok"et"al."2011,"Tsai"et"al."2003)."

Régulation!postJtranscriptionnelle!de!VEGFJC! Les"microGARNs"constituent"un"mécanisme"de"plus"en"plus"étudié"dans"la"régulation"de" l’expression" des" gènes." Ainsi" de" récentes" études" ont" montré" que" les" microGARNs" mir1862,"mir27b"et"mir128"inhibent"l’expression"de"VEGFGC"et"la"progression"tumorale" (Ye"et"al."2013,"Hu"et"al."2014)."Un"autre"exemple"de"régulation"postGtranscriptionnelle"a" été" décrit" récemment" avec" une" surexpression" de" VEGFGC" en" hypoxie" via" un" «"Internal" Ribosomal"Entry" Site"»" sur" l’ARNm" de" VEGFGC" (Morfoisse" et" al." 2014)" et" de" VEGFGD" (Morfoisse"et"al"Cancer"Res"2016)."En"plus"de"toutes"ces"étapes"de"régulation"de"VEGFGC," la" protéine" subit" un" ensemble" de" processus" protéolytiques" permettant" sa" maturation" (Joukov"et"al"EMBO"J"1996)."

Maturation!de!la!protéine!VEGFJC!!! Une"fois"produite,"la"protéine"VEGFGC"subit"un"processus"de"protéolyse"complexe."Ces" modifications"sont"assurées"par"des"proGprotéines"convertases"(PC)":"furine,"PC5"et"PC7" (Siegfried"et"al."2003)."L’ARNm"de"VEGFGC"code"pour"une"protéine"de"419"acides"aminés" d’environ"59"kDa"(Joukov"et"al."1996)."Ce"proGVEGFGC"ainsi"synthétisé"est"composé";"i)" d’une"séquence"signal"des"acides"aminés"1"à"12";"ii)"d’un"domaine"NGterminal"des"acides" aminés"13"à"102";"iii)""du"«"VEGF"homology"domain"»"situé"entre"les"acides"aminés"103"et" 227";"iiii)."Un"domaine"cGterminal"allant"des"acides"aminés"228"à"419."Le"proGVEGFGC"est" produit" et" sécrété" sous" forme" d’homodimère." Il" est" ensuite" clivé" au" niveau" du" site" HSIIRR227SL"(Figure"11)."Ce"clivage"donne"naissance"à"une"partie"de"31kDa"qui"est"clivée"

53" " à"nouveau"par"une/des"protéases"inconnues"pour"obtenir"un"VEGFGC"mature"de"21kDa." Le" site" de" clivage" HSIIRR227SL" est" un" site" universel" de" clivage" reconnu" par" les" proG protéines" convertases." Il" faut" noter" le" clivage" du" proGVEGFGC" n’est" pas" nécessaire" à" la" sécrétion"de"la"protéine."Ainsi,"une"inhibition"chimique"des"PCs,"ou"l’expression"d’une" version"mutante"non"clivable"du"VEGFGC"résulte"en"une"accumulation"de"la"molécule"pro" VEGFGC" dans" le" milieu" de" culture" in" vitro" (Siegfried" et" al." 2003)." La" forme" mature" du" VEGFGC"a"plus"d’affinité"avec"le"VEGFRG3"même"si"la"forme"31"kDa"est"aussi"capable"de"se" fixer"au"récepteur.""

Figure!11!:!Maturation!et!mode!d’action!du!VEGFBC!

54" " La!fonction!du!VEGFBC!

Fonction!canonique!! Le" VEGFGC" code" pour" une" protéine" qui" est" sécrétée." Il" atteint" sa" forme" mature" après" clivage"par"les"PCs."Au"cours"de"l’embryogenèse,"il"est"indispensable"à"la"migration"des" cellules"endothéliales"du"sac"lymphatique"primaire."L’invalidation"du"gène"vegfOc"chez"la" souris"induit"la"mort"embryonnaire"à"cause"de"problèmes"au"niveau"de"la"formation"du" système"lymphatique."Pendant"le"développement"du"système"lymphatique,"le"VEGFRG3" exprimé" à" la" surface" des" cellules" endothéliales" lymphatiques" se" lie" au" VEGFGC." Cette" interaction"entraîne"la"prolifération,"la"survie"et"la"migration"des"LECs"(Mäkinen"et"al." 2001)." Le" VEGFGC" se" lie" également" au" VEGFRG2" mais" cette" voie" est" plutôt" liée" à" l’angiogenèse."En"plus"des"VEGFRs,"le"VEGFGC"se"lie"aussi"au"neuropilines"(Figure"12)." Les" neuropilines" sont" une" famille" de" corécepteurs" (NRP1" et" NRP2)" des" VEGFRs" qui" régulent"la"signalisation"de"ces"derniers."Le"VEGFGC"a"plus"d’affinité"avec"NRP2"et"cette" interaction" stimule" l’activation" de" VEGFRG2" ou" 3" (Favier" et" al." 2006)." NRP2" est" nécessaire" à" la" formation" des" petits" vaisseaux" lymphatiques" et" des" capillaires" et" son" activation" potentialise" les" effets" de" la" voie" VEGFGC/VEGFRG3" au" cours" de" la" lymphangiogenèse"(Xu"et"al."2010)."La"liaison"du"VEGFGC"au"VEGFRs"peut"induire"à"la" fois"la"lymphangiogenèse"et"l’angiogenèse"montrant"la"complexité"de"cette"voie."Chez"des" patients" atteints" de" divers" cancers," une" augmentation" de" la" quantité" de" VEGFGC" a" été" décrite" (Decio" et" al." 2014)." Cette" augmentation" de" VEGFGC" est" associée" à" une" augmentation"de"la"densité"de"vaisseaux"lymphatiques"et"à"un"mauvais"pronostic"dans" les"cancers"du"sein,"du"colon"et"de"la"prostate"(Mohammed"et"al."2007,"Soumaoro"et"al." 2006)."Comme"durant"l’embryogenèse,"le"VEGFGC"est"un"facteur"de"croissance"induisant" principalement" la" lymphangiogenèse" dans" les" cancers." Cependant," cette" lymphangiogenèse" associée" à" l’expression" de" VEGFGC" est" responsable" de" métastases" ganglionnaires"régionales"et"aussi"de"métastases"plus"éloignées"de"la"tumeur"primaire" (Skobe"et"al."2001,"Rinderknecht"and"Detmar"2008).""

55" " ""

Figure!12!:!Représentation!de!la!famille!VEGF!et!leurs!récepteurs!!

Fonctions!non!canoniques!du!VEGFJC! Au"cours"de"l’embryogenèse,"les"VEGFGC"se"lient"aux"VEGFRs."Ces"récepteurs"du"VEGFGC" se"trouvent"principalement"au"niveau"des"LECs."Cependant,"dans"un"contexte"tumoral," de" plus" en" plus" d’études" mettent" en" évidence" l’existence" d’une" fonction" autocrine" du" VEGFGC."Ainsi,"dans"les"leucémies"aigues"myéloblastiques,"une"coGexpression"de"VEGFGC" et" de" VEGFRG3" a" été" décrite" (Fielder" et" al." 1997)," suggérant" une" activité" autocrine" du" VEGFGC" sur" les" cellules" cancéreuses." Ensuite," l’axe" VEGFGC/VEGFRG3" stimule" la" production"de"la"protéine"de"survie"BclG2"qui"limiterait"la"réponse"aux"chimiothérapies"" (Dias"et"al."2000)."Le"VEGFGC"est"également"impliqué"dans"les"phénomènes"de"migration," d’invasion"et"de"prolifération"des"cellules"de"cancer"du"sein,"du"poumon,"de"la"vessie"et" de"l’œsophage"(Su"et"al."2006,"Timoshenko,"Rastogi,"and"Lala"2007)."Par"ailleurs,"l’axe" VEGFGC/NRP2"joue"aussi"un"rôle"dans"la"survie"des"cellules"de"cancer"de"la"prostate"en" condition"de"stress"oxydatif"(Muders"et"al."2009)."L’expression"de"VEGFGC"en"condition" de" stress" induit" l’activation" de" mTOR" via" la" phosphorylation" d’Akt" et" NRP2" est" aussi" impliquée" dans" cette" voie." VEGFGC" et" NRP2" joue" aussi" un" rôle" dans" la" résistance" " des" cellules"tumorale"en"activant"l’autophagie"dans"les"cellules"cancéreuses"via"«"Lysosomal"

56" " Associated" Membrane" Protein" 2" (LAMPG2)"»" et" «"WD" Repeat" and" FYVE" Domain" Containing" (WDFYG1)"»" (Stanton" et" al." 2013)." Cependant," NRP2" n’a" pas" de" domaine" catalytique"dans"sa"partie"cytoplasmique"et"dont"il"n’existe"aucune"preuve"que"NRP2"est" capable"de"transmettre"un"signal"seul."Cependant,"le"domaine"plasmatique"de"NRP1"peut" stimuler"de"manière"indépendante"du"VEGFRG2"la"voie"de"signalisation"MAP"Kinase"et" PI3"kinase/AKT"(Cao"et"al."2008)."NRP2"se"lie"normalement"aux"récepteurs"VEGF"et"aux" plexins" mais" il" se" lie" aussi" aux" intégrines." En" effet," la" transition" epithélioG mésenchymateuse"des"cellules"de"cancer"du"poumon"induite"par"TGFβ"se"fait"via"NRP2" qui" est" aussi" impliqué" dans" " les" métastases" via" l’intégrine" 5" (Cao" et" al." 2013)." Ces" fonctions"de"NRP2"ne"sont"pas"forcément"liées"à"VEGFGC."Par"contre,"le"VEGFGC"joue"un" rôle" majeur" dans" les" processus" d’EMT" et" d’initiation" tumorale" associés" aux" cellules" souches"tumorales."Dans"le"cancer"du"poumon,"l’inhibition"du"VEGFGC"induit"une"baisse" des" marqueurs" mésenchymateux" et" une" augmentation" des" marqueurs" épithéliaux" accompagnées" d’un" affaiblissement" des" caractéristiques" souches" (Khromova" et" al." 2012)."Cependant,"les"mécanismes"expliquant"comment"VEGFGC"et"NRP2"induisent"les" caractéristiques"souches"et"l’EMT"sont"encore"inconnus."En"plus"des"cellules"tumorales," d’autres" cellules" sécrètent" du" VEGFGC." Les" cytokines" inflammatoires" produites" par" les" cellules" tumorales," stimulent" la" production" de" VEGFGC" par" des" cellules" immunitaires" comme"les"macrophages"dans"le"microenvironnement"tumoral"(Figure"13)."L’expression" de" VEGFGC" dans" ce" microenvironnement" régule" aussi" la" réponse" immunitaire." Ainsi," l’expression" de" VEGFGC" dans" les" cancers" gastriques" est" inversement" corrélée" à" l’infiltration" des" cellules" dendritiques" (Takahashi" et" al." 2002)." De" plus," l’inhibition" du" VEGFRG3" des" «"Natural" Killers" (NK)"»" dans" les" leucémies" aigues" myéloïdes" restore" l’activité"cytotoxique"des"NKs"(Lee"et"al."2013)"et"comme"vu"plus"haut,"le"VEGFGC"induit" une" immunotolérance" en" supprimant" les" LT8" au" niveau" des" ganglions" lymphatiques" drainant" " les" mélanomes" (Lund" et" al." 2012)." L’ensemble" de" ces" études" montre" l’importance"du"VEGFGC"dans"l’immunité"tumorale,"l’EMT"et"les"caractéristiques"souches.""

57" " ""

Figure!13!:!Les!cibles!de!VEGFBC!dans!le!microenvironnement!tumoral!!

EMT!!

La!transition!épithéliumBmésenchyme! La" transition" épithéliumGmésenchyme" (EMT)" est" un" processus" par" lequel" des" cellules" perdent" leurs" caractéristiques" épithéliales" et" acquièrent" des" caractéristiques" mésenchymateuses"(Figure"14)."L’épithélium"est"un"ensemble"de"cellules"qui"constituent" en" général" une" barrière." Ces" cellules" ont" aussi" une" fonction" sécrétrice." Elles" se" caractérisent" par" une" polarité" basoGlatérale" et" l’expression" de" protéines" de" jonctions" serrées." Cependant" les" cellules" mésenchymateuses" servent" plutôt" de" soutient" ou" d’ancrage."Elles"assurent"une"fonction"dans"plusieurs"processus"dont"la"cicatrisation"et"la" réparation" tissulaire." L’EMT" est" un" processus" réversible";" des" cellules" mésenchymateuses"peuvent"aussi"acquérir"des"caractéristiques"épithéliales"(Figure"14)." On" parle" alors" de" transition" mésenchymeGépithélium" ou" «"mesenchymal" epithelial" transition"»"(MET)."L’EMT"a"été"décrite"en"premier"durant"l’embryogenèse"(Hay"1995)." L’EMT" correspond" au" fait" que" des" cellules" épithéliales" perdent" leurs" jonctions," leur" polarité"basoGapicale,"réorganisent"leur"cytosquelette"et"acquièrent"une"polarité"avantG

58" " arrière."Des"changements"au"niveau"de"certaines"voies"de"signalisation"aboutissent"à"une" augmentation" de" leurs" capacités" invasives" (Thiery" et" al." 2009)." Ces" changements" ont" pour"conséquence"la"modification"du"profil"d’expression"génique."L’EMT"est"cependant" réactivée"chez"l’adulte"durant"les"processus"de"cicatrisation,"lors"de"la"fibrose"et"même" dans" les" cancers" (Thiery" et" al." 2009," Chapman" 2011)." Selon" le" contexte" et" le" type" cellulaire," les" cellules" épithéliales" peuvent" avoir" de" manière" concomitante" des" propriétés"épithéliales"et"mésenchymateuses."On"parle"alors"d’EMT"partielle.""

Figure!14!:!Représentation!schématique!des!changements!lors!de!l’EMT!et!la!MET!!

Les!marqueurs!de!l’EMT!

Modifications!des!jonctions!! L’un" des" signes" d’EMT" est" la" perte" de" l’expression" de" la" protéine" EGcadhérine" qui" déstabilise"les"jonctions"adhérentes."Toutes"les"protéines"impliquées"dans""le"maintien" des"jonctions"apicales"serrées"et"des"desmosomes""vont"être"réprimées"(Huang,"Guilford," and" Thiery" 2012)." La" fonction" de" barrière" des" cellules" épithéliales" n’est" ainsi" plus"

59" " assurée." En" revanche," les" protéines" qui" supportent" les" adhésions" mésenchymateuses" sont"exprimées."Ainsi"la"NGcadhérine"qui"participe"à"la"réorganisation"du"réseau"d’actine" est"surexprimée"(Yilmaz"and"Christofori"2009).""

Modification!du!cytosquelette!et!motilité! Les"gènes"codant"des"protéines"du"cytosquelette"sont"aussi"modifiés"durant"l’EMT."Ainsi," la"composition"des"filaments"intermédiaires"change"avec"le"remplacement"de"kératine" par" la" vimentine" (Huang," Guilford," and" Thiery" 2012)." La" keratine" et" la" vimentine" régulent" le" transfert" des" protéines" membranaires" et" le" trafic" des" organelles" mais" chacune" transfère" différentes" protéines" à" la" membrane." Par" exemple," la" kératine" transporte"l’EGcadhérine"vers"la"membrane"mais"pas"la"vimentine"(Toivola"et"al."2005)." Les" changements" au" niveau" des" filaments" intermédiaires" augmentent" la" motilité" cellulaire" à" cause" probablement" d’interactions" entre" ces" filaments" et" les" protéines" motrices"(Mendez,"Kojima,"and"Goldman"2010)."De"plus,"les"projections"membranaires" grâce" à" la" production" de& novo" d’actine" facilitent" la" motilité." Ces" projections" correspondent" à" des" lamellipodes" ou" des" pseudopodes." Grâce" à" leur" fonction" protéolytiques," ces" projections" dégradent" la" matrice" extracellulaire" et" facilitent" l’invasion"(Ridley"2011)."Une"augmentation"de"la"contractilité"et"la"formation"des"fibres" de"stress"d’actine"accompagnent"l’EMT."Ces"modifications"d’expression"et"d’activité"de" l’actine" sont" régulées" par" les" protéines" RHO" GTPases." RHOA" active" la" formation" des" fibres" de" stress" alors" que" RAC1" et" CDC42" induisent" la" formation" des" pseudopodes" et" lamellipodes"(Ridley"2011).""

Les!facteurs!de!transcription!de!l’EMT!!

SNAILS! Les" changements" observés" durant" une" EMT" sont" régulés" principalement" par" des" facteurs" de" transcriptions" comme" SNAIL," TWIST" et" «"zing" finger" EGbox" binding" (ZEB)"».""Il"existe"plusieurs"SNAIL"mais"seul"SNAIL1"appelé"SNAIL"et"SNAIL2"aussi"connu" sous"le"nom"de"SLUG"sont"impliqués"dans"le"processus"d’EMT."Elles"inhibent"l’expression" des"gènes"épithéliaux"comme"EGCatherine."SNAIL1"via"son"domaine"«"zinc"finger"»"se"fixe" sur"la"séquence"EGbox"située"sur"le"promoteur"d’EGcadhérine"(Batlle"et"al."2000)."SNAIl1" induit" des" modifications" épigénétiques" sur" le" promoteur" de" ces" gènes" cibles," principalement" des" méthylation" et" acétylation." Ces" modifications" entrainent" soit" une"

60" " activation" soit" une" répression" du" gène" selon" le" type" de" méthylation" (Lin," Ponn," et" al." 2010,"Dong"et"al."2012)"au"niveau"de"«"domaines"bivalents"»"retrouvés"sur"le"promoteur" d’EGcadhérine." SNAIL1" active" d’autre" part" des" gènes" codant" des" protéines" «"mésenchymateuses"».""

Les!facteurs!de!transcription!basal!helixJloopJhelix!(bHLH)! Parmi"les"facteurs"de"transcription"bHLH,"E12,"E47,"TWIST1,"TWIST2"jouent"un"rôle"très" important" durant" l’EMT." Par" exemple" l’expression" de" TWIST1" entraîne" la" répression"" d’EGcadhérine"et"l’expression"de"la"NGcadhérine"(Yang,"Sun,"et"al."2012)."TWIST"est"induit" via"l’activation"de"plusieurs"voies"de"signalisation."HIF1"induit"l’expression"de"TWIST"en" condition"hypoxique"et"provoque"ainsi"l’EMT"et"la"dissémination"des"cellules"tumorales." TWIST1" et" TWIST2" forment" des" homodimères" et" peuvent" aussi" former" des" hétérodimères" avec" E12" et" E47" pour" une" régulation" de" transcription" de" gènes" spécifiques.""Cependant,"la"protéine"«"inhibitor"of"differentiation"(ID)"»"peut"former"un" hétérodimére" avec" ces" facteurs" de" transcription." Vu" que" la" protéine" ID" n’a" pas" de" domaine" de" liaison" à" l’ADN," elle" constitue" un" inhibiteur" naturel" de" ces" facteurs" de" transcription."La"répression"d’ID"par"le"facteur"de"croissance"TGF"entraine"l’activation" de"TWIST"et"d’autres"protéines"bHLH"et"ainsi"induit"l’EMT"(Kang,"Chen,"and"Massagué" 2003)." Enfin," comme" SNAIL," TWIST" est" aussi" régulé" par" phosphorylation" (Figure" 15)." TWIST"peut"être"phosphorylé"par"sur"la"sérine"68"par"les"MAPKs"ce"qui"le"protège"de"la" dégradation"(Hong"et"al."2011).""

61" " " Figure!15!:!Régulation!des!facteurs!de!transcription!lors!de!l’EMT!

Les!facteurs!de!transcription!ZEB! ZEB1"et"ZEB2"se"fixent"aux"séquences"EGbox"afin"de"réprimer"ou"d’induire"l’expression" des"gènes"(Peinado,"Olmeda,"and"Cano"2007,"Xu,"Lamouille,"and"Derynck"2009)."ZEB1" réprime" l’expression" des" gènes" en" se" liant" à" un" corépresseur," le" «"CGterminal" binding" protein"»" (CTBP)." Cependant," il" inhibe" la" transcription" indépendamment" de" CTBP" (SánchezGTilló" et" al." 2010)." Par" ailleurs," ZEB1" active" l’expression" de" gènes" en" interagissant"avec"des"coGactivateurs"comme"p300"(Postigo"et"al."2003)."Comme"SNAIL" et" TWIST," ZEB" se" lie" au" EGbox" des" promoteurs" et" active" l’expression" des" gènes" mésenchymateuses""ou"réprime"l’expression"des"gènes"épithéliaux."L’expression"de"ZEB" suit"souvent"celle"de"SNAIL"suggérant"une"régulation"de"ZEB"par"SNAIL."TWIST"et"SNAIL" coopèrent"pour"induire"l’expression"de"ZEB"(Dave"et"al."2011)."TGFβ,"les"protéines"WNT" et" les" facteurs" de" croissance" activant" les" MAPKs" activent" l’expression" de" ZEB" (Xu," Lamouille,"and"Derynck"2009)."D’autres"facteurs"de"transcription"jouent"un"rôle"dans"le" processus" d’EMT." Ils" appartiennent" à" la" familles" des" «"forkhead" box"»" (FOX)" (Eijkelenboom" and" Burgering" 2013)" ou" à" celle" des" GATA" (Guanine" Adenine" Thymine"

62" " Adenine)"(Bresnick"et"al."2010)"qui"comme"SNAIL,"TWIST"et"ZEB,"régulent"l’expression" des" protéines" impliquées" dans" les" jonctions" épithéliales" et" la" polarité" des" cellules" épithéliales" (Campbell" et" al." 2011)." Parmi" cette" diversité" de" facteurs" de" transcription," certains"sont"nécessaires"et"d’autres"ont"une"activité"restreinte"sur"l’EMT.""

Les!inducteurs!d’EMT!

Les!protéines!de!la!famille!du!TGF! La"famille"des"TGFβ"est"composée"de"trois"TGFβ,"deux"activines"et"plusieurs"BMPs."Ils" sont"les"ligands"de"récepteurs"kinase"transmembranaires."TGFβ1"induit"l’EMT"lors"de"la" cicatrisation," le" développement" des" fibroses" et" des" cancers." Dans" les" carcinomes," l’expression"du"TGFβ1"induit"une"plasticité"des"cellules"épithéliales"résultant"en"l’EMT" indispensable" à" l’invasion" et" la" dissémination" des" cellules" tumorales" (Katsuno," Lamouille,"and"Derynck"2013)."D’autres"membres"de"la"famille"du"TGFβ"induisent"l’EMT." La" voie" de" signalisation" de" BMP" avec" WNT" et" FGF" participe" à" l’induction" de" la" dissémination" des" cellules" de" la" crête" neurale" (SaukaGSpengler" and" BronnerGFraser" 2008)." Pareillement," BMP2," 4" et" 7" induisent" l’EMT," l’invasion" des" cellules" épithéliales" pancréatiques"associée"à"la"répression"d’EGcadhérine"et"l’expression"de"MMP2"(Gordon" et"al."2009)."L’EMT"est"induite"par"le"TGFβ"via"un"contrôle"subtil"de"transcription"par"les" protéines"de"la"famille"SMAD"(Figure"16)."En"effet,"TGFβ"entraine"l’expression"de"SNAIL1" via" SMAD3" (Hoot" et" al." 2008)." De" plus," le" complexe" SMAD3GSMAD4" participe" à" la" répression" de" la" protéine" ID1" induite" par" TGFβ" et" entraine" l’expression" de" TWIST1" (Kang,"Chen,"and"Massagué"2003)."Enfin,"l’expression"de"certains"gènes"lors"de"l’EMT"est" contrôlée" par" les" protéines" SMADs" de" manière" indépendante" de" la" transcription." Parallèlement"aux"SMADs,"le"TGFβ"active"les"voies"des"RHO"likeGGTPase,"PI3K"et"MAPKs" afin" d’induire" l’EMT" (Xu," Lamouille," and" Derynck" 2009," Zavadil" and" Böttinger" 2005," Zavadil" et" al." 2001)." L’activation" de" RHO," RAC1" et" CDC42" est" responsable" de" la" réorganisation"des"filaments"d’actine"et"la"formation"des"pseudopodes"et"lamellipodes" (Ridley"2011)."Ensuite,"AKT"est"activée"via"PI3K"et""mTOR"dans"les"cellules"épithéliales" entrant" en" EMT" (Bakin" et" al." 2000," Lamouille" et" al." 2012)." Dans" ce" processus," le" complexe" mTORC1" contribue" à" l’augmentation" de" la" taille" des" cellules," de" la" synthèse" protéique," de" la" motilité" et" de" l’invasion" (Lamouille" and" Derynck" 2007)" alors" que" mTORC2" agit" plutôt" au" niveau" de" la" transition" du" phénotype" épithélial" au" phénotype" mésenchymateuse" (Lamouille" et" al." 2012)." Enfin," le" TGFβ" active" aussi" les" voies" de"

63" " signalisation" des" MAPKs" ERK," p38" et" JUN" «"NGterminal" Kinase"»." Cela" va" permettre" de" soutenir" les" effets" de" l’EMT" déjà" en" place" en" stabilisant" SNAIL1" par" exemple" via" inhibition"de"GDK3β,"augmentant"ainsi"l’activité"de"SNAIL"(Marchetti"et"al."2008).""

" Figure!16!:!la!voie!du!TGFβ!et!son!mode!d’action!lors!de!l’EMT!

Les!récepteurs!tyrosine!kinase!(RTK)! Les" RTKs" sont" plutôt" connus" pour" induire" la" prolifération" cellulaire." Mais" en" interagissant" avec" leur" ligand," ils" activent" plusieurs" voies" de" signalisation" notamment" celle"des"MAPKs"et"PI3K/Akt"dont"les"rôles"respectifs"viennent"d’être"discutés."Ces"voies" sont"souvent"activées"par""les"facteurs"de"croissance."Par"ailleurs"le"FGF1"induit"l’EMT" dans"les"cancers"de"la"vessie"par"l’induction"de"l’expression"de"SNAIL1,"la"déstabilisation" des"desmosomes,"l’expression"de"l’intégrine"α2β1"et"MMP13"(Vallés"et"al."1996)."Comme" le" FGF1," plusieurs" autres" facteurs" de" croissance" induisent" l’EMT." Ainsi" l’"«"insulin" like" growth"factor"1(IGF1)"»"induit"l’EMT"dans"certains"modèles"de"cellules"en"culture."Outre" les"RTKs,"l’activation"des"voies"de"signalisation"Notch"et"Hedgehog"résulte"en"l’induction" d’EMT."Les"conditions"environnementales"comme"l’hypoxie"via"l’expression"de"HIF1α"et" conséquemment" de" SNAIL1" induisent" également" l’EMT" (Yang" and" Wu" 2008)."

64" " L’implication" de" tous" ces" mécanismes" montre" combien" l’EMT" est" un" processus" complexe."Il"doit"être"vu"comme"un"phénomène"transitoire"en"réponse"à"un"stimulus."Les" modèles"de"culture"cellulaire,"les"études"du"développement"et"des"cancers"ont"permis" d’effectuer"de"grandes"avancées"dans"la"compréhension"de"l’EMT."Dans"le"cancer,""l’EMT" n’est" pas" un" phénomène" isolé." Il" fait" partis" d’un" programme" plus" complexe" lié" entre" autre"à"l’établissement"et"au"maintien"des"cellules"souches"cancéreuses.""

Les!cellules!souches!cancéreuses!(CSC)! Les"CSCs"représentent"une"minorité"de"cellules"dans"la"tumeur."Elles"sont"définies"par" leurs"caractéristiques"souches."Les"autres"cellules"de"la"tumeur"sont"dépourvues"de"ces" caractéristiques"(AlGHajj"et"al."2003)."L’origine"de"ces"cellules"est"controversée"à"savoir"si" les" CSCs" viennent" des" cellules" souches" ou" progénitrices" normales" qui" ont" acquis" des" propriétés" cancéreuses" ou" de" cellules" " pleinement" différenciées" qui" se" seraient" dédifférenciées""pour"réacquérir"des"propriétés"souches"à"cause"de"mutations"(Cabillic" and" Corlu" 2016)." Plusieurs" études" ont" été" réalisées" afin" d’élucider" cette" énigme." Finalement,"les"deux"voies"peuvent"être"la"source"des"CSCs"dépendant"du"contexte"et"du" type"cellulaire.""

Caractéristiques!des!CSCs!! Elles" sont" multi" ou" pluripotentes" et" dotées" d’un" potentiel" d’autoGrenouvèlement" et" différenciation"en"plusieurs"types"cellulaires."Ainsi,"elles"prolifèrent"indéfiniment."Elles" subissent" des" divisions" symétriques" pour" donner" deux" cellules" filles" avec" des" caractéristiques"souches"identiques"à"la"cellule"mère."Elles"peuvent"aussi"se"diviser"de" façon"asymétrique"pour"donner"une"cellule"souche"et"une"cellule"plus"différenciée."Elles" parviennent" ainsi" à" maintenir" leur" nombre" dans" la" tumeur" tout" en" générant" d’autres" types"cellulaires"(Pardal,"Clarke,"and"Morrison"2003)."Les"études"sur"les"AML"ont"mis"en" évidence" des" protéines" de" surface" exclusivement" exprimées" par" les" CSCs" (Bonnet" and" Dick"1997,"Lapidot"et"al."1994)."Ainsi"des"cellules""CD34+"CD38G"ont"la"capacité"de"former" des" grandes" colonies" dans" les" souris" SCID" immunoGdéficientes" et" de" présenter" un" potentiel" d’autoGrenouvèlement" et" de" propriétés" multipotentes" comme" les" souches" normales." D’autres" marqueurs" de" cellules" souches" ont" été" décrits" dans" les" tumeurs" solides"comme"le"CD44+,"CD24"low"(AlGHajj"et"al."2003)."Ces"cellules"isolées"reconstituent" l’hétérogénéité" tumorale." L’expression" de" CD133" a" été" observée" sur" les" cellules"

65" " initiatrices"de"tumeur"dans"les"cancers"du"cerveau"(Singh"et"al."2004),"du"colon"(O'Brien" et"al."2007)"et"du"pancréas"(Hermann"et"al."2007)."Une"protéine"d’adhésion"EPCAM"a"été" identifiée"non"seulement"sur"les"cellules"épithéliales"progénitrices"normales"(Yamashita" et" al." 2009)" mais" aussi" sur" les" cellules" initiatrices" de" tumeur" dans" les" carcinomes" hépatocellulaires." Cependant" il" existe" des" marqueurs" de" CSC" spécifiques" d’un" type" cellulaire" comme" l’aldéhyde" déshydrogénase" 1" (ALDH1)" pour" le" cancer" du" sein" (Ginestier"et"al."2007)."" Les" transporteurs" ABC" (ATP" Binding" Cassette)" (ABCB1" et" ABCG2)" composent" une" famille" de" protéines" transmembranaires" surexprimées" à" la" surface" des" CSCs." Ces" transporteurs"ont"pour"fonction"l’efflux"de"substances"(déchets"métaboliques)"hors"de"la" cellule." Sur" les" CSCs," ils" contribuent" à" la" résistance" aux" chimiothérapies." Les" cellules" CD133+"isolées" de" cancers" du" cerveau" surGexprimant" la" protéine" BCRP1" (ABCG2)" sont" résistantes"à"la"chimiothérapie"alors"que"les"mêmes"cellules"exprimant"des"taux"faibles" de" BCRP1" sont" sensibles" aux" même" drogues" (Liu" et" al." 2006)." Les" CSCs" génèrent" des" tumeurs" résistantes" aux" thérapies" (Singh" and" Settleman" 2010)." Les" traitements" anticancéreux"standard"ciblant"la"masse"tumorale"n’ont"pas"d’effet"sur"les"CSCs"(Bao"et" al." 2006," Diehn" et" al." 2009)." Grâce" à" leur" capacité" initiatrice" de" tumeur" elles" peuvent" reformer"une"masse"tumorale"provoquant"une"récidive"(Figure"17).""

" Figure!17!:!Schéma!théorique!des!propriétés!des!cellules!souches!cancéreuses!

66" " Les!voies!de!signalisation!!

La!voie!hedgehog! "La"voie"hedgehog"a"été"mise"en"évidence"chez"la"drosophile."Elle"est"activée"durant"le" développement" embryonnaire" notamment" lors" de" polarisation" des" segments" chez" la" drosophile."Cette"voie"régule"aussi"la"prolifération,"la"migration"et"la"différenciation"de" plusieurs"types"cellulaires."Elle"induit"la"sécrétion"de"VEGF"et"d’IGF."Dans"le"cancer,"elle" active"la"voie"Akt"primordiale"pour"l’entrée"dans"le"cycle"cellulaire"des"CSCs"(Merchant" and" Matsui" 2010)." BMI1" («"B" cellGspecific" Moloney"murine" leukemia" virus"integration" site"1"»),"une"autre"cible"de"la"voie"hedgehog,"inhibe"les"suppresseurs"de"tumeur"comme" P16INK4a" et" PTEN." L’inhibition" de" PTEN" active" la" voie" Akt" résultant" en" l’EMT" via" l’expression"de"SNAIL"et"l’inhibition"d’EGcadhérine"(Wang"et"al."2013)"

La!voie!du!TGF! TGFβ1," β2" et" β3" forment" la" famille" des" protéines" TGFβ." Ils" activent" deux" types" de" récepteurs" TβR1" et" TβR2" pour" induire" une" fonction" biologique." Les" principaux" effecteurs"de"cette"voie"sont"les"protéines"SMADs."La"phosphorylation"de"R1"provoque" l’activation"de"SMAD"2"et"3"qui"se"lient"à"SMAD4."Le"complexe"SMAD2/3/4"s’associe"à" d’autres"facteurs"de"transcription"afin"de"réguler"l’expression"des"gènes"impliqués"dans" la" différenciation" cellulaire." Ainsi" les" SMADs" interagissent" avec" les" facteurs" de" transcription"ZEB"(ZEB1et"ZEB2/SIP1)"(Comijn"et"al."2001)"pour"réprimer"l’expression" d’EGcadhérine"lors"de"l’EMT.""«"Smad"Interacting"Protein"1"(SIP1)"»"qui"est"essentielle"au" maintien"des""cellules"souches"embryonnaires"humaines"vient"renforcer"le"lien"entre"la" voie"TGFβ,"l’EMT"et"les"cellules"souches"(Chng"et"al."2010)."Dans"le"cancer"du"sein,"la"voie" du"TGFβ"est"suractivée"dans"les"cellules"métastatiques"CD44+"et"CD24G"démontrant"une" fois"de"plus"le"rôle"du"TGFβ"dans"l’EMT"et"la"promotion"de"l’état"souche"(Shipitsin"et"al." 2007).""

La!voie!Wnt! La"stimulation"de"cette"voie"entraîne"l’activation"de"la"protéine"βGcaténine."La"liaison"de" Wnt"avec"son"récepteur"Frizzled"provoque"la"fixation"des"corécepteurs"«"LDL"Receptor" Related"Protein"(LRP5"et"LRP6)"»."En"absence"de"ligand,"βGcaténine"est"dégradée"par"le" protéasome"(Liu"et"al."2010)."En"présence"de"ligand,"Gcaténine"est"stable"et"transloque" dans"le"noyau"afin"d’activer"les"facteurs"de"transcription"TCF/LEF."Ces"facteurs"induisent"

67" " la"prolifération"et"la"différenciation"des"cellules."La"voie"Wnt"joue"un"rôle"très"important" dans"l’auto"renouvellement"des"cellules"souches"cancéreuses"dans"le"cancer"du"sein."La" surexpression" de" LPR6" est" associée" aux" tumeurs" les" plus" agressives" (Liu" et" al." 2010)." D’autres"voies"de"signalisation"sont"impliquées"dans"le"processus"de"mise"en"place"et/ou" de"maintien"des"cellules"souches"cancéreuses."La"voie"Notch"par"exemple"régule"l’autoG renouvèlement"et"la"différenciation"des"cellules"souches"neurales"et"CSCs."Il"faut"surtout" prendre"conscience"que"toutes"ces"voies"sont"interconnectées"ce"qui"rend"très"complexe" l’éradication"des"CSCs.""

Résistance!aux!traitements,!EMT!et!CSCs! Les" thérapies" antiGtumorales" sont" souvent" sujettes" à" résistance." Cette" résistance" peut" être"intrinsèque."Ainsi,"le"patient"est"réfractaire"d’emblée"au"traitement."De"manière"plus" fréquente,"le"patient"peut"répondre"de"manière"transitoire"mais"rechute"à"cause"d’une" résistance"acquise."Une"partie"de"ces"résistances"acquises"est"due"à"des"mutations"sur" certains" gènes" (Engelman" et" al." 2007)." Une" autre" partie" est" la" conséquence" de" modifications" épigénétiques" réversibles" contrairement" aux" mutations" acquises." Ces" modifications"sont"associées"à"des"changements"d’état"de"différenciation"et"à"l’EMT"qui" provoquent"la"dédifférenciation"des"cellules"en"CSCs"(Voulgari"and"Pintzas"2009)."L’état" de"différenciation"de"la"tumeur"contribue"également"à"la"résistance"de&novo."Lorsque"les" tumeurs" sont" traitées" avec" des" inhibiteurs" d’EGFR," les" cellules" épithéliales" surG exprimant"EGcadhérine"sont"plus"sensibles"alors"que"les"cellules"résistantes"présentent" des" propriétés" mésenchymateuses" (Witta" et" al." 2006)" Les" cancers" urothéliaux" musculaires" invasifs," par" opposition" cancers" urothéliaux" superficiels" sont" de" mauvais" pronostic"avec"un"taux"de"mortalité"élevé."Les"cellules"de"ces"tumeurs"présentent"des" propriétés" mésenchymateuses" associées" à" une" expression" de" ZEB" 2" et" SIP1" qui" les" protègent" de" l’apoptose" induite" par" les" dommages" à" l’ADN" (Sayan" et" al." 2009)." Les" cancers"qui"ont"propriétés"mésenchymateuses"sont"associés"à"une"mortalité"plus"élevée." Cependant," même" dans" les" cancers" du" sein" épithéliaux" avec" une" amplification" de" HER2/neu,"les"cellules"CD44+"CD24G"sont"résistantes"aux"inhibiteurs"de"HER2"(Li"et"al." 2008)." Par" contre," les" cancers" «"mésenchymateux"»" sont" plus" sensibles" aux" chimiothérapies"néoGadjuvantes"que"les"cancers"«"épithéliaux"»"(Carey"et"al."2007,"Paik" et" al." 2006," Liedtke" et" al." 2008)." Malgré" une" réponse" initiale" supérieure," ces" patients"" présentent" un" pronostic" vital" plus" péjoratif" suggérant" que" les" cellules" mésenchymateuses" sont" plus" aptes" à" développer" des" résistances." Les" cancers" du" type" 68" " épithélial"sont"initialement"plus"sensibles"aux"thérapies"ciblées,"comme"les"inhibiteurs" d'EGFR"et"d'HER2"alors"que"les"cancers"mésenchymateux"sont"plus"sensibles"aux"agents" induisant"des"dommages"à"l'ADN."" ""

69" " " RESULTATS"

" " " " " " " " " " " " " " " " " " " " " " " " " " " " "

70" " Article!1!: CXCL7 is a predictive marker of sunitinib efficacy in clear cell renal cell carcinomas

" Malgré"le"développement"de"plusieurs"agents"thérapeutiques"contre"les"ccRCC,"aucune" de" ces" thérapies" n’est" curative" de" nos" jours." De" plus," l’hétérogénéité" de" la" tumeur"" primaire"et"des"métastases"(Gerlinger"et"al."2014)"entraine"une"grande"variabilité"dans" les"réponses"des"patients"aux"traitements"comme"le"sunitinib."En"plus"d’être"inefficace" sur"certains"patients"le"sunitinib"a"un"coût"non"négligeable."Afin"de"remédier"à"cela,"la" mise" en" évidence" de" marqueurs" de" prédictifs" de" sensibilité" aux" traitements" est" essentielle." Les" patients" seraient" mieux" orientés" vers" le" traitement" adéquat" dans" l’arsenal" thérapeutique" disponible" contre" les" ccRCCs." Les" chimiokines" " ELR" CXCL" (chimiokine"CXC"contenant"un"motif"acide"glutamique,"leucine""argine"(ELR))"sont"des" cibles" thérapeutiques" pertinentes" dans" les" ccRCCs" (Giuliano" et" al." 2014," Grépin," Ambrosetti,"et"al."2014)."Dans"ce"contexte,"la"cytokine"CXCL7"et"ses"récepteurs""CXCR1"et" CXCR2"qui"sont"exprimés"par"les"cellules"tumorales","les"plaquettes"(von"Hundelshausen," Petersen," and" Brandt" 2007)," les" cellules" endothéliales" et" les" cellules" immunitaires" (Galliera," Corsi," and" Banfi" 2012)" forment" un" axe" capable" de" modifier" l’environnement" tumoral."Nous"avons"donc"étudié"le"rôle"prédictif"de"CXCL7"sur"l’efficacité"du"sunitinib,"le" traitement"de"référence"des"ccRCC"métastatiques"(Motzer"and"Bukowski"2006).""" " !!

71" " FULL PAPER

British Journal of Cancer (2017), 1–7 | doi: 10.1038/bjc.2017.276

Keywords: CXCL7; ccRCC; sunitinib; predictive marker; plasma

CXCL7 is a predictive marker of sunitinib efficacy in clear cell renal cell carcinomas

Maeva Dufies1,2,12, Sandy Giuliano1,2,12, Julien Viotti3, Delphine Borchiellini4, Linsay S Cooley5, Damien Ambrosetti6, Me´lanie Guyot1,13, Papa Diogop Ndiaye1, Julien Parola1, Audrey Claren1,4, Renaud Schiappa3, Jocelyn Gal3, Antoine Frangeul4, Arnaud Jacquel7, Ophe´lie Cassuto4, Renaud Gre´pin2, Patrick Auberger7, Andre´as Bikfalvi5, Ge´rard Milano4, Bernard Escudier8, Nathalie Rioux-Leclercq9, Camillo Porta10, Sylvie Negrier11, Emmanuel Chamorey3, Jean-Marc Ferrero4 and Gilles Page`s*,1 1University of Nice Sophia Antipolis, Institute for Research on Cancer and Aging of Nice, CNRS UMR 7284, INSERM U1081, 33 av. Valombrose, Centre Antoine Lacassagne, France; 2Centre Scientifique de Monaco, Biomedical Department, 8 Quai Antoine Ier, Monaco MC-98000, Principality of Monaco; 3Centre Antoine Lacassagne, Statistic Department, Nice, France; 4Centre Antoine Lacassagne, Clinical Research Department, Nice, France; 5University of Bordeaux, INSERM U1029, Pessac, France; 6Department of Pathology, Nice University Hospital, University of Nice Sophia Antipolis, Nice, France; 7University of Nice Sophia Antipolis, Centre Me´ diterrane´ en de Me´ decine Mole´ culaire (C3M), INSERM U1065, Nice, France; 8Institut Gustave Roussy, Villejuif, France; 9Department of Pathology, University Hospital, Rennes, France; 10Medical Oncology, I.R.C.C.S. San Matteo University Hospital Foundation, Pavia, Italy and 11Centre Le´ on Be´ rard, Lyon, France

Background: Sunitinib is one of the first-line standard treatments for metastatic clear cell renal cell carcinoma (ccRCC) with a median time to progression shorter than 1 year. The objective is to discover predictive markers of response to adapt the treatment at diagnosis.

Methods: Prospective phase 2 multi-centre trials were conducted in ccRCC patients initiating sunitinib (54 patients) or bevacizumab (45 patients) in the first-line metastatic setting (SUVEGIL and TORAVA trials). The plasmatic level of CXCL7 at baseline was correlated with progression-free survival (PFS).

À Results: The cut-off value of CXCL7 for PFS was 250 ng ml 1. Patients with CXCL7 plasmatic levels above the cut-off at baseline À (250 ng ml 1) had a significantly longer PFS (hazard ratio 0.323 (95% confidence interval 0.147–0.707), P ¼ 0.001). These results were confirmed in a retrospective validation cohort. The levels of CXCL7 did not influence PFS of the bevacizumab-treated patients.

Conclusions: CXCL7 may be considered as a predictive marker of sunitinib efficacy for ccRCC patients.

Metastatic clear cell renal cell carcinoma (ccRCC) is a highly for the use of anti-angiogenic treatments targeting the VEGF/ angiogenic tumour. Most of the cases harbour a mutation, deletion VEGFR pathway. Several tyrosine kinase inhibitors (TKIs) have or methylation of the von Hippel Lindau gene, leading to prolonged progression-free survival (PFS) in patients with overexpression of VEGF. Therefore, ccRCC represent a paradigm metastatic disease. Sunitinib, a multi kinase inhibitor targeting

*Correspondence: Dr G Page` s; E-mail: [email protected] 12These authors contributed equally to this work. 13Present address: Institut de Pharmacologie Mole´ culaire et Cellulaire, Unite´ Mixte de Recherche 7275, Centre National de la Recherche Scientifique, Paris, France. Received 27 April 2017; revised 4 July 2017; accepted 24 July 2017 r 2017 Cancer Research UK. All rights reserved 0007 – 0920/17 www.bjcancer.com | DOI:10.1038/bjc.2017.276 Advance Online Publication: 29 August 2017 1 BRITISH JOURNAL OF CANCER CXCL7 is a predictive marker of sunitinib efficacy in RCC

VEGF, PDGF, CSF1 receptors, c-KIT, FLT3 and RET, was the first or uncontrolled brain metastases, an estimated lifetime less than 3 to be approved in the first-line setting, with a median PFS of 11 months, uncontrolled hypertension or clinically significant cardi- months. (Motzer et al, 2009). Other targeted therapies have been ovascular events (heart failure, prolongation of the QT interval), developed, including TKIs such as axitinib (Motzer et al, 2013) or history of other primary cancer. All patients gave written informed pazopanib (Escudier et al, 2014), which inhibit VEGFR1, 2, 3, consent. Tumours were assessed at baseline and then every 12 PDGFR, c-KIT and the VEGF-directed humanised monoclonal weeks by thoracic, abdominal, pelvic and bone CT scans. Brain CT antibody bevacizumab (used in combination with interferon scans were performed in case of symptoms. (IFN)-a) (Escudier et al, 2010), and the mTOR inhibitors Study design (SUVEGIL and TORAVA trials). The prospective everolimus (Motzer et al, 2010) and temsirolimus (Motzer et al, cohort includes patients from the SUVEGIL (38 patients) and 2008). More recently, cabozantinib, which inhibits VEGFR, c-MET TORAVA (16 patients) trials. and AXL (Choueiri et al, 2015), and the anti-PD1 immune The SUVEGIL trial (clinicaltrial.gov, NCT00943839) was a checkpoint inhibitor nivolumab, were proposed as other alter- multicentre prospective single-arm study. The goal of the trial is to natives (Motzer et al, 2015). Although these treatments have finally determine whether a link exists between the effectiveness of improved the clinical outcome of metastatic ccRCC, efforts are therapy with sunitinib malate and development of blood needed to identify the best treatment regimen according to patient biomarkers in patients with kidney cancer. Patients received oral profile. The current therapeutic practices are based on two sunitinib (50 mg per day) once daily for 4 weeks (on days 1 to 28), assumptions: (i) the treatment must destroy blood vessels to followed by 2 weeks without treatment. Courses repeat every 6 eliminate the tumours and (ii) endothelial cells are normal cells weeks in the absence of disease progression or unacceptable that cannot adapt to the selection pressure exerted by the anti- toxicity. angiogenic treatments. However, tumour cells also express The TORAVA trial (clinicaltrial.gov, NCT00619268) was a receptors targeted by anti-angiogenic drugs and this may randomised prospective study. Patient characteristics and results contribute to tumour adaptation and relapse. Another unexpected have been previously described (Negrier et al, 2011). Briefly, aspect associated with the use of anti-angiogenesis treatments is the patients aged 18 years or older with untreated metastatic ccRCC heterogeneity of the patients’ response. Some patients are were randomly assigned (2 : 1 : 1) to receive the combination of refractory right away and die rapidly, although others have a À bevacizumab (10 mg kg 1 iv every 2 weeks) and temsirolimus transient response and then relapse. A minority of patients are (25 mg iv weekly), or the combination of IFN-a (9 mIU iv three responders for a very long period of time (Motzer et al, 2009; À times per week) and bevacizumab (10 mg kg 1 iv every 2 weeks), Escudier et al, 2010). These results indicate that ccRCC is a or one of the standard treatments: sunitinib (50 mg per day orally heterogeneous disease with variable clinical evolution (Gerlinger for 4 weeks followed by 2 weeks off) (Negrier et al, 2011). et al, 2012). If the treatment targeted only the genetically stable These studies were approved by the ethic committee at each vasculature, a more homogeneous response would be predicted. participating centre and run in agreement with the International Therefore, the treatments induced a ‘Darwinian’ adaptation of Conference on Harmonization of Good Clinical Practice Guideline. tumour cells in relation to the microenvironment. This conclusion lead to two observations: (1) the necessity to identify predictive Retrospective validation cohort. Thirty-one patients from the markers of efficacy; (2) the need for the identification of University hospitals of Rennes (France) and Pavia (Italy), treated ‘druggable’ targets participating in progression on anti-angiogenic with sunitinib (50 mg per day, once daily for 4 weeks), were treatments that should be independent of the VEGF/VEGFR axis. analysed retrospectively. Preclinical studies have suggested that ELR þ CXCL chemokines Efficacy and safety. Blood samples were collected during the (C-X-C motif Chemokine ligand containing the glutamic acid, inclusion visit (baseline) and at the end of the four weeks of leucine arginine (ELR) motif) may represent prognosis markers of sunitinib administration at each cycle for biochemical analysis. survival in ccRCC patients and may constitute relevant therapeutic Plasmatic level of CXCL7 was correlated with OS and PFS, targets as evidenced in experimental tumours in mice (Grepin et al, defined respectively as the time from inclusion in the trial to death 2012; Grepin et al, 2014). As VEGF/VEGFR, the ELR þ CXCL from all causes (for OS) and to progression, treatment cessation or cytokine CXCL7 and its specific receptors CXCR1 and CXCR2 are death (for PFS), censored at last follow-up for those still alive or produced by tumour cells, platelets (von Hundelshausen et al, who have not progressed. 2007), endothelial and immune cells (neutrophils and macrophages) (Galliera et al, 2012). They are involved in inflammation Memorial Sloan-Kettering Cancer Centre (MSKCC/Motzer) and angiogenesis. This generates simultaneously autocrine and score. Memorial Sloan-Kettering Cancer Centre score predicts paracrine loops that impact the microenvironment. survival based on clinical and laboratory data in metastatic RCC The current standard of care for ccRCC is to administer anti- patients. It is a combination of: (1) performance status (Karnofsky angiogenic therapies such as sunitinib. ELR þ CXCL chemokines score)o80%, (2) time from diagnosis to systemic treatmento12 may represent relevant predictive markers for sunitinib efficacy. months, (3) haemoglobin less than the lower limit of normal, (4) We investigated whether CXCL7 chemokine easily measured in lactate dehydrogenase41.5 Â upper limit of normal, (5) corrected plasma samples, is a relevant predictive markers of sunitinib calcium410 mg dl À 1 (2.5 mmol l À 1). Score 0, 1–2 or X3 efficacy. corresponding respectively to Good, Intermediate or Bad Risk Group (Motzer et al, 1999). Biochemical analysis. Blood samples were centrifuged (10 000 g MATERIALS AND METHODS for 10 min) and the plasmas were collected and conserved at À 80 1C. Plasmatic level of CXCL7 (1/100 dilution) was determined by ELISA using Peprotech kits (reference 900-K40). Patients. Eligible patients for SUVEGIL and TORAVA trials were at least 18 years of age and had metastatic ccRCC histologically Quantitative real-time PCR experiments. One microgram of confirmed, with the presence of measurable disease according to total RNA was used for the reverse transcription, using the Response Evaluation Criteria in Solid Tumors v1.1. Patients had QuantiTect Reverse Transcription kit (QIAGEN, Hilden, Germany), not received previous systemic therapy for RCC and were eligible with blend of oligo (dT) and random primers to prime first-strand for sunitinib or bevacizumab combined with IFN treatment in the synthesis. SYBR master mix plus (Eurogentec, Liege, Belgium) was first-line setting. Patients were ineligible if they had symptomatic used for quantitative real-time PCR (qPCR). The oligonucleotides

2 www.bjcancer.com | DOI:10.1038/bjc.2017.276 CXCL7 is a predictive marker of sunitinib efficacy in RCC BRITISH JOURNAL OF CANCER used for qPCR experiments are described in Supplementary Table 1. Characteristics of the patients included in the study Table S2. Treatment Sunitinib Bevacizumab Sunitinib Statistical analysis. Progression-free survival was defined as the Number of patients 54 45 31 time between blood sample collection and progression, or death Gender from any cause, censoring those alive and progression free at last Female 10 (18.52%) 12 (26.67%) 7 (22.58%) follow-up. OS was defined as the time from blood sample Male 44(81.48%) 33(73.33%) 24(77.42%) collection to the date of death from any cause, censoring those Primary tumour À 1 alive at last follow-up. The CXCL7 cut-off point (250 ng ml ) for Fuhrman grade PFS was determined using spline curves analysis. T-test was 1 1 (1.85%) 1 (2.22%) applied to compare continuous variables and chi-square test, or 2 11(20.37%) 9(20%) 11(35.48%) Fisher’s exact test (when application condition of w2-test were not 3 23 (42.59%) 17 (37.78%) 8 (25.8%) 4 12 (22.22%) 11 (24.44%) 1 (3.22%) fulfilled), were used for categorical variables. Kaplan–Meier NA 7 (12.96%) 7 (15.56%) 11 (35.48%) method was used to produce survival curves and analyses of censored data were performed using log-rank test. To guarantee pT 1 10(18.52%) 8(17.78%) 4(12.9%) the independence of CXCL7 as a predictive factor from validate 2 12(22.22%) 7(15.56%) 7(22.58%) predictive factor, multivariate analysis were performed using cox X3 25(46.30%) 22(48.89%) 20(64.52%) regression adjusted on MSKCC score. Adjusted hazard ratio (HR) NA 7 (12.96%) 8 (17.78%) and 95% confidence interval (95% CI) were calculated. pN Smoothing splines curves for HR were used to determined cut- 0 17 (31.48%) 21 (67.74%) off for censored data. 1 2 (3.7%) 1 (3.22%) The predictive cut-off determined in the prospective cohort was 2 4 (7.4%) 6 (19.35%) validated by using same cut-off, same definition for PFS and OS, NA 31(57.41%) 45(100%) 3(9.68%) and same statistical models in data from retrospective cohort. All Beginning of the treatment analyses were performed using R software, version 3.2.2 (Vienna, Metastatic from the 21(38.39%) 16(35.56%) 11(35.48%) Austria, https://www.r-project.org/). diagnosis Time from diagnosis Healthy donor plasma. Plasmas were obtained from healthy to treatment donors with informed consent following the Declaration of o1 yr 27 (50%) 26 (57.78%) 14 (45.16%) Helsinki according to recommendations of an independent X1 yr 27 (50%) 19 (42.22%) 17 (54.84%) scientific review board. Number of metastatic sites 1 27 (50%) 22 (48.89%) 11 (35.48%) 2 17 (31.48%) 13 (28.89%) 6 (19.35%) RESULTS X3 10(18.52%) 10(22.22%) 14(45.16%) Risk factor (MSKCC) Patient and pathological parameters. Fifty-four patients were Good 18(43.90%) 6(14.29%) 7(22.58%) prospectively enrolled in the SUVEGIL (clinicaltrial.gov, Intermediate 15 (36.59%) 13 (30.95%) 5 (16.13%) NCT00943839) and TORAVA (clinicaltrial.gov, NCT00619268) Bad 8 (19.51%) 23 (54.76%) 19 (61.29%) trials, and treated by sunitinib. At diagnosis, the median age was NA 13 3 62.96 years. All patients were nephrectomised. Twenty-one Median follow-up 27.7 (23.4–34.4) 24 (23.1–26.2) 39.4 (28.9–50.6) patients (38%) had metastasis at diagnosis. Twelve (22%) patients (month) had Fuhrman grade 1–2 and 35 (65%) patients had Fuhrman grade CXCL7 À 3–4 tumours. The time from diagnosis to apparition of metastasis Average (ng ml 1) 264.02 283.8 286.57 (139.3–485.2) (148–381) (171.8–490) was inferior to 1 year for 27 patients (50%) and superior to 1 year À o250 ng ml 1 22 (40.74%) 9 (20%) 9 (29.03%) for 27 patients (50%). Forty-one of 54 patients were evaluated for À X250 ng ml 1 32 (59.26%) 36 (80%) 22 (70.97%) the MSKCC score. Eighteen patients (43.9%) had a good MSKCC ¼ score, 15 patients (36.59%) had intermediate MSKCC score and 8 Abbreviation: MSKCC Memorial Sloan-Kettering Cancer Centre. patients (19.51%) had a poor MSKCC score. Forty-five patients were prospectively enrolled in the TORAVA clinical trial and treated by bevacizumab in combination with IFN-a or temsir- (for PFS), censored at last follow-up for those still alive or who olimus. Forty-two of 45 patients (93.3%) were evaluated for the have not progressed. MSKCC score. Six patients (14.29%) had a good, 13 (30.95%) had The median PFS was 20.5 months and the median OS was 38.4 and intermediate and 23 (54.76%) had a bad MSKCC score. The months (Supplementary Figure S1). The plasmatic level of CXCL7 population characteristics and pathological parameters are sum- was measured at diagnosis and throughout the different cycles of marised in Table 1. sunitinib treatments. No correlation was observed between disease Thirty-one patients from the University hospitals of Rennes progression and the variation in cytokine plasmatic levels along time. Only plasmatic levels at diagnosis were correlated to PFS. The (France) and Pavia (Italy) were included in a validation retro- À 1 spective cohort. Seven patients (22.58%) had a good, 5 patients CXCL7 cut-off point (250 ng ml ) for PFS was determined using spline curves analysis (R software, version 3.2.2). (16.13%) had an intermediate and 19 patients (61.29%) had a bad À Patients with CXCL7 plasma levels below 250 ng ml 1 (range MSKCC score. The patients’ characteristics are also summarised in À 139.2–250 ng ml 1) had a shorter median PFS (12.6 months) Table 1. À compared with patients with plasma levels above 250 ng ml 1 À Progression-free survival in prospective/validation cohorts (range 250–485.2 ng ml 1, 27.7 months, P ¼ 0.001; HR 0.285 (CI 95% 0.161–0.504)) (Figure 1A). (sunitinib group) and correlation to CXCL7 plasmatic level. À The levels of CXCL7 (cut-off: 250 ng ml 1) were also correlated Plasmatic level of CXCL7 was correlated with OS and PFS, defined À respectively as the time from inclusion in the trial to death from all with OS. Indeed, patients with plasma levels below 250 ng ml 1 causes (for OS) and to progression, treatment cessation or death had a lower median OS (23.5 months) compared with patients with www.bjcancer.com | DOI:10.1038/bjc.2017.276 3 BRITISH JOURNAL OF CANCER CXCL7 is a predictive marker of sunitinib efficacy in RCC

ABProspective cohort Retrospective cohort sunitinib group sunitinib group 100 100 CXCL7 <250 ng ml–1 CXCL7 <250 ng ml–1 CXCL7 >250 ng ml–1 CXCL7 >250 ng ml–1 80 80

60 60

40 40

20 20 P P

Progression-free survival (%) Progression-free =0.001 survival (%) Progression-free =0.0002 0 0 0 12 24 36 48 0 12 24 36 48 Months Months Median survival Median survival Plasma levels Plasma levels (months) (months) CXCL7 high 27.7 CXCL7 high 14.1

CXCL7 low 12.6 CXCL7 low 10.4

Figure 1. Relationship between plasmatic levels of CXCL7 and PFS of ccRCC patients treated with sunitinib in prospective and retrospective cohort. (A and B) Kaplan–Meier analysis of PFS of patients with ccRCC treated with sunitinib. Progression-free survival (PFS) was calculated from À patient subgroups with plasmatic level for CXCL7 at the diagnosis that were less or greater than a cut-off value of 250 ng ml 1, for SUVEGIL and TORAVA trials—prospective analysis (A) or for retrospective analysis (B). Statistical significance (P value) and the time of the median disease free are indicated.

Table 2. Clinical and biological parameters and multivariate Prospective cohort analysis of patients 100 Bevacizumub group CXCL7 <250 ng ml–1 P Variable Description Risk ratio (IC 95% OR) -value CXCL7 >250 ng ml–1 À Biological o250 ng ml 1 10.030 80 parameter À CXCL7 4250 ng ml 1 0.341 (0.126–0.923) 60 Clinical Good 1 parameters MSKCCscore Intermediate 1.37(0.472–3.979) 0.55 40 Bad 3.13 (0.993–9.865) 0.04 Abbreviations: MSKCC ¼ Memorial Sloan-Kettering Cancer Centre; PFS ¼ progression-free 20 P survival. Multivariate analysis between CXCL7, MSKCC score and PFS. survival (%) Progression-free =NS

0 À 1 P ¼ plasmatic levels above 250 ng ml (not reached, 0.047; HR 0 12 24 36 0.374 (CI 95% 0.177–0.79)) (Supplementary Figure S3A). Months These results were confirmed in a validation retrospective À 1 cohort. Indeed, patients with plasma levels below 250 ng ml had Plasma levels Median survival a shorter median PFS (10.4 vs 14.1 months, P ¼ 0.0002; HR 0.207 (months) (CI 95% 0.371–0.313)) (Figure 1B). CXCL7 high 10.6 To guaranty the independence of our biological parameter, it was important to show that CXCL7 was not a surrogate marker of clinical CXCL7 low 11.4 parameters. As shown in Supplementary Table S1, the levels of CXCL7 (inferior or superior at 250 ng ml À 1) and clinical parameters Figure 2. Relationship between plasmatic levels of CXCL7 and PFS of þ of patients in the prospective cohort (age, gender, Fuhrman grade, ccRCC patients treated with bevacizumab INF in prospective pT, pN, pM or MSKCC score) are not correlated. The biological and cohort. Kaplan–Meier analysis of PFS of patients with ccRCC treated þ clinical parameters (levels of CXCL7 and MSKCC scores, the with bevacizumab INF. Progression-free survival (PFS) was calculated from patient subgroups with plasmatic level for CXCL7 at the diagnosis standard score used for patient evaluation in clinical practices) were À 1 then analysed in a multivariate Cox regression model on PFS that were less or greater than a cut-off value of 250 ng ml , for P (Table 2). CXCL7 expression was identified as an independent TORAVA trial—prospective analysis. Statistical significance ( value) ¼ prognostic parameter for PFS (P ¼ 0.03, HR 0.341 (CI 95% 0.126– and the time of the median disease free are indicated. NR not 0.926)). A similar results was obtained for a bad MSKCC score with reached. respect to PFS (P ¼ 0.04, HR 3.13 (CI 95% 0.993–9.865), Table 2). TORAVA clinical trial that were treated with bevacizumab þ IFN. Progression-free survival in prospective cohort (bevacizumab þ The median PFS was 10.6 months and the median OS was 24.1 IFN group) and correlation to CXCL7 plasmatic levels. To months (Supplementary Figure S2). The CXCL7 plasmatic levels determine the predictive role of CXCL7 for sunitinib efficacy, we did not discriminate patients with a long or a short PFS tested the plasmatic levels of CXCL7 in a subset of patients of the (Figure 2A) or OS (Supplementary Figure 3B).

4 www.bjcancer.com | DOI:10.1038/bjc.2017.276 CXCL7 is a predictive marker of sunitinib efficacy in RCC BRITISH JOURNAL OF CANCER

Additional exploratory analyses. Post-hoc analysis was conducted poor prognosis). The intra-tumour CXCL7 mRNA amounts and to better define the role of CXCL7 in the tumour. the plasmatic CXCL7 levels are inversely correlated (correlation In silico available transcriptomic data showed that high CXCL7 coefficient (CC) À 0.79, Figure 3C). Physiologically, CXCL7 are mRNA levels correlated with tumour stage (Supplementary Figure produced by platelets (von Hundelshausen et al, 2007) and S4A) and PFS (Supplementary Figure S4B). We compared the level immune cells (neutrophils and macrophages) (Galliera et al, 2012), of CXCL7 in the plasma of 24 healthy donors and 37 metastatic and are involved in inflammation and angiogenesis. Hence, we ccRCC patients (following surgical removal of the primary analysed the correlation between intra-tumour CXCL7 amounts tumour). The level of CXCL7 is lower in ccRCC patients and the amounts of immune cells. Intra-tumour CXCL7 is (Figure 3A). Moreover, patients that relapse on sunitinib have positively correlated with neutrophils (N, CC ¼ 0.8) and M2 decreased plasmatic CXCL7 levels compared to responsive macrophages invasion (M2, CC ¼ 0.85; Figure 3D). Colonisation of patients. Their CXCL7 plasmatic concentrations are not signifi- ccRCC by these cells has previously been associated with a poor cantly different as compared to healthy donors (Figure 3B). prognosis (Santoni et al, 2014; Song et al, 2015). The above-mentioned results were apparently discordant (high To explain the sequence of events during tumour development, intra-tumour levels and low plasmatic levels both correlated with we tested plasmatic and intra-tumour CXCL7 levels in

AB * * 800 800 NS ) ) –1 600 –1 600

400 400

200 200 CXCL7 (ng ml CXCL7 (ng ml

0 0 Healthy donors RCC patients Healthy Responsive Relapse donors patients patients C 160 140 120 D 100 80 CC N M1 M2 (U.A) 60 C7 0.80.46 0.85 40 20 Correlation coefficient : –0.79 CXCL7 mRNA tumour 0 200 250 300 350 CXCL7 plasma (ng ml–1) E Patients with good prognosis Patients with poor prognosis Sunitinib response +++ Sunitinib response +/–

CXCL7 CXCL7

CXCL7

CXCL7 Tumour Plasma Tumour Plasma

CXCL7 level

Tumour cells Neutrophils Macrophages M1 Chemokine CXCL7 Macrophages M2 Blood vessels

Figure 3. Correlation between tumour (mRNA) and plasmatic (protein) CXCL7 levels. (A and B) The plasmatic levels of CXCL7 in healthy donors or ccRCC patients were determined by ELISA. (C) The plasmatic and tumour levels of CXCL7 were determined respectively by ELISA and by qPCR. The CC between the two values was calculated. (D) The intra-tumour CXCL7 mRNA levels (7 ccRCC patients), neutrophils (N, LCN2 and ELANE mRNA), M1 macrophages (M1, iNOS and IL1b mRNA) and M2 macrophages (M2, ARG1 and MRC1 mRNA) were determined by qPCR. The CC between each value is indicated. (E) Recapitulative schema: during tumour initiation, the amount of CXCL7-producing cells (neutrophils and M1 macrophages) is more important in the blood stream than in the tumour. Hence, CXCL7 amounts are greater in the blood than in the tumour. Then, the anti-tumour response is linked to an attraction of monocytes to the tumour were they polarised towards M1 macrophages. As tumour cells also express CXCL7, neutrophils are attracted to the tumours. During the tumour development phase, monocytes are polarised towards M2 macrophages that produced CXCL7 in addition to those produced by tumour cells. Attraction of CXCL7-producing cells to the tumour creates an exhaustion of the cytokines in the plasma and an overproduction in the tumour. *Po0.05. www.bjcancer.com | DOI:10.1038/bjc.2017.276 5 BRITISH JOURNAL OF CANCER CXCL7 is a predictive marker of sunitinib efficacy in RCC experimental tumours of mice xenografted with human ccRCC et al, 2007; Motzer et al, 2009). However, some patients respond cells. The experimental tumours obtained by xenografting ccRCC for a long period of time to sunitinib reaching 4–5 years (Chara cells developed as we previously described (Grepin et al, 2014; et al, 2011). Therefore, the long PFS and OS probably reflect the Dufies et al, 2017) (Supplementary Figure S5A). Mouse CXCL7 recruitment by serendipity of good responders. In a study that had plasmatic levels were decreased in mice with ‘human’ tumours enrolled 4543 patients, a good MSKCC score was correlated to a (Supplementary Figure S5B), which was consistent with the results long PFS (15 months) and a long OS (54.6 months) (Gore et al, obtained with plasma from patients. HES staining of the 2015). Therefore, these results show in an independent cohort that experimental tumours showed an important infiltration of immune some patients benefit for a long period of time of sunitinib. cells identified as natural killer (NK), macrophages and neutrophils However, the important message is the identification of a biological (Supplementary Figure S5C). A strong correlation was observed in marker that allows identification of very good responders and these experimental tumours between human CXCL7 (H CXCL7) subsequently patients who benefit of the treatment in terms of PFS and mouse intra-tumour NK cells (CC ¼ 0.71) suggesting an and OS. immune response triggered by CXCL7. A strong correlation was We were puzzled by the inverse correlation between intra- also observed between mouse CXCL7 (M CXCL7) and the tumour and plasmatic levels and their relative significance as a presence of neutrophils (N, CC ¼ 0.81). CXCL7 was not correlated prognostic determinant (Grepin et al, 2014). These experiments with M1 but was correlated with M2 macrophages (M2, suggested that during the initial phase of tumour development, CC ¼ 0.78), which is consistent with the correlation between N immune cells that participate in the physiological plasmatic level of and M2 (CC ¼ 0.9). Human and mouse CXCL7 were also CXCL7 are targeted to the tumour to mount an anti-tumour correlated with high levels of CXCR1 (CC ¼ 0.57 and 0.9 response, probably via the recruitment of NK cells. However, respectively) and CXCR2 (CC ¼ 0.85 and 0.55 respectively) which tumour cells that also produce CXCL7 may participate in immune are physiologically expressed on neutrophils and macrophages tolerance, inflammation and angiogenesis via attraction of (correlation between CXCR1 and N, CC ¼ 0.6 and correlation neutrophils, differentiation of monocytes towards M2 macro- between CXCR1 and M2, CC ¼ 0.58). A correlation was observed phages and proliferation of endothelial cells (neutrophils, mono- between CXCR2 and NK cells even though NK cells do not express cytes and endothelial cells physiologically expressed CXCL7 this receptor physiologically (Supplementary Figure S5D). receptors). Consequently, CXCL7 levels increase in the tumour These results suggest that immune cells producing CXCL7 in whereas they decrease in the plasma (see Figure 3E). Such the plasma are attracted towards the tumours and participate in the disequilibrium is probably characteristic of aggressive tumours/ pro-inflammatory/pro-proliferative response contributing in metastasis. Such aggressiveness may be attributed to the CXCL7- tumour growth (see the schematic representation in Figure 3E). dependent production of VEGFC and VEGFD, leading to the development of a lymphatic network (Yu et al, 2010), a known feature of metastatic dissemination (Karaman and Detmar, 2014). It would be important to determine whether CXCL7 is a predictive DISCUSSION marker of efficacy for other anti-angiogenic drugs targeting the VEGF/VEGFR axis currently used as second or third-line Despite the development of several therapies for ccRCC and the treatments. increase of PFS, no curative treatment currently exists. Moreover, because of the heterogeneity of the initial tumours and their subsequent metastasis (Gerlinger et al, 2014), the response to the current first line therapy with sunitinib is highly variable. There are CONCLUSION two major constraints. The first evident one is related to the improvement of patient survival. The second is economical and Our study highlighted CXCL7 as a relevant predictive marker of related to the high costs of targeted therapies. In order to reconcile sunitinib efficacy, the first line treatment of metastatic ccRCC. We the therapeutic and the economic imperatives, robust predictive also established a threshold value for patients for low or high risk markers of treatment efficacy have to be identified. To be of relapse in two independent cohorts. These results must be manageable in the clinical practices, protocols must be easy to confirmed on a larger prospective cohort to introduce CXCL7 transfer to technical platforms of hospitals, must be non-invasive measurements into clinical practice to rapidly adapt the treatment and applicable to small blood or urine samples. at the diagnosis. We believe that we have identified such a marker in the frame of prospective multicentre clinical trials (SUVEGIL and TORAVA), the angiogenic and pro-inflammatory cytokine CXCL7. This was corroborated in a retrospective cohort of patients highlighting the ACKNOWLEDGEMENTS relevance of our results. Moreover, CXCL7 was not indicative of PFS for patients treated with bevacizumab þ IFN. These results are This work was supported by the French association for cancer strongly indicative of the predictive value of CXCL7 for sunitinib research (ARC), IRIS Pharma (PDN), the Fondation de France (SG efficacy. Whereas sunitinib induces a significant decrease of cell and MD financial supports), the French National Institute for viability in vitro, bevacizumab þ INF did not (Supplementary Cancer Research (INCA), Foundation Franc¸ois Xavier Mora (SG Figure S6). This result may explain, at least in part, the difference financial support), the Re´gion Provence Alpes Cote d’Azur (PND in the predictive value of CXCL7 for sunitinib but not for financial support) and the Fondation Cordon de Vie Monaco. We bevacizumab efficacy. The challenge for a relevant marker is to thank the IRCAN core facilities (animal facility) for technical help. precisely identify a threshold value. A standard value has been We also thank the Department of Pathology, especially Arnaud determined through the statistical analysis but must be refined Borderie and Sandrine Destree, for technical help. especially for patients whose CXCL7 levels are borderline compared with the identified 250 ng ml À 1 threshold value. We were surprised by the very long PFS (27.7 months) and OS (38.4 months) that we recorded in the prospective clinical trials as CONFLICT OF INTEREST compared with those described in the pivotal trial that lead to sunitinib approval (PFS ¼ 11 months; OS ¼ 20.5 months) (Motzer The authors declare no conflict of interest.

6 www.bjcancer.com | DOI:10.1038/bjc.2017.276 CXCL7 is a predictive marker of sunitinib efficacy in RCC BRITISH JOURNAL OF CANCER

REFERENCES bevacizumab/Avastin treatment: the role of CXCL cytokines. Oncogene 31(13): 1683–1694. Karaman S, Detmar M (2014) Mechanisms of lymphatic metastasis. J Clin Chara L, Rodriguez B, Holgado E, Ramirez N, Fernandez-Ranada I, Invest 124(3): 922–928. Mohedano N, Arcediano A, Garcia I, Cassinello J (2011) An unusual Motzer RJ, Escudier B, McDermott DF, George S, Hammers HJ, Srinivas S, metastatic renal cell carcinoma with maintained complete response to Tykodi SS, Sosman JA, Procopio G, Plimack ER, Castellano D, Choueiri TK, sunitinib treatment. Case Rep Oncol 4(3): 583–586. Gurney H, Donskov F, Bono P, Wagstaff J, Gauler TC, Ueda T, Tomita Y, Choueiri TK, Escudier B, Powles T, Mainwaring PN, Rini BI, Donskov F, Schutz FA, Kollmannsberger C, Larkin J, Ravaud A, Simon JS, Xu LA, Hammers H, Hutson TE, Lee JL, Peltola K, Roth BJ, Bjarnason GA, Waxman IM, Sharma P (2015) Nivolumab versus everolimus in advanced Geczi L, Keam B, Maroto P, Heng DY, Schmidinger M, Kantoff PW, renal-cell carcinoma. N Engl J Med 373(19): 1803–1813. Borgman-Hagey A, Hessel C, Scheffold C, Schwab GM, Tannir NM, Motzer RJ, Escudier B, Oudard S, Hutson TE, Porta C, Bracarda S, Grunwald V, Motzer RJ (2015) Cabozantinib versus everolimus in advanced renal-cell Thompson JA, Figlin RA, Hollaender N, Kay A, Ravaud A (2010) Phase 3 carcinoma. N Engl J Med 373(19): 1814–1823. trial of everolimus for metastatic renal cell carcinoma: final results and Dufies M, Giuliano S, Ambrosetti D, Claren A, Diogop Ndiaye P, Mastri M, analysis of prognostic factors. Cancer 116(18): 4256–4265. Moghrabi W, Cooley LS, Ettaiche M, Chamorey E, Parola J, Vial V, Motzer RJ, Escudier B, Oudard S, Hutson TE, Porta C, Bracarda S, Grunwald V, Lupu-Plesu M, Bernhard JC, Ravaud A, Borchiellini D, Ferrero JM, Thompson JA, Figlin RA, Hollaender N, Urbanowitz G, Berg WJ, Kay A, Bikfalvi A, Ebos JM, Khabar KS, Grepin R, Pages G (2017) Sunitinib Lebwohl D, Ravaud A (2008) Efficacy of everolimus in advanced renal cell stimulates expression of VEGFC by tumor cells and promotes carcinoma: a double-blind, randomised, placebo-controlled phase III trial. lymphangiogenesis in clear cell renal cell carcinomas. Cancer Res 77(5): Lancet 372(9637): 449–456. 1212–1226. Motzer RJ, Escudier B, Tomczak P, Hutson TE, Michaelson MD, Negrier S, Escudier B, Bellmunt J, Negrier S, Bajetta E, Melichar B, Bracarda S, Oudard S, Gore ME, Tarazi J, Hariharan S, Chen C, Rosbrook B, Kim S, Ravaud A, Golding S, Jethwa S, Sneller V (2010) Phase III trial of Rini BI (2013) Axitinib versus sorafenib as second-line treatment for bevacizumab plus interferon alfa-2a in patients with metastatic renal cell advanced renal cell carcinoma: overall survival analysis and updated carcinoma (AVOREN): final analysis of overall survival. J Clin Oncol results from a randomised phase 3 trial. Lancet Oncol 14(6): 552–562. 28(13): 2144–2150. Motzer RJ, Hutson TE, Tomczak P, Michaelson MD, Bukowski RM, Oudard S, Escudier B, Porta C, Bono P, Powles T, Eisen T, Sternberg CN, Gschwend JE, Negrier S, Szczylik C, Pili R, Bjarnason GA, Garcia-del-Muro X, Sosman JA, De Giorgi U, Parikh O, Hawkins R, Sevin E, Negrier S, Khan S, Diaz J, Solska E, Wilding G, Thompson JA, Kim ST, Chen I, Huang X, Figlin RA Redhu S, Mehmud F, Cella D (2014) Randomized, controlled, double- (2009) Overall survival and updated results for sunitinib compared with blind, cross-over trial assessing treatment preference for pazopanib versus interferon alfa in patients with metastatic renal cell carcinoma. J Clin 27 sunitinib in patients with metastatic renal cell carcinoma: PISCES Study. Oncol (22): 3584–3590. J Clin Oncol 32(14): 1412–1418. Motzer RJ, Hutson TE, Tomczak P, Michaelson MD, Bukowski RM, Rixe O, Galliera E, Corsi MM, Banfi G (2012) Platelet rich plasma therapy: Oudard S, Negrier S, Szczylik C, Kim ST, Chen I, Bycott PW, Baum CM, inflammatory molecules involved in tissue healing. J Biol Regul Homeost Figlin RA (2007) Sunitinib versus interferon alfa in metastatic renal-cell 356 Agents 26(2 Suppl 1): 35S–42S. carcinoma. N Engl J Med (2): 115–124. Gerlinger M, Horswell S, Larkin J, Rowan AJ, Salm MP, Varela I, Fisher R, Motzer RJ, Mazumdar M, Bacik J, Berg W, Amsterdam A, Ferrara J (1999) McGranahan N, Matthews N, Santos CR, Martinez P, Phillimore B, Survival and prognostic stratification of 670 patients with advanced renal 17 Begum S, Rabinowitz A, Spencer-Dene B, Gulati S, Bates PA, Stamp G, cell carcinoma. J Clin Oncol (8): 2530–2540. Pickering L, Gore M, Nicol DL, Hazell S, Futreal PA, Stewart A, Negrier S, Gravis G, Perol D, Chevreau C, Delva R, Bay JO, Blanc E, Ferlay C, Swanton C (2014) Genomic architecture and evolution of clear cell renal Geoffrois L, Rolland F, Legouffe E, Sevin E, Laguerre B, Escudier B (2011) cell carcinomas defined by multiregion sequencing. Nat Genet 46(3): Temsirolimus and bevacizumab, or sunitinib, or interferon alfa and 225–233. bevacizumab for patients with advanced renal cell carcinoma (TORAVA): 12 Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, a randomised phase 2 trial. Lancet Oncol (7): 673–680. Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Santoni M, Berardi R, Amantini C, Burattini L, Santini D, Santoni G, Cascinu S Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, (2014) Role of natural and adaptive immunity in renal cell carcinoma 134 Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, response to VEGFR-TKIs and mTOR inhibitor. Int J Cancer (12): Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, 2772–2777. Swanton C (2012) Intratumor heterogeneity and branched evolution Song W, Yeh CR, He D, Wang Y, Xie H, Pang ST, Chang LS, Li L, Yeh S revealed by multiregion sequencing. N Engl J Med 366(10): 883–892. (2015) Infiltrating neutrophils promote renal cell carcinoma progression Gore ME, Szczylik C, Porta C, Bracarda S, Bjarnason GA, Oudard S, Lee SH, via VEGFa/HIF2alpha and estrogen receptor beta signals. Oncotarget 6(22): 19290–19304. Haanen J, Castellano D, Vrdoljak E, Schoffski P, Mainwaring P, von Hundelshausen P, Petersen F, Brandt E (2007) Platelet-derived Hawkins RE, Crino L, Kim TM, Carteni G, Eberhardt WE, Zhang K, chemokines in vascular biology. Thromb Haemost 97(5): 704–713. Fly K, Matczak E, Lechuga MJ, Hariharan S, Bukowski R (2015) Final Yu M, Berk R, Kosir MA (2010) CXCL7-mediated stimulation of results from the large sunitinib global expanded-access trial in metastatic lymphangiogenic factors VEGF-C, VEGF-D in human breast cancer cells. renal cell carcinoma. Br J Cancer 113(1): 12–19. J Oncol 2010: 939407. Grepin R, Guyot M, Giuliano S, Boncompagni M, Ambrosetti D, Chamorey E, Scoazec JY, Negrier S, Simonnet H, Pages G (2014) The CXCL7/CXCR1/2 axis is a key driver in the growth of clear cell renal cell carcinoma. Cancer Res 74(3): 873–883. This work is published under the standard license to publish agree- Grepin R, Guyot M, Jacquin M, Durivault J, Chamorey E, Sudaka A, ment. After 12 months the work will become freely available and Serdjebi C, Lacarelle B, Scoazec JY, Negrier S, Simonnet H, Pages G (2012) the license terms will switch to a Creative Commons Attribution- Acceleration of clear cell renal cell carcinoma growth in mice following NonCommercial-Share Alike 4.0 Unported License.

Supplementary Information accompanies this paper on British Journal of Cancer website (http://www.nature.com/bjc)

www.bjcancer.com | DOI:10.1038/bjc.2017.276 7 CXCL7 <250 ng/ml CXCL7 >250 ng/ml P value Age 0.567 Mean (SD) 64.1 (9.3) 62.5 (11) Gender 1 Female 4 (19) 6 (20) Male 17 (81) 24 (80) Fuhrman grade 0.101 1 1 (5.3) 0 (0) 2 6 (31.6) 5 (20) 3 10 (52.6) 10 (40) 4 2 (10.5) 10 (40) pT 0.744 1 5 (31.2) 5 (17.9) 2 3 (18.8) 8 (28.6) ≥ 3 8 (50) 15 (53.6) pN 0.55 0 7 (77.8) 10 (76.9) 1 1 (11.1) 0 (0) 2 1 (11.1) 3 (23.1) pM 0.717 0 10 (62.5) 14 (51.9) 1 6 (37.5) 13 (48.1) Risk factor (MSKCC) 0.38 Good 5 (35.7) 12 (48) Intermediate 7 (50) 7 (28) Bad 2 (14.3) 6 (24)

Supplementary Table S1: Dufies, M, Giuliano, S et al FORWARD REVERSE

Human genes 36B4 CAGATTGGCTACCCAACTGTT GGCCAGGACTCGTTTGTACC CXCL7 GTACGGGGACTTCAGACAGAT CGCGACAGACACTGCATAAC

Mouse genes 36B4 AGATTCGGGATATGCTGTTGGC TCGGGTCCTAGACCAGTGTTC CXCL7 TACAGCTGGAAAATCTGATG GAATTGAATGGGGTTCCAGAG KLRD1 (Natural killer) AAGTCTTGGAAAAGAAGCAG GCATTCCAATCCAGAAAAAG CD69 (Natural killer) AAAAGGACATGACGTTTCTG CAGCTGTTAAATTCTTTGCC LCN2 (Neutrophil) ATATGCACAGGTATCCTCAG GAAACGTTCCTTCAGTTCAG ELANE (Neutrophil) GTGAACGGCCTAAATTTCC ATAATCACAATGTCGTTCAGC miNOS (M1 macrophages) TCACCTTCGAGGGCAGCCGA TCCGTGGCAAAGCGAGCCAG IL1β (M1 macrophages) GCAACTGTTCCTGAACTCAACT ATCTTTTGGGGTCCGTCAACT mARG1 (M2 macrophages) GATTATCGGAGCGCCTTTCT CCACACTGACTCTTCCATTCTT MRC1 (M2 macrophages) TGCCGGCGTTGCAGCCTATT GCTCATTCTGCTCGATGTTGCCCA CXCR1 CCTCAGATCAAACAATGGC ACCAGTGCATAAAAGACAAC CXCR2 CTACTGCAGGATTAAGTTTACC GACGTATATTACAACCACAGC

Supplementary Table S2: Dufies, M, Giuliano, S et al Prospective cohort Retrospective cohort A Sunitinib group C Sunitinib group 100 100 (%) (%) 80 80

Survival 60 Survival 60 Free 40 Free 40

20 20 Progression- 0 Progression- 0 0 12 24 36 48 0 12 24 36 48 Months Months

Median survival (months) Median survival (months)

20.5 (17.7-27.7) 15.2 (11.4-24.3) B D 100 100

80 80 (%) (%) 60 60

40 40

Overall Survival 20 Overall Survival 20

0 0 0 12 24 36 48 0 12 2436 4860 72 Months Months

Median survival (months) Median survival (months)

38.4 (23.5-NR) 39.3 (28.9-50.6)

Supplementary Figure S1: Dufies, M, Giuliano, S et al Prospective cohort

A 100 Bevacizumub group

(%) 80

60 Survival

40 Free

0 Progression- 0 12 24 36 Months Median survival (months)

10.6 (7.4-14.3)

B 100

80 (%) 60

Overall Survival 20

0 0 12 24 36 Months Median survival (months)

24.1 (18.4-NR)

Supplementary Figure S2: Dufies, M, Giuliano, S et al Prospective cohort Prospective cohort Sunitinib group Bevacizumub group A 100 CXCL7 <250 ng/ml B 100 CXCL7 <250 ng/ml CXCL7 >250 ng/ml CXCL7 >250 ng/ml 80 80 (%) (%) 60 60

40 40 Overall Survival Overall Survival 20 20 p=0.047 p=NS

0 0 0 12 24 36 48 0 12 24 36 Months Months

Median survival Median survival Plasma levels Plasma levels (months) (months) CXCL7 high NR CXCL7 high 24.1

CXCL7 low 23.5 CXCL7 low NR

Supplementary Figure S3: Dufies, M, Giuliano, S et al AB ** ** 100 CXCL7 low

(%) CXCL7 high 80 p=0.0026 60 Survival mRNA (A.U.)

Free 40

CXCL7 stage1 stage2 stage3 20

Progression- 0 C 0 20 40 60 80 100 120 140 *** Months

Intratumor mRNA Median survival (months) CXCL7 low 106.8 mRNA (A.U.) CXCL7 high 11.6 CXCL7

M0 M1

Supplementary Figure S4: Dufies, M, Giuliano, S et al A 1.2 B 55 1 50

0.8 /ml) ng 0.6 45

0.4 40 *** Tumor volume (cm3)volume Tumor 0.2 ( mCXCL7 35

0 30 35 45 55 65 75 Without With tumor Days after injection tumor C

N M1 M2 NK D CC H CXCL7 M CXCL7 M CXCR1 M CXCR2 (LCN2) (IL1β) (ARG1) (KLRD1) H CXCL7 0.38 0.08 -0.17 0.12 0.71 0.57 0.85 M CXCL7 0.81 -0.24 0.78 0.03 0.9 0.55 N 0.15 0.9 0.06 0.6 0.01 (LCN2) M1 -0.44 0.36 -0.3 -0.11 (IL1β) M2 -0.05 0.58 -0.02 (ARG1) NK 0.29 0.52 (KLRD1) M CXCR1 0.62 M CXCR2 Supplementary Figure S5: Dufies, M, Giuliano, S et al 120

100

60 *** 40 Cell viability (%)

0 CT suni BVZ

Supplementary Figure S6: Dufies, M, Giuliano, S et al " Article!2: Sunitinib Stimulates Expression of VEGFC by Tumor Cells and Promotes Lymphangiogenesis in Clear Cell Renal Cell Carcinomas

! La"mise"en"évidence"de"nouveaux"marqueurs"prédictifs"des"réponses"au"sunitinib"règle" une" partie" du" problème" lié" à" la" variabilité" des" réponses." En" effet" alors" que" certains" patients" sont" d’emblée" réfractaires" au" traitement," son" efficacité" est" transitoire" chez" d’autres" patients" qui" développent" des" résistances" (Escudier" 2010a)." L’étude" des" mécanismes" de" résistance" prend" alors" une" grande" importance" notamment" pour" la" découverte"de"nouvelles"cibles"thérapeutiques."La"compensation"de"l’inhibition"de"l’axe" VEGFGA/VEGFR"par"l’expression"d’autres"facteurs"est"l’un"des"mécanismes"proposé"afin" d’expliquer"ces"résistances"(Welti"et"al."2013).""Des"études"précliniques"chez"la"souris" ont" montré" que" le" traitement" des" " ccRCC" avec" un" AAG" (bevacizumab)" induit" le" développement"de"vaisseaux"lymphatiques"au"sein"de"la"tumeur"(Grepin"et"al."2012)."Le" développement"du"système"lymphatique"est"associé"à"l’expression"du"VEGFGC."Dans"ce" contexte," je" me" suis" intéressé" à" l’implication" du" VEGFGC" et" de" la" lymphangiogenèse" associée" dans" la" progression" tumorale" après" traitement" avec" le" sunitinib." Nous" avons" cherché" à" comprendre" comment" le" sunitinib" et" d’autres" AAG" utilisés" dans" le" ccRCC" régulent"l’expression"de"VEGFGC"dans"les"ccRCCs.""

72" " Published OnlineFirst January 13, 2017; DOI: 10.1158/0008-5472.CAN-16-3088

Cancer Therapeutics, Targets, and Chemical Biology Research

Sunitinib Stimulates Expression of VEGFC by Tumor Cells and Promotes Lymphangiogenesis in Clear Cell Renal Cell Carcinomas Maeva Duﬁes1, Sandy Giuliano1,2, Damien Ambrosetti3, Audrey Claren1,4, Papa Diogop Ndiaye1, Michalis Mastri5, Walid Moghrabi6, Lindsay S. Cooley7, Marc Ettaiche8, Emmanuel Chamorey8, Julien Parola1, Valerie Vial2, Marilena Lupu-Plesu1, Jean Christophe Bernhard9, Alain Ravaud10, Delphine Borchiellini11, Jean-Marc Ferrero11, Andreas Bikfalvi7, John M. Ebos5, Khalid Saad Khabar6, Renaud Grepin 2, and Gilles Pages 1

Abstract

Sunitinib is an antiangiogenic therapy given as a ﬁrst-line tristetraprolin, two AU-rich element–binding proteins. Suni- treatment for renal cell carcinoma (RCC). While treatment tinib stimulated a VEGFC-dependent development of lym- improves progression-free survival, most patients relapse. We phatic vessels in experimental tumors. This may explain hypothesized that patient relapse can stem from the devel- our ﬁndings of increased lymph node invasion and new opment of a lymphatic network driven by the production of metastatic sites in 30% of sunitinib-treated patients and the main growth factor for lymphatic endothelial cells, increased lymphatic vessels found in 70% of neoadjuvant VEGFC. In this study, we found that sunitinib can stimulate treated patients. In summary, a therapy dedicated to destroy- vegfc gene transcription and increase VEGFC mRNA half-life. ing tumor blood vessels induced the development of lym- In addition, sunitinib activated p38 MAPK, which resulted phatic vessels, which may have contributed to the treatment in the upregulation/activity of HuR and inactivation of failure. Cancer Res; 77(5); 1212–26. Ó2017 AACR.

Introduction However, metastatic RCC has a very poor prognosis because of intrinsic resistance to radio- and chemotherapy. The main feature Renal cell carcinoma (RCC) represents 85% of kidney cancers of RCC is hypervascularization explained by overexpression of and 3% of adult cancers. However, its incidence has steadily VEGF, which is linked to mutation/inactivation of the von Hippel- increased. If diagnosed early, the main treatment is surgery. Lindau (vhl) gene, an E3 ubiquitin ligase of the hypoxia inducible factor 1a (HIF-1a). The most widely used systemic therapy for first-line metastatic RCC is sunitinib, a tyrosine kinase inhibitor 1 University of Nice Sophia Antipolis, Institute for Research on Cancer and Aging (TKI) with activity against the VEGF receptors (VEGFR1/2/3), of Nice, CNRS UMR 7284, INSERM U1081, Centre Antoine Lacassagne, Nice, PDGFR, CSF1R, and c-Kit (1). However, treatment benefits are France. 2Biomedical Department, Centre Scientifique de Monaco, Monaco, Principality of Monaco. 3Central Laboratory of Pathology, Centre Hospitalier transitory in most cases and the majority of patients develop Universitaire (CHU) de Nice, Hopital^ Pasteur, Nice, France. 4Radiotherapy resistance after one year (2). While the primary target of sunitinib Department, Centre Antoine Lacassagne, Nice, France. 5Center for Genetics is the host blood vessels via inhibition of VEGF receptors, and Pharmacology, Roswell Park Cancer Institute, Buffalo, New York. 6King mechanisms of resistance have been shown to stem from complex 7 Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia. University interactions between tumor and stromal cell populations (3). 8 of Bordeaux, INSERM U1029, Bordeaux, France. Statistics Department, Centre Indeed, several mechanisms have been proposed and can include Antoine Lacassagne, Nice, France. 9Service d'Urologie, Centre Hospitalier Uni- versitaire (CHU) de Bordeaux, Bordeaux, France. 10Service d'Oncologie compensatory growth factor stimulation (i.e., FGF), suppression Medicale, Centre Hospitalier Universitaire (CHU) de Bordeaux, Bordeaux, of immunoregulatory cells, or even co-option of existing blood France. 11Oncology Department, Centre Antoine Lacassagne, Nice, France. vessels, all of which together (or separately) could negate the – Note: Supplementary data for this article are available at Cancer Research impact of antivascular treatment strategies (3 6). Perhaps most Online (http://cancerres.aacrjournals.org/). provocatively, recent preclinical evidence has shown that some antiangiogenic treatments may elicit metastatic cell phenotypes, S. Giuliano, D. Ambrosetti, A. Claren, P.D. Ndiaye contributed equally to this fi article. which, in turn, may also compromise tumor-reducing bene ts (7, 8). Currently, more than 25 preclinical studies have confirmed R. Grepin and G. Pages codirected the work. this phenomena (9); however, the clinical impact is not known. Corresponding Author: Gilles Pages, University Nice-Sophia Antipolis CNRS Therapy-induced metastasis may explain disease progression UMR 7284, INSERM U1081, 28 Avenue Valrose, Nice 06103, France. Phone/Fax: following perioperative (adjuvant) treatment or rebound tumor 334-9203-1231; E-mail: [email protected] growth after treatment withdrawal (10, 11). doi: 10.1158/0008-5472.CAN-16-3088 Many studies have shown a close relationship between Ó2017 American Association for Cancer Research. lymphangiogenesis, invasion, and metastasis. During tumor

1212 Cancer Res; 77(5) March 1, 2017

Resistance to Sunitinib in Renal Cell Carcinoma

development, the lymphatic system is considered one of the (22). DAPI, DMSO, and 5,6-Dichlorobenzimidazole 1-b-D- primary routes of tumor cell dissemination that leads to ribofuranoside (DRB) were purchased from Sigma-Aldrich. distant metastatic growth. VEGFC is currently the best characterized lymphangiogenic factor that acts via VEGFR3 (12). Cell culture In normal adult tissues, VEGFR3 expression is largely restrict- RCC4 (R4), ACHN (A), Caki-2 (C2), 786-0 (786) and A-498 ed to lymphatic endothelial cells (LEC) and its activation is (498) RCC cell lines, human embryonic kidney (HEK293), RAW responsible for LEC proliferation, migration, and survival. 264.7 (RAW) macrophage cell lines were purchased from the However, VEGFR3 is also expressed on angiogenic blood ATCC (March 3, 2013). Stocks were made at the original date of vessels (13). Several reports indicate that VEGFC expression obtaining the cells, and were usually passaged for no more than 4 in cancer cells correlates with accelerated tumor progression months. These cell lines have been authenticated by DNA pro- and/or an unfavorable clinical outcome (14). VEGFC over- filing using 8 different and highly polymorphic short tandem expression in breast cancers has been shown to correlate repeat loci (DSMZ). RCC10 (R10) were a kind gift from Dr. W.H. with lymphangiogenesis and metastasis (15). In preclinical Kaelin (Dana-Farber Cancer Institute, Boston, MA). Primary cells models of RCC, endothelial cells chronically exposed to an were already described and cultured in a medium specific for renal anti-VEGF antibody proliferate in response to VEGFC stim- cells (PromoCell; ref. 23). ulation, whereas na€ve endothelial cells are unable to do so (16). Moreover, in a preclinical model of lung cancer, Immunoblotting resistance to aflibercept (a decoy receptor for VEGF and Cells were lysed in buffer containing 3% SDS, 10% glycerol, PlGF) is related to an increase in VEGFC (17). The VEGFC and 0.825 mmol/L Na2HPO4. Thirty to 50 mg of proteins were mRNA has a long 30 untranslated domain (30UTR) contain- separated on 10% SDS-PAGE, transferred onto a polyvinylidene fl ing an adenylate and uridylate-rich element (ARE). ARE di uoride membrane (Immobilon, Millipore) and then exposed elements are binding sites for the embryonic lethal abnormal to the appropriate antibodies. Proteins were visualized with the vision (ELAV) protein, also named HuR (Hu antigen R) and ECL system using HRP-conjugated anti-rabbit or anti-mouse tristetraprolin (TTP), also named ZFP36 (zinc finger protein secondary antibodies. 36). These mRNA-binding proteins have been described previously as mRNA-stabilizing and -destabilizing factors, Quantitative real-time PCR experiments respectively. HuR stabilizes the mRNA of cell-cycle regula- One microgram of total RNA was used for the reverse tran- tors, growth, inflammatory, and angiogenesis factors includ- scription, using the QuantiTect Reverse Transcription Kit (Qia- ing VEGF (18). These features give HuR an oncogenic status gen), with blend of oligo (dT) and random primers to prime first- (19). TTP has an opposite role (destabilization of these ARE- strand synthesis. SYBR Master Mix Plus (Eurogentec) was used for mRNA including VEGF), hence acting as a potent tumor quantitative real-time PCR (qPCR). The mRNA level was normal- suppressor (20). The ERK and MAPK p38 phosphorylate ized to 36B4 mRNA. For oligo sequences, also see Supplementary HuR and TTP. While ERK and p38-dependent phosphoryla- Materials. tion activate the mRNA-stabilizing activity of HuR, phosphorylation has an opposite effect on TTP. The balance Tumor xenograft experiment between TTP and HuR will determine mRNA stabilization Ectopic model of RCC. Five million 786-O cells were injected or degradation (21). Although VEGFC plays a causal role in subcutaneously into the flank of 5-week-old nude (nu/nu) female lymphangiogenesis and lymphatic metastasis, little is known mice (Janvier). The tumor volume was determined with a caliper about VEGFC regulation in tumor cells in response to cancer (v ¼ L Â l2 Â 0.5). When the tumor reached 100 mm3, mice were therapies. treated 5 days a week for 4 weeks, by gavage with placebo In this study, we describe a molecular mechanism linking (dextrose water vehicle) or sunitinib (40 mg/kg). This study was sunitinib treatment to lymphangiogenesis activation through carried out in strict accordance with the recommendations in the the stimulation of vegfc gene transcription and stabilization Guide for the Care and Use of Laboratory Animals. Our experi- of VEGFC mRNA. We found that VEGFC upregulation corre- ments were approved by the "Comite National Institutionnel lated with the development of a lymphatic network both in d'Ethique pour l'Animal de Laboratoire (CIEPAL)" (reference: fi tumors in mice and in patients. Our ndings suggest that NCE/2013-97). antiangiogenic benefits of sunitinib may be compromised by stimulation of lymphatic vessel formation and explain com- Orthotopic model of RCC. Tumor samples were obtained from pensatory prometastatic behaviors that may compromise previously published studies involving neoadjuvant sunitinib treatment efficacy. treatment in an ortho-surgical model (orthotopic tumor cell implantation followed by surgical tumor removal) of RCC (ani- Materials and Methods mal protocols, approvals, cell origins, and results have been previously; ref. 24). Briefly, human kidney SN12PM6LUCþ cells Reagents and antibodies (2 Â 106), were implanted into the kidney (subcapsular space) of Sunitinib, axitinib, everolimus, pazopanib, regorafenib, 6 to 8weekold female CB17 SCID and treated for 14 days with sorafenib, SB203580, and PD184352 were purchased from sunitinib (60 mg/kg/day) prior to nephrectomy. Selleckchem. Anti-HSP90 and anti-HSP60 antibodies were purchased from Santa Cruz Biotechnology. Anti-p38, anti- Immunofluorescence phospho-p38, anti-ERK, and anti-phospho-ERK antibodies Tumor sections were handled as described previously were from Cell Signaling Technology. TTP and HuR antibodies (22, 25). Sections were incubated with DAPI, anti-mouse are home-made and were generated as described previously LYVE-1 polyclonal (Ab 14817, Abcam), or monoclonal

www.aacrjournals.org Cancer Res; 77(5) March 1, 2017 1213

Duﬁes et al.

anti-a-smooth muscle actin (a-SMA, A2547, Sigma), and rat RNA immunoprecipitation monoclonal anti-mouse CD31 (clone MEC 13.3, BD Phar- HEK-293 cells were transfected overnight with 2 mgvector mingen) antibodies. expressing HA-tagged TTP or myc-tagged HuR. Cells were lysed in RNA IP buffer [100 mmol/L KCl, 5 mmol/L MgCl2,10 IHC mmol/L HEPES (pH 7.0), 0.5% NP40], freshly supplemented Samples were collected with the approval of the Local before use with 1 mmol/L DTT, 5 mL/mL units RNase Out Ethics committee. Sections from blocks of formalin-fixed and (Invitrogen) and protease inhibitor cocktail 1Â (Roche). The paraffin-embedded tissue were examined for immunostaining lysate was centrifuged for 10 minutes at 12,000 rpm, and the for podoplanin, CD31, p-p38, aSMA, and LYVE1. After depar- supernatant was transferred to new tubes with either mono- affinization, hydration, and heat-induced antigen retrieval, the clonal anti-myc antibody or monoclonal anti-HA antibody tissue sections were incubated for 20 minutes at room temper- (coupled with Protein G-sepharose beads). The beads were ature with monoclonal anti-podoplanin, anti p-p38, anti washed with RNA IP buffer. Aliquots were collected for immu- aSMA, and anti LYVE1 antibodies diluted at 1:100. Biotiny- noblotting and the remaining beads were subjected to total lated secondary antibody (DAKO) was applied and binding RNA extraction using TRI Reagent (Sigma), followed by chlo- was detected with the substratediaminobenzidineagainsta roform and isopropanol precipitation. Preswollen protein-G hematoxylin counterstain. agarose beads (GE Healthcare) were prepared by washing in PBS buffer and incubating with the antibody, followed by – Measurement of cytokines PBS washing. For HA-tagged TTP lysates, anti-myc coupled beads were used as a negative control. For myc-tagged HuR After stimulation, cell supernatant was recovered for – VEGFC measurement using the Human DuoSet ELISA kit lysates, anti-HA coupled beads were used as a negative control. cDNA was synthesized from 500 ng RNA using SuperScript II (R&D Systems). Reverse Transcriptase (Invitrogen). qPCR was performed in multiplex reaction using the C1000 thermal cycler (Bio-Rad). Luciferase assays FAM-labeled TaqMan probes (Metabion) for human VEGF-A Transient transfections were performed using 2 mLoflipo- (forward primer, 50-AGAAGGAGGAGGGCAGAATC-30;reverse fectamine (Gibco BRL) and 0.5 mgoftotalplasmidDNA- primer, 50-TCTCGATTGGATGGCAGTAG-30; and TaqMan Renilla fi fi fl luciferase in a 500 mL nal volume. The re ycontrol probe, 50-Fam-CATCCAT GAACTTCACC ACTTCGTGA-BHQ- plasmid was cotransfected with the test plasmids to control 1-30 and for human VEGF-C (forward primer: 50-GGATGCTG- fi for the transfection ef ciency. Twenty-four hours after trans- GAGAT GACTCAA-30;reverseprimer:50-TTCATCCAGCTCC- Renilla fi fl fection, cell lysates were tested for and re yluciferase. TTGTTTG-30 and TaqMan probe: 50-Fam-TCCACAGATGTCAT- All transfections were repeated four times using different GGAATCCATCTG- BHQ-1-30 were used. VIC-labeled Ribosom- plasmid preparations. LightSwitch Promoter Reporter VEGFC 0 al Protein (PO) probe was multiplexed with FAM-labeled (S710378) and LightSwitch 3 UTR reporter VEGFC (S803537) probes as the endogenous control to normalize for the levels were purchased from Active Motif. The short and long forms of of the genes of interest. the VEGFC promoter are a kind gift of Dr. Heide L. Ford (University of Colorado School of Medicine, Aurora, CO) and 5,6-Dichlorobenzimidazole riboside pulse chase experiments Kari Alitalo (Faculty of Medicine, Biomedicum Helsinki, Uni- DRB (25 mg/mL) was added to the cells and RNAs were versity of Helsinki, Helsinki, Finland; ref. 15). prepared from 0 to 4 hours thereafter. The level of VEGFC was determined by qPCR and was normalized to 36B4 mRNA. The Fluorescence assays relative amounts of VEGFC mRNA at time 0 before DRB addition ARE reporter constructs and reporter activity. RPS30 promoter- were set to 100%. linked EGFP reporter expression vectors containing the 30 UTR with VEGFC ARE (50GATTTCTTTAAAAGAATGACTATATAATT- siRNA assay TATTTCC-30) was constructed by inserting annealed synthetic siRNA transfection was performed using Lipofectamine RNAi- complementary oligonucleotides with BamHI and XbaI overhangs MAX (Invitrogen). Cells were transfected with either 50 nmol/L of into the same sites of the stable control bovine growth hormone si-HuR (Ambion, 4390824, s4610) or si-Control (Ambion, (BGH) 30 UTR of the plasmid. The mutant ARE form was similarly 4390843). After 48 hours, cells were stimulated with 5 mmol/L constructed in which ATTTA was mutated to ATCTA (26). of sunitinib or 5 mmol/L of sorafenib. Two days later, qPCR was performed, as described above. Functional response of the VEGFC ARE. Tetracycline-inducible (Tet-On) TTP-expressing constructs were used as described Gene expression microarray analysis previously (26). HEK293 Tet-On Advanced cells (Clontech) Normalized RNA sequencing data produced by The Cancer were transfected with 50 ng of either the wild-type or mutant Genome Atlas (TCGA) were downloaded from cBiopotal VEGFC 30UTR reporters along with a normalization control (www.cbioportal.org, TCGA Provisional; RNA-Seq V2). Data represented by red fluorescent protein expression plasmid, and were available for 503 of the 536 RCC tumor samples TCGA 10 ng of the TetO-TTP constructs. Transfections were per- subjected to mRNA expression profiling. The subtype classifica- formed using Lipofectamine 2000 (Invitrogen) according to tions were obtained through cBioPortal for Cancer Genomics the manufacturer's instructions. Doxycycline (0.25 mg/mL) was and the 33 samples lacking classifications were discarded. The added to the transfected cells for 16 hours and fluorescence nonmetastatic group contained 424 patients and the metastatic was acquired by imaging and quantified by the ProXcell group contained 79 patients. The results published here are in imaging segmentation and quantification software. whole or in part based upon data generated by the TCGA

1214 Cancer Res; 77(5) March 1, 2017 Cancer Research

Resistance to Sunitinib in Renal Cell Carcinoma

Research Network: http://cancergenome.nih.gov/ (27, 28). The RCC cell lines that we developed in collaboration with the surgery Kaplan–Meier method was used to produce overall survival department of the Nice Hospital (Nice, France; ref. 23) and curves. The VEGFC z-score cut-off point for the overall survival stratified by aggressiveness. This was determined by (i) the ability was determined with the spline analysis. The effect of VEGFC to form tumors in mice (for established cell lines) and (ii) the and its OR was estimated using a Cox model adjusted to the time of overall survival of patients (for primary tumor cells). expression of other genes and important patient characteristics. We observed that increases in aggressiveness corresponded to higher levels of VEGFC mRNA (Fig. 1A) and VEGFC production Patients and association studies (Fig. 1B) in cell lines and primary cells. The reported intratumor This was a retrospective study with all patients (312) consulted concentrations of sunitinib in mice and patients were 5.5–13 for a renal mass between 2008 and 2015 in Centre Antoine mmol/L (29, 30). According to these results, we specifically used Lacassagne (Nice, France). Of these 312 patients, 87 only had this range of concentrations in our experiments. Sunitinib been analyzed (cause of elimination of analyzable patients; no induced an increase in the VEGFC mRNA level in six RCC cell follow-up, renal metastasis from another cancer, RCC treated by lines (Fig. 1C) and in four primary tumor cells (Fig. 1D), while it surgery without metastasis, patients without progression ...). The did not change VEGFC mRNA in normal primary renal cells 87 patients had metastatic RCC treated in the first line with derived from three independent patients (Fig. 1E). Sunitinib therapies including IFNa Æ bevacizumab, sunitinib, temsiroli- increased VEGFC protein in the conditioned medium in the six mus. Only the patients that relapsed and patients without lymph RCC cell lines (Fig. 1F) and in the four independent primary cells node metastases before the treatment were included to test for the (Fig. 1G) while it had no effect on normal cells (Fig. 1H). Our presence of lymph node metastases and new sites of metastasis on results show sunitinib-induced effects are transient as VEGFC treatment [20 patients treated with sunitinib and 11 patients mRNA amounts return to their basal levels following treatment treated with other drugs (essentially with IFNa Æ bevacizumab)]. cessation (Supplementary Fig. S1A). RCC cells do not express Lymph node metastases and new sites of metastasis are two VEGFR1/2/3 (qPCR DCt >35) but highly express CSF1R (DCt 29), independent factors in comparison to age, gender, and metastatic PDGFR (DCt 22) and c-Kit (DCt 25). Hence, aberrant expression of stage at diagnostic (M0 or M1). Neoadjuvant patient samples were these receptors on tumor cells that was already reported (31–33) obtained from Nice, Bordeaux, and Monaco Hospitals. Patients may explain the observed increase in VEGFC expression following were treated for at least two months before surgery (Supplemen- sunitinib exposure. Moreover, we examined imatinib, a well- tary Table S1A). known inhibitor of c-Kit and PDGFR (like sunitinib), could similarly stimulate VEGFC expression (Supplementary Fig. Statistical analysis S1B). Other VEGFR TKIs (i.e., axitinib, pazopanib, sorafenib) For in vitro and in vivo analysis. All data are expressed as the were also found to increase VEGFC mRNA in 786-O cells while mean Æ SEM. Statistical significance and P values were deter- bevacizumab/IFNa and the mTOR inhibitor everolimus did not mined by the two-tailed Student t test. One-way ANOVA was (Supplementary Fig. S1C). Sorafenib, used as second-line therapy used for statistical comparisons. Data were analyzed with in RCC, upregulated VEGFC mRNA and protein levels in both Prism 5.0b (GraphPad Software) by one-way ANOVA with established and primary tumor cells but had no effect on normal Bonferroni post hoc. renal epithelial cells (Supplementary Fig. S1D to S1I). In cells adapted to a high concentration of sunitinib (10 mmol/L; For patient analysis. All categorical data were described using Supplementary Fig. S2A and S2B; ref. 34), basal VEGFC mRNA frequencies and percentages. Quantitative data were presented and protein amounts were increased compared to na€ve cells using median and range or mean and SD. Censored data were (Supplementary Fig. S2C and S2D). These results suggest a general described using Kaplan–Meier estimation median survival and mechanism by which drugs that directly or indirectly target 95% confidence interval (CI). Statistical analyses were two angiogenesis induce VEGFC expression by tumor cells. sided and were considered to be significant if P 0.05 using R3.2.2. Sunitinib stimulates VEGFC promoter activity Sunitinib-mediated induction of VEGFC mRNA suggested Univariate analysis. Statistical comparisons were made using x2 stimulation of transcription, stabilization of mRNA, or a combi- test or Fisher exact test for categorical data, t test, or Wilcoxon test nation of both mechanisms. Hence, we first investigated the for quantitative data and log-rank test for censored data. Smooth- activity of the VEGFC promoter after sunitinib treatment. Suni- ing spline curves were used to predict death risk versus VEGFC tinib stimulated the activity of the VEGFC promoter in RCC cell mRNA expression. lines (Fig. 2A) and in primary tumor cells (Fig. 2B). The transcription factor sine oculis 1 (SIX1) participates in VEGFC tran- Multivariate analysis. Multivariate analysis was carried out by scription (15). We generated two reporter constructs, one with a creating a Cox model. Choice of the final model was made VEGFC promoter containing SIX1-binding sites (long form) and performing backward stepwise model selection. All variables one deleted of the SIX1 consensus site (short form, Fig. 2C). The associated with P 0.1 on univariate analysis were included in long form was stimulated by sunitinib, while the short one was the model. not, in four independent cell lines (Fig. 2D) and two primary tumor cells (Fig. 2E). VEGFC promoter activity was also higher in Results sunitinib-resistant cells (Supplementary Fig. S2F) generated previously by chronic exposure to the drug (34). As suggested by the Sunitinib stimulates VEGFC expression increase in the mRNA levels (Supplementary Fig. S1C), sorafenib To examine the relationship between tumor growth potential also stimulated activity of the VEGFC promoter in cell lines and and VEGFC production, we used established and primary patient primary cells (Supplementary Fig. S3A and S3B). These results

www.aacrjournals.org Cancer Res; 77(5) March 1, 2017 1215

Duﬁes et al.

A Cell lines Primary cells B Cell lines Primary cells 600 1,400

500 1,200 1,000 400 800 300 600 200 400 100 200 VEGFC (% of control) of (% VEGFC 0 0 VEGFC mRNA (% of control) of (% mRNA VEGFC

Aggressivness Aggressivness Aggressivness Aggressivness C F 3,500 900 CT CT 800 suni 3,000 suni 700 2,500 600 500 2,000 400 1,500 300 1,000 (% of control) of (%

VEGFC mRNA VEGFC 200 500

100 control) of (% VEGFC 0 0 R4 R10 786 498 R4 R10 786 498 D G 500 CT 2,500 CT

400 suni 2,000 suni

300 1,500

200 1,000 (% of control) VEGFC mRNA 100 500 VEGFC (% of control) 0 0 E TF M CC H TF M CC 120 CT suni 120 CT suni 100 100

80 80

60 60

40 40 (% of control) VEGFC mRNA 20 20 VEGFC (% of control) 0 0 14S 15S 18S 14S 15S 18S Normal cells Normal cells

Figure 1. Sunitinib increased VEGFC expression. A–H, Different RCC cell lines [RCC4 (R4), RCC10 (R10), ACHN (A), Caki-2 (C2), 786-O (786) and A498 (498)] and primary RCC cells (TF, M, CC, and M2) and normal renal cells (14S, 15S, and 18S) were evaluated for VEGFC mRNA levels by qPCR (A, C, D, E)andfor VEGFC protein in cell supernatants by ELISA (B, F, G, H). C and F, RCC cell lines were treated with 5 mmol/L sunitinib (suni) for 48 hours. D and G, Primary cells were treated with sunitinib (10 mmol/L for TF, and 5 mmol/L for M and CC) for 48 hours. E and H, Normal kidney cells were treated with 5 mmol/L sunitinib for 48 hours. For A, C,andD,themRNAlevelofR4isconsideredasthereferencevalue(100%).Dataarerepresentedasmeanof three independent experiments Æ SEM. Ã, P < 0.05; ÃÃ, P < 0.01; ÃÃÃ, P < 0.001.

1216 Cancer Res; 77(5) March 1, 2017 Cancer Research

Resistance to Sunitinib in Renal Cell Carcinoma

A B 400 300 CT CT ** suni *** suni 300 *** *** 200 200

100 (% of control) of (% (% of control) of (% 100 Luciferase activity Luciferase activity Luciferase

0 0 786 498 TF CC

ATG C Xho I Sac l

Figure 2. VEGFC promoter Sunitinib increased VEGFC promoter activity. A and B, RCC cell lines (A)and –4 kb –1,755 +1 +206 primary cells (B) were transfected LUC Long form with a Renilla luciferase reporter gene SIX BS under the control of the VEGFC –201 Reporter luciferase promoter and treated with sunitinib LUC Short form (suni) 2.5 mmol/L for cell lines or 5 mmol/L for primary cells for 24 hours. The Renilla luciferase activity D 300 400 CT R4 * CT R10 normalized to the firefly luciferase ** (control vector)wasthereadoutofthe suni 300 suni VEGFC promoter activity. C, Schemas 200 of truncated short and long forms of 200 VEGFC promoter activity reporter 100 genes used in D and E. D and E, RCC 100 (% of control) of (% (% of control) of (% cell lines (D) and primary cells (E) were Luciferase activity Luciferase Luciferase activity Luciferase transfected with a firefly luciferase 0 0 reporter gene under the control of the Short form Long form Short form Long form truncated short or long form VEGFC promoter and treated with sunitinib 300 ** 300 *** CT 786 498 2.5 mmol/L for cell lines or 5 mmol/L CT for primary cells for 24 hours. suni suni Normalized luciferase activity 200 200 (control vector) was the readout of the VEGFC promoter activity. Data are represented as mean of three 100 100 Æ control) of (% independent experiments SEM. control) of (% Ã ÃÃ ÃÃÃ Luciferase activity , P < 0.05; , P < 0.01; , P < 0.001. activity Luciferase 0 0 Short form Long form Short form Long form

E 200 TF ** 300 CC CT CT ** suni suni 200 100 100 (% of control) of (% (% of control) Luciferase activity Luciferase activity Luciferase 0 0 Short form Long form Short form Long form indicate transcriptional-dependent induction of VEGFC expres- Sorafenib also increased the VEGFC mRNA half-life (4.4 Æ 2.8 sion by antiangiogenic treatments. hours vs. 14.8 Æ 5.6 hours; Supplementary Fig. S3C). A reporter gene in which the VEGFC-30UTR was inserted downstream of the Sunitinib induces a 30UTR-dependent increases in the luciferase gene (Fig. 3C) was also induced by sunitinib in two cell VEGFC mRNA half-life lines (Fig. 3D), two primary tumor cells (Fig. 3E), and in sunitinib- A second mechanism that may explain treatment-induced resistant cells (Supplementary Fig. S2G). Similarly, sorafenib 0 VEGFC mRNA increases is the stabilization of VEGFC mRNA. stimulated the activity of the VEGFC-3 UTR reporter gene in both Indeed, sunitinib increased VEGFC mRNA half-life 6-fold in RCC established and primary tumor cells (Supplementary Fig. S3D cell lines (4.4 Æ 2.8 hours vs. 26.5 Æ 12.9 hours; Fig. 3A) and by 4- and S3E). These results demonstrate that two antiangiogenic fold in primary cells (4.1 Æ 3.9 hours vs. 16.7 Æ 6.3 hours; Fig. 3B). treatments enhance VEGFC expression through the stabilization

www.aacrjournals.org Cancer Res; 77(5) March 1, 2017 1217

Duﬁes et al.

A B Cell lines Primary cells 100 100

80 ** 80 ** 60 60

40 CT 40 CT suni suni Figure 3. 20 20 Sunitinib increased the half-life of VEGFC mRNA via its 30UTR. A and B, % of Remaining VEGFC mRNA 0 % of Remaining VEGFC mRNA 0 The RCC cell line (786-O; A)or 02460246primary cells (CC; B) were treated DRB (h) DRB (h) with 5 mmol/L sunitinib (suni) for 48 hours. Cells were then incubated t t Treatment 1/2 (h) Treatment 1/2 (h) with DRB for 2, 4, or 6 hours. The CT 4.4 (±1.8) CT 4.1 (±1.9) remaining VEGFC mRNA was evaluated by qPCR and its mRNA Sunitinib 26.5 (±6.9) Sunitinib 16.7 (±5.3) half-life was calculated. C, Schema of the VEGFC luciferase 30UTR reporter gene. D and E, RCC cell lines (D)or C VEGFC-luciferase 3’UTR reporter F VEGFC-GFP 3’UTR reporter primary cells (E) were transfected 0 Promoter Promoter with the VEGFC 3 UTR reporter gene and treated with sunitinib 5 mmol/L luciferase 3’UTR VEGFC-ARE GFP VEGFC-ARE (or 10 mmol/L for TF) for 24 hours. The normalized luciferase activity was the 1497 1528 GATTTCTTTAAAAGAATGACTATATAATTTATTTCC readout of the reporter gene mRNA D 300 CT *** ** GATTTCTTTAAAAGAATGACTATATAATGTATTTCC half-life. F, Schemas of VEGFC GFP suni 30UTR wild-type or mutated for the *** ARE element reporter genes. G, Cells 200 GFP VEGFC-mutant-ARE were transfected with the VEGFC + GFP 30UTR wild-type or mutated Tet Promoter reporter vectors for the VEGFC 30UTR 100 (% of control) ARE. TTP expression was induced

Luciferase activity TTP with 0.25 mg/mL doxycycline (dox). The ﬂuorescence level was the 0 readout of the reporter gene half-life. 786 498 Data are represented as mean of three independent experiments Æ E 300 G CT SEM. Ã, P < 0.05; ÃÃ, P < 0.01; ÃÃÃ suni * 100 , P < 0.001. * 200 **

50 100 (% of control) of (% Luciferase activity Luciferase Fluorescence level

0 ofcontrol) % (Dox+/Dox-, 0 TF CC CT ARE mutant-ARE

of its mRNA and via its 30UTR. As for VEGFA, the VEGFC mRNA the relative level of the target mRNA half-life (Fig. 4C). The 30UTR contains an adenylate and uridylate-rich element (ARE), a phosphorylation of HuR stimulated its ability to stabilize a given binding site for HuR and TTP. We used a wild-type or a VEGFC mRNA, while phosphorylation of TTP played an opposite role 30UTR mutated for the ARE site coupled to the EGFP reporter gene (35). The expression of TTP is concomitant with a modification in (Fig. 3F). Expression of TTP, using a doxycycline-regulated con- molecular weight from a 36 kDa form (the form with maximal struct, decreased the level of fluorescence only when the ARE site mRNA-destabilizing activity, only observed in LPS-stimulated was present (Fig. 3G). These results suggest that TTP decreases RAW cells), to intermediate forms and finally to a low-mobility VEGFC mRNA half-life by binding to its ARE in the 30UTR. form of approximately 45 kDa. These different modifications highly depend on p38-dependent phosphorylation (36). While TTP and HuR bind VEGFC mRNA and modulate its half-life in nonstimulated 786-O cells, high-mobility and intermediate Using RNA immunoprecipitation with antibodies to tagged- forms of TTP are present, sunitinib stimulation results in the HuR and tagged-TTP, we found that HuR and TTP bound directly accumulation of low-mobility forms corresponding to the max- to VEGFC mRNA (Fig. 4A and B). TNFa and VEGFA mRNA imally phosphorylated protein with impaired mRNA-destabiliz- binding served as positive controls and GFP and GAPDH mRNA ing activity (37). HuR is not expressed in unstimulated or LPS- as negative controls (Fig. 4A and B and Supplementary Fig. S3F stimulated RAW cells. However, sunitinib stimulated HuR expres- and S3G). The balance between TTP and HuR activity determined sion and the accumulation of its low-mobility/active forms

1218 Cancer Res; 77(5) March 1, 2017 Cancer Research

Resistance to Sunitinib in Renal Cell Carcinoma

A B 1,500 2,500 ** ** 2,000 1,000 Figure 4. 1,500 TTP and HuR bind VEGFC mRNA and 1,000 modulate its half-life. A and B, HuR/ 500 (% of control) of (% (% of control) of (% VEGFC mRNA VEGFC TTP interactions with VEGFC mRNA mRNA VEGFC 500 were analyzed by RIP-Chip (RNA-IP). 0 0 HEK-293 cells were transfected with HA-CT Myc-HuR Myc-CT HA-TTP Myc-HuR or HA-CT (used as a negative control; A)orwithHA-TTPandMyc-CT 150 150 (used as a negative control; B). Exogenous TTP and HuR crosslinked to NS mRNA were immunoprecipitated with 100 100 NS anti-HA and anti-Myc antibodies. The levels of immunoprecipitated VEGFC 50 50 or GFP mRNA (used as a negative GFP mRNA (% of control) of (% GFP mRNA control) were assessed by qPCR. C, control) of (% Schematic balance between TTP and 0 0 HuR, and its effect on mRNA stability. HA-CT Myc-HuR Myc-CT HA-TTP D, RAW cells were stimulated with 10 mmol/L lipopolysaccharide (LPS) for 6 CD hours and were used as a positive HuR RAW 786 control for active (unphosphorylated) TTP and inactive [partially (intermediate) ARE mRNA MW (kDa) CT LPS CT suni and fully phosphorylated form] forms 45 TTP of TTP. 786-O cells were stimulated (inactive form) with 5 mol/L sunitinib (suni) for 6 HuR m mRNA Intermediate hours. TTP and HuR expression was TTP degradation analyzed by Western blot analysis. 36 TTP HSP90 served as a loading control. E (active form) and F, 786-O cells were transfected 0 with a VEGFC luciferase 3 UTR TTP mRNA HuR reporter gene in the presence of stabilization expression vectors for HuR, TTP, or HuR HSP90 TTPvar (a mutation that induces a decrease in TTP mRNA translation and serves as a negative control; ref. 38) EF and treated or not with sunitinib 5 160 ** 300 CT suni mmol/L for 24 hours. The normalized 140 *** *** luciferase counts served as readout of 120 the reporter gene mRNA half-life. Data 200 100 are represented as mean of three ** independent experiments Æ SEM. 80 NS ÃÃ P ÃÃÃ P 60

, < 0.01; , < 0.001. NS, (% of control) 100 (% of control) nonsigniﬁcant. 40 Luciferase activity Luciferase activity 20 0 0 CT TTP HuR CT TTP TTPvar

(Fig. 4D). Overexpression of TTP decreased VEGFC 30UTR report- to its cytoplasmic translocation (active forms) in established er activity while overexpression of HuR increased it (Fig. 4E). RCC cell lines (Fig. 5A; Supplementary Fig. S4A) and in primary Moreover, overexpression of TTP inhibited the sunitinib-depen- tumor cells (Supplementary Fig. S4B). Inhibition of ERK dent increase in the VEGFC 30UTR reporter activity while over- (PD184352) did not modify the sunitinib-dependent increase expression of a mutated form of TTP that is poorly translated of the VEGFC mRNA, while inhibition of p38 (SB203580) (TTPvar; ref. 38) did not (Fig. 4F). These results suggest that a strongly reduced it in an established cell line (Fig. 5B; Supple- regulated balance of active TTP and HuR plays a key role in the mentary Fig. S4C) and in primary tumor cells (Supplementary regulation of the VEGFC mRNA half-life induced by sunitinib. Fig. S4D). Phospho-p38 basal activity increased in sunitinib- resistant cells and was correlated with enhanced VEGFC expres- p38 and HuR are required for sunitinib-dependent increases sion (Supplementary Fig. S2C–S2E). Inhibition of p38 reduced in VEGFC expression the sunitinib-dependent increase in the VEGFC mRNA half-life ERK and p38 pathways are critical for the modulation of the and the VEGFC 30UTR reporter gene activity in an established TTP and HuR activity (39). Sunitinib induced rapid activation RCC cell line (Fig. 5C and D) and in primary tumor cells of the ERK and p38 pathways determined by the presence of (Supplementary Fig. S4E). Moreover, SB203580 inhibited the their phosphorylated forms (p-ERK and p-p38). Activation of sunitinib-dependent induction of HuR (Fig. 4E). These results ERK and p38 correlated with an increase in HuR amounts and suggested that VEGFC upregulation of the mRNA half-life is

www.aacrjournals.org Cancer Res; 77(5) March 1, 2017 1219

Duﬁes et al.

ABsuni (h) 0 1 2 4 6 *** 600 HuR CT 500 suni p-ERK 400 ERK 300

p-p38 200 Figure 5. (% of control) of (% VEGFC mRNA VEGFC p38 and HuR were required for the p38 100 induction of VEGFC expression by 0 sunitinib. A, 786-O cells were treated HSP60 CT PD SB with 5 mmol/L sunitinib (suni) for 1 to 6 hours. HuR, p-ERK, ERK, p-p38, and p38 C D expression was analyzed by 120 immunoblotting. HSP60 served as a 300 loading control. B, 786-O cells were CT 100 *** treated with 5 mmol/L sunitinib in the suni presence of 10 mmol/L PD184352 (PD, 80 ERK inhibitor) or 10 mmol/L SB203580 200 (SB, p38 inhibitor) for 48 hours. The 60 VEGFC mRNA level was determined by ** qPCR. C, 786-O cells were treated with 5 40 mmol/L sunitinib with or without 10 CT control) of (% 100

Luciferase activity Luciferase mmol/L SB203580 for 48 hours. Cells suni were then treated with DRB for 2, 4, or 6 20 suni + SB hours. The remaining VEGFC mRNA was % of Remaining VEGFC mRNA 0 0 evaluated by qPCR. D, 786-O cells were 0246 transfected with a VEGFC luciferase CT SB 0 DRB (h) 3 UTR reporter gene and treated or not with 5 mmol/L sunitinib and/or with 10 E suni – ++– mmol/L SB203580 for 24 hours. The normalized luciferase counts served as a SB ––++ readout of the reporter gene mRNA HuR half-life. E, 786-O cells were treated p-p38 with 5 mmol/L sunitinib and/or 10 mmol/L SB203580. HuR, p-p38, and p38 p38 expression was analyzed by immunoblotting. HSP60 served as a HSP60 loading control. F and G, 786-O cells were transfected with CT or HuR siRNA. FG** Twenty-four hours later, cells were 400 siCT CT treated with 5 mmol/L sunitinib for 48 120 hours. HuR (F) and VEGFC (G)mRNA suni 100 siHUR levels were determined by qPCR. Data 200 are represented as mean of three 80 independent experiments Æ SEM. ÃÃ, P 0.01; ÃÃÃ, P 0.001. 60 < < 40 100

20 *** *** HuR mRNA (% of control) of (% mRNA HuR

0 control) of (% mRNA VEGFC 0 CT suni siCT siHuR

dependent on HuR induction via stimulation of the p38 path- treated with sunitinib after the tumors reached 100 mm3 for way. Hence, downregulation ofHuRwithsiRNAinhibitedthe approximately one month before analysis. Sunitinib stimulated sunitinib-dependent increase of VEGFC mRNA in a RCC cell lymphangiogenesis was confirmed by the presence of LYVE1- line (Fig. 5F and G) and in primary tumor cells (Supplementary positive lymphatic endothelial cells (LEC) in the core of the Fig. S4F and S4G). These results confirmed that sunitinib implanted tumors after treatment while no staining was induces a cascade of events starting from early activation of observed in the core of control tumors (Fig. 6A and B). Phos- p38 to activation of HuR that finally results in induction of pho-p38 labeling was also increased in tumors of sunitinib- VEGFC mRNA expression. treated mice, which confirmed in vitro observations (Fig. 6C and D). The levels of human (produced by xenotransplanted tumor Sunitinib induces lymphangiogenesis in vivo cells) and mouse (stromal cells) VEGFC and HuR mRNA were To correlate the sunitinib-dependent induction of VEGFC increased in the tumors of mice treated with sunitinib in expression to lymphangiogenesis, we evaluated the presence of comparison with the levels of both factors in the tumors of lymphatic vessels in experimental RCC tumors obtained by nontreated mice (Supplementary Fig. S5A). Sunitinib also subcutaneous injection of 786-O cells in nude mice. Mice were induced the expression of genes involved in lymphangiogenesis

1220 Cancer Res; 77(5) March 1, 2017 Cancer Research

Resistance to Sunitinib in Renal Cell Carcinoma

A LYVE1 C p-p38 CT suni CT suni

x10 x10

BD7 100 6 *** **

2 80 5 4 60 3 40 2 % of p-p38 20 1 tumors Positive vessel per mm 0

Number of lymphatic lymphatic of Number 0 CT suni CT suni E F G Orthotopic tumor Primary tumor implantation resection 6 6 P = P = 0.0242 0.0257 100 NeoAdi Suni 4 4

50 Low VEGFC 2 2 High VEGFC Survival (%) P = 0.023 0 0 VEGFC mRNA (% of CT) of (% mRNA VEGFC 0 0120 40 60 80 00 CT suni CT) of (% mRNA m-PROX1 CT suni Days post implantation H I

* CT 80

LYVE1 40 (%)

20 suni 0

Tumor with lymphatic vessels CT suni

Figure 6. Sunitinib induced lymphangiogenesis in vivo. A–E, Experimental tumors obtained after subcutaneous injection of 5 Â 106 786-O cells were analyzed for LYVE1 expression. A, Lymphatic endothelial cells were detected by LYVE1 immunolabeling (green). Scale bar, 20 mm. B, Quantification of the LYVE1-positive vessels per mm2 (n ¼ 20). C, p-p38 was detected by IHC. Scale bar, 20 mm. D, Quantification of p-p38–labeled cells (n ¼ 20). E–I, SN12PM6LUCþ (2 Â 106 cells) were implanted into the kidney (subcapsular space) of 6 to 8-week-old female CB17 SCID. E, VEGFC mRNA mRNA levels in tumors from control mice and mice treated in a neoadjuvant setting (n ¼ 24). F, The Kaplan–Meier analysis of mice treated or not with sunitinib and the prognostic role of VEGFC mRNA levels (n ¼ 24). G, PROX1 mRNA levels in tumors from control mice and mice treated in a neoadjuvant setting (n ¼ 24). H, Lymphatic endothelial cells were detected by LYVE1 IHC. Scale bar, 20 mm. I, Quantification of the LYVE1-positive vessels (n ¼ 24). Ã, P < 0.05; ÃÃ, P < 0.01; ÃÃÃ, P < 0.001.

www.aacrjournals.org Cancer Res; 77(5) March 1, 2017 1221

Duﬁes et al.

including vegfr3, nrp2, prox1 by cells of the microenvironment S6C and S6D). We next compared the presence of lymph node (mouse; Supplementary Fig. S5A). Increased expression of metastasis in nonresponder patients treated with sunitinib (20 of proangiogenic genes upon sunitinib treatment (vegfa, vegfr1, 87 patients with metastatic RCC; see Methods) or other thera- nrp1, and a-sma)wasobserved(SupplementaryFig.S5A)but peutic options including IFNa, bevacizumab, and a mTOR inhib- blood vessel maturation attested by coverage of endothelial itor (temsirolimus, 11 of 87 patients; Supplementary Table S1B). cells (CD31 labeling) by pericytes (a-SMA labeling) was equiv- Thirty-five percent (7/20) of patients treated with sunitinib alent in control or sunitinib-treated mice tumors (Supplemen- developed lymph node metastasis and new metastatic sites, tary Fig. S5B). Together, these results suggest that sunitinib did whereas patients treated with other therapeutic options did not not alter the vascular network function but induced lymphatic (0/11; P ¼ 0.033; Fig. 7D). As expected, patients with lymph node network development, potentially increasing tumor metastatic invastion (N1) and new metastatic sites had a shorter overall potential. To examine this, we performed a retrospective anal- survival (P ¼ 0.012; Fig. 7D). ysis of lymphatic marker expression in human RCC cells This result reinforces our conclusion that sunitinib stimu- (SN12-PM6LUCþ) implanted orthotopically in SCID mice, trea- lates the development of a lymphatic network. Finally, ted neoadjuvantly with sunitinib, and then surgically resected cBioPortal analysis showed that high amounts of VEGFC as described by Ebos and colleagues (24). In these studies, correlated with decreased overall survival (P ¼ 0.0026; Sup- neoadjuvant sunitinib treatment was found to have no benefit plementary Fig. S6A) and an increased proportion of meta- in reducing primary tumor growth but, following surgery static patients ((M1 patients) P ¼ 0.019; Supplementary Fig. and sunitinib treatment withdrawal, increased metastasis and S6B). The level of VEGFC, the tumor stage, the metastatic reduced overall survival. Our results show VEGFC level in- status (M1) and the lymph node invasion (N1) were analyzed creases in excised tumors from sunitinib-treated mice in a multivariate Cox regression model on overall survival. (Fig. 6E) and a majority of mice that were treated in a neoad- VEGFC expression was identified as an independent prognos- juvant setting developed metastasis as shown previously (24). tic parameter for overall survival (P ¼ 0.000253; Supplemen- Strikingly, high VEGFC levels in tumors from sunitinib-treated tary Fig. S6C). mice were correlated with shorter survival (Fig. 6F). This observation is consistent with the development of a lymphatic Discussion network shown by the increases in PROX1 levels (Fig. 6G) and the presence of LYVE1-positive lymphatic vessels (35% in Resistance to antiangiogenic treatments are classified into: (i) control mice vs. 70% in sunitinib-treated mice; Fig. 6H and I). intrinsic resistance; tumors fail to respond from the outset of Together, these results suggest a strong correlation between treatment, (e.g., through sequestration of sunitinib in lyso- sunitinib treatment and a VEGFC/lymphatic vessel–dependent, somes; ref. 34), and (ii) acquired resistance; induction of which may, in turn, impact metastatic progression and reduce compensatory pathways (41) including an increase in growth survival. factor expression (e.g., Increased VEGFC and resulting lymphangiogenesis), as we have showninthisstudy.Eightypercent Sunitinib is associated with increased lymphangiogenesis of metastases of solid cancers are estimated to disseminate and lymph node metastasis in RCC patients through the lymphatic system, while 20% of metastases may There are several benefits to neoadjuvant treatment including occur through the blood vasculature or by direct seeding (12). (i) the downsizing of renal tumors to facilitate surgery or ablative Lymph node metastasis is the first sign of tumor progression in approaches (which can preserve renal function), (ii) to assess the majority of epithelial malignancies (12) (invasion of sen- patient sensitivity to treatment (if recurrence occurs), and (iii) to tinel lymph nodes in breast cancers is directly linked to prog- prevent metastatic spread (thereby improving post-surgical sur- nosis). An increased density of peri- and intratumor lymphatic vival; ref. 40). However, the benefits of neoadjuvant antiangio- vessels, as we observed in tumors from sunitinib-treated mice, genic treatments have yet to be validated clinically (40), thus indicates activation of lymphangiogenesis. A significant corre- examination of patient materials is rarely performed. However, lation has been observed between lymphatic vessel density and we obtained 13 patient samples (from a total of 3,000) that were lymph node and organ metastasis in clinical studies. High treated with antiangiogenic therapy in a neoadjuvant setting in expression levels of the lymphangiogenic factor VEGFC corre- different French hospitals (Supplementary Table S1A). Treated lates with lymph node metastasis in numerous tumor types. tumor samples were compared with tumors of untreated patients Overexpression of VEGFC or VEGFD in mouse models for the presence of lymphatic vessels. While lymphatic vessels increases the lymphatic vessel density and diameter of lymph were detected in 4 of 20 tumors (20%) from untreated patients, nodes and organ metastasis (12). Intratumor lymphatic vessels we found that lymphatic vessels in 9 of 13 tumors (69.2%, P ¼ and increased metastasis had been observed in VEGFC-over- 0.005) from patients treated in a neoadjuvant setting (Fig. 7A expressing tumors implanted in mice (12). Furthermore, lym- and B). The presence of lymphatic vessels correlated with an phatic invasion by RCC cells was the only independent risk increase in VEGFR3, NRP2, PROX1, LYVE1, and HuR mRNA factor for lymph node metastasis (42). Whereas sunitinib and levels (Fig. 7C). VEGFC levels (Fig. 7C) and p-p38 were not sorafenib increased VEGFC transcription and mRNA stabiliza- modified, which we speculate to be due to sunitinib treatment tion, they did not increase VEGFD expression in our RCC 0 that was stopped for more than one month before surgical model systems. Strikingly, VEGFD mRNA 3 UTR does not resection to allow wound-healing processes. This result was contain ARE. Sunitinib increased VEGFC expression only in consistent with the transient effect of sunitinib of VEGFC expres- tumor cells and not in normal cells (kidney cells and LEC), sion that we observed in vitro (Supplementary Fig. S1A). Sunitinib suggesting specificgeneticparticularitiesthatmediatetumor in a neoadjuvant setting was not associated with tumor blood cell adaptation to a toxic drug. Our results are consistent with vascular changes (CD31 and a-SMA labeling; Supplementary Fig. those of Sennino and colleagues who described VEGFC

1222 Cancer Res; 77(5) March 1, 2017 Cancer Research

Resistance to Sunitinib in Renal Cell Carcinoma

A CT suni B

80 M1 ** 70 M0

20 Figure 7. Sunitinib induced lymphangiogenesis 10 and metastasis in lymph nodes of RCC Tumors with lymphatic vessels (%) vessels lymphatic with Tumors patients. Tumors from untreated RCC 0 patients and tumors from patients CT suni/axi treated with sunitinib or axitinib in a neoadjuvant setting were compared (see Supplementary Table S1A). A, Lymphatic endothelial cells were detected by podoplanin (PDN) immunolabeling (brown).Scale bar, 25 CD 400 mm. B, Quantiﬁcation of the CT *** with lymph node metastasis and new metastatic sites podoplanine (PDN)-positive vessels. *** suni *** ** without lymph node metastasis and new metastatic sites C, The levels of indicated mRNA were 300 * 120 P = 0.033 determined by qPCR. D, RCC patients that progressed on treatment were 200 100 analyzed for the presence of lymph 80 100 node metastasis and new metastatic 60 sites (two or more new sites). Patients 40 were stratiﬁed for sunitinib or other control) of (% mRNA 0

vegfc vegfr3 nrp2 prox1 lyve1 hur (%) Patients 20 antiangiogenic treatments (see CT 100 100 100 100 100 100 Supplementary Table S1B). E, The suni 153 327 211 272 258 262 0 Kaplan–Meier analysis of patients that sunitinib other had relapsed on sunitinib with treatment treatment noninvaded or de novo invaded E 100 lymph nodes. N0 N1 80 Lymph node 60 metastasis (N) - (N0) + (N1) 40 Median survival 32.8 12.7 20 (months) Overall survival (%) P = 0.012 0 0 5 10 15 20 25 30 35 Months induction in response to antiangiogenic drugs in a mouse VEGFC promoter deleted from the domain containing SIX1- model of pancreatic tumors (43). Regorafenib, another multi- binding sites, suggesting that it may play a role. However, the kinase TKI with similar targets as sunitinib, did not stimulate deleted domain also contains binding sites for NF-kB, GATA-2 VEGFC expression. Unlike sunitinib, regorafenib inhibits p38 and 3, Erg-1, and p53 transcription factors. Hence, further activity, reinforcing the speciﬁcroleofp38inthesunitinib- experiments are needed to determine the transcription factors dependent increase of VEGFC expression. Unfortunately, this implicated in the sunitinib-dependent stimulation of vegfc gene treatment has not been approved for clinical use because of its transcription. Analysis of available online databases (TGCA) side effects (44). Regulation of VEGFC expression has been with cBioPortal (http://www.cbioportal.org) showed that over- poorly addressed (45). Vegfc gene transcription depends on the expression of SIX1 does not correlate with short progression- transcription factor SIX1, which is overexpressed in metastatic free or overall survival for patients with metastatic RCC unlike breast cancers (15). The critical role for SIX1 in lymphatic for breast cancers patients, whereas VEGFC is a factor correlated dissemination of breast cancer cells provides a direct mecha- with poor prognosis, suggesting that SIX1 is not the major nistic explanation linking VEGFC expression, lymphangiogen- driver of vegfc gene transcription in these tumors. VEGFC esis, and metastasis (15). Sunitinib did not stimulate the expression is regulated at the level of its mRNA half-life by a

www.aacrjournals.org Cancer Res; 77(5) March 1, 2017 1223

Duﬁes et al.

subtle balance between HuR and TTP.Byincreasingthehalf-life these results could be that sunitinib is detrimental for patients, of a specific mRNA, HuR enhances the levels of proteins that which is absolutely not the case (Supplementary Table S1B; promote cell proliferation, increase cell survival and local refs. 10, 11). Our results confirm that sunitinib compared with angiogenesis, help the cancer cells to evade immune recogni- other treatments prolonged survival but suggest a fraction of the tion, and facilitate cancer cell invasion and metastasis (46). nonresponder patients had more invaded lymph nodes and new High levels of cytoplasmic HuR have been found in oral, metastatic sites compared with nonresponder patients treated colorectal, gastric, lung, breast, ovarian, renal, skin carcinomas, with other therapeutic agents, a potential reason for progression and mesothelioma. Stromal cells and adjacent nontumor tis- on treatment. Our results suggest that combining VEGFC inhibi- sues do not show cytoplasmic expression of HuR (47). Cyto- tors with sunitinib could limit the progression of the disease. In an plasmic HuR expression is associated with reduced RCC sur- analyzed cohort, 28% of patients stopped sunitinib treatment vival (48) and downregulation of HuR inhibits cell prolifera- because of intolerance. New TKIs like axitinib or pazopanib are tion and induces apoptosis of RCC cells (49). Cytoplasmic better tolerated or provide a better health-related quality of life expression of HuR is associated with lymph node metastasis (53). Moreover, they have an acidic pKa that prevents their and advanced disease in non–small cell lung, colon, and upper sequestration in lysosomes, a mechanism associated with resis- urinary tract urothelial carcinomas. The cytoplasmic levels of tance to sunitinib (34). However, both inhibitors also target HuR are increased in tumors with lymphatic/vascular invasion PDGFR and cKit and similarly stimulated VEGFC (Supplementary compared with tumors without vessel invasion in cervical, Fig. S1). This is another argument for the combination of TKI with colon, and in situ breast ductal carcinomas (47). These results VEGFC inhibitors. To conclude, overexpression of VEGFC repre- are consistent with sunitinib-dependent activation of HuR, sents an extrinsic mechanism of adaptation of RCC leading to increased VEGFC mRNA half-life, and lymphangiogenesis. drug resistance. Sunitinib was shown to diminish the postsurgical benefits of For multikinase TKIs such as sunitinib and sorafenib, VEGFC is neoadjuvant sunitinib treatment in mice, including the pro- induced in an VEGFR-independent manner even in vitro, but for motion of metastasis (and decrease in survival) of select RCC VEGFR-selective TKIs, VEGFC is induced only in vivo, suggesting models. Initial observations suggested that sunitinib modified that in those cases the target cells are stromal. A recapitulated the localization of the metastases and could increase incidence schema is shown in Supplementary Fig. S7. Although sunitinib in the lymphatic vessels of spleen and stomach, two organs has revolutionized the care of patients, its efficacy may be drained to a large extent by lymphatic vessels (24). These improved by targeting VEGFC-dependent development of the findings are consistent with our results showing that sunitinib lymphatic network, a major route of spread of tumor cells when increased VEGFC expression and lymphangiogenesis. Sunitinib the patients become resistant to therapy. also induced VEGFR2 and VEGFR3 expression in LEC in vitro and in stromal cells in vivo.Expressionmayfavortheparacrine Disclosure of Potential Conflicts of Interest action of VEGFC overexpressed by tumor cells after sunitinib No potential conflicts of interest were disclosed. treatment on LEC and the development of a lymphatic network in the tumor. Whereas sunitinibinhibitsVEGFR2andVEGFR3, Authors' Contributions the accumulation of VEGFC may act during the treatment break Conception and design: M. Dufies, S. Giuliano, K.S. Khabar, R. Grepin, G. Pages that is part of the sunitinib regimen (four weeks of treatment Development of methodology: M. Dufies, S. Giuliano, P.D. Ndiaye, followed by a two week holiday to limit toxicity). Alternatively, E. Chamorey, J. Parola, J.M. Ferrero, G. Pages VEGFC may stimulate neuropilin-2 (NRP2), the coreceptor of Acquisition of data (provided animals, acquired and managed patients, fi VEGFR3, which is overexpressed on RCC cells (50). The impor- provided facilities, etc.): M. Du es, D. Ambrosetti, A. Claren, L.S. Cooley, E. Chamorey, V. Vial, J.C. Bernhard, A. Ravaud, D. Borchiellini, J.M. Ferrero, tance of NRP2 during the initiation of new lymphatic vessel R. Grepin, G. Pagès sprouts was detected in hypoplastic lymphatic vessels observed Analysis and interpretation of data (e.g., statistical analysis, biostatistics, in nrp2 gene–targeted mice. Among other structures, NRP2 is computational analysis): M. Dufies, S. Giuliano, D. Ambrosetti, M. Mastri, expressed on veins and upregulated in tumor-associated lymphat- M. Ettaiche, E. Chamorey, J.M. Ferrero, J.M. Ebos, G. Pages ic vessels where it binds VEGFC, VEGFA, and partially processed Writing, review, and/or revision of the manuscript: M. Dufies, D. Ambrosetti, VEGFD (12). Moreover, blocking NRP2 function inhibits tumor M. Lupu-Plesu, A. Ravaud, D. Borchiellini, J.M. Ferrero, A. Bikfalvi, J.M. Ebos, K.S. Khabar, G. Pages cell metastasis (51). NRP2 expressed on cancer cells interacts with Administrative, technical, or material support (i.e., reporting or organizing alpha 5 integrin on endothelial cells to mediate vascular extrav- data, constructing databases): W. Moghrabi, J. Parola, J.M. Ferrero, G. Pages asation and promotion of metastasis in zebrafish and murine Study supervision: R. Grepin, G. Pages xenograft models of RCC and pancreatic adenocarcinoma (50). Moreover, NRP2 correlates with poor prognosis in patients with Acknowledgments advanced RCC. The median overall survival was longer for We thank Dr. Baharia Mograbi for help with cBioPortal analysis, Dr. patients with low levels of NRP2 (26 months) compared with Christiane Brahimi-Horn for editorial assistance, Dr. Ophelie Cassuto for help patients overexpressing NRP2 (13 months; ref. 52). NRP2 mRNA with analysis of patient records, and Maher Al-Saif for technical assistance in is increased in tumors of mice treated with sunitinib. For this ARE cloning. We also thank Dr. Heide L. Ford and Dr. Kari Alitalo for the kind gift of VEGFC promoter reporter genes. reason, cotreatment with sunitinib and VEGFC-blocking agents may be a good option for RCC patients to reduce the progression of the disease. This cotreatment could result in the reduction of the Grant Support sunitinib dose, avoiding intercure during which lymphatic vessels This work was supported by the French association for cancer research (ARC; to G. Pages), the French National Institute for Cancer may develop. Research (INCA; to G. Pages), the Fondation de France (S. Giuliano and Our study showed that sunitinib correlated to lymphatic inva- M. Dufies), the Fondation d'entreprise Groupe Pasteur Mutualite sion in experimental and human tumors. A quick appraisal of (M. Dufies), the Framework Program 7 of the European Commission –

1224 Cancer Res; 77(5) March 1, 2017 Cancer Research

Resistance to Sunitinib in Renal Cell Carcinoma

Marie Curie Intra-European grant "VELYMPH" (M. Lupu-Plesu), the The costs of publication of this article were defrayed in part by the U.S. Department of Defense under Award no. W81XWH-14-1-0210 payment of page charges. This article must therefore be hereby marked (J.M. Ebos), "Cordon de Vie" Monaco (S. Giuliano), and IRIS Pharma. advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate The Bordeaux samples and associated data were collected, selected, and this fact. made available under the project clinicobiological National Cancer Data- base Kidney UroCCR supported by l'Institut National du Cancer (INCA; to Received November 15, 2016; revised December 7, 2016; accepted December A. Ravaud and J.-C. Bernhard). 8, 2016; published OnlineFirst January 13, 2017.

References 1. Eisen T, Sternberg CN, Robert C, Mulders P, Pyle L, Zbinden S, et al. 23. Grepin R, Ambrosetti D, Marsaud A, Gastaud L, Amiel J, Pedeutour F, et al. Targeted therapies for renal cell carcinoma: review of adverse event man- The relevance of testing the efficacy of anti-angiogenesis treatments on cells agement strategies. J Natl Cancer Inst 2012;104:93–113. derived from primary tumors: a new method for the personalized treat- 2. Escudier B. Sunitinib for the management of advanced renal cell carcino- ment of renal cell carcinoma. PLoS ONE 2014;9:e89449. ma. Expert Rev Anticancer Ther 2010;10:305–17. 24. Ebos JM, Mastri M, Lee CR, Tracz A, Hudson JM, Attwood K, et al. 3. Giuliano S, Pages G. Mechanisms of resistance to anti-angiogenesis ther- Neoadjuvant antiangiogenic therapy reveals contrasts in primary and apies. Biochimie 2013;95:1110–9. metastatic tumor efficacy. EMBO Mol Med 2014;6:1561–76. 4. Ribatti D. Tumor refractoriness to anti-VEGF therapy. Oncotarget 2016;7: 25. GrepinR,GuyotM,GiulianoS,BoncompagniM,AmbrosettiD, Chamorey 46668–77. E, et al. The CXCL7/CXCR1/2 axis is a key driver in the growth of clear cell 5. Wang Z, Dabrosin C, Yin X, Fuster MM, Arreola A, Rathmell WK, et al. Broad renal cell carcinoma. Cancer Res 2014;74:873–83. targeting of angiogenesis for cancer prevention and therapy. Semin Cancer 26. al-Haj L, Al-Ahmadi W, Al-Saif M, Demirkaya O, Khabar KS. Cloning-free Biol 2015;35:S224–43. regulated monitoring of reporter and gene expression. BMC Mol Biol 6. Welti J, Loges S, Dimmeler S, Carmeliet P. Recent molecular discoveries in 2009;10:20. angiogenesis and antiangiogenic therapies in cancer. J Clin Invest 2013; 27. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. 123:3190–200. Integrative analysis of complex cancer genomics and clinical profiles using 7. EbosJM,LeeCR,Cruz-MunozW,BjarnasonGA,ChristensenJG, Kerbel RS. the cBioPortal. Sci Signal 2013;6:pl1. Accelerated metastasis after short-term treatment with a potent inhibitor of 28. CeramiE,GaoJ,DogrusozU,GrossBE,SumerSO,AksoyBA,etal. The cBio tumor angiogenesis. Cancer Cell 2009;15:232–9. cancer genomics portal: an open platform for exploring multidimensional 8. Paez-Ribes M, Allen E, Hudock J, Takeda T, Okuyama H, Vinals F, et al. cancer genomics data. Cancer Discov 2012;2:401–4. Antiangiogenic therapy elicits malignant progression of tumors to increased 29. Gotink KJ, Broxterman HJ, Honeywell RJ, Dekker H, de Haas RR, Miles local invasion and distant metastasis. Cancer Cell 2009;15:220–31. KM, et al. Acquired tumor cell resistance to sunitinib causes resistance in 9. Ebos JM. Prodding the beast: assessing the impact of treatment-induced aHT-29humancoloncancerxenograftmousemodelwithoutaffecting metastasis. Cancer Res 2015;75:3427–35. sunitinib biodistribution or the tumor microvasculature. Oncoscience 10. Blagoev KB, Wilkerson J, Stein WD, Motzer RJ, Bates SE, Fojo AT. Sunitinib 2014;1:844–53. does not accelerate tumor growth in patients with metastatic renal cell 30. Gotink KJ, Broxterman HJ, Labots M, de Haas RR, Dekker H, Honeywell RJ, carcinoma. Cell Rep 2013;3:277–81. et al. Lysosomal sequestration of sunitinib: a novel mechanism of drug 11. Miles D, Harbeck N, Escudier B, Hurwitz H, Saltz L, Van Cutsem E, et al. resistance. Clin Cancer Res 2011;17:7337–46. Disease course patterns after discontinuation of bevacizumab: pooled 31. Shim M, Song C, Park S, Choi SK, Cho YM, Kim CS, et al. Prognostic analysis of randomized phase III trials. J Clin Oncol 2011;29:83–8. significance of platelet-derived growth factor receptor-beta expression in 12. Alitalo A, Detmar M. Interaction of tumor cells and lymphatic vessels in localized clear cell renal cell carcinoma. J Cancer Res Clin Oncol cancer progression. Oncogene 2012;31:4499–508. 2015;141:2213–20. 13. Tammela T, Zarkada G, Wallgard E, Murtomaki A, Suchting S, Wirzenius M, 32. Schiefer AI, Mesteri I, Berghoff AS, Haitel A, Schmidinger M, Preusser M, et al. Blocking VEGFR-3 suppresses angiogenic sprouting and vascular et al. Evaluation of tyrosine kinase receptors in brain metastases of clear cell network formation. Nature 2008;454:656–60. renal cell carcinoma reveals cMet as a negative prognostic factor. Histo- 14. SuJL,YenCJ,ChenPS,ChuangSE,HongCC,KuoIH,etal.Therole of the pathology 2015;67:799–805. VEGF-C/VEGFR-3 axis in cancer progression. Br J Cancer 2007;96:541–5. 33. MenkeJ,KriegsmannJ,SchimanskiCC,SchwartzMM,SchwartingA,Kelley 15. Wang CA, Jedlicka P, Patrick AN, Micalizzi DS, Lemmer KC, Deitsch E, et al. VR. Autocrine CSF-1 and CSF-1 receptor coexpression promotes renal cell SIX1 induces lymphangiogenesis and metastasis via upregulation of VEGF- carcinoma growth. Cancer Res 2012;72:187–200. C in mouse models of breast cancer. J Clin Invest 2012;122:1895–906. 34. Giuliano S, Cormerais Y, Dufies M, Grepin R, Colosetti P, Belaid A, et al. 16. Grau S, Thorsteinsdottir J, von Baumgarten L, Winkler F, Tonn JC, Schichor Resistance to sunitinib in renal clear cell carcinoma results from seques- C. Bevacizumab can induce reactivity to VEGF-C and -D in human brain tration in lysosomes and inhibition of the autophagic flux. Autophagy and tumour derived endothelial cells. J Neurooncol 2011;104:103–12. 2015;11:1891–904. 17. LiD,XieK,DingG,LiJ,ChenK,LiH,etal.Tumorresistanceto anti-VEGF 35. Tiedje C, Ronkina N, Tehrani M, Dhamija S, Laass K, Holtmann H, et al. therapy through up-regulation of VEGF-C expression. Cancer Lett The p38/MK2-driven exchange between tristetraprolin and HuR regu- 2014;346:45–52. lates AU-rich element-dependent translation. PLoS Genet 2012;8: 18. Levy NS, Chung S, Furneaux H, Levy AP. Hypoxic stabilization of vascular e1002977. endothelial growth factor mRNA by the RNA-binding protein HuR. J Biol 36. Mahtani KR, Brook M, Dean JL, Sully G, Saklatvala J, Clark AR. Mitogen- Chem 1998;273:6417–23. activated protein kinase p38 controls the expression and posttranslational 19. WangJ,GuoY,ChuH,GuanY,BiJ,WangB.Multiplefunctionsof the RNA- modification of tristetraprolin, a regulator of tumor necrosis factor alpha binding protein HuR in cancer progression, treatment responses and mRNA stability. Mol Cell Biol 2001;21:6461–9. prognosis. Int J Mol Sci 2013;14:10015–41. 37. Hitti E, Iakovleva T, Brook M, Deppenmeier S, Gruber AD, Radzioch D, 20. Brennan SE, Kuwano Y, Alkharouf N, Blackshear PJ, Gorospe M, Wilson et al. Mitogen-activated protein kinase-activated protein kinase 2 regulates GM. The mRNA-destabilizing protein tristetraprolin is suppressed in many tumor necrosis factor mRNA stability and translation mainly by altering cancers, altering tumorigenic phenotypes and patient prognosis. Cancer tristetraprolin expression, stability, and binding to adenine/uridine-rich Res 2009;69:5168–76. element. Mol Cell Biol 2006;26:2399–407. 21. Griseri P, Pages G. Regulation of the mRNA half-life in breast cancer. World 38. Griseri P, Bourcier C, Hieblot C, Essafi-Benkhadir K, Chamorey E, Touriol JClinOncol2014;5:323–34. C, et al. A synonymous polymorphism of the Tristetraprolin (TTP) gene, an 22. Essafi-Benkhadir K, Onesto C, Stebe E, Moroni C, Pages G. Tristetraprolin AU-rich mRNA-binding protein, affects translation efficiency and response inhibits Ras-dependent tumor vascularization by inducing vascular endo- to Herceptin treatment in breast cancer patients. Hum Mol Genet 2011; thelial growth factor mRNA degradation. Mol Biol Cell 2007;18:4648–58. 20:4556–68.

www.aacrjournals.org Cancer Res; 77(5) March 1, 2017 1225

Duﬁes et al.

39. Sandler H, Stoecklin G. Control of mRNA decay by phosphorylation of 47. Govindaraju S, Lee BS. Adaptive and maladaptive expression of tristetraprolin. Biochem Soc Trans 2008;36(Pt 3):491–6. the mRNA regulatory protein HuR. World J Biol Chem 2013; 40. Bex A, Kroon BK, de Bruijn R. Is there a role for neoadjuvant targeted 4:111–8. therapy to downsize primary tumors for organ sparing strategies in renal 48. Ronkainen H, Vaarala MH, Hirvikoski P, Ristimaki A. HuR expression is a cell carcinoma? Int J Surg Oncol 2012;2012:250479. marker of poor prognosis in renal cell carcinoma. Tumour Biol 2011; 41. Sennino B, Ishiguro-Oonuma T, Wei Y, Naylor RM, Williamson CW, 32:481–7. Bhagwandin V, et al. Suppression of tumor invasion and metastasis by 49. Danilin S, Sourbier C, Thomas L, Lindner V, Rothhut S, Dormoy V, et al. concurrent inhibition of c-Met and VEGF signaling in pancreatic neuro- Role of the RNA-binding protein HuR in human renal cell carcinoma. endocrine tumors. Cancer Discov 2012;2:270–87. Carcinogenesis 2010;31:1018–26. 42. Ishikawa Y, Aida S, Tamai S, Akasaka Y, Kiguchi H, Akishima-Fukasawa Y, 50. Cao Y, Hoeppner LH, Bach S, E G, Guo Y, Wang E, et al. Neuropilin-2 et al. Signiﬁcance of lymphatic invasion and proliferation on regional promotes extravasation and metastasis by interacting with endothelial lymph node metastasis in renal cell carcinoma. Am J Clin Pathol 2007; alpha5 integrin. Cancer Res 2013;73:4579–90. 128:198–207. 51. Caunt M, Mak J, Liang WC, Stawicki S, Pan Q, Tong RK, et al. Blocking 43. Sennino B, Ishiguro-Oonuma T, Schriver BJ, Christensen JG, McDonald neuropilin-2 function inhibits tumor cell metastasis. Cancer Cell 2008; DM. Inhibition of c-Met reduces lymphatic metastasis in RIP-Tag2 trans- 13:331–42. genic mice. Cancer Res 2013;73:3692–703. 52. CetinB, GonulII,Buyukberber S,Afsar B, Gumusay O,Algin E, et al. The 44. Belum VR, Wu S, Lacouture ME. Risk of hand-foot skin reaction with impact of immunohistochemical staining with ezrin-carbonic anhy- the novel multikinase inhibitor regorafenib: a meta-analysis. Invest New drase IX and neuropilin-2 on prognosis in patients with metastatic renal Drugs 2013;31:1078–86. cell cancer receiving tyrosine kinase inhibitors. Tumour Biol 2015;36: 45. Enholm B, Paavonen K, Ristim€aki A, Kumar V, Gunji Y, Klefstrom J, et al. 8471–8. Comparison of VEGF, VEGF-B, VEGF-C and Ang-1 mRNA regulation by 53. Escudier B, Porta C, Bono P, Powles T, Eisen T, Sternberg CN, et al. serum, growth factors, oncoproteins and hypoxia. Oncogene 1997;14: Randomized, controlled, double-blind, cross-over trial assessing 2475–83. treatment preference for pazopanib versus sunitinib in patients with 46. Abdelmohsen K, Gorospe M. Posttranscriptional regulation of cancer metastatic renal cell carcinoma:PISCESStudy.JClinOncol2014;32: traits by HuR. Wiley Interdiscip Rev RNA 2010;1:214–29. 1412–8.

1226 Cancer Res; 77(5) March 1, 2017 Cancer Research

Sunitinib Stimulates Expression of VEGFC by Tumor Cells and Promotes Lymphangiogenesis in Clear Cell Renal Cell Carcinomas

Maeva Dufies, Sandy Giuliano, Damien Ambrosetti, et al.

Cancer Res 2017;77:1212-1226. Published OnlineFirst January 13, 2017.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-16-3088

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2017/01/13/0008-5472.CAN-16-3088.DC1

Cited articles This article cites 53 articles, 22 of which you can access for free at: http://cancerres.aacrjournals.org/content/77/5/1212.full.html#ref-list-1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected].

Downloaded from cancerres.aacrjournals.org on April 28, 2017. © 2017 American Association for Cancer Research. Article!3!: VEGFC: a double-edged sword in the aggressiveness and resistance to treatment of renal cell carcinoma!

! L’étude" précédente" démontre" que" le" sunitinib" induit" in&vitro&et" in&vivo" l’expression" de" VEGFGC" et" la" lymphangiogenèse" dans" les" ccRCCs" (Dufies" et" al." 2017)." Cependant," nos" collègues"cliniciens"nous"ont"rapporté"l’existence"de"zones"hypoxiques"dans"les"ccRCCs" traités" avec" AAG." L’inhibition" de" la" néovascularisation" tumorale" provoque" un" défaut" d’apport" en" oxygène" qui" induit" des" zones" d’hypoxie" dans" la" tumeur" (Blagosklonny" 2004)." L’hypoxie" constitue" une" condition" de" stress" pour" les" cellules" tumorales."" Plusieurs" modifications" d’expression" génique" des" cellules" tumorales" et" de" celles" du" microenvironnement" vont" se" produire" pour" remédier" à" ce" stress." L’hypoxie" stimule" l’expression" d’autres" facteurs" proGangiogéniques" en" cas" de" blocage" de" l’axe" VEGFG A/VEGFR"(Kerbel"2001)."L’hypoxie"induit"aussi"la"sélection"de"cellules"résistantes"aux" AAGs" dont" le" potentiel" métastatique" est" augmenté" via" l’expression" de" protéines" proG migratoires"comme"le"SDFG1"(Finger"and"Giaccia"2010,"Semenza"2014)."Par"ailleurs,"le" réseau"lymphatique"associé"à"l’expression"de"VEGFGC"est"une"voie"de"dissémination"des" cellules" tumorale" (Dufies" et" al." 2017)." Nous" nous" sommes" donc" intérrogés" sur" une" possible" régulation" d’expression" de" VEGFGC" par" l’hypoxie" Et" par" quels" mécanismes" moléculaires"cette"régulation"se"produisait."

73" " VEGFC: a double-edged sword in the aggressiveness and resistance to treatment of renal cell carcinoma Papa Diogop Ndiaye1, Maeva Dufies2, Sandy Giuliano2, Laetitia Douguet1, Renaud Grépin2, Jérôme Durivault2, Philippe Lenormand1, Natacha Glisse1, Janita Mintcheva1, Valérie Vouret- Craviari1, Baharia Mograbi1, Nathalie Rioux-Leclercq3, Pascal Maire4 and Gilles Pagès1, 2 $

1- University of Nice Sophia Antipolis, Institute for research on cancer and aging of Nice, CNRS UMR 7284; INSERM U1081, Centre Antoine Lacassagne, France 2- Centre Scientifique de Monaco, Biomedical Department, 8 Quai Antoine Ier, MC-98000 Monaco, Principality of Monaco. 3- Rennes University, Rennes University Hospital, Department of Pathology, Rennes, France 4- INSERM U1016, Institut Cochin, Paris, 75014 France ; CNRS UMR 8104, Paris, 75014 France ; Université Paris Descartes, Sorbonne Paris Cité, Paris, 75014 France. Corresponding author: Gilles Pagès: [email protected]

Key words: VEGFC, lymphangiogenesis, sunitinib, angiogenesis, renal cell carcinoma, metastasis, CRISPR/Cas9

Running Title: VEGFC beneficial/pejorative role in tumor development

Conflict of interest: The authors have declared that no conflict of interest exists.

1" " Abstract

Hypoxic zones are common features of metastatic tumors. Renal cell carcinoma (RCC) expressing the Hypoxia Inducible Factor (HIF) because of inactivation of the von Hippel

Lindau gene (vhl), represent models of chronic hypoxia. Their outcome depends on the extent of their dissemination at diagnosis. Therefore, deciphering the mechanisms of metastasis is a major concern. The Vascular Endothelial Growth Factor C (VEGFC)-dependent development of a lymphatic network is in front line of metastatic spreading. To address the role of VEGFC in RCC dissemination, we studied its expression in hypoxia and we invalidated its gene in human and mouse model cell lines of RCC. Hypoxia down-regulates VEGFC mRNA through a decrease in transcription and mRNA stability but concomitantly induced VEGFC protein expression. Increased proliferation and migration abilities, over-activation of the AKT signaling pathway and enhanced expression of mesenchymal and stem cell markers characterized vegfc-/- cells. Whereas vegfc-/- cells do not form tumors in immuno-deficient mice, they develop aggressive tumors in immuno-competent mice. Moreover, mouse RCC cells generate fast-growing tumors in mice invalidated for six1 or eya2, two major regulators of VEGFC expression. Lymphangiogenic markers overexpression including VEGFC is linked to increased disease-free and overall survival in patients with non-metastatic tumors whereas decreased progression-free and overall survival is observed for metastatic patients. Our experiments describe a subtle regulation of VEGFC by hypoxia and highlight its beneficial or pejorative role. Therefore, targeting VEGFC for therapy must be considered with caution.

2" " Introduction

The incidence of renal cell carcinoma (RCC) has increased during the last decade. Surgical resection of small and non-invasive tumors detected by serendipity is curative most of the time. However, metastatic RCC are intrinsically resistant to radio- and chemo-therapy which worsened the prognosis. Mutation/inactivation of the von Hippel-Lindau (vhl) gene, an E3 ubiquitin ligase of the Hypoxia Inducible Factor 1 or 2α (HIF-1α, HIF-2α), occurs in almost eighty percent of RCC. The subsequent stabilization of HIF-1α leads to overexpression of

VEGF and exacerbated vascularization. Therefore, RCC represent a paradigm for the use of anti-angiogenic therapies (AAG). They have changed the survival of patients with metastatic disease from a few months with previous treatment with interleukin 2 or interferon to several years in the most favorable cases (1). However, in case of relapse, death generally occurs within a few months. The first line reference treatment is sunitinib, an inhibitor of tyrosine kinase receptors (TKI) including the VEGF (VEGFR1/2/3), PDGF, CSF1 receptors and c-Kit

(2). At relapse on sunitinib, other TKIs or mTOR inhibitors are available for second or third line including axitinib (3), pazopanib (4), cabozantinib (5), lenvatinib (6) and everolimus (7).

Immunotherapies have also demonstrated promising results in recent clinical trials (8).

However, these treatments are not curative. Therefore, understanding the mechanism of metastatic propagation from an indolent disease in non-metastatic patients (M0) to a more aggressive pathology in metastatic patients (M1) and at relapse on treatment may highlight new therapeutic strategies.

The lymphatic network has long been considered an inert system for the return of interstitial fluids to the bloodstream and the drainage of leukocytes and antigens to the lymph nodes (9).

It transports the tumor cells to the lymph nodes where they are eliminated by immune cells.

This system therefore constitutes a natural barrier to metastatic dissemination. However, tumor cells saturating the lymph nodes produce VEGFC, a growth factor for lymphatic

3" " endothelial cells (10). The neo-formed lymphatic vessels bypass the lymph node and accelerate the propagation of the tumor cells to the next lymph nodes and then to other organs

(11). Therefore, the lymphatic network represents an important route of dissemination of tumor cells. The number of invaded lymph nodes is correlated with the severity of the disease.

Therefore, the detection of the sentinel lymph node has become a routine of hospital practice.

At this stage, clinicians are faced with a serious situation limiting the therapeutic options. This observation highlights two antagonistic roles of the lymphatic vessels: a beneficial effect in the initial phase of tumor development and a pejorative effect when the lymphatic network is saturated (12).

This antagonism has not been address in depth and deserved to be investigated. We recently published that different TKIs used for the treatment of metastatic RCC, including sunitinib, stimulate the development of lymphatic vessels in experimental tumors and in tumors from patients treated in a neo-adjuvant setting (13). These results suggest that TKIs reduced the tumor burden but favor metastatic dissemination. Our paper reconciles in part the results of preclinical studies showing that AAG may elicit metastatic cell phenotypes compromising tumor-reducing benefits (14,15). However, these results were not confirmed by clinical trials showing that TKI did not stimulate tumor growth in patients with metastatic RCC (16).

Understanding the different roles of the lymphatic network and studying the molecular actors involved in its development represent major therapeutic issues. The growth factor of lymphatic endothelial cells is VEGFC whose expression is stimulated by TKIs (13). A close correlation exists between reduced survival, presence of hypoxic zones and high levels of

VEGFC in these areas (17). Conventional or targeted chemotherapy and radiotherapy induce intra-tumor hypoxia (18) and production of VEGFC (13,19). Hypoxia is a patho/physiological condition for the selection of aggressive tumor cells which is dependent on HIF1 and/or HIF2.

HIF1 has tumor suppressor characteristics whereas HIF2 has oncogenic properties in RCC

4" " (20). Testing the role of hypoxia in RCC cells and the involvement of HIF1 or HIF2 appeared inappropriate. However, tumors with active VHL have a darker prognosis (21).

The presence of lymphatic vessels and the metastatic potential of tumors have been studied extensively but these investigations have mainly been performed on advanced tumors. The role of lymphatic vessels on non-metastatic (M0)/metastatic (M1) tumor aggressiveness has not been investigated. In addition, knowledge of the molecular mechanisms responsible for the presence of VEGFC at diagnosis and in response to treatments is a major research issue.

Controlling VEGFC's action on lymphatic vessel development would improve the effectiveness of current treatments.

In this study, we observe that basal expression of VEGFC depends on HIF2 in cells representative of aggressive RCC. However, we found that hypoxia further stimulated

VEGFC expression. Although, VEGFC protein levels increased, mRNA levels were downregulated by hypoxia. This down-regulation depends on a transcriptional and posttranscriptional mechanism for which NF kappa B is involved (NFκB). To further address the role of VEGFC in vivo xenograft experiments were performed in immuno-compromised and immuno-proficient mice wild-type or knock-out for the transcriptional positive regulator of

VEGFC, the sine oculis (six1) gene and its co-activator eye absent homolog 2 (eya2) (22,23).

Whereas tumors developed rapidly and metastasized in immuno-proficient mice, their growth was greatly inhibited in immuno-deficient mice. Our findings suggest that VEGFC regulation by hypoxia is subtle and highly depends on hypoxia in a HIF2 dependent manner. VEGFC appears as beneficial or detrimental for tumor growth. Targeting VEGFC should be considered with caution for the treatment of RCC patients.

5" " Materials and Methods

Reagents and antibodies

Sunitinib, was purchased from Selleckchem (Houston, USA). Anti-ARD1 antibodies were home-made and previously described (24). Ant-Myc and anti-Twist antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-Slug, anti-Nanog, anti-Oct-4, anti-Slug antibodies were from Cell Signaling Technology (Beverly, MA, USA).

Cell culture

786-0 (786) and RENCA (498) RCC cell lines were purchased from the American Tissue

Culture Collection. RCC10 (R10) cells were a kind gift from Dr. W.H. Kaelin (Dana-Farber

Cancer Institute, Boston, MA).

Immunoblotting

Cells were lysed in buffer containing 3% SDS, 10% glycerol and 0.825 mM Na2HPO4. 30 to

50 µg of proteins were separated on 10% SDS-PAGE, transferred onto a PVDF membrane

(Immobilon, Millipore, France) and then exposed to the appropriate antibodies. Proteins were visualized with the ECL system using horseradish peroxidase-conjugated anti-rabbit or anti- mouse secondary antibodies.

Quantitative Real-Time PCR (qPCR) experiments

One microgram of total RNA was used for the reverse transcription, using the QuantiTect

Reverse Transcription kit (QIAGEN, Hilden, Germany), with blend of oligo (dT) and random primers to prime first-strand synthesis. SYBR master mix plus (Eurogentec, Liege, Belgium) was used for qPCR. The mRNA level was normalized to 36B4 mRNA. Oligo sequences of the VEGFC mRNA were already described (13).

6" " Genomic disruption of vegfc using CRISPR-Cas9

786-O, RCC10 or RENCA cells were transfected with PX458 plasmids containing CRISPR-

Cas9 targeting regions of the first exon of the vegfc gene using JetPRIME (Polyplus). The pSpCas9 (BB)-2A-GFP (PX458) plasmid was a gift from Dr. Feng Zhang (Addgene plasmid

# 48138) (25). The sgRNA sequence that we cloned into the vector to target vegfc gene was:

5’-TGTAGATGAACTCATGACT-3’ for the human gene and 5’-

ACCGTCGCCGCCTTCGAGTC -3’. As the PX458 plasmid contains GFP, cells were first sorted using flow cytometry to obtain cells containing the CRISPR-Cas9 and followed by clonal selection and screening. Sequencing of human genomic DNA to confirm the mutations leading to vegfc invalidation was performed using the following primers; Forward: 5’-

CAGCTCTCGTTTCCAATGC-3’; Reverse: 5’-GGAGCCTCAACAGTAGGTAG-3’.

Tumor xenograft experiments

Ectopic model of RCC: Five million 786-O or 105 RENCA cells were injected subcutaneously into the flank of 5-week-old nude (nu/nu), Balb-C mice (Janvier, France). six1 (23) or eya2

(26) knock-out (KO) and corresponding littermates were injected equivalently. The tumor volume was determined with a caliper (v = L*l2*0.5). This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory

Animals. Our experiments were approved by the ‘‘Comité national institutionnel d’éthique pour l’animal de laboratoire (CIEPAL)’’ (reference: NCE/2013-97).

Measurement of cytokines

After stimulation, cells supernatant was recovered for VEGFC measurement using the Human

DuoSet ELISA kit (R&D Systems, MN, USA).

7" " Luciferase assays

Transient transfections were performed using 2 µl of lipofectamine (GIBCO BRL) and 0.5 µg of total plasmid DNA-renilla luciferase in a 500 µl final volume. The firefly control plasmid was co-transfected with the test plasmids to control for the transfection efficiency. 24 hours after transfection, cell lysates were tested for renilla and firefly luciferase. All transfections were repeated four times using different plasmid preparations. LightSwitch™ Promoter

Reporter VEGFC (S710378) and LightSwitch™ 3´UTR reporter VEGFC (S803537) were purchased from Active motif (CA, USA). The short and long forms of the VEGFC promoter are a kind gift of Dr. Heide L Ford and Kari Alitalo (22). Mutation of the NFκB binding site in the ϖεγφχ promoter was obtained as already described (wild-type sequence of the NFkB site; 5’-GGGAAACGGGGAGCT-3’; mutated; 5’-GGGAAACAAGGAGCT-3’, (27)).

Chromatine immunoprecipitation (ChIp)

These experiments were performed as already described (28). Briefly, cells were grown in normoxia or hypoxia for 24 h (5–10#×#106 cells were used per condition). Cells were then fixed with 1% (v/v) formaldehyde (final concentration) for 10 min at 37°C and the action of the formaldehyde then stopped by the addition of 125 mM glycine (final concentration). Next, cells were washed in cold PBS containing a protease inhibitor cocktail (Roche, Basel,

Switzerland), scrapped into the same buffer and centrifuged. The pellets were re-suspended in lysis buffer, incubated on ice for 10 min, and sonicated to shear the DNA into fragments of between 200 and 1,000 base pairs. Insoluble material was removed by centrifugation and the supernatant was diluted 10-fold by addition of ChIP dilution buffer and pre-cleared by addition of salmon sperm DNA/protein A agarose 50% slurry during 1 h at 4°C. About 5% of the diluted samples was stored and constituted the input material. Immunoprecipitation was then performed by addition of anti-HIF-2α or anti-tubulin as IgG control antibodies for 24 h at

8" " 4°C. Immunocomplexes were recovered by adding 50% of salmon sperm DNA/protein A agarose and washed sequentially with low salt buffer, high salt buffer, LiCl buffer and TE.

DNA complexes were extracted in elution buffer, and cross-linking was reversed by incubating overnight at 65°C in the presence of 200 mM NaCl (final concentration). Proteins were removed by incubating for 2 h at 42°C with proteinase K and the DNA was extracted with phenol/chloroform and precipitated with ethanol. Immuno-precipitated DNA was amplified by PCR with the following primers: HIF primers: First couple of primers, promoter region –458/-344 Forward primer: 5’-GGACAAGAACTCGGGAGTGG-3’; reverse primer:

5’-ACCGGCTTTAGAGGTGATGC; second couple of primers, promoter region -457/-342;

5’-GACAAGAACTCGGGAGTGGC-3’; 5’-GGACCGGCTTTAGAGGTGAT-3’; NFκB primers, promoter region -364/-105 ; Forward primer: 5’-GCATCACCTCTAAAGCCGGT-

3’; reverse primer : 5’-TGCCTGCGCTTATGTGAGAG-3’.

5,6-Dichlorobenzimidazole riboside (DRB) pulse chase experiments

25 µg/ml of DRB was added to the cells and RNAs were prepared from 0 to 4 hr thereafter.

The level of VEGFC was determined by qPCR and was normalized to 36B4 mRNA. The relative amounts of VEGFC mRNA at time 0 before DRB addition were set to 100%.

siRNA assay siRNA transfection was performed using Lipofectamine RNAiMAX (Invitrogen). Cells were transfected with either 50 nM of si-HIF2a (siHIF2-sense: 5'-

CAGCAUCUUUGAUAGCAGU-3';siHIF2-antisense: 5'-ACUGCUAUCAAAGAUGCUG-3' or si-Control (Ambion, 4390843). Two days later, qPCR was performed, as described above.

9" " Single cell suspension procedures

Spleen were mechanically dissociated, homogenized, and passed through a 100 µM cell strainer in PBS with 5% FCS and 0.5% EDTA. Tumors were mechanically dissociated and digested with 1 mg ml-1 collagenase A (Roche) and 0.1 mg ml-1 DNase I (Roche) for 30 min

37°C.

Cell staining and flow cytometry

Surface staining was performed by incubating cells on ice, for 20 min, with saturating concentrations of labeled Abs in PBS, 5% FCS and 0.5% EDTA. Mouse cell-staining reactions were preceded by 15-min incubation with purified anti-CD16/32 Abs (2.4G2).. The following anti-mouse antibodies were used: FITC – conjugated anti-B220 (RA3-632), PE – conjugated anti-δTCR (GL3), APC- conjugated anti-CD11b (M1/70), PerCP-Cy5.5 – conjugated anti-CD3 (145-2C11), V450-conjugated anti-Ly6C (AL-21), PE-Cy7-conjugated anti-CD11c (HL3), AF700-conjugated anti-Ly6G (1A8), BV786-conjugated anti-CD45.2

(104), BV711-conjugated anti-CD4 (RM4-5), BV650-conjugated anti-CD8 (53-6.7).

Antibodies were purchased from BD Biosciences except anti-CD11b and anti-B220 from eBioscience. Data files were acquired on Aria II and analyzed using Diva software (BD

Biosciences).

Gene expression microarray analysis

Normalized RNA sequencing (RNA-Seq) data produced by The Cancer Genome Atlas

(TCGA) were downloaded from cBiopotal (www.cbioportal.org, TCGA Provisional; RNA-

Seq V2) and were analyzed as previously described (13). The results published here are in whole or in part based upon data generated by the TCGA Research

Network: http://cancergenome.nih.gov/ (29,30).

10" " Patients and association studies

Primary tumor samples of M0 RCC patients were obtained from the Rennes University hospital (31). The disease-free survival (DFS) and overall survival (OS) were calculated from patient subgroups with VEGFC mRNA levels that were less or greater than the third quartile value (Figure 1 and Supplementary Table S1). DFS was defined as the time from surgery to the appearance of metastasis. PFS was defined as the time between surgery and subsequent blood sampling and progression, or death from any cause, censoring live patients and progression free at last follow-up. OS was defined as the time from blood sample collection to the date of death from any cause, censoring those alive at last follow-up. The Kaplan Meier method was used to produce survival curves and analyses of censored data were performed using Cox models.

Statistical analysis

All data are expressed as the mean ± the standard error (SEM). Statistical significance and p values were determined by the two-tailed Student’s t-test or Mann-Whitney test. One-way

ANOVA was used for statistical comparisons. Data were analyzed with Prism 5.0b

(GraphPad Software) by one-way ANOVA with Bonferroni post hoc.

Results

VEGFC expression depends on HIF2 in RCC

The relative expression of HIF1α and HIF2α had already been investigated in different RCC cell lines and we confirmed that some RCC cells inactivated for vhl express the two HIF isoforms (RCC4) or only HIF2 for the most aggressive (RCC10 and 786-O, Fig. S1A, (32)).

Therefore, we focused on the most aggressive cells only expressing HIF2α (33). We already demonstrated that VEGFC expression is correlated to the relative aggressiveness of RCC cells

11" " according to their ability to form tumors in nude mice (13). VEGFC expression is induced by hypoxia in cells representative of breast, pancreatic cancers and melanoma but such induction is independent of HIF1 (17). However, the role of HIF2, the “oncogenic form of HIF” in

VEGFC expression has not been investigated. Therefore, we hypothesized that HIF2 should be one of the driver of VEGFC expression in aggressive cells mimicking a chronic hypoxia through vhl inactivation. Consistently, HIF2 knocked-down by siRNA (Fig. 1A and B, Fig.

S1B) decreased VEGFC protein levels in two independent cell lines only expressing HIF2α

(Fig. 1C, 786-O cells and Fig. S1C, RCC10 cells). However, HIF2α knocked-down resulted in an increase in mRNA levels in these two independent cell lines (Fig. 1D, Fig S1D).

Moreover, HIF1α and HIF2α knocked-down in cells expressing both proteins (RCC4) also resulted in VEGFC mRNA increase (Fig. S1E). These results suggest that HIF1 and HIF2 are involved in a negative control of VEGFC mRNA expression probably via an inhibition of transcription and/or mRNA stability.

Opposite effects of hypoxia on vegfc gene transcription, mRNA half-life and protein expression

Generally, mRNA and protein levels are directly correlated. However, this paradigm cannot be applied to VEGFC. Morfoisse and co-worker have elegantly shown that hypoxia stimulated VEGFC protein production but induced a down-regulation of mRNA amounts

(17). Considering the negative role of HIF2 in the control of mRNA amounts but the positive role in VEGFC protein expression (see above), we hypothesized that an experimental condition resulting in an increase of HIF2 level would down-regulate VEGFC mRNA amounts and up-regulate protein amounts. Although 786-O cells constitutively express HIF2 because of vhl inactivation, hypoxia further induced HIF2α expression (Fig. 2A).

Consistently, as HIF2 acts as a negative regulator on mRNA levels (Fig. 1D), VEGFC mRNA

12" " amounts decreased in hypoxia (Fig. 2B). To understand the molecular mechanism associated with this down-regulation, we investigated the vegfc promoter activity as a read out of its transcriptional control. A schema of this promoter containing functional hypoxia (HRE (34)) and NFκB response elements (35), is shown in Fig. 2C. The luciferase activity representing vegfc promoter activity is down-regulated by hypoxia which strongly suggests that the decrease of mRNA levels in hypoxia at least relies on an inhibition of transcription (Fig. 2D).

To address the effect of hypoxia on mRNA half-life which mainly relies on its 3’UTR, we used a reporter vector in which the luciferase mRNA half-life/activity is controlled by the

VEGFC mRNA 3’UTR (Fig. 2E) (13). The luciferase activity which is a read out of the

3’UTR-dependent mRNA half-life was inhibited by hypoxia (Fig. 2F). The decrease in mRNA amounts (Fig. S2A; inhibition of transcription (Fig. S2C), and mRNA stability (Fig.

S2D)) and increase in protein amounts (Fig. S2B) were confirmed in an independent cell line

(R10). Whereas VEGFC protein production is dependent on HIF2 in normoxia, HIF2 downregulation by siRNA did not affect VEGFC production in hypoxia (Fig. 2G). Equivalent results were obtained on an independent cell line (Fig. S2E). These results suggest that, chronic hypoxia mediated by vhl inactivation and acute hypoxia induced by incubation in low oxygen concentrations, regulate VEGFC expression through independent mechanisms.

Inhibitory effects of hypoxia on vegfc gene transcription depend on a cross talk between

HIF2 and NFκB.

The NFκB-dependent transcriptional regulation of vegfc by tumor necrosis factor alpha

(TNFα) was already described (35). Considering that NFκB and HIF2 conversely control each other’s transcription (36), we analyzed their respective expression and effect on vegfc transcription in hypoxia. ChIP experiments clearly demonstrated that NFκB and HIF2, bound to the vegfc promoter (Fig. 3A). Moreover, Guo et al showed that HIF1 stimulates VEGFC

13" " expression via its direct binding to hypoxia responsive elements present in the vegfc promoter region (34). We demonstrated that HIF2 also binds to the same domain of the vegfc promoter.

HIF2 recruitment increased and NFκB binding decreased in hypoxia (Fig. 3A). These results and the ones presented in the previous figures strongly suggest that the inhibition in vegfc transcription observed in hypoxia relies on an inhibitory mechanism mediated by enhanced

HIF2 binding and decreased NFκB binding to the promoter. The basal vegfc promoter activity depends on the integrity of the NFκB binding site (Fig. 3B, Fig. S3). Moreover, vegfc promoter activity is stimulated by the inflammatory cytokine TNFα and this induction is dependent on the NFκB binding site as demonstrated in gallbladder carcinoma confirming the functionality of NFκB in RCC cells (Fig. 3C, (35)). These results strongly suggest that the inhibition in vegfc transcription in hypoxia involves a dissociation of NFκB from its binding domain. Since NFκB activity is dependent on phosphorylation, we investigated its post translational modification in hypoxia. Stimulation by TNFα was used as a positive control of

NFκB activation/phosphorylation. As expected, TNFα stimulated p65 expression and phosphorylation. Although p65 amount was increased by hypoxia, probably reflecting the

HIF-dependent transcriptional up-regulation of NFκB, the phosphorylated active form of

NFκB was down-regulated. The TNFα-mediated increased expression and phosphorylation of

NFκB was also decreased by hypoxia. In the same experimental conditions, HIF2 was not affected by TNFα although its expression was up-regulated by hypoxia (Fig. 3D). Therefore, these results strongly suggest that hypoxia inhibits the activity/phosphorylation of NFκB limiting its affinity to the vegfc promoter, thereby participating in the down-regulation of vegfc transcription.

14" " vegfc invalidation resulted in increased proliferation, migration and stem cell makers expression

To further address the role of VEGFC in hypoxic environment mediated along tumor development, we invalidated its gene in human (786-O) and mouse (RENCA) RCC cells.

RENCA express low whereas 786-O cells express high VEGFC levels (Fig. S4A). Two independent invalidated clones were obtained and characterized for 786-O (Fig. 4A-C), and for RENCA cells (Fig. S4B and C). VEGFC mRNA and protein were not detected in 786-O

(Fig. 4A and B) and RENCA clones (Fig. S4B and D). The vegfc genomic locus was sequenced and showed small deletions in 786-O cells (Fig. 4C) demonstrating the specificity of the CRISPR/Cas9 invalidation. 786-O invalidated clones (Cl1, Cl2) presented higher proliferation abilities assessed by clonogenic assays (bigger size of the colonies, Fig. 4D) and measurement of lived cell counts after six and seven days (Fig. 4E). Increased proliferation is associated with enhanced AKT activity (Fig. 4F). vegfc-invalidated 786-O clones also presented higher invasive properties (Fig. 4G and H) which was consistent with an increase in the mesenchymal markers Slug and Twist (Fig. 4I) (37). Moreover, they also expressed increased levels of the stem cell markers c-Myc, Nanog and Oct-4, (Fig. 4I). These results suggest that VEGFC lowers the intrinsic aggressiveness of RCC cells.

vegfc-invalidated cells do not form tumors in immuno-deficient mice whereas they formed highly fast-growing and invasive tumors in immuno-competent mice

To further address the role of VEGFC on the microenvironment and subsequently on tumor aggressiveness, we subcutaneously implanted the invalidated cells in different mouse models: i) immuno-deficient nude mice were implanted with human cells; ii) immuno-proficient Balb-

C mice were implanted with mouse cells. vegfc-/- 786-O cells invalidated for vegfc formed very small tumors in only 30% of inoculated nude mice (Fig. 5A and Fig. S5A). This result is

15" " consistent with the stem cells properties of vegfc-/- cells and the lower ability of stem cells to form tumors compared to highly proliferative and differentiated tumor cells (Fig. 4I). In an opposite way, vegfc-/- RENCA cells formed bigger tumors more rapidly (Fig. 5B and Fig.

5SB). Moreover, only mice injected with vegfc-/- cells presented early metastatic dissemination (Fig. 5C). This inverse result in immuno-competent and immuno-deficient mice suggests that VEGFC produced by tumor cells may educate immune cells of the tumor microenvironment. Therefore, at the end of the experiment, we analyzed the composition of immune cells infiltrate within tumors. Recruitment of immune cells were equivalent in tumors generated with wild-type, vegfc-/+ or vegfc-/- RENCA cells. No modulation in lymphoid compartment was observed (CD4+T, CD8+T, γδT and B cells). While recruitment of total dendritic cells (CD11c+) and monocytic myeloid derived suppressor cells (M-MDSC) were not affected by the presence or absence of VEGFC, we found a significant decrease in polymorphonuclear myeloid derived suppressor cells (PMN-MDSC) in tumor obtained with vegfc-/- cells (Fig. 5D). This observation suggests that VEGFC produced by tumor cells is responsible for PMN-MDSC recruitment. This result may be in contradiction with tumor aggressiveness obtained with vegfc-/- RENCA cells since PMN-MDSC are generally associated with a poor prognosis. However, we also observed an increase of tumor associated macrophages (TAM) (Fig. 5E), which may compensate for the decreased infiltration of PMN-

MDSC. These results (initial aggressiveness, decrease and increase of different immune cells synonymous of poor prognosis) illustrate the ambivalence of VEGFC activity during the initiation of tumor development and during the phase of accelerated growth and dissemination.

16" " VEGFC produced by the microenvironment lowers tumor growth

To address the role of VEGFC produced by the microenvironment, we injected wild-type

RENCA cells in mice invalidated for genes coding the two major regulators of vegfc transcription, SIX1 and its co-activator EYA2. Tumors generated with RENCA cells developed faster and were bigger in the six1-/- mice (Fig. 6A). The incidence of tumor formation is equivalent in wild-type or six1-/- mice (almost 70% of mice developed a tumor).

In eya2-/- mice, bigger tumors appeared rapidly in eya2-/- mice fifteen days after injection but reached the same incidence and volume in wild-type and invalidated mice at the end of the experiment (Fig. 6B and Fig. S5C). These results strongly suggest that VEGFC produced by tumor cells (above) or cells of the microenvironment slows-down tumor initiation.

Angiogenic and lymphangiogenic genes have an opposite prognostic role in non- metastatic and metastatic patients

The ambivalence observed for the VEGFC role in animal models has prompted us to analyze the prognostic role of lymphangiogenic genes in the cohort of M0 patients from the Rennes

University hospital (Fig. 7A for the patients’ characteristics). High levels of VEGFC mRNA in the tumors of M0 patients (third quartile cut-off, n=25/38) correlated with a shorter DFS

(82 months versus not reach, p = 0.0222, Fig. 7B left) and OS (122 months versus not reach, p

= 0.0244, Fig. 7B right). These results suggest that VEGFC reduces tumor aggressiveness and they are in agreement with mice tumors. Then, 535 patients of The Cancer Genome Atlas

(TCGA) were analyzed for mRNA levels of lymphangiogenic genes and subsequent follow- up in term of disease free (DFS), progression free (PFS) and overall survival (OS). We correlated a panel of lymphangiogenic genes ((cxcl9 (C9) ; cxcl 10 (C10) ; neuropilin 2 (N2) ; podoplanine (PODO) ; prox 1 (PX1) ; vegfc (VC) ; vegfd (VD) ; vegfr2 (R2) ; vegfr3 (R3)) to these parameters. High VEGFC levels were associated with a longer DFS and OS in M0

17" " patients whereas high VEGFC levels were associated with shorter PFS and OS in M0 patients.

The same trend was observed for other lymphangiogenic genes including C9, C10, N2, VD and R3. This trend was not observed for PODO and PX1. According to this result, we established a prognosis score according to lymphangiogenic genes. Addition of the relative weight we attributed to each gene depending on a trend or a significant correlation with DFS,

PFS or OS highlighted that high levels of pro-lymphangiogenic genes reflect a longer

DFS/OS in M0 patients whereas it was exactly the contrary in M1 patients (Fig. 7C).

Discussion

The comparison of the levels of VEGFC in normal and tumor tissues and the regulation of its expression has been poorly investigated (38). We have shown that VEGFC levels are very high in primary cells derived from the healthy kidney portion. The levels of VEGFC produced by the tumor cells are highly variable; only the cells of metastatic tumors reach levels comparable to those of healthy cells (13). This situation evokes a beneficial effect of VEGFC in the primary stages of tumor development while high levels become pejorative during the stages of metastatic dissemination. This ambiguity was addressed by implanting tumor cells in immuno-deficient mice mimicking a pathological situation in which the immune system has been bypassed and no more exerts a tumor control. In this situation, VEGFC promotes tumor vascularization and tumor cell spread through the development of a lymphatic network.

Therefore, inhibiting its expression or activity may lower tumor growth. This hypothesis was confirmed by our experiments. However, the first role of VEGFC is to induce lymphatic vessel expansion in primary tumors. Dendritic cells present in the primary tumor will capture tumor antigens and will migrate to the draining lymph nodes where they will activate specific lymphocytes (39). Hence, an efficient lymphatic network participates in tumor control by the immune system (40). Considering this early primary event, VEGFC activity may contribute

18" " to keep tumor expansion in check and therefore appears as beneficial. Therefore, limiting

VEGFC production would probably result in a local tumor growth and selection of tumor cells with high proliferative and invasive abilities. VEGFC expressed by tumor cells saturating the sentinel lymph nodes will act directly on preexisting lymphatic vessels to induce lymphangiogenesis (12) and tumor dissemination. The local inflammation resulting from this rapid growth will educate inflammatory cells for local production of VEGFC and finally tumor cells spreading via the newly formed lymphatic network (41). Moreover,

VEGFC was shown to induce immune tolerance and protection against preexisting antitumor immunity which is consistent with the presence of PMN-MDSC observed in our study (42).

The presence of tumor associated lymphoid structures in generally associated with a good prognosis because they educate intratumor CD8+ T cells. CD8+ T cells are beneficial for most of the tumors. However, for RCC two types of CD8+ T cells linked to a good or poor prognosis and the absence of fully functional mature dendritic cells were described (43). It is possible that these two populations were selected according to the relative levels of VEGFC present in the microenvironment. This second step was addressed in immuno-competent mice by reducing VEGFC production in tumor cells or in cells of the microenvironment. These two ambivalent roles of VEGFC probably explain discrepant results of the literature showing that

VEGFC may have or not prognostic significance of pejorative evolution in different cancers

(44-46). Therefore, the presence of lymphatic vessels within a tumor specimen or the presence of VEGFC in the plasma as a surrogate marker of lymphangiogenesis and metastasis may highlight two opposite situations. Thus, correlating tumor or plasmatic VEGFC to prognosis may show opposite results and puzzle the investigator. Hence, targeting VEGFC right away for metastatic patients considering that it participates in the dissemination of tumor cells should be considered with caution. The role of VEGFC as a predictive marker of antiangiogenic drugs may also represent an interesting tool for patient follow-up considering the

19" " development of lymphatic vessels on treatment (13). Lower baseline levels of VEGF-C were associated with longer PFS and OS for patients treated by sunitinib and the drug induced a down-regulation of plasmatic levels in a cohort of sixty one patients (47). However, in an independent cohort of patients recently described we were unable to demonstrate a correlation between plasmatic VEGFC and PFS or OS in patients treated by sunitinib (48). These contradictory results also highlight our insufficient knowledge of how VEGFC expression is controlled. The hypoxia-HIF-dependent regulation of VEGFC expression is currently highly debated and probably involves different partners as we demonstrated in this study. We showed that HIF2 is probably the major regulator of vegfc transcription in a pathological situation mimicking a chronic hypoxia in RCC cells invalidated for pvhl. However, the HIF2 binding site on the vegfc promoter described in this study in tumor cells (proximal to the transcription initiation site) is different from those described for HIF1 in inflamed macrophages (34). HIF1 may play a role in the induction of vegfc expression in hypoxia in normal cells but not in tumor cells as it was recently shown (17). Moreover, the partners involved in vegfc transcription may also depend on acute or chronic hypoxia in tumor cells.

The decrease in VEGFC mRNA levels and the concomitant up-regulation in protein levels described in our study were consistent with previous published results (17). However, it was really surprising to observe an induction of HIF2α expression in cells mutated for vhl. The decrease in transcriptional activity observed in acute hypoxia mainly relies on the decrease in

NFκB binding to the vegfc promoter although HIF2α is up-regulated. The decrease in mRNA stability depends on the VEGFC mRNA 3’UTR. We recently demonstrated the role of an equilibrated balance of tristetraprolin and HuR in enhanced VEGFC mRNA half-life in response to anti-angiogenic drugs (13). We tested both proteins in hypoxia but their levels and phosphorylation were not modified. Additional work is needed to discover mRNA binding factors involved in the decreased mRNA half-life observed in hypoxia.

20" " In conclusion, VEGFC regulation is a concert of transcription factors and mRNA binding proteins (see recapitulative schema, Fig. S6). They act differently of VEGFC expression depending on chronic or acute hypoxia. Deciphering this subtle regulation may allow understanding the sequence of events leading to a beneficial or pejorative role of lymphangiogenesis.

21" " Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Authors' Contributions

Conception and design: P.Diogop Ndiaye, M.Dufies, S. Giuliano, K.S., L.Douguet, R.Grépin,

G. Pagès

Development of methodology: P.Diogop Ndiaye, G. Pagès

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): P.Diogop Ndiaye, M.Dufies, N.Glisse, J.Mintcheva, L.Douguet, G.

Pagès

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P.Diogop Ndiaye, M.Dufies, N.Glisse, J.Mintcheva, G. Pagès

Writing, review, and/or revision of the manuscript: G. Pagès

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): V.Vouret-Craviari, B. Mograbi, G. Pagès

Study supervision: G. Pagès

Acknowledgments

We thank Dr. Heide L. Ford and Dr. Kari Alitalo for the kind gift of VEGFC promoter reporter genes.

Grant Support

This work was supported by the French Association for Cancer Research (ARC; to G. Pagès), the French National Institute for Cancer Research (INCA; to G. Pagès), the Fondation de

France (S. Giuliano and M. Dufies), the Fondation d'entreprise Groupe Pasteur Mutualité

22" " (M. Dufies), the Framework Program 7 of the European Commission –Marie Curie Intra-

European grant! "VELYMPH", Fondation François Xavier Mora (S Giuliano) and IRIS

Pharma (P. Diogop NDiaye).

References

1. Gore ME, Szczylik C, Porta C, Bracarda S, Bjarnason GA, Oudard S, et al. Final results

from the large sunitinib global expanded-access trial in metastatic renal cell carcinoma.

Br J Cancer 2015;113:12-9

2. Eisen T, Sternberg CN, Robert C, Mulders P, Pyle L, Zbinden S, et al. Targeted therapies

for renal cell carcinoma: review of adverse event management strategies. J Natl Cancer

Inst 2012;104:93-113

3. Motzer RJ, Escudier B, Tomczak P, Hutson TE, Michaelson MD, Negrier S, et al.

Axitinib versus sorafenib as second-line treatment for advanced renal cell carcinoma:

overall survival analysis and updated results from a randomised phase 3 trial. Lancet

Oncol 2013;14:552-62

4. Motzer RJ, Hutson TE, Cella D, Reeves J, Hawkins R, Guo J, et al. Pazopanib versus

sunitinib in metastatic renal-cell carcinoma. N Engl J Med 2013;369:722-31

5. Choueiri TK, Escudier B, Powles T, Mainwaring PN, Rini BI, Donskov F, et al.

Cabozantinib versus Everolimus in Advanced Renal-Cell Carcinoma. N Engl J Med

2015;373:1814-23

6. Motzer RJ, Hutson TE, Ren M, Dutcus C, Larkin J. Independent assessment of lenvatinib

plus everolimus in patients with metastatic renal cell carcinoma. Lancet Oncol

2016;17:e4-5

23" " 7. Motzer RJ, Escudier B, Oudard S, Hutson TE, Porta C, Bracarda S, et al. Efficacy of

everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-

controlled phase III trial. Lancet 2008;372:449-56

8. Escudier B, Sharma P, McDermott DF, George S, Hammers HJ, Srinivas S, et al.

CheckMate 025 Randomized Phase 3 Study: Outcomes by Key Baseline Factors and

Prior Therapy for Nivolumab Versus Everolimus in Advanced Renal Cell Carcinoma.

Eur Urol 2017

9. Aebischer D, Iolyeva M, Halin C. The inflammatory response of lymphatic endothelium.

Angiogenesis 2014;17:383-93

10. Hirakawa S, Brown LF, Kodama S, Paavonen K, Alitalo K, Detmar M. VEGF-C-induced

lymphangiogenesis in sentinel lymph nodes promotes tumor metastasis to distant sites.

Blood 2007;109:1010-7

11. Alitalo A, Detmar M. Interaction of tumor cells and lymphatic vessels in cancer

progression. Oncogene 2012;31:4499-508

12. Karaman S, Detmar M. Mechanisms of lymphatic metastasis. J Clin Invest

2014;124:922-8

13. Dufies M, Giuliano S, Ambrosetti D, Claren A, Ndiaye PD, Mastri M, et al. Sunitinib

Stimulates Expression of VEGFC by Tumor Cells and Promotes Lymphangiogenesis in

Clear Cell Renal Cell Carcinomas. Cancer Res 2017;77:1212-26

14. Paez-Ribes M, Allen E, Hudock J, Takeda T, Okuyama H, Vinals F, et al. Antiangiogenic

therapy elicits malignant progression of tumors to increased local invasion and distant

metastasis. Cancer Cell 2009;15:220-31

15. Ebos JM, Lee CR, Cruz-Munoz W, Bjarnason GA, Christensen JG, Kerbel RS.

Accelerated metastasis after short-term treatment with a potent inhibitor of tumor

angiogenesis. Cancer Cell 2009;15:232-9

24" " 16. Blagoev KB, Wilkerson J, Stein WD, Motzer RJ, Bates SE, Fojo AT. Sunitinib does not

accelerate tumor growth in patients with metastatic renal cell carcinoma. Cell Rep

2013;3:277-81

17. Morfoisse F, Kuchnio A, Frainay C, Gomez-Brouchet A, Delisle MB, Marzi S, et al.

Hypoxia induces VEGF-C expression in metastatic tumor cells via a HIF-1alpha-

independent translation-mediated mechanism. Cell Rep 2014;6:155-67

18. Varna M, Gapihan G, Feugeas JP, Ratajczak P, Tan S, Ferreira I, et al. Stem cells

increase in numbers in perinecrotic areas in human renal cancer. Clin Cancer Res

2015;21:916-24

19. Lupu-Plesu M, Claren A, Martial S, N'Diaye PD, Lebrigand K, Pons N, et al. Effects of

proton versus photon irradiation on (lymph)angiogenic, inflammatory, proliferative and

anti-tumor immune responses in head and neck squamous cell carcinoma. Oncogenesis

2017;6:e354

20. Shen C, Kaelin WG, Jr. The VHL/HIF axis in clear cell renal carcinoma. Semin Cancer

Biol 2013;23:18-25

21. Kammerer-Jacquet SF, Crouzet L, Brunot A, Dagher J, Pladys A, Edeline J, et al.

Independent association of PD-L1 expression with noninactivated VHL clear cell renal

cell carcinoma-A finding with therapeutic potential. Int J Cancer 2017;140:142-8

22. Wang CA, Jedlicka P, Patrick AN, Micalizzi DS, Lemmer KC, Deitsch E, et al. SIX1

induces lymphangiogenesis and metastasis via upregulation of VEGF-C in mouse models

of breast cancer. J Clin Invest 2012;122:1895-906

23. Le Grand F, Grifone R, Mourikis P, Houbron C, Gigaud C, Pujol J, et al. Six1 regulates

stem cell repair potential and self-renewal during skeletal muscle regeneration. J Cell

Biol 2012;198:815-32

25" " 24. Bilton R, Mazure N, Trottier E, Hattab M, Dery MA, Richard DE, et al. Arrest-defective-

1 protein, an acetyltransferase, does not alter stability of hypoxia-inducible factor (HIF)-

1alpha and is not induced by hypoxia or HIF. J Biol Chem 2005;280:31132-40

25. Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F. Genome engineering using

the CRISPR-Cas9 system. Nat Protoc 2013;8:2281-308

26. Grifone R, Demignon J, Giordani J, Niro C, Souil E, Bertin F, et al. Eya1 and Eya2

proteins are required for hypaxial somitic myogenesis in the mouse embryo. Dev Biol

2007;302:602-16

27. Lidke DS, Huang F, Post JN, Rieger B, Wilsbacher J, Thomas JL, et al. ERK nuclear

translocation is dimerization-independent but controlled by the rate of phosphorylation. J

Biol Chem 2010;285:3092-102

28. Simon MP, Tournaire R, Pouyssegur J. The angiopoietin-2 gene of endothelial cells is

up-regulated in hypoxia by a HIF binding site located in its first intron and by the central

factors GATA-2 and Ets-1. J Cell Physiol 2008;217:809-18

29. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative

analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci

Signal 2013;6:pl1

30. Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer

genomics portal: an open platform for exploring multidimensional cancer genomics data.

Cancer Discov 2012;2:401-4

31. Kammerer-Jacquet SF, Brunot A, Pladys A, Bouzille G, Dagher J, Medane S, et al.

Synchronous Metastatic Clear-Cell Renal Cell Carcinoma: A Distinct Morphologic,

Immunohistochemical, and Molecular Phenotype. Clin Genitourin Cancer 2017;15:e1-e7

26" " 32. Shinojima T, Oya M, Takayanagi A, Mizuno R, Shimizu N, Murai M. Renal cancer cells

lacking hypoxia inducible factor (HIF)-1alpha expression maintain vascular endothelial

growth factor expression through HIF-2alpha. Carcinogenesis 2007;28:529-36

33. Schodel J, Grampp S, Maher ER, Moch H, Ratcliffe PJ, Russo P, et al. Hypoxia,

Hypoxia-inducible Transcription Factors, and Renal Cancer. Eur Urol 2016;69:646-57

34. Guo YC, Zhang M, Wang FX, Pei GC, Sun F, Zhang Y, et al. Macrophages Regulate

Unilateral Ureteral Obstruction-Induced Renal Lymphangiogenesis through C-C Motif

Chemokine Receptor 2-Dependent Phosphatidylinositol 3-Kinase-AKT-Mechanistic

Target of Rapamycin Signaling and Hypoxia-Inducible Factor-1alpha/Vascular

Endothelial Growth Factor-C Expression. Am J Pathol 2017;187:1736-49

35. Du Q, Jiang L, Wang X, Wang M, She F, Chen Y. Tumor necrosis factor-alpha promotes

the lymphangiogenesis of gallbladder carcinoma through nuclear factor-kappaB-mediated

upregulation of vascular endothelial growth factor-C. Cancer Sci 2014;105:1261-71

36. Szade A, Grochot-Przeczek A, Florczyk U, Jozkowicz A, Dulak J. Cellular and molecular

mechanisms of inflammation-induced angiogenesis. IUBMB Life 2015;67:145-59

37. Lamouille S, Xu J, Derynck R. Molecular mechanisms of epithelial-mesenchymal

transition. Nat Rev Mol Cell Biol 2014;15:178-96

38. Enholm B, Paavonen K, Ristimäki A, Kumar V, Gunji Y, Klefstrom J, et al. Comparison

of VEGF, VEGF-B, VEGF-C and Ang-1 mRNA regulation by serum, growth factors,

oncoproteins and hypoxia. Oncogene 1997;14:2475-83

39. Dieu-Nosjean MC, Giraldo NA, Kaplon H, Germain C, Fridman WH, Sautes-Fridman C.

Tertiary lymphoid structures, drivers of the anti-tumor responses in human cancers.

Immunol Rev 2016;271:260-75

40. Mellman I, Coukos G, Dranoff G. Cancer immunotherapy comes of age. Nature

2011;480:480-9

27" " 41. Riabov V, Gudima A, Wang N, Mickley A, Orekhov A, Kzhyshkowska J. Role of tumor

associated macrophages in tumor angiogenesis and lymphangiogenesis. Front Physiol

2014;5:75

42. Lund AW, Duraes FV, Hirosue S, Raghavan VR, Nembrini C, Thomas SN, et al. VEGF-

C promotes immune tolerance in B16 melanomas and cross-presentation of tumor antigen

by lymph node lymphatics. Cell Rep 2012;1:191-9

43. Giraldo NA, Becht E, Pages F, Skliris G, Verkarre V, Vano Y, et al. Orchestration and

Prognostic Significance of Immune Checkpoints in the Microenvironment of Primary and

Metastatic Renal Cell Cancer. Clin Cancer Res 2015;21:3031-40

44. Liang B, Li Y. Prognostic Significance of VEGF-C Expression in Patients with Breast

Cancer: A Meta-Analysis. Iran J Public Health 2014;43:128-35

45. Bierer S, Herrmann E, Kopke T, Neumann J, Eltze E, Hertle L, et al. Lymphangiogenesis

in kidney cancer: expression of VEGF-C, VEGF-D and VEGFR-3 in clear cell and

papillary renal cell carcinoma. Oncol Rep 2008;20:721-5

46. Kyzas PA, Cunha IW, Ioannidis JP. Prognostic significance of vascular endothelial

growth factor immunohistochemical expression in head and neck squamous cell

carcinoma: a meta-analysis. Clin Cancer Res 2005;11:1434-40

47. Rini BI, Michaelson MD, Rosenberg JE, Bukowski RM, Sosman JA, Stadler WM, et al.

Antitumor activity and biomarker analysis of sunitinib in patients with bevacizumab-

refractory metastatic renal cell carcinoma. J Clin Oncol 2008;26:3743-8

48. Dufies M, Giuliano S, Viotti J, Borchiellini D, Cooley LS, Ambrosetti D, et al. CXCL7 is

a predictive marker of sunitinib efficacy in clear cell renal cell carcinomas. Br J Cancer

2017;117:947-53

28" " Figure Legends

A C ***

siCTL siHIF2α 100

80 HIF2α 60 40

Tubulin control) of (% VEGFC protein protein VEGFC 20 0 CTL HIF2 B D *** 160 ***

100

120 80 80

mRNA mRNA 60 a

40 control) of (% VEGFC mRNA mRNA VEGFC

HIF2 40 (% control) of (% 20 0 0 CTL HIF2 CTL HIF2

Figure 1: Ndiaye et al

Figure 1. HIF2α participates in the high basal level of VEGFC mRNA in 786-O cells. A, HIF2α expression was down-regulated by siRNA; tubulin is shown as a loading control. B, quantification of the blot shown in A. C, VEGFC protein expression was assessed by ELISA in siRNA control (CTL) and HIF2α-directed siRNA (HIF2). D, VEGFC mRNA levels were evaluated by qPCR in siRNA control (CTL) and HIF2α-directed siRNA (HIF2), Results are represented as mean of three independent experiments ± SEM. *** p<0.001.

29" " A B *** CTL 100

NX HX 80 60

HIF2α 40 20 (% control) of (% Tubulin mRNA VEGFC 0 NX HX C E GCGTG GGGGAGCTC NFKB VEGFC-luciferase 3’UTR reporter -498 -236 HRE Promoter VEGFC luciferase 3’UTR VEGFC Luciferase

3’UTR D Promoter F *** 100 *** 100 80 80 60 60 40 40 20 20 (% (% of control) (% (% of control) 0 0 Luciferase activity Luciferase activity

Luciferase activity Luciferase activity NX HX NX HX NS G ** 150 ** *

100

50 (% control) of (% VEGFC protein protein VEGFC 0 NX HX NX HX

siCTL siHIF2a Figure 2. Hypoxia induces a down-regulation of VEGFC mRNA and an increase in VEGFC protein. A, Hypoxia stimulates HIF2α expression in 786-O control (CTL) and 786-O cells re-expressing VH (VHL) in hypoxia (HX) versus normoxic conditions (NX) as shown by immune-blot; tubulin is shown as a loading control. B, VEGFC mRNA levels in 786-O cells were evaluated by qPCR in normoxia (NX) and hypoxia (HX). C, Schematic representation of the vegfc promoter showing the hypoxia response element (HRE) and the NFκB binding site and their respective localization according to the transcription initiation start site . D, 786-O cells were transfected with a renilla luciferase reporter gene under the control of the vegfc promoter and incubated in normoxia (NX) or hypoxia (HX) for 48hr. The renilla luciferase activity normalized to the firefly luciferase (control vector) was the readout of the VEGFC promoter activity. E, Schematic representation of the VEGFC mRNA 3’UTR down-stream of the renilla luciferase reporter gene under the control of the cytomegalo virus promoter. F, 786-O cells were transfected with the above-mentioned reporter gene and incubated in normoxia (NX) or hypoxia (HX) for 48hr. The renilla luciferase activity normalized to the firefly luciferase (control vector) was the readout of the VEGFC mRNA half-life. G, VEGFC protein expression was assessed by ELISA in siRNA control (CTL) and HIF2α-directed siRNA (HIF2) and in normoxic (NX) or hypoxic (HX) conditions. Results are represented as mean of at least three independent experiments ± SEM. * p<0.05, ** p<0.01, *** p<0.001, NS, non-significant.

30" "

A B *** INPUT CTL Spe.Ab 100 NX HX NX HX NX HX 80 HIF2a 60

NFkB 40 20 (% (% of control)

Luciferase activity Luciferase activity 0 WT MUT

*** C D -TNFα +TNFα *** 160 NX HX NX HX 120 pP65 NS 80 P65 40 (% (% of control) HIF2α Luciferase activity Luciferase activity 0 CTL CTL + MUT MUT + TNF TNF Tubulin

Figure 3: Ndiaye et al

Figure 3. NFκB plays a key role in the control of vegfc transcription in normoxia and hypoxia. A, Chromatin immuno-precipitations from cells incubated in normoxia (NX) or hypoxia (HX) were performed by using control (CTL) or specific antibodies (Spe.Ab) for HIF2α or NFκB. The specific DNA domains containing the HIF2α or NFκB binding site in the vegfc promoter amplified by PCR were shown. B, 786-O cells were transfected with a renilla luciferase reporter gene under the control of the vegfc promoter that was wild-type (WT) or mutated for the NFκB binding site. Luciferase activity was measured 48hr post transfection. C, The same transfections were performed but the cells were stimulated or not with TNFα (150 ng/ml) for 48hr. D, 786-O cells were incubated in normoxia (NX) or hypoxia (HX) and concomitantly stimulated with TNFα (150 ng/ml). The total (P65) and phosphorylated (pP65) form of NFκB and HIF2α were detected by immuno-blot; tubulin is shown as a loading control. When indicated, results are represented as mean of at least three independent experiments ± SEM. *** p<0.001, NS, non-significant.

31" "

A 120 B 1,2 C CRISPR/Cas9 TARGET CTL GAGCAGTTACGGTCTGTGTCCAGTGTAGATGAACTCATGACTGTA

80 cells) 0,8 6 Cl1 GAGCAGTTACGGTCTGTGTCCAGTG-AGATGAACTCATGACTGTA 40 0,4 VEGFC VEGFC /ml/10 Cl2 GAGCAGTTACGGTCTGTGTCCAGTG-A---GAAGCTCATGACTGTA VEGFC mRNA mRNA VEGFC *** *** ng *** *** (Arbitrary units) (Arbitrary 0 ( 0 CTL Cl1 Cl2 CTL Cl1 Cl2 D E F 100 ** 400 CTL Cl1 Cl2 CTL ** 80 300 CTL ** Cl 1 ** 60 Cl1 Cl2 200 40 Cl2 20 /AKT %Control) 100 G increase) (Fold Proliferation index index Proliferation CTL Cl1 Cl2 0 pAKT 0 0 1 2 3 4 5 6 7 CTL Cl 1 Cl2 Time (Days)

I CTL Cl1 Cl2 c-Myc

H 200 ** Nanog Stem 150 *** Oct-4 100 50 Slug Mesenchymal 0 (% control) of (% Twist Migration ability ability Migration CTL Cl1 Cl2 et al Tubulin Figure 4: Ndiaye

Figure 4. 786-O cells invalidated for vegfc present increased proliferation and migration abilities. A, VEGFC mRNA levels were tested by qPCR in control (CTL) and two independent clones (Cl1, Cl2) invalidated for vegfc. B, VEGFC proteins levels were tested by ELISA in the supernatant of control (CTL) and two independent clones (Cl1, Cl2) invalidated for vegfc. C, The locus of vegfc was sequenced in control (CTL) and two independent clones (Cl1, Cl2) invalidated for vegfc; single or multi-bases deletions were detected in the vegfc locus. D and E, The proliferation abilities of control (CTL) or invalidated (Cl1, Cl2) cells were tested by clonogenic assays (D) or quantification of alive cells (E). F, Increased proliferation abilities were correlated with enhanced AKT activity evaluated by measuring the ratio of phosphorylated on total AKT amounts (pAKT/AKT). G and H, vegfc invalidated cells presented increased migration ability in Boyden chambers assays (G). H, Quantification of the results shown in G. I, vegfc invalidated cells presented increased levels of stem cell (c-Myc, Nanog, Oct-4) and mesenchymal (Slug, Twist) markers detected by immunoblots; tubulin is shown as a loading control. When indicated, results are represented as mean of at least three independent experiments ± SEM. ** p<0.01, *** p<0.001.

32" " A B C 1000 1600

) ) WT 3 3 800 vegfc-/- 1200 600 WT 800 400 vegfc-/- ** 400 ** 200 ** *** Tumor volume (mm volume Tumor Tumor volume (mm volume Tumor 0 0 40 60 80 100 120 140 20 25 30 35 40 Time (days) Time (days) D E 50 100 *

40 ** 80

30 60

20 40

10 20 TAM (% cells) CD45+ of (% TAM

PMN-MDSC (% cells) CD45+ of (% PMN-MDSC 0 0 WT vegfc-/- WT vegfc-/-

Figure 5: Ndiaye et al

Figure 5. Invalidation of vegfc in tumor cells results in opposite abilities to form tumors in immune-deficient or immune-proficient mice. A, Experimental tumors in nude mice (20 mice per conditions) were obtained after subcutaneous injection of 5x106 wild-type (WT) or vegfc-invalidated 786-O cells. B, Experimental tumors in immuno-competent Balb-C mice (10 mice per conditions) were obtained after subcutaneous injection of 105 wild-type (WT) or invalidated RENCA cells (vegfc -/-). C, Representative image of the peritoneal metastasis observed a few days after infection of vegfc-/- RENCA cells. D, detection by flow cytometry of PMN-MDSC cells in tumors generated with wild-type (WT) and invalidated RENCA cells (vegfc -/-). E, detection by flow cytometry of TAM in tumors generated with wild-type (WT) and invalidated RENCA cells (vegfc -/-). When indicated, results are represented as mean ± SEM. ** p<0.01, *** p<0.001.

33" " A B *** 1600 600

) ) 3 3 1200 WT WT 400 eya2-/- six1-/- 800 *

200 * 400 Tumor volume (cm volume Tumor (cm volume Tumor 0 0 0 5 10 15 20 0 5 10 15 20 Time (days) Time (days)

Figure 6: Ndiaye et al

Figure 6. VEGFC produces by the microenvironment slow-down tumor growth. A and B Experimental tumors were performed in wild-type (WT), six1 (six-/-) or eya2 (eya-/-) invalidated mice (10 mice per conditions) by subcutaneous injection of 105 RENCA cells; A, Comparison of tumor growth in WT and six1-/- mice; B, Comparison of tumor growth in WT and eya2-/- mice. Tumor volume are represented as mean ± SEM. * p<0.05, ** p<0.01, *** p<0.001.

34" " A GROUP Number of patients 38 ccRCC 38 (100%) Sex - Woman 14 (36.8%) - Man 24 (63.2%) Age 62 (42-82) Furhman grade - 2 17 (44.7%) - 3 16 (42.1%) - 4 5 (13.2%) Metastatic status pM - M0 38 (100%) Lymph node status pN - N0 34 (89.5%) - N1 4 (10.5%) Size status pT - 1 17 (44.7%) - 2 6 (15.8%) - 3 15 (39.5%) B

VEGFC high VEGFC high

VEGFC low VEGFC low

L Survival C9 C10 N1 N2 PODO PX1 VA VC VD R1 R2 R3 SCORE M0

DFS (T) (T) (T) 0.02 0.03 (T) 0.03 0.01 4

OS (T) (T) (T) 0.04 0.01 (T) 0.02 0.001 (T) 5

PFS (T) 0.01 0.04 0.04 0.01 (T) 0.03 -12 OS 0.04 0.01 0.01 -6

Figure 7: Ndiaye et al

Figure 7. Different patient outcome depending on the expression of VEGFC and lymphangiogenic genes. A, Characteristics of the patient included in the study. B, Kaplan–Meier analysis of DFS or OS of M0 patients. DFS and OS were calculated from patient subgroups with VEGFC mRNA levels that were less or greater than the third quartile. Statistical significance (p value) is indicated. C, Correlation between lymphangiogenic genes and survival (DFS/PFS/OS) in M0 and M1 patients. The tested lymphangiogenic genes were the following : cxcl9 (C9) cxcl10 (C10), neuropilin 2 (N2), podoplanin (PODO), prox1 (PX1), vegfc (VC), vegfd (VD), vegfr2 (R2) and vegfr3 (R3). Genes associated with poor prognosis appear in white on a black background; the genes associated with good prognosis appear in black on a gray background. A significant p value is given; a tendency to significance is mentioned by T (trend) in black on a gray background. A score was established from the following sequence; a positive point is given for a gene with a good prognosis trend; two positive points for genes with a significant p value; two negative points are given for a gene with a significant p value. Positive scores were found for DFS and OS for M0 patients. The scores were negative for PFS and OS for M1 patients.

35" " Legends of Supplemental Figures

A R4 R10 786

HIF2a

HIF1a

HSP90

300 B C *** D E 100 140 ** ** 250 120

80 siCTL siHIF2α 200 100 ** ** 60 80 150 HIF2α 40 60 100 (% control) of (% control) of (% (% control) of (%

VEGFC mRNA mRNA VEGFC 40 mRNA VEGFC VEGFC protein protein VEGFC Tubulin 20 50 20 0 0 0 CTL HIF2 CTL HIF2 CTL HIF1 HIF2 HIF1HIF2

Figure S1: Ndiaye et al

Supplementary Figure S1. HIF2α participates in the high basal level of VEGFC mRNA in RCC cells. A, HIF1α and HIF2α expression was tested by immunoblot in different RCC cell lines (RCC4 (R4), RCC10 (R10) and 786-O (786) cells); HSP90 is shown as a loading control. B, HIF2α expression was down-regulated by siRNA in RCC10 cells; tubulin is shown as a loading control. C, VEGFC protein levels were evaluated by ELISA in siRNA control (CTL) and HIF2α-directed siRNA (HIF2) transfected RCC10 cells. D, VEGFC mRNA levels were evaluated by qPCR in siRNA control (CTL) and HIF2α-directed siRNA (HIF2) transfected RCC10 cells. E, VEGFC mRNA levels were evaluated by qPCR in siRNA control (CTL), HIF1α-directed siRNA (HIF1), HIF2α-directed siRNA (HIF2) or both siRNA (HIF1 HIF2) transfected RCC4 cells. When indicated, results are represented as mean of three independent experiments ± SEM. ** p<0.01, *** p<0.001. * A *** B 140 100 120

80 100 80 60 60 40 40 (% control) of (% (% control) of (% VEGFC mRNA mRNA VEGFC 20 protein VEGFC 20 0 0 NX HX NX HX D C Promoter 3’UTR ** *** 100 100 80 80 60 60 40 40 (% (% of control) 20 (% of control) 20 Luciferase activity Luciferase activity Luciferase activity Luciferase activity 0 0 NX HX NX HX NS E * 150 * * * * 100

50 (% control) of (% VEGFC protein protein VEGFC 0 NX H NX H siCTLX siHIF2aX Figure S2: Ndiaye et al

Supplementary Figure S2. Hypoxia induces a down-regulation of VEGFC mRNA and an increase in VEGFC protein. A, VEGFC mRNA levels in RCC10 cells were evaluated by qPCR in normoxia (NX) and hypoxia (HX). B, VEGFC protein expression in RCC10 cells was assessed by ELISA in normoxic (NX) or hypoxic (HX) conditions. C, RCC10 cells were transfected with a renilla luciferase reporter gene under the control of the vegfc promoter and incubated in normoxia (NX) or hypoxia (HX) for 48hr. The renilla luciferase activity normalized to the firefly luciferase (control vector) was the readout of the VEGFC promoter activity. D, RCC10 cells were transfected with the VEGFC mRNA 3’UTR reporter gene and incubated in normoxia (NX) or hypoxia (HX) for 48hr. The renilla luciferase activity normalized to the firefly luciferase (control vector) was the readout of the VEGFC mRNA half-life. E, VEGFC protein expression was assessed by ELISA in siRNA control (CTL) and HIF2α-directed siRNA (HIF2) and in normoxic (NX) or hypoxic (HX) conditions. Results are represented as mean of at least three independent experiments ± SEM. * p<0.05, ** p<0.01, *** p<0.001, NS, non-significant. ** *** 100

60 NS 40 (% (% of control) 20 Luciferase activity Luciferase activity

0 CTL NX MUT NX CTL HX MUT HX

Figure S3: Ndiaye et al

Supplementary Figure S3. NFκB plays a key role in the control of vegfc transcription in normoxia and hypoxia. RCC10 cells were transfected with a renilla luciferase reporter gene under the control of the vegfc promoter that was wild-type (WT) or mutated for the NFκB binding site. Cells were incubated in normaoxia (NX) or hypoxia (HX) for 48hr. Results are represented as mean of at least three independent experiments ± SEM. ** p<0.01, *** p<0.001, NS, non-significant.

A 900

cells) 600 6 VEGFC VEGFC 300 (pg/ml/10

0 786 RENCA Human Mouse

B C 80 120

100 60 80 cells) 6 60 40

40 VEGFC 20 VEGFC mRNA mRNA VEGFC (Arbitrary units) (Arbitrary 20 (pg/ml/10 0 0 CTL Cl1 Cl2 CTL Cl1 Cl2 Figure S4: Ndiaye et al

Supplementary Figure S4. VEGFC levels in human and mouse cells; invalidation of vegfc in mouse (RENCA) RCC cells. A, VEGFC expression levels in human (786) or mouse (RENCA) RCC cells assessed by ELISA. B and C, invalidation of vegfc in RENCA cells. VEGFC mRNA (B) and protein (C) levels were assessed in control (CTL) or different clones of RENCA cells invalidated for vegfc (Cl1, Cl2) by qPCR or ELISA respectively. A B 100 50

80 40 tumor) tumor) tumor) tumor) 60 30 with with 40 20 Incidence Incidence Incidence Incidence 20 10 (% mouse of (%

0 mouse of (% 0 WT vegfc-/- C WT vegfc-/- 80

60 tumor) tumor)

with 40

Incidence Incidence 20

(% mouse of (% 0 WT eya2-/- WT eya2-/- 15 D 18 D Figure S5: Ndiaye et al

Supplementary Figure S5. vegfc invalidation resulted in a lower incidence of tumor formation in immuno-deficient mice but an increased incidence in when vegfc is invalidated/down-regulated in tumor cells or in cells of the microenvironment. A, Experimental tumors were performed in nude mice with wild-type (WT), or vegfc-invalidated 786-O cells (vegfc-/-, 20 mice per conditions) by subcutaneous injection of 5x106 cells. The percentage of mice with tumors at sacrifice was plotted. B, Experimental tumors were performed in Balb-C mice with wild-type (WT), or invalidated RENCA cells (vegfc-/-, 10 mice per conditions) by subcutaneous injection of 0.1x106 cells. The percentage of mice with tumors twenty days after injection was plotted. C, Experimental tumors were performed in wild-type (WT), or eya2-invalidated mice (8 mice per conditions) by subcutaneous injection of 0.5x106 RENCA cells. The percentage of mice with tumors following fifteens days (15 D) or eighteen days (18 D) post injection sacrifice was plotted.!! ! ! (early stage) (Advanced stage)

VEGFC Protein Tumor cells

Microenvironment POST TRANSCRIPTION Normoxia Hypoxia ?

AAAAAAAA VEGFC mRNA

TRANSCRIPTION HIF2α Normoxia Hypoxia

NFkB

VEGFC

Figure S6: Ndiaye et al ! Supplementary Figure S6: Recapitulative schema showing the complex regulation of VEGF-C in normoxia and hypoxia and the correlation between VEGF-C levels and tumor aggressiveness! CONCLUSION(!

" " " " " " " " " " " " " " " " " " " " " " " " " " " " "

74" " Le"cancer"du"rein"a"une"incidence"assez"faible"(3%)"par"rapport"au"cancer"du"poumon" (24%)" par" exemple." Cependant," pour" les" cancers" du" rein" métastatiques" comme" pour" tous"les"cancers"métastatiques,"malgré"les"efforts"fournis"pour"trouver"des"traitements" de"plus"en"plus"efficaces,"les"patients"guéris"sont"extrêmement"rares"voire"inexistants" (Hsieh" et" al." 2017)." " Ils" récidivent" presque" systématiquement" et" développent" des" résistances" qui" vont" mener" au" décès." Ainsi" avec" tous" les" membres" de" l’équipe" du" Dr" Pagès"nous"oeuvrons"afin"d’améliorer"les"connaissances"dans"le"domaine"du"cancer"du" rein" et" les" mécanismes" de" résistance" aux" traitements." La" connaissance" de" ces" mécanismes"permettra"de"mieux"lutter"contre"ce"type"de"cancers."" Les"RCCs"constituent"le"type"de"cancer"du"rein"le"plus"représenté.""Les"RCCs"peuvent"être" subdivisés"en"4"groupes"selon"le"fond"génétique"des"cellules"tumorales."Mais"les"ccRRCs" représentent"environ"75%"des"RCCs""et"constituent"le"type"de"RCC"sur"lequel"j’ai"effectué" mes" recherches." Les" ccRCCs" se" caractérisent" le" plus" souvent" par" l’absence" de" l’expression"de"pVHL"(Hakimi,"Pham,"and"Hsieh"2013)."Cette"protéine"est"la"principale" régulatrice" " des" protéines" HIFα" dont" elle" provoque" la" dégradation" par" le" protéasome" (Semenza"2013)."Les"HIFs"sont"des"protéines"majeures"impliquées"dans"la"réponse"aux" stress" hypoxique." Lorsque" la" cellule" est" en" condition" hypoxique," l’expression" de" HIF" permet"l’induction"de"VEGFGA"qui"est"un"des"facteurs"de"croissance"majeurs"impliqués" dans"les"processus"de"néoGvascularisation."Dans"les"ccRCCs,"l’absence"de"pVHL"entraine" une" vascularisation" tumorale" importante" au" sein" de" la" tumeur" via" l’axe" HIF/VEGFGA." Ainsi," devant" le" manque" d’efficacité" des" traitements" antérieurs" et" un" réel" besoin" d’amélioration" de" la" prise" en" charge" des" ccRRCs," des" molécules" bloquant" la" vascularisation"tumorale"ont"été"développées":"les"AAGs."Le"principe"étant"de"détruire"le" réseau"vasculaire"de"la"tumeur"et"ainsi"empêcher"l’apport"en"nutriment"et"oxygène"vers" les"cellules"tumorales."Les"cellules"tumorales"ont"une"grande"capacité"de"prolifération." De" ce" fait," elles" ont" besoin" de" beaucoup" d’énergie" dont" la" source" est" transportée" essentiellement" " par" les" vaisseaux" sanguins." Ainsi" en" inhibant" la" formation" de" ces" vaisseaux,"la"tumeur"aurait"dû"être"éradiquée"par"manque"de"nutriment"et"d’oxygène." Malheureusement" le" traitement" des" ccRCC" par" un" AAG" entrainent" dans" un" premier" temps"un"réduction"de"taille"de"la"tumeur"comme"attendu"mais"à"terme"tous"les"patients" traités" rechutent." Dans" ce" contexte," il" était" très" important" de":" i)" découvrir" des" marqueurs" prédisant" la" réponse" des" patients" au" AAGs";" ii)" de" comprendre" les" mécanismes" de" résistance" afin" de" mettre" en" évidence" de" nouvelles" cibles"

75" " thérapeutiques."Durant"ma"thèse,"j’ai"donc"participé"à"plusieurs"projets"visant"à"apporter" des"solutions"à"ces"problèmes."J’ai"donc"participé"à"la"mise"en"évidence"du"CXCL7"comme" marqueur"prédictif"de"réponse"au"sunitinib"dans"les"RCCs."" Le" CXCL7" fait" partie" de" la" famille" des" chimiokines" CXCL" +ELR." Elle" possède" des" propriétés" chimioGattractante" et" proGangiogénique" faisant" d’elle" une" molécule" importante"dans"l’inflammation."Les"cibles"de"CXCL7"sont"donc"les"cellules"endothéliales" (angiogenèse)"et"les"cellules"immunitaires"(granulocytes,"neutrophiles)"qui"expriment"à" leur" surface" ses" récepteurs" CXCR1" et/ou" CXCR2" (von" Hundelshausen," Petersen," and" Brandt" 2007)." Dans" ce" premier" article," les" résultats" ont" montré" que" CXCL7" est" un" marqueur" prédictif" de" réponse" au" sunitinib" chez" les" patients" avec" des" ccRCC" métastatiques." Les" patients" qui" présentent" un" taux" de" CXCL7" plasmatique" " inférieur" à" 250ng/mL"ont"une"survie"sans"progression"(PFS)"plus"courte"(12,6"mois)"par"rapport" aux" patients" avec" un" taux" de" CXCL7" supérieur" à" 250ng/mL." Ces" derniers" patients"" présentent" une" PFS" de" 27,7" mois." La" plus" grande" difficulté" dans" cette" étude" a" été" d’identifier" le" seuil" de" concentration" plasmatique" (250ng/mL)" permettant" de" déterminer"les"patients"en"échec"thérapeutique."Pour"y"parvenir,"nous"avons"alors"utilisé" des"analyses"statistiques"en"collaboration"étroite"avec"le"département"de"statistiques"du" Centre" Antoine" Lacassagne." Ces" résultats" ont" été" obtenus" à" partir" d’une" étude" prospective" portant" sur" 54" patients"" et" confirmés" par" une" étude" rétrospective" sur" 31" patients."Dans"ce"papier,"la"quantité"de"CXCL7"plasmatique"est"plus"élevée"chez"les"sujets" sains"que"chez"les"patients"atteints"de"ccRCC."En"plus"lorsque"les"patients"sont"traités" avec"du"sunitinib"le"taux"de"CXCL7"a"tendance"à"revenir"à"la"normale"chez"les"patients" qui" répondent" au" traitement" par" rapport" aux" patients" qui" récidivent" dont" le" taux" de" CXCL7"plasmatique"reste"bas."Cependant,"au"sein"de"la"tumeur,"la"quantité"de"CXCL7"est" inversement" corrélée" au" taux" plasmatique." Des" taux" intraGtumoraux" élevés" de" CXCL7" sont" associés" à" l’agressivité" tumorale" (Grépin," Guyot," et" al." 2014)." Ce" résultat" peut" paraitre" contradictoire." Cependant," la" source" physiologique" de" CXCL7" dont" l’autre" appellation" est" le" «"Pro" Platelet" Basic" Protein"»" reste" les" plaquettes" et" aussi" certaines" cellules"immunitaires"comme"les"macrophages."Ainsi,"lors"du"développement"tumoral," un"site"inflammatoire"se"crée"et"les"cellules"de"l’inflammation"sont"recrutées"au"niveau" de" ce" site." De" plus,""les" macrophages," spécifiquement" les" M2" induisent" la" tolérance" immunitaire"et"supportent"la"progression"tumorale."Le"recrutement"de"ces"cellules"de" l’inflammation"est"donc"associé"à"un"mauvais"pronostic."Dans"ce"papier,"CXCL7"a"aussi"

76" " été"testé"comme"marqueur"prédictif"de"réponse"à"un"autre"AAG,"le"bevacizumab,"mais" aucune"corrélation"n’a" été"mise"en"évidence"entre"le"taux"plasmatique"de"CXCL7"et"la" PFS" ou" même" la" survie" globale" (OS)." L’identification" de" ce" marqueur" apporte" des" éléments" thérapeutiques" importants" dans" la" lutte" contre" les" ccRCCs." La" possibilité" de" discriminer"les"patients"qui"vont"répondre"au"sunitinib"par"rapport"aux"nonGrépondeurs" améliorerait"la"qualité"de"vie"des"patients"en"évitant"un"traitement"inutile"dont"les"effets" secondaires"sont"péjoratifs"pour"les"patients"(hypertension,"syndrome"main"pied)."Les" économies"des"dépenses"de"santé"doivent"également"être"prises"en"compte."Les"patients" en" rechute" pourront" être" orientés" vers" d’autres" traitements" qui" sont" actuellement" administrés"en"deuxième"ou"troisième"ligne"lorsque"le"sunitinib"est"inefficace."" La"résistance"de"ccRCC"aux"traitements"est"une"autre"question"adressée"dans"l’équipe." Les"résistances"aux"AAGs"peuvent"être"issues"des"mécanismes"suivants":"les"résistances" intrinsèques"correspondant"lorsque"la"tumeur"ne"répond"pas"du"tout"au"traitement"par" exemple"en"séquestrant"l’agent"thérapeutique"dans"les"lysosomes"(Giuliano"et"al."2015)";" les" résistances" acquises" avec" la" compensation" par" d’autres" voies" de" signalisation" (Sennino"et"al."2012)"notamment"la"voie"du"récepteur"de"l’EGF"(Grépin"et"al"2012);"enfin" l’induction" de" facteur" de" croissance" comme" le" VEGFGC" qui" est" traitée" dans" le" second" article"présenté"dans"ma"thèse."Cet"article"a"démontré"que"le"VEGFGC"dont"l’expression" est"associée"à"la"lymphangiogenèse"(Joukov"et"al."1996),"est"surexprimé"dans"les"ccRCCs" après" traitement" avec" le" sunitinib." L’induction" du" VEGFGC" par" le" sunitinib" est" dépendante""de"la"MAP"Kinase"p38"et"de"la"balance"entre"deux"régulateurs"du"temps"de" demiGvie"de"l’ARNm"VEGFGC,"la"tristetraproline"(TTP,"régulateur"négatif)"et"Hu"antigen"R" (HuR,"régulateur"positif)."Le"sunitinib"active"p38"qui"phosphoryle"TTP"et"HuR"entrainant" la" désactivation" et" l’activation" respective" de" ces" deux" protéines" exprimées" dans" le" cytoplasme"des"cellules"de"ccRCC."HuR"activé"par"p38"se"fixe"sur"le"3’UTR"de"l’ARNm"du" VEGFGC"et"le"stabilise."TTP"phosphorylé"peut"se"fixer"à"l’ARNm"VEGFGC"mais"ne"peut"plus" recruter"la"machinerie"de"dégradation"de"l’ARNm."De"plus,"le"sunitinib"stimule"l’activité" du"promoteur"de"vegfOc."Ces"deux"mécanismes"combinés"expliquent"l’induction"du"VEGFG C" après" traitement" par" le" sunitinib" des" ccRCCs." Ces" résultats" obtenus" sur" des" lignées" cellulaires"ont"été"confirmés"sur"des"lignées"primaires"et"in&vivo."Les"cellules"tumorales" traitées"par"le"sunitinib"sont"en"condition"de"stress"très"sévère"qui,"à"long"terme,"peut" induire"leur"mort."Ainsi,"elles"ont"le"choix"entre"s’adapter"à"ce"stress"ou"s’échapper"de"cet" environnement" délétère." Dans" ce" cas," les" cellules" de" ccRCC" font" appel" a" plusieurs"

77" " stratégies" en" changeant" le" profil" d’expression" de" leurs" gènes" afin" d’aboutir" à" la" surexpression"du"VEGFGC"et"l’induction"d’un"réseau"lymphatique"qui"constitue"une"voie" de" dissémination" métastatique" (He" et" al." 2005)." Loin" de" douter" de" l’efficacité" du" sunitinib"qui"a"révolutionné"la"prise"en"charge"des"patients,"la"mise"en"évidence"du"rôle" du" VEGFGC" suggérerait" qu’un" ciblage" concomitant" avec" les" AAGs" comme" le" sunitinib" pourrait"s’avérer"séduisant"en"thérapie."Intellectuellement"parlant,"cette"étude"met"aussi" en"évidence"un"mécanisme"de"régulation"subtil"par"le"sunitinib"impliquant"la"protéine"de" stress"p38"et"une"régulation"transcriptionnelle"et"postGtranscriptionnelle"impliquant"la" balance"TTP/HuR,"deux"acteurs"connus"pour"leur"propriété"de"suppresseur"de"tumeur" ou"d’oncogène"respectivement."Ces"résultats"montrent"l’importance"de"l’expression"de" VEGFGC"dans"les"récidives"des"ccRCCs"et"prouvent" que"l’étude"de"la"régulation"de"son" expression" mérite" des" investigations" plus" approfondies." Ces" études" ont" été" réalisées" dans" le" troisième" article" dans" lequel" la" régulation" du" VEGFGC" par" l’hypoxie" dans" les" ccRCCs" a" été" investiguée." Ces" tumeurs" qui" ont" la" particularité" d’exprimer" de" façon" constitutive"la"protéine"HIFα"à"cause"de"l’absence"de"VHL"ont"un"rapport"particulier"à" l’hypoxie."En"effet,"cette"expression"de"HIFα"entraine"une"surexpression"de"VEGFGA"qui" est"le"facteur"de"croissance"inducteur"principal"de"l’angiogenèse"provoquant"une"hyper" vascularisation"des"ccRCC."Les"AAGs,"ont"donc"été"développés"pour"traiter"ces"tumeurs" dont" la" croissance" semblait" dépendante" de" l’angiogenèse." Evidemment," les" AAGs," en" détruisant"les"vaisseaux"sanguins,"induisent"de"l’hypoxie"dans"l’environnement"tumoral." Si"la"régulation"par"l’hypoxie"du"VEGFGA"a"été"bien"étudiée,"la"régulation"du"VEGFGC"par" l’hypoxie"reste"mal"connue."Ainsi,"comme"certains"AAGs"induisent"l’expression"du"VEGFG C"et"que"l’hypoxie"résulte"du"blocage"de"l’angiogenèse,"j’ai"voulu"déterminer"si"l’hypoxie" consécutive"au"traitement"dans"les"ccRCCs"pouvait"être"la"cause"de"la"surexpression"du" VEGFGC.""Lorsque"des"cellules"de"ccRCC"sont"placées"en"condition"hypoxique,"elles"surG expriment" la" protéine" VEGFGC." Cependant," de" façon" surprenante,"l’augmentation"de"la" production"de"VEGFGC"en"hypoxie"est"accompagnée"d’une"diminution"de"la"quantité"de" son"ARNm."Ce"résultat"est"en"accord"avec"les"résultats"obtenus"dans"différents"cancers" avec"une"baisse"de"l’ARNm"et"une"augmentation"de"la"protéine"en"condition"hypoxique" (Morfoisse" et" al." 2014)." Dans" cette" étude," l’expression" de" VEGFGC" en" hypoxie" est" indépendante"de"HIF1α."Mon"travail"montre"qu’HIF2α"est"impliqué"et"est"déterminant" pour"l’expression"de"VEGFGC"dans"les"cellules"de"ccRCC."Dans"ces"cellules,"les"facteurs"de" transcription" NFκB" et" HIF2α" sont" les" acteurs" majeurs" de" l‘expression" de" VEGFGC" au"

78" " niveau"transcriptionnel"et"postGtranscriptionnel"(figure"18)."Le"rôle"du"VEGFGC"dans"la" tumorigenèse"des"ccRCC"a"été"testé"en"générant"des"cellules"tumorales"de"ccRCC"humain" ou" de" souris" invalidées" pour" le" vegfOc" par" la" technologie" des" CRISPRGCas9." Cet" outil" a" permis"de"montrer"que"le"VEGFGC"est"une"«"arme"à"double"tranchant"»"dans"le"ccRCC"qui" possède" des" spécificités" immunologiques" différentes" d’autres" cancers" (Şenbabaoğlu" et" al."2016)."Le"VEGFGC"peut"être"bénéfique"ou"péjoratif"selon"le"stade"précoce"ou"avancé"de" la" tumeur." Ces" résultats" viennent" attirer" l’attention" sur" les" précautions" qu’il" faudra" prendre"quant"au"développement"de"potentielles"thérapies"antiGVEGFGC."" "

" Figure!18!:!Schéma!récapitulatif!!de!la!régulation!du!VEGFBC!par!l’hypoxie!et!ses! différentes!selon!l’étape!durant!la!tumorigenèse!des!ccRCC! " !

79" " DISCUSSION(!

80" " Les" résultats" obtenus" au" cours" de" cette" thèse" viennent" indéniablement" apporter" une" pierre"à"l’édifice"de"la"compréhension"de"la"biologie"des"ccRCC."Cependant,"il"est"toujours" possible"de"l‘améliorer."Il"est"donc"important"de"pouvoir"porter"un"regard"critique"sur"ce" travail."Cela"doit"être"à"mon"avis"une"des"qualités"principales"de"tout"chercheur."

Pour"commencer"avec!le!premier!article,"un"nombre"plus"important"de"patients"en"plus" de"ceux"décrits"dans"les"essais"cliniques"SUVEGIL"et"TORAVA"(Article"1)"donnerait"un" plus"gros"impact"à"ce"travail."De"plus,"il"s’agit"de"deux"études"prospectives."Les"patients" ont"été"recrutés"pour"déterminer"s’il"existe"un"lien"entre"l’efficacité"du"traitement"par" sunitinib"et"le"développement"de"biomarqueurs"sanguins"pour"SUVEGIL."Le"nombre"de" patients"était"déjà"déterminé"lorsqu’on"a"commencé"notre"étude."Cependant,"notre"étude" a"le"mérite"d’avoir"mis"en"place"un"essai"clinique"multicentrique"démontrant"que"CXCL7" pouvait" être" un" marqueur" prédictif" de" l’efficacité" du" sunitinib." Ces" résultats" ont" été" confirmés" dans" une" cohorte" rétrospective" indépendante." Le" CXCL7" fait" partie" d’une" famille" de" facteurs" inflammatoires" exprimés" lors" du" développement" tumoral." L’expression" de" ces" facteurs" est" corrélée" avec" différentes" caractéristiques" cliniques." Ainsi," les" cytokines" CXCL7," l’interleukineG25" (IL25)" et" le" «"Macrophage" migration" Inhibitory" Factor" (MIF)"»" ont" été" ensuite" mises" en" évidence" comme" marqueurs" pronostiques"et"prédictifs""d’une"évolution"péjorative"dans"les"cancers"du"foie""(Guo,"Ru," et"al."2017)."Auparavant,"le"CXCL7"avait"déjà"été"décrit"comme"marqueur"pronostique" dans"les"syndromes"myélodysplasiques"(Aivado"et"al."2007)."Tous"ces"articles"montrent" que"CXCL7"a"un"fort"potentiel"en"tant"que"marqueur"biologique."Sa"détection"est"facile" par" un" simple" test" " ELISA" sur" un" échantillon" sanguin" de" quelques" microlitres." Cette" technique" non" invasive" pourrait" être" facilement" introduite" en" pratiques" hospitalières." Nos"résultats"sur"CXCL7"viennent"s’inscrire"dans"le"processus"de"mise"en"place"de"tests" compagnons"associés"à"l’utilisation"de"médicaments."D’ailleurs"une"étude"de"plus"grande" envergure" est" en" cours" de" réalisation" au" Centre" Antoine" Lacassagne" qui" abrite" nos" locaux"sur"les"cytokines"CXC"ELR+"et"leur"potentiel"rôle"comme"marqueur"prédictif"de" réponse"au"sunitinib"des"cancers"du"rein"métastatiques."

Dans! le! deuxième! article," nous" avons" identifié" une" voie" menant" aux" récidives" des" ccRCCs" sous" sunitinib" et" conduisant" à" des" tumeurs" plus" agressives." Le" sunitinib," le" traitement"de"référence"de"première"ligne"peut"induire"le"développement"de"vaisseaux" lymphatiques."Evidemment"ces"résultats"ne"remettent"pas"en"cause"l’effet"thérapeutique"

81" " du"sunitinib"mais"suggèrent"que"dans"certaines"récidives,"le"développement"du"réseau" lymphatique"peut"constituer"une"voie"de"dissémination"des"cellules"tumorales"et"donc" l’apparition" de" nouveaux" foyers" tumoraux." L’arrivée" du" sunitinib" dans" l’arsenal" des" traitements" AAG" a" prolongé" la" survie" des" patients" atteints" de" ccRCC" métastatiques." Même"si"les"effets"secondaires"peuvent"être"sévères"entrainant"l’arrêt"du"traitement"et" que"certains"patients"sont"insensibles,"de"très"longues"PFS"(>"50"mois)"ont"été"observées" (Tannir" et" al." 2017," Gore" et" al." 2015)." AuGdelà" des" effets" directs" du" sunitinib" sur" le" patient," la" recherche" permettrait" d’améliorer" son" mode" d’administration" afin" d’éviter" certains" effets" secondaires" et" des" marqueurs" prédictifs" permettant" de" mieux" sélectionner"les"patients"répondeurs"ont"été"mis"en"évidence."Tout"ceci"a"contribué"à"une" meilleure" compréhension" des" ccRCCs." Nos" résultats" montrent" que" le" VEGFGC" est" un" marqueur"d’agressivité"dans"les"ccRCCs"et"cela"est"en"accord"avec"une"grande"partie"de" la" littérature." Des" études" utilisant" différentes" méthodes" ont" permis" de" comprendre" le" rôle" du" VEGFGC" dans" la" tumorigenèse." En" surGexprimant" du" VEGFGC" dans" les" cellules" cancéreuses"injectées"à"des"souris"ou"en"développant"des"souris"transgéniques,"le"VEGFG C"contribue"au"développement"de"la"lymphangiogenèse"et"la"croissance"tumorale"(Skobe" et" al." 2001)." Les" tumeurs" exprimant" VEGFGC" présentent" plus" de" métastases" lymphatiques" et" pulmonaires" alors" que" les" tumeurs" n’exprimant" pas" VEGFGC" n’en" présentent" pas." Des" résultats" similaires" ont" été" décrits" dans" les" cancers" de" cellules" β" pancréatiques" (Mandriota" et" al." 2001)" démontrant" définitivement" le" rôle" du" VEGFGC" dans" la" lymphangiogenèse" et" les" métastases" tumorales." Cependant," le" fait" que" le" sunitinib"puisse"induire"l’expression"de"VEGFGC"est"un"résultat"totalement"contre"intuitif" sachant" que" le" sunitinib" peut" entrainer" l’inhibition" de" la" fonction" des" cellules" endothéliales"lymphatiques"et"les"métastases"lymphatiques"dans"un"model"de"cancer"du" sein"(Kodera"et"al."2011)."Un"article"publié"en"2012"par"Shojaei"et&al"a"réconcilié"ces"deux" observations" qui" semblent" contradictoires." Dans" cet" article," les" auteurs" étudient" les" voies" de" résistance" au" sunitinib" de" lignées" cellulaires" de" plusieurs" cancers." Leur" conclusion" est" que" l’induction" des" métastases" lymphatiques" par" le" sunitinib" est" dépendante"du"type"cellulaire"(Shojaei"et"al."2012)."En"effet"parmi"les"lignées"testées,"les" 4T1"(cancer"du"sein)"et"les"colo205"(cancer"colorectal)"présentent"plus"de"métastases" après" traitement" avec" le" sunitinib" alors" que" cet" effet" est" absent" sur" la" lignée" H460" (cancer" du" poumon)." Cet" article" indique" l’induction" métastatique" induite" suite" au" traitement" par" sunitinib" se" fait" via" l’axe" HGF/cGMet." De" manière" plus" intéressante," les"

82" " expressions"de"HGFα,"de"cGMET,"de"VEGFGC"sont"corrélées"entre"elles"et"à"l’induction"de" la"lymphangiogenèse"dans"un"modèle"d’inflammation"de"la"cornée"(Cao"et"al."2006)."AuG delà"de"l’action"du"sunitinib"sur"la"lymphangiogenèse,"nous"avons"mis"en"évidence"un" mécanisme"de"régulation"de"VEGFGC"impliquant"p38,"TTP"et"HuR"dans"les"ccRCCs."HuR" fait" partie" de" la" famille" des" protéines" «"Embryonic" Lethal" Abnormal" Vision"»" (ELAV)" connues" pour" leur" fonction" dans" la" régulation" de" la" stabilité" des" ARNm." Les" ARNm" codant"VEGFGA"et"COXG2"sont"stabilisés"par"HuR"en"se"fixant"sur"leurs"domaines"riches" en"AU"(Kurosu"et"al."2011)."Cependant,"la"stabilisation"de"l’ARNm"de"VEGFGC"par"HuR" par"un"mécanisme"impliquant"p38"n’a"jamais"été"décrite"auparavant."Seul"un"article"de" 2005" avait" décrit" la" stabilisation" de" l’ARNm" de" VEGFGC" par" HuR" dans" les" cancers" du" poumon"non"à"petites"cellules"(Wang"et"al."2009)"mais"il"est"nullement"mentionné"p38" dans"cet"article."Notre"article"décrit"ainsi"un"mécanisme"inédit"de"régulation"du"VEGFGC" dans" les" ccRCCs" après" traitement" avec" le" sunitinib." L’utilisation" des" AAGs" contre" les" cancers"a"permis"aux"patients"de"gagner"quelques"mois"de"PFS."Cependant"des"cas"de" rémissions" longues" après" un" traitement" AAG" sont" très" rares." Afin" d’obtenir" des" effets" curatifs,"ne"faudraitGil"pas"changer"de"stratégie"?"Les"AAGs"entrainent"une"diminution"de" la" perfusion" des" tumeurs," donc" de" l’hypoxie" et" l’acidification" de" l’environnement" tumoral."Ces"deux"phénomènes"entrainent"une"inhibition"du"système"immunitaire"par" l’inhibition"de""capacité"des"cellules"dendritiques"à"présenter"les"antigènes"tumoraux"aux" lymphocytes"T"et"donc"par"la"même"occasion"les"lymphocytes"T"(Barsoum"et"al."2014," Calcinotto"et"al."2012)."De"plus,"l’hypoxie"entraine"l’expression"de"PDLG1"et"donc"induire" une"immunotolérance"et"un"échappement"tumoral"(Noman"et"al."2014)."Non"seulement" l’hypoxie"peut"entrainer"une"immunotolérance"mais"en"plus"elle"permet"de"sélectionner" les" cellules" tumorales" les" plus" agressives." Les" cellules" répondant" aux" signaux" physiologiques" entrent" en" apoptose" sous" hypoxie" (Wilson" and" Hay" 2011)." De" plus," l’hypoxie"augmente""les"capacités"migratoires"des"cellules"en"provoquant"l’expression"de" protéines"proGmigratoires"comme"le"VEGF"et"le"SDF1α"(Finger"and"Giaccia"2010)."Enfin," l’hypoxie"constitue"une"niche"pour"les"cellules"souches"cancéreuses"leur"conférant"une" résistance" aux" thérapies" (figure" 19)." Toutes" ces" observations" font" que" l’hypoxie" et" l’acidité"intratumorale"est"fortement"corrélée"à"un"mauvais"pronostic"(Wilson"and"Hay" 2011,"Brand"et"al."2016)."

83" " "

Figure! 19!:! les! conséquences! de! l’hypoxie! et! de! l’acidité! intraBtumorale! dans! le! développement!de!la!tumeur!et!sa!réponse!aux!thérapies.!!

Pour"aller"plus"loin,"nous"avons"cherché"à"savoir"si"le"VEGFGC"était"régulé"par"l’hypoxie." Le"principal"but"des"traitements"AAG"est"de"bloquer"la"néoGvascularisation"tumorale"ce" qui"a"pour"conséquence"logique"une"hypoxie"intra"tumorale."Nous"nous"sommes"alors" posé"la"question"à"savoir"si"l’expression"de"VEGFGC"observée"après"traitement"avec"le" bevacizumab"(Grepin"et"al."2012)"est"due"à"l’AAG"luiGmême"ou"à"l’hypoxie"induite"par"le" traitement." Le" dernier" article" démontre" que" l’expression" de" VEGFGC" est" effectivement" régulée" par" l’hypoxie" via" un" processus" impliquant" HIF2α" et" NFκB." Ces" deux" facteurs" régulent" l’expression" de" VEGFGC" au" niveau" transcriptionel" et" postGtranscriptionel." Cependant,"le"mécanisme"par"lequel"HIF2α"régule"la"stabilité"de"l’ARNm"de"VEGFGC"n’est" pas" décrit." Nous" avons" regardé" les" protéines" qui" se" fixent" sur" l’ARNm" de" VEGFGC" en" normoxie" comme" en" hypoxie" mais" HIF2α" n’y" figure" pas." HIF2α" régulerait" donc" la" stabilité" de" l’ARNm" de" VEGFGC" de" manière" indirecte." HuR" et" TTP" sont" des" candidats" évidents"puisque"nous"avons"montré"leur"importance"dans""la"stabilité"de"l’ARNm"VEGFG C"après"traitement"au"sunitinib"(Dufies"et"al."2017)."L’activation/activité"de"HuR"et"de" TTP"n’est"pas"modifiée"par"hypoxie."La"stabilité"de"l’ARNm"VEGFGC"par"HIF2α"se"fait"de" manière"indirecte"via"une"protéine"non"encore"identifiée."Depuis"ces"dernières"années," la" régulation" du" VEGFGC" a" fait" l’objet" de" plusieurs" études." L’accumulation" de" ces" publications"démontre"un"intérêt"grandissant"pour"le"système"lymphatique"et"son"rôle"

84" " physiologique"et"pathologique,"dans"le"cancer."Ainsi,"plusieurs"facteurs"de"transcriptions" ont"été"décrits"pour"réguler"l’expression"de"VEGFGC,"notamment"STAT3"dans"les"cancer" du" côlon" (Zhu" et" al." 2016)," SIX1" dans" le" cancer" du" sein" (Wang" et" al." 2012)" ou" encore" NFκB" dans" les" cancers" de" la" vésicule" biliaire" (Du" et" al." 2014)." Par" ailleurs," le" rôle" des" HIFs" dans" la" régulation" de" VEGFGC" n’est" pas" bien" connu." En" effet," la" régulation" traductionnelle" de" VEGFGC" a" été" décrite" comme" étant" indépendante" de" HIF1α" dans" plusieurs" cancers" métastatiques" (Morfoisse" et" al." 2014)." Mon" étude" est" donc" une" des" premières"à"décortiquer"précisément"le"mécanisme"de"régulation"du"VEGFGC"en"hypoxie." Dans" notre" article," nous" mettons" aussi" en" évidence" le" rôle" du" VEGFGC" dans" la" tumorigenèse" des" ccRCCs." Lorsque" le" vegfOc& est" inactivé," les" cellules" tumorales" deviennent"plus"agressives"in&vitro."Une"surGactivation"de"la"voie"Akt"sur"les"KO"VEGFGC"a" été" observée." La" voie" Akt" est" associée" à" l’agressivité" tumorale" se" traduisant" par" l’induction" d’un" état" souche" et" de" l’EMT" dans" les" RCC" (Lin" et" al." 2015)." La" phosphorylation"d’Akt"stimule"la"prolifération"dont"le"niveau"est"augmenté"dans"les"KO" VEGFGC."Les"marqueurs"souches"sont"associés"à"une"croissance"tumorale"in&vivo."Malgré" leurs"marqueurs"d’agressivité"les"cellules"VEGFGC"KO"ne"forment"des"tumeurs"qu’après" un" long" délai" par" rapport" aux" cellules" sauvages" quand" elles" sont" injectées" dans" des" souris"nude."Notre"hypothèse"stipule"que"le"VEGFGC"joue"un"rôle"dans"la"mise"en"place" d’une" niche" adéquate" qui" permet" aux" cellules" tumorales" de" se" développer." Dans" les" souris"nude,"aucun"autre"type"cellulaire"ne"compense"ce"manque."Cependant,"dans"les" souris"immunocompétentes"il"existe"certainement"des"cellules"capables"de"compenser" ce"manque"afin"de"permettre"une"mise"en"place"tumorale"assez"rapide"comme"le"montre" nos"résultats"avec"les"cellules"syngéniques"RENCA."Par"ailleurs,"les"CSCs"ont"capacité""de" rentrer"en"dormance."Elle"peuvent"alors"se"réactiver"à"n’importe"quel"moment"(Chang" 2016)" comme" chez" les" patients" qui" récidivent" après" une" longue" période" de" latence." L’invalidation" du" VEGFGC" dans" les" cellules" cancéreuses" est" surtout" associée" à" une" réduction"de"la"croissance"tumorale,"de"la"lymphangiogenèse"ainsi"que"des"métastases" (Ye" et" al." 2013," Decio" et" al." 2014)." Nos" résultats" mettent" en" évidence" un" important" problème" lié" au" choix" du" modèle" d’étude." Les" résultats" obtenus" sur" les" cellules" 786G0" vegfOc"KO"suggèrent"une"agressivité"accrue"de"ces"cellules."Cependant,"elles"ne"forment" pas"de"tumeurs"in&vivo."C’est"un"cas"typique"qui"montre"la"nécessité"d’avoir"des"modèles" pertinents"in&vivo"permettant"de"générer"des"résultats"cohérents"entre"études"in&vitro"et" in& vivo& (Delfortrie& et& al.& 2011)." Dans" cet" article" nous" utilisons" aussi" deux" lignées"

85" " différentes" pour" les" expériences" in& vivo." Ces" lignées" de" RCC" ont" un" fond" génétique" différent"et"n’expriment"pas"les"mêmes"protéines."Les"RENCAs"utilisées"pour"injecter"les" souris"BALBGC"sont"des"cellules"murines"alors"que"les"786G0"sont"des"cellules"humaines." De"plus,"les"RENCA"expriment"VHL"alors"que"les"786G0"sont"mutées"VHL."Les"résultats" obtenus" à" partir" de" ces" deux" lignées" sont" donc" difficilement" comparables." Des" souris" humanisées"auraient"été"un"modèle"plus"adéquat"pour"l’injection"de"cellules"humaines." Dans" l’idéal," il" faudrait" prendre" la" même" souche" de" souris" (BALBCs" qui" sont" immunocompétentes),"détruire"leur"système"immunitaire"pour"leur"injecter"les"cellules" RENCA"avec"ou"sans"VEGFGC."Une"expérience"visant"en"injecter"les"cellules"RENCA"WT"et" VEGFGC"KO"dans"des"souris"nude"est"en"cours"de"préparation."Cette"expérience"viendra" compléter"les"résultats"présentés"ici."Ainsi"on"pourra"comparer"les"résultats"obtenus"lors" de" cette" expérience" avec" celles" obtenus" sur" les" souris" immunocompétentes." Nous" pourrons" alors" tirer" des" conclusions" définitives" et" plus" solides" quant" à" la" fonction" du" système" immunitaire" dans" la" tumorigenèse" des" ccRCC" en" fonction" de" la" présence" de" VEGFGC.""

Un" autre" point" important" de" notre" étude" est" que" lors" de" nos" expériences" in& vivo," les" cellules"tumorales"ont"été"injectées"en"sous"cutané."Evidemment,"il"serait"préférable"de" réaliser"des"injections"orthotopiques"suivies"d’une"chirurgie"ablative"de"la"tumeur"serait" un" modèle" plus" pertinent" à" mettre" en" place" au" laboratoire." Le" développement" d’un" modèle" murin" de" cancer" du" rein" qui" se" développe" spontanément" ferait" avancer" la" recherche"dans"ce"domaine."vhl"semblait"être"un"candidat"idéal"à"muter"pour"les"ccRCCs" vu"que"son"inactivation"caractérise"les"ccRCC."Cependant"cette"inactivation"n’induit"que" très"rarement"des"tumeurs"rénales"chez"la"souris."Dans"ce"contexte,"il"est"extrêmement" difficile" de" se" rapprocher" de" la" pathologie" humaine." Une" double" inactivation" vhl& p53" semble" être" plus" pertinente" (Nyhan," O'Sullivan," and" McKenna" 2008)." L’obtention" de" plusieurs"modèles"mimant"au"mieux"la"pathologie"humaine"permettra"la"caractérisation" d’outils"thérapeutiques"de"plus"en"plus"pertinents."

86" " PERSPECTIVES)"

87" " "

Les"résultats"obtenus"au" cours" de"cette" thèse"apportent"certain"éclaircissement"sur" la" régulation"du"VEGFGC"dans"les"ccRCCs"et"sur"les"RCCs"en"général"également."Cependant," ils" mettent" l’accent" sur" un" certain" nombre" d’inconnues" qui" démontrent" notre" méconnaissance" de" l’évolution" de" ces" cancers." La" fonction" du" VEGFGC" dans" la" tumorigenèse" et" son" rapport" avec" les" cellules" du" système" immunitaire" mériteraient" d’être"étudiée"plus"en"profondeur."L’implication"de"NFκB"dans"la"régulation"de"VEGFGC" indique" que" les" conditions" inflammatoires" qui" règnent" au" sein" de" la" tumeur" sont" fortement"liées"à"l’induction"de"VEGFGC"et"donc"à"la"lymphangiogenèse"et"à"la"réponse" immunitaire"tumorale."Mieux"comprendre"les"différentes"fonctions"du"VEGFGC"au"cours" du"développement"de"la"tumeur"serait"d’une"grande"utilité"dans"la"lutte"contre"le"cancer." Des"études"expliquant"la"différence"entre"le"phénotype"in&vivo"et"in&vitro&des"786G0"KO" VEGFGC"seraient"très"intéressantes"à"mener.""

Bien"que"l’induction"de"VEGFGC"ait"été"décrite"comme"indépendante"de"HIF1α"dans"les" cancers"du"sein,"je"pense"dans"les"ccRCCs,"il"serait"intéressant"de"voir"si"la"réexpression" de"HIF1α"dans"les"lignées"n’exprimant"qu’HIF2α"pourrait"changer"l’expression"de"VEGFG C.""Des"publications"décrivent"notamment"une"corrélation"entre"l’expression"d’HIF1α"et" VEGFGC."L’induction"de"VEGFGC"par"HIF1α"par"interaction"directe"avec"le"promoteur"de" vegfOc&a"été"décrite"dans"des"cellules"normales"(Guo,"Zhang,"et"al."2017)."Je"pense"que"la" compréhension"du"mode"d’action"des"HIFs"dans"les"cancers"est"un"enjeu"majeur"dans"la" lutte" contre" les" cancers," surtout" les" ccRCCs." En" effet," les" traitements" AAGs" contre" les" ccRCCs" induisent" indéniablement" une" hypoxie" tumorale." L’hypoxie" entraine" une" agressivité" des" cellules" tumorales" par" la" sélection" des" cellules" les" plus" agressives." L’hypoxie"entraine"également"la"dédifférenciation"des"cellules"tumorales"en"CSCs"qui"est" associée"à"l’EMT."L’EMT"est"un"processus"nécessaire"à"la"dissémination"métastatique,"un" autre" marqueur" d’agressivité." Les" principaux" acteurs" de" l’adaptation" des" cellules" à" l’hypoxie"sont"HIF1α"et"HIF2α."Les"deux"protéines"sont"alors"des"cibles"de"choix"pour" lutter" contre" le" cancer." Sachant" l’expression" HIF2α" est" associé" aux" formes" tumorales" plus" agressives," une" meilleure" compréhension" de" son" mode" d’action" pourrait" révolutionner" la" lutte" contre" les" ccRCC" en" particulier" et" les" cancers" en" général." " Le" ciblage"de"l’activité"des"HIFs"est"actuellement"en"plein"essor"(Guan"et"al."2010,"Sowter"et" al."2003)."

88" " En"comparant"la"normoxie"et"l’hypoxie"avec"nos"collaborateurs,"nous"avons"regardé"le" différentiel"des"protéines"qui"se"fixent"sur"l’ARNm"de"VEGFGC"dans"ces"deux"conditions." HIF2α," TTP" et" HuR" n’ont" pas" été" détectés."D’autres" protéines" de" liaison" à" l’ARNm" pourraient""être"l’élément"manquant"pour"expliquer"comment"HIF2α"régule"la"stabilité" de"l’ARNm"VEGFGC"en"hypoxie."L’identification"d’une"ou"plusieurs"protéines"dans"cette" liste"dont"l’expression"est"liée"à"HIF2α"permettrait"de"démontrer"de"façon"définitive"le" mécanisme"de"régulation"de"la"stabilité"de"l’ARNm"de"VEGFGC"par"HIF2α"en"hypoxie.""

Comme"discuter"ciGdessus,"je"pense"que"les"cellules,"RENCAs"VEGFGC"KO,"peuvent"être" utilisées"pour"comprendre"le"rôle"et"le"lien"entre"l’expression"de"VEGFGC"et"la"réponse" immunitaire"tumorale"lors""de"la"tumorigenèse"des"ccRCCs."Par"irradiation,"le"système" immunitaire"de"souris"BALBGC"peut"être"détruit."Ainsi,"les"cellules"RENCA"exprimant"ou" pas"le"VEGFGC"peuvent"être"injectées"aux"souris."Cette"étude"constitue"en"elleGmême"un" projet" que" l’équipe" développera." " Avec" le" retour" des" thérapies" visant" les" points" de" contrôle" immunitaires" contre" les" cancers" du" rein," un" tel" projet" prend" tout" son" importance"et"est"totalement"d’actualité.""

Maintenant"que"l’importance"du"VEGFGC"est"avérée,"le"développement"d’anticorps"antiG VEGFGC"est"une"seconde"étape."Au"regard"de"nos"résultats,"la"combinaison"d’AAG"avec"un" antiGVEGFGC" serait" envisageable." En" continuité" avec" mon" travail," ce" projet" pourrait" apporter" des" réponses" définitives" à" toutes" les" questions" soulevées" dans" cette" thèse." Ainsi,"la"toxicité"potentielle"et"l’efficacité"thérapeutique"d’une"telle"molécule"pourraient" être"testées"in&vivo.ar"La"balance"entre"une"efficacité"optimale"et"un"minimum"de"d’effet" secondaire"est"très"difficile"a"établir"en"tenant"compte"de"la"variabilité"interGpatients."

Les" résultats" générés" durant" ces" trois" années" de" thèse" mettent" surtout" en" valeur" l’importance" du" microenvironnement" durant" le" développement" tumoral." Pour" mieux" combattre" les" cancers," la" compréhension" du" rôle" de" ce" microenvironnement" est" indispensable." Mieux" appréhender" ses" effets," permettra" de" cibler" ses" composants" péjoratifs" et" donc" freiner" ses" effets" sur" la" croissance" tumorale" et" la" dissémination" métastatique.""" !

89" " REFERENCES'!

" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "

90" " Achen,"M."G.,"B."K."McColl,"and"S."A."Stacker."2005.""Focus"on"lymphangiogenesis"in"tumor" metastasis."""Cancer&Cell"7"(2):121G7."doi:"10.1016/j.ccr.2005.01.017."

Acker,"T.,"A."DiezGJuan,"J."Aragones,"M."Tjwa,"K."Brusselmans,"L."Moons,"D."Fukumura,"M." P." MorenoGMurciano," J." M." Herbert," A." Burger," J." Riedel," G." Elvert," I." Flamme," P." H." Maxwell,"D."Collen,"M."Dewerchin,"R."K."Jain,"K."H."Plate,"and"P."Carmeliet."2005.""Genetic" evidence" for" a" tumor" suppressor" role" of" HIFG2alpha."" " Cancer& Cell" 8" (2):131G41." doi:" 10.1016/j.ccr.2005.07.003."

Addison,"C."L.,"T."O."Daniel,"M."D."Burdick,"H."Liu,"J."E."Ehlert,"Y."Y."Xue,"L."Buechi,"A."Walz," A." Richmond," and" R." M." Strieter." 2000." "The" CXC" chemokine" receptor" 2," CXCR2," is" the" putative"receptor"for"ELR+"CXC"chemokineGinduced"angiogenic"activity."""J&Immunol"165" (9):5269G77."

Aivado,"M.,"D."Spentzos,"U."Germing,"G."Alterovitz,"X."Y."Meng,"F."Grall,"A."A."Giagounidis,"G." Klement," U." Steidl," H." H." Otu," A." Czibere," W." C." Prall,"C."IkingGKonert," M." Shayne," M." F." Ramoni,"N."Gattermann,"R."Haas,"C."S."Mitsiades,"E."T."Fung,"and"T."A."Libermann."2007." "Serum" proteome" profiling" detects" myelodysplastic" syndromes" and" identifies" CXC" chemokine"ligands"4"and"7"as"markers"for"advanced"disease."""Proc&Natl&Acad&Sci&U&S&A" 104"(4):1307G12."doi:"10.1073/pnas.0610330104."

AlGHajj," M.," M." S." Wicha," A." BenitoGHernandez," S." J." Morrison," and" M." F." Clarke." 2003." "Prospective"identification"of"tumorigenic"breast"cancer"cells."""Proc&Natl&Acad&Sci&U&S&A" 100"(7):3983G8."doi:"10.1073/pnas.0530291100."

Alitalo,"K."2011.""The"lymphatic"vasculature"in"disease."""Nat&Med"17"(11):1371G80."doi:" 10.1038/nm.2545."

Alongi," F.," S." Arcangeli," L." Triggiani," R." Mazzola," M." Buglione" di" Monale" E" Bastia," S." Fersino," A." Baiguini," B." A." JereczekGFossa," S." M." Magrini," and" on" the" behalf" of" Italian" Association"of"Radiation"Oncology"[AIRO]."2017.""Stereotactic"ablative"radiation"therapy" in" renal" cell" carcinoma:" From" oligometastatic" to" localized" disease."" " Crit& Rev& Oncol& Hematol"117:48G56."doi:"10.1016/j.critrevonc.2017.07.004."

Alt,"A."L.,"S."A."Boorjian,"C."M."Lohse,"B."A."Costello,"B."C."Leibovich,"and"M."L."Blute."2011." "Survival" after" complete" surgical" resection" of" multiple" metastases" from" renal" cell" carcinoma."""Cancer"117"(13):2873G82."doi:"10.1002/cncr.25836."

Arenberg," D." A.," M." P." Keane," B." DiGiovine," S." L." Kunkel," S." B." Morris," Y." Y." Xue," M." D." Burdick," M." C." Glass," M." D." Iannettoni," and" R." M." Strieter." 1998." "EpithelialGneutrophil" activating" peptide" (ENAG78)" is" an" important" angiogenic" factor" in" nonGsmall" cell" lung" cancer."""J&Clin&Invest"102"(3):465G72."doi:"10.1172/JCI3145."

Bakin," A." V.," A." K." Tomlinson," N." A." Bhowmick," H." L." Moses," and" C." L." Arteaga." 2000." "Phosphatidylinositol"3Gkinase"function"is"required"for"transforming"growth"factor"betaG mediated" epithelial" to" mesenchymal" transition" and" cell" migration."" " J& Biol& Chem" 275" (47):36803G10."doi:"10.1074/jbc.M005912200."

Banerji,"S.,"J."Ni,"S."X."Wang,"S."Clasper,"J."Su,"R."Tammi,"M."Jones,"and"D."G."Jackson."1999." "LYVEG1," a" new" homologue" of" the" CD44" glycoprotein," is" a" lymphGspecific" receptor" for" hyaluronan."""J&Cell&Biol"144"(4):789G801."

91" " Bao,"S.,"Q."Wu,"R."E."McLendon,"Y."Hao,"Q."Shi,"A."B." Hjelmeland," M." W." Dewhirst," D." D." Bigner,"and"J."N."Rich."2006.""Glioma"stem"cells"promote"radioresistance"by"preferential" activation" of" the" DNA" damage" response."" " Nature" 444" (7120):756G60." doi:" 10.1038/nature05236."

Barsoum,"I."B.,"C."A."Smallwood,"D."R."Siemens,"and"C."H."Graham."2014.""A"mechanism"of" hypoxiaGmediated" escape" from" adaptive" immunity" in" cancer" cells."" " Cancer& Res" 74" (3):665G74."doi:"10.1158/0008G5472.CANG13G0992."

Batlle," E.," E." Sancho," C." Francí," D." Domínguez," M." Monfar," J." Baulida," and" A." García" De" Herreros." 2000." "The" transcription" factor" snail" is" a" repressor" of" EGcadherin" gene" expression"in"epithelial"tumour"cells."""Nat&Cell&Biol"2"(2):84G9."doi:"10.1038/35000034."

Battelli,"C.,"and"D."C."Cho."2011.""mTOR"inhibitors"in"renal"cell"carcinoma."""Therapy"8" (4):359G367."doi:"10.2217/thy.11.32."

Beerepoot,"L."V.,"N."Mehra,"J."S."Vermaat,"B."A."Zonnenberg,"M."F."Gebbink,"and"E."E."Voest." 2004." "Increased" levels" of" viable" circulating" endothelial" cells" are" an" indicator" of" progressive"disease"in"cancer"patients."""Ann&Oncol"15"(1):139G45."

Bensalah," K.," L." Albiges," J." C." Bernhard," P." Bigot," T." Bodin," R." Boissier," J." M." Corréas," P." Gimel,"J."A."Long,"F."X."Nouhaud,"I."Ouzaïd,"P."Paparel,"N."RiouxGLeclercq,"and"A."Méjean." 2016." "[CCAFU" french" national" guidelines" 2016G2018" on" renal" cancer]."" " Prog&Urol" 27" Suppl"1:S27GS51."doi:"10.1016/S1166G7087(16)30702G3."

Bergers,"G.,"and"D."Hanahan."2008.""Modes"of"resistance"to"antiGangiogenic"therapy."""Nat& Rev&Cancer"8"(8):592G603."doi:"10.1038/nrc2442."

Bergers,"G.,"S."Song,"N."MeyerGMorse,"E."Bergsland,"and"D."Hanahan."2003.""Benefits"of" targeting" both" pericytes" and" endothelial" cells" in" the" tumor" vasculature" with" kinase" inhibitors."""J&Clin&Invest"111"(9):1287G95."doi:"10.1172/JCI17929."

Berra," E.," E." Benizri," A." Ginouvès," V." Volmat," D." Roux," and" J." Pouysségur." 2003." "HIF" prolylGhydroxylase" 2" is" the" key" oxygen" sensor" setting" low" steadyGstate" levels" of" HIFG 1alpha"in"normoxia."""EMBO&J"22"(16):4082G90."doi:"10.1093/emboj/cdg392."

Bertozzi,"C."C.,"A."A."Schmaier,"P."Mericko,"P."R."Hess,"Z."Zou,"M."Chen,"C."Y."Chen,"B."Xu,"M." M."Lu,"D."Zhou,"E."Sebzda,"M."T."Santore,"D."J."Merianos,"M."Stadtfeld,"A."W."Flake,"T."Graf," R." Skoda," J." S." Maltzman," G." A." Koretzky," and" M." L." Kahn." 2010." "Platelets" regulate" lymphatic"vascular"development"through"CLECG2GSLPG76"signaling."""Blood"116"(4):661G 70."doi:"10.1182/bloodG2010G02G270876."

Bhatt," R." S.," A." J." Zurita," A." O'Neill," A." NordenGZfoni," L." Zhang," H." K." Wu," P." Y." Wen," D." George,"V."P."Sukhatme,"M."B."Atkins,"and"J."V."Heymach."2011.""Increased"mobilisation"of" circulating" endothelial" progenitors" in" von" HippelGLindau" disease" and" renal" cell" carcinoma."""Br&J&Cancer"105"(1):112G7."doi:"10.1038/bjc.2011.186."

Blagosklonny,"M."V."2004.""Antiangiogenic"therapy"and"tumor"progression."""Cancer&Cell" 5"(1):13G7."

92" " Bonnet," D.," and" J." E." Dick." 1997." "Human" acute" myeloid" leukemia" is" organized" as" a" hierarchy"that"originates"from"a"primitive"hematopoietic"cell."""Nat&Med"3"(7):730G7."

Bos,"F."L.,"M."Caunt,"J."PetersonGMaduro,"L."PlanasGPaz,"J."Kowalski,"T."Karpanen,"A."van" Impel," R." Tong," J." A." Ernst," J." Korving," J." H." van" Es," E." Lammert," H." J." Duckers," and" S." SchulteGMerker." 2011." "CCBE1" is" essential" for" mammalian" lymphatic" vascular" development"and"enhances"the"lymphangiogenic"effect"of" vascular" endothelial" growth" factorGC"in"vivo."""Circ&Res"109"(5):486G91."doi:"10.1161/CIRCRESAHA.111.250738."

Bracher," A.," A." S." Cardona," S." Tauber," A." M." Fink," A." Steiner," H." Pehamberger," H." Niederleithner,"P."Petzelbauer,"M."Gröger,"and"R."Loewe."2013.""Epidermal"growth"factor" facilitates"melanoma"lymph"node"metastasis"by"influencing"tumor"lymphangiogenesis.""" J&Invest&Dermatol"133"(1):230G8."doi:"10.1038/jid.2012.272."

Brand," A.," K." Singer," G." E." Koehl," M." Kolitzus," G." Schoenhammer," A." Thiel," C." Matos," C." Bruss,"S."Klobuch,"K."Peter,"M."Kastenberger,"C."Bogdan,"U."Schleicher,"A."Mackensen,"E." Ullrich," S." FichtnerGFeigl," R." Kesselring," M." Mack," U." Ritter," M." Schmid," C." Blank," K." Dettmer,"P."J."Oefner,"P."Hoffmann,"S."Walenta,"E."K."Geissler,"J."Pouyssegur,"A."Villunger," A."Steven,"B."Seliger,"S."Schreml,"S."Haferkamp,"E."Kohl,"S."Karrer,"M."Berneburg,"W."Herr," W." MuellerGKlieser," K." Renner," and" M." Kreutz." 2016." "LDHAGAssociated" Lactic" Acid" Production" Blunts" Tumor" Immunosurveillance" by" T" and" NK" Cells."" " Cell& Metab" 24" (5):657G671."doi:"10.1016/j.cmet.2016.08.011."

Bresnick,"E."H.,"H."Y."Lee,"T."Fujiwara,"K."D."Johnson,"and"S."Keles."2010.""GATA"switches" as" developmental" drivers."" " J& Biol& Chem" 285" (41):31087G93." doi:" 10.1074/jbc.R110.159079."

Cabillic," F.," and" A." Corlu." 2016." "Regulation" of" Transdifferentiation" and" Retrodifferentiation" by"Inflammatory" Cytokines" in" Hepatocellular" Carcinoma.""" Gastroenterology"151"(4):607G15."doi:"10.1053/j.gastro.2016.06.052."

Calcinotto,"A.,"P."Filipazzi,"M."Grioni,"M."Iero,"A."De"Milito,"A."Ricupito,"A."Cova,"R."Canese," E." Jachetti," M." Rossetti," V." Huber," G." Parmiani," L." Generoso," M." Santinami," M." Borghi," S." Fais," M." Bellone," and" L." Rivoltini." 2012." "Modulation" of" microenvironment" acidity" reverses"anergy"in"human"and"murine"tumorGinfiltrating"T"lymphocytes."""Cancer&Res"72" (11):2746G56."doi:"10.1158/0008G5472.CANG11G1272."

Campbell," K.," G." Whissell," X." FranchGMarro," E." Batlle," and" J." Casanova." 2011." "Specific" GATA"factors"act"as"conserved"inducers"of"an"endodermalGEMT."""Dev&Cell"21"(6):1051G 61."doi:"10.1016/j.devcel.2011.10.005."

Cao,"R.,"M."A."Björndahl,"M."I."Gallego,"S."Chen,"P."Religa,"A."J."Hansen,"and"Y."Cao."2006." "Hepatocyte"growth"factor"is"a"lymphangiogenic"factor"with"an"indirect"mechanism"of" action."""Blood"107"(9):3531G6."doi:"10.1182/bloodG2005G06G2538."

Cao,"R.,"H."Ji,"N."Feng,"Y."Zhang,"X."Yang,"P."Andersson,"Y."Sun,"K."Tritsaris,"A."J."Hansen,"S." Dissing,"and"Y."Cao."2012.""Collaborative"interplay"between"FGFG2"and"VEGFGC"promotes" lymphangiogenesis" and" metastasis."" " Proc&Natl&Acad&Sci&U&S&A" 109" (39):15894G9." doi:" 10.1073/pnas.1208324109."

93" " Cao,"Y.,"L."H."Hoeppner,"S."Bach,"G."E,"Y."Guo,"E."Wang,"J."Wu,"M."J."Cowley,"D."K."Chang,"N." Waddell,"S."M."Grimmond,"A."V."Biankin,"R."J."Daly,"X."Zhang,"and"D."Mukhopadhyay."2013." "NeuropilinG2"promotes"extravasation"and"metastasis"by"interacting"with"endothelial"α5" integrin."""Cancer&Res"73"(14):4579G4590."doi:"10.1158/0008G5472.CANG13G0529."

Cao,"Y.,"L."Wang,"D."Nandy,"Y."Zhang,"A."Basu,"D."Radisky,"and"D."Mukhopadhyay."2008." "NeuropilinG1" upholds" dedifferentiation" and" propagation" phenotypes" of" renal" cell" carcinoma"cells"by"activating"Akt"and"sonic"hedgehog"axes."""Cancer&Res"68"(21):8667G 72."doi:"10.1158/0008G5472.CANG08G2614."

Carey,"L."A.,"E."C."Dees,"L."Sawyer,"L."Gatti,"D."T."Moore,"F."Collichio,"D."W."Ollila,"C."I."Sartor," M." L." Graham," and" C." M." Perou." 2007." "The" triple" negative" paradox:" primary" tumor" chemosensitivity" of" breast" cancer" subtypes."" " Clin& Cancer& Res" 13" (8):2329G34." doi:" 10.1158/1078G0432.CCRG06G1109."

Carracedo,"A.,"L."Ma,"J."TeruyaGFeldstein,"F."Rojo,"L."Salmena,"A."Alimonti,"A."Egia,"A."T." Sasaki,"G."Thomas,"S."C."Kozma,"A."Papa,"C."Nardella," L."C."Cantley,"J."Baselga,"and"P."P." Pandolfi." 2008." "Inhibition" of" mTORC1" leads" to" MAPK" pathway" activation" through" a" PI3KGdependent" feedback" loop" in" human" cancer."" " J& Clin& Invest" 118" (9):3065G74." doi:" 10.1172/JCI34739."

Casanovas," O.," D." J." Hicklin," G." Bergers," and" D." Hanahan." 2005." "Drug" resistance" by" evasion" of" antiangiogenic" targeting" of" VEGF" signaling" in" lateGstage" pancreatic" islet" tumors."""Cancer&Cell"8"(4):299G309."doi:"10.1016/j.ccr.2005.09.005."

Caunt," M.," J." Mak," W." C." Liang," S." Stawicki," Q." Pan," R." K." Tong," J." Kowalski," C." Ho," H." B." Reslan," J." Ross," L." Berry," I." Kasman," C." Zlot," Z." Cheng," J." Le" Couter," E." H." Filvaroff," G." Plowman," F." Peale," D." French," R." Carano," A." W." Koch," Y." Wu," R." J." Watts," M." TessierG Lavigne," and" A." Bagri." 2008." "Blocking" neuropilinG2" function" inhibits" tumor" cell" metastasis."""Cancer&Cell"13"(4):331G42."doi:"10.1016/j.ccr.2008.01.029."

Chang,"J."C."2016.""Cancer"stem"cells:"Role"in"tumor"growth,"recurrence,"metastasis,"and" treatment" resistance."" " Medicine& (Baltimore)" 95" (1" Suppl" 1):S20G5." doi:" 10.1097/MD.0000000000004766."

Chapman," H." A." 2011." "EpithelialGmesenchymal" interactions" in" pulmonary" fibrosis.""" Annu&Rev&Physiol"73:413G35."doi:"10.1146/annurevGphysiolG012110G142225."

Chen,"C."G.,"P."Tang,"and"Z."T."Yu."2012.""[Effect"of"HMGB1"on"the"VEGFGC"expression"and" proliferation" of" esophageal" squamous" cancer" cells].""" Zhonghua& Zhong& Liu& Za& Zhi" 34" (8):566G70."

Chng,"Z.,"A."Teo,"R."A."Pedersen,"and"L."Vallier."2010.""SIP1"mediates"cellGfate"decisions" between"neuroectoderm"and"mesendoderm"in"human"pluripotent"stem"cells."""Cell&Stem& Cell"6"(1):59G70."doi:"10.1016/j.stem.2009.11.015."

Choueiri," T." K.," B." Escudier," T." Powles," N." M." Tannir," P." N." Mainwaring," B." I." Rini," H." J." Hammers,"F."Donskov,"B."J."Roth,"K."Peltola,"J."L."Lee,"D."Y."C."Heng,"M."Schmidinger,"N." Agarwal,"C."N."Sternberg,"D."F."McDermott,"D."T."Aftab,"C."Hessel,"C."Scheffold,"G."Schwab," T."E."Hutson,"S."Pal,"R."J."Motzer,"and"METEOR"investigators."2016.""Cabozantinib"versus" everolimus" in" advanced" renal" cell" carcinoma" (METEOR):" final" results" from" a"

94" " randomised," openGlabel," phase" 3" trial."" " Lancet& Oncol" 17" (7):917G927." doi:" 10.1016/S1470G2045(16)30107G3."

Chow," W." H.," L." M." Dong," and" S." S." Devesa." 2010." "Epidemiology" and" risk" factors" for" kidney"cancer."""Nat&Rev&Urol"7"(5):245G57."doi:"10.1038/nrurol.2010.46."

CohenGKaplan," V.," I." Naroditsky," A." Zetser," N." Ilan," I." Vlodavsky," and" I." Doweck." 2008." "Heparanase"induces"VEGF"C"and"facilitates"tumor"lymphangiogenesis."""Int&J&Cancer"123" (11):2566G73."doi:"10.1002/ijc.23898."

Comijn," J.," G." Berx," P." Vermassen," K." Verschueren," L." van" Grunsven," E." Bruyneel," M." Mareel," D." Huylebroeck," and" F." van" Roy." 2001." "The" twoGhanded" E" box" binding" zinc" finger" protein" SIP1" downregulates" EGcadherin" and" induces" invasion."" " Mol& Cell" 7" (6):1267G78."

Dadras,"S."S.,"T."Paul,"J."Bertoncini,"L."F."Brown,"A."Muzikansky,"D."G."Jackson,"U."Ellwanger," C." Garbe," M." C." Mihm," and" M." Detmar." 2003." "Tumor" lymphangiogenesis:" a" novel" prognostic"indicator"for"cutaneous"melanoma"metastasis"and"survival."""Am&J&Pathol"162" (6):1951G60."doi:"10.1016/S0002G9440(10)64328G3."

Das,"S.,"E."Sarrou,"S."Podgrabinska,"M."Cassella,"S."K."Mungamuri,"N."Feirt,"R."Gordon,"C."S." Nagi,"Y."Wang,"D."Entenberg,"J."Condeelis,"and"M."Skobe."2013.""Tumor"cell"entry"into"the" lymph" node" is" controlled" by" CCL1" chemokine" expressed" by" lymph" node" lymphatic" sinuses."""J&Exp&Med"210"(8):1509G28."doi:"10.1084/jem.20111627."

Dave," N.," S." GuaitaGEsteruelas," S." Gutarra," À" Frias," M." Beltran," S." Peiró," and" A." G." de" Herreros."2011.""Functional"cooperation"between"Snail1"and"twist"in"the"regulation"of" ZEB1" expression" during" epithelial" to" mesenchymal" transition."" " J& Biol& Chem" 286" (14):12024G32."doi:"10.1074/jbc.M110.168625."

Decio,"A.,"G."Taraboletti,"V."Patton,"R."Alzani,"P."Perego,"R."Fruscio,"J."M."Jürgensmeier,"R." Giavazzi,"and"D."Belotti."2014.""Vascular"endothelial"growth"factor"c"promotes"ovarian" carcinoma"progression"through"paracrine"and"autocrine"mechanisms."""Am&J&Pathol"184" (4):1050G61."doi:"10.1016/j.ajpath.2013.12.030."

Delfortrie,"S.,"S."Pinte,"V."Mattot,"C."Samson,"G."Villain,"B."Caetano,"G."LauridantGPhilippin," M." C." Baranzelli," J." Bonneterre," F." Trottein," C." Faveeuw," and" F." Soncin." 2011." "Egfl7" promotes" tumor" escape" from" immunity" by" repressing" endothelial" cell" activation.""" Cancer&Res"71"(23):7176G86."doi:"10.1158/0008G5472.CANG11G1301."

Deschavanne,"P."J.,"and"B."Fertil."1996.""A"review"of"human"cell"radiosensitivity"in"vitro.""" Int&J&Radiat&Oncol&Biol&Phys"34"(1):251G66."

Dias,"S.,"K."Hattori,"Z."Zhu,"B."Heissig,"M."Choy,"W."Lane,"Y."Wu,"A."Chadburn,"E."Hyjek,"M." Gill," D." J." Hicklin," L." Witte," M." A." Moore," and" S." Rafii." 2000." "Autocrine" stimulation" of" VEGFRG2" activates" human" leukemic" cell" growth" and" migration.""" J& Clin& Invest" 106" (4):511G21."doi:"10.1172/JCI8978."

Diehn,"M.,"R."W."Cho,"N."A."Lobo,"T."Kalisky,"M."J."Dorie,"A."N."Kulp,"D."Qian,"J."S."Lam,"L."E." Ailles,"M."Wong,"B."Joshua,"M."J."Kaplan,"I."Wapnir,"F."M."Dirbas,"G."Somlo,"C."Garberoglio," B."Paz,"J."Shen,"S."K."Lau,"S."R."Quake,"J."M."Brown,"I."L."Weissman,"and"M."F."Clarke."2009."

95" " "Association"of"reactive"oxygen"species"levels"and"radioresistance"in"cancer"stem"cells.""" Nature"458"(7239):780G3."doi:"10.1038/nature07733."

Dong,"C.,"Y."Wu,"J."Yao,"Y."Wang,"Y."Yu,"P."G."Rychahou,"B."M."Evers,"and"B."P."Zhou."2012." "G9a" interacts" with" Snail" and" is" critical" for" SnailGmediated" EGcadherin" repression" in" human"breast"cancer."""J&Clin&Invest"122"(4):1469G86."doi:"10.1172/JCI57349."

Du,"Q.,"L."Jiang,"X."Wang,"M."Wang,"F."She,"and"Y."Chen."2014.""Tumor"necrosis"factorGα" promotes" the" lymphangiogenesis" of" gallbladder" carcinoma" through" nuclear" factorGκBG mediated" upregulation" of" vascular" endothelial" growth" factorGC."" " Cancer& Sci" 105" (10):1261G71."doi:"10.1111/cas.12504."

Du,"R.,"K."V."Lu,"C."Petritsch,"P."Liu,"R."Ganss,"E."Passegué,"H."Song,"S."Vandenberg,"R."S." Johnson," Z." Werb," and" G." Bergers." 2008." "HIF1alpha" induces" the" recruitment" of" bone" marrowGderived" vascular" modulatory" cells" to" regulate" tumor" angiogenesis" and" invasion."""Cancer&Cell"13"(3):206G20."doi:"10.1016/j.ccr.2008.01.034."

Dufies,"M.,"S."Giuliano,"D."Ambrosetti,"A."Claren,"P."D."Ndiaye,"M."Mastri,"W."Moghrabi,"L."S." Cooley," M." Ettaiche," E." Chamorey," J." Parola," V." Vial," M." LupuGPlesu," J." C." Bernhard," A." Ravaud,"D."Borchiellini,"J."M."Ferrero,"A."Bikfalvi,"J."M."Ebos,"K."S."Khabar,"R."Grépin,"and"G." Pagès."2017.""Sunitinib"Stimulates"Expression"of"VEGFC"by"Tumor"Cells"and"Promotes" Lymphangiogenesis"in"Clear"Cell"Renal"Cell"Carcinomas."""Cancer&Res"77"(5):1212G1226." doi:"10.1158/0008G5472.CANG16G3088."

Dumont,"D."J.,"L."Jussila,"J."Taipale,"A."Lymboussaki,"T."Mustonen,"K."Pajusola,"M."Breitman," and" K." Alitalo." 1998." "Cardiovascular" failure" in" mouse" embryos" deficient" in" VEGF" receptorG3."""Science"282"(5390):946G9."

Dunworth,"W."P.,"J."CardonaGCosta,"E."C."Bozkulak,"J."D."Kim,"S."Meadows,"J."C."Fischer,"Y." Wang,"O."Cleaver,"Y."Qyang,"E."A."Ober,"and"S."W."Jin."2014.""Bone"morphogenetic"protein" 2"signaling"negatively"modulates"lymphatic"development"in"vertebrate"embryos."""Circ& Res"114"(1):56G66."doi:"10.1161/CIRCRESAHA.114.302452."

Eijkelenboom," A.," and" B." M." Burgering." 2013." "FOXOs:" signalling" integrators" for" homeostasis"maintenance."""Nat&Rev&Mol&Cell&Biol"14"(2):83G97."doi:"10.1038/nrm3507."

Engelman,"J."A.,"K."Zejnullahu,"T."Mitsudomi,"Y."Song,"C."Hyland,"J."O."Park,"N."Lindeman,"C." M."Gale,"X."Zhao,"J."Christensen,"T."Kosaka,"A."J."Holmes,"A."M."Rogers,"F."Cappuzzo,"T."Mok," C." Lee," B." E." Johnson," L." C." Cantley," and" P." A." Jänne." 2007." "MET" amplification" leads" to" gefitinib" resistance" in" lung" cancer" by" activating" ERBB3" signaling."" " Science" 316" (5827):1039G43."doi:"10.1126/science.1141478."

Enholm," B.," K." Paavonen," A." Ristimäki," V." Kumar," Y." Gunji," J." Klefstrom," L." Kivinen," M." Laiho," B." Olofsson," V." Joukov," U." Eriksson," and" K." Alitalo." 1997." "Comparison" of" VEGF," VEGFGB," VEGFGC" and" AngG1" mRNA" regulation" by" serum," growth" factors," oncoproteins" and"hypoxia."""Oncogene"14"(20):2475G83."doi:"10.1038/sj.onc.1201090."

Epstein," A." C.," J." M." Gleadle," L." A." McNeill," K." S." Hewitson," J." O'Rourke," D." R." Mole," M." Mukherji,"E."Metzen,"M."I."Wilson,"A."Dhanda,"Y."M." Tian," N." Masson," D." L." Hamilton," P." Jaakkola," R." Barstead," J." Hodgkin," P." H." Maxwell," C." W." Pugh," C." J." Schofield," and" P." J."

96" " Ratcliffe." 2001." "C." elegans" EGLG9" and" mammalian" homologs" define" a" family" of" dioxygenases"that"regulate"HIF"by"prolyl"hydroxylation."""Cell"107"(1):43G54."

Escudier," B." 2010a." "Sunitinib" for" the" management" of" advanced" renal" cell" carcinoma.""" Expert&Rev&Anticancer&Ther"10"(3):305G17."doi:"10.1586/era.10.26."

Escudier," B.," T." Eisen," W." M." Stadler," C." Szczylik," S." Oudard," M." Siebels," S." Negrier," C." Chevreau,"E."Solska,"A."A."Desai,"F."Rolland,"T."Demkow,"T."E."Hutson,"M."Gore,"S."Freeman," B." Schwartz," M." Shan," R." Simantov," R." M." Bukowski," and" TARGET" Study" Group." 2007." "Sorafenib" in" advanced" clearGcell" renalGcell" carcinoma."" " N&Engl&J&Med" 356" (2):125G34." doi:"10.1056/NEJMoa060655."

Escudier," B." J." 2010b." "Combination" therapy" in" patients" with" metastatic" renal" cell" carcinoma."""Clin&Adv&Hematol&Oncol"8"(10):665G6."

Escudier,"B.,"A."Pluzanska,"P."Koralewski,"A."Ravaud,"S."Bracarda,"C."Szczylik,"C."Chevreau," M."Filipek,"B."Melichar,"E."Bajetta,"V."Gorbunova,"J."O."Bay,"I."Bodrogi,"A."JagielloGGruszfeld," N."Moore,"and"AVOREN"Trial"investigators."2007.""Bevacizumab"plus"interferon"alfaG2a" for"treatment"of"metastatic"renal"cell"carcinoma:"a"randomised,"doubleGblind"phase"III" trial."""Lancet"370"(9605):2103G11."doi:"10.1016/S0140G6736(07)61904G7."

Escudier," B.," P." Sharma," D." F." McDermott," S." George," H." J." Hammers," S." Srinivas," S." S." Tykodi,"J."A."Sosman,"G."Procopio,"E."R."Plimack,"D."Castellano,"H."Gurney,"F."Donskov,"K." Peltola," J." Wagstaff," T." C." Gauler," T." Ueda," H." Zhao," I." M." Waxman," R." J." Motzer," and" CheckMate" 025" investigators." 2017." "CheckMate" 025" Randomized" Phase" 3" Study:" Outcomes"by"Key"Baseline"Factors"and"Prior"Therapy"for"Nivolumab"Versus"Everolimus" in"Advanced"Renal"Cell"Carcinoma."""Eur&Urol."doi:"10.1016/j.eururo.2017.02.010."

Farace," F.," M." GrossGGoupil," E." Tournay," M." Taylor," N." Vimond," N." Jacques," F." Billiot," A." Mauguen," C." Hill," and" B." Escudier." 2011." "Levels" of" circulating" CD45(dim)CD34(+)VEGFR2(+)" progenitor" cells" correlate" with" outcome" in" metastatic" renal"cell"carcinoma"patients"treated"with"tyrosine"kinase"inhibitors."""Br&J&Cancer"104" (7):1144G50."doi:"10.1038/bjc.2011.72."

Farnsworth," R." H.," M." G." Achen," and" S." A." Stacker." 2006." "Lymphatic" endothelium:" an" important"interactive"surface"for"malignant"cells."""Pulm&Pharmacol&Ther"19"(1):51G60." doi:"10.1016/j.pupt.2005.02.003."

Favier,"B.,"A."Alam,"P."Barron,"J."Bonnin,"P."Laboudie,"P."Fons,"M."Mandron,"J."P."Herault,"G." Neufeld,"P."Savi,"J."M."Herbert,"and"F."Bono."2006.""NeuropilinG2"interacts"with"VEGFRG2" and"VEGFRG3"and"promotes"human"endothelial"cell"survival"and"migration."""Blood"108" (4):1243G50."doi:"10.1182/bloodG2005G11G4447."

Fielder,"W.,"U."Graeven,"S."Ergün,"S."Verago,"N."Kilic,"M."Stockschläder,"and"D."K."Hossfeld." 1997.""Expression"of"FLT4"and"its"ligand"VEGFGC"in"acute"myeloid"leukemia."""Leukemia" 11"(8):1234G7."

Finger," E." C.," and" A." J." Giaccia." 2010." "Hypoxia," inflammation," and" the" tumor" microenvironment" in" metastatic" disease."" " Cancer& Metastasis& Rev" 29" (2):285G93." doi:" 10.1007/s10555G010G9224G5."

97" " Folkman,"J.,"and"Y."Shing."1992.""Control"of"angiogenesis"by"heparin"and"other"sulfated" polysaccharides."""Adv&Exp&Med&Biol"313:355G64."

Francois," M.," N." L." Harvey," and" B." M." Hogan." 2011." "The" transcriptional" control" of" lymphatic" vascular" development."" " Physiology& (Bethesda)" 26" (3):146G55." doi:" 10.1152/physiol.00053.2010."

François,"M.,"A."Caprini,"B."Hosking,"F."Orsenigo,"D."Wilhelm,"C."Browne,"K."Paavonen,"T." Karnezis," R." Shayan," M." Downes," T." Davidson," D." Tutt,"K."S."Cheah,"S."A."Stacker,"G."E." Muscat,"M."G."Achen,"E."Dejana,"and"P."Koopman."2008.""Sox18"induces"development"of" the" lymphatic" vasculature" in" mice."" " Nature" 456" (7222):643G7." doi:" 10.1038/nature07391."

Förster," R.," A." C." DavalosGMisslitz," and" A." Rot." 2008." "CCR7" and" its" ligands:" balancing" immunity"and"tolerance."""Nat&Rev&Immunol"8"(5):362G71."doi:"10.1038/nri2297."

Gale,"N."W.,"R."Prevo,"J."Espinosa,"D."J."Ferguson,"M."G."Dominguez,"G."D."Yancopoulos,"G." Thurston,"and"D."G."Jackson."2007.""Normal"lymphatic"development"and"function"in"mice" deficient"for"the"lymphatic"hyaluronan"receptor"LYVEG1."""Mol&Cell&Biol"27"(2):595G604." doi:"10.1128/MCB.01503G06."

Galliera,"E.,"M."M."Corsi,"and"G."Banfi."2012.""Platelet"rich"plasma"therapy:"inflammatory" molecules"involved"in"tissue"healing."""J&Biol&Regul&Homeost&Agents"26"(2"Suppl"1):35SG 42S."

GarmyGSusini," B.," C." J." Avraamides," M." C." Schmid," P." Foubert," L." G." Ellies," L." Barnes," C." Feral,"T."Papayannopoulou,"A."Lowy,"S."L."Blair,"D."Cheresh,"M."Ginsberg,"and"J."A."Varner." 2010." "Integrin" alpha4beta1" signaling" is" required" for" lymphangiogenesis" and" tumor" metastasis."""Cancer&Res"70"(8):3042G51."doi:"10.1158/0008G5472.CANG09G3761."

Gerlinger," M.," S." Horswell," J." Larkin," A." J." Rowan," M." P." Salm," I." Varela," R." Fisher," N." McGranahan," N." Matthews," C." R." Santos," P." Martinez," B." Phillimore," S." Begum," A." Rabinowitz,"B."SpencerGDene,"S."Gulati,"P."A."Bates,"G."Stamp,"L."Pickering,"M."Gore,"D."L." Nicol,"S."Hazell,"P."A."Futreal,"A."Stewart,"and"C."Swanton."2014.""Genomic"architecture" and" evolution" of" clear" cell" renal" cell" carcinomas" defined" by" multiregion" sequencing.""" Nat&Genet"46"(3):225G233."doi:"10.1038/ng.2891."

Ginestier," C.," M." H." Hur," E." CharafeGJauffret," F." Monville," J." Dutcher," M." Brown," J." Jacquemier,"P."Viens,"C."G."Kleer,"S."Liu,"A."Schott,"D."Hayes,"D."Birnbaum,"M."S."Wicha,"and" G." Dontu." 2007." "ALDH1" is" a" marker" of" normal" and" malignant" human" mammary" stem" cells" and" a" predictor" of" poor" clinical" outcome."" " Cell& Stem& Cell" 1" (5):555G67." doi:" 10.1016/j.stem.2007.08.014."

Giuliano,"S.,"Y."Cormerais,"M."Dufies,"R."Grépin,"P."Colosetti,"A."Belaid,"J."Parola,"A."Martin," S."LacasGGervais,"N."M."Mazure,"R."Benhida,"P."Auberger,"B."Mograbi,"and"G."Pagès."2015." "Resistance" to" sunitinib" in" renal" clear" cell" carcinoma" results" from" sequestration" in" lysosomes" and" inhibition" of" the" autophagic" flux."" " Autophagy" 11" (10):1891G904." doi:" 10.1080/15548627.2015.1085742."

98" " Giuliano,"S.,"M."Guyot,"R."Grépin,"and"G."Pagès."2014.""The"ELR(+)CXCL"chemokines"and" their"receptors"CXCR1/CXCR2:"A"signaling"axis"and"new"target"for"the"treatment"of"renal" cell"carcinoma."""Oncoimmunology"3:e28399."doi:"10.4161/onci.28399."

Gnarra,"J."R.,"K."Tory,"Y."Weng,"L."Schmidt,"M."H."Wei,"H."Li,"F."Latif,"S."Liu,"F."Chen,"and"F." M."Duh."1994.""Mutations"of"the"VHL"tumour"suppressor"gene"in"renal"carcinoma."""Nat& Genet"7"(1):85G90."doi:"10.1038/ng0594G85."

Gogineni," A.," M." Caunt," A." Crow," C." V." Lee," G." Fuh," N." van" Bruggen," W." Ye," and" R." M." Weimer." 2013." "Inhibition" of" VEGFGC" modulates" distal" lymphatic" remodeling" and" secondary"metastasis."""PLoS&One"8"(7):e68755."doi:"10.1371/journal.pone.0068755."

Gordan,"J."D.,"P."Lal,"V."R."Dondeti,"R."Letrero,"K."N."Parekh,"C."E."Oquendo,"R."A."Greenberg," K."T."Flaherty,"W."K."Rathmell,"B."Keith,"M."C."Simon," and" K." L." Nathanson." 2008." "HIFG alpha" effects" on" cGMyc" distinguish" two" subtypes" of" sporadic" VHLGdeficient" clear" cell" renal"carcinoma."""Cancer&Cell"14"(6):435G46."doi:"10.1016/j.ccr.2008.10.016."

Gordon," K." J.," K." C." Kirkbride," T." How," and" G." C." Blobe." 2009." "Bone" morphogenetic" proteins" induce" pancreatic" cancer" cell" invasiveness" through" a" Smad1Gdependent" mechanism" that" involves" matrix" metalloproteinaseG2."" " Carcinogenesis" 30" (2):238G48." doi:"10.1093/carcin/bgn274."

Gore," M." E.," C." Szczylik," C." Porta," S." Bracarda," G." A." Bjarnason," S." Oudard," S." H." Lee," J." Haanen,"D."Castellano,"E."Vrdoljak,"P."Schöffski,"P."Mainwaring,"R."E."Hawkins,"L."Crinò,"T." M." Kim," G." Carteni," W." E." Eberhardt," K." Zhang," K." Fly," E." Matczak," M." J." Lechuga," S." Hariharan," and" R." Bukowski." 2015." "Final" results" from" the" large" sunitinib" global" expandedGaccess"trial"in"metastatic"renal"cell"carcinoma."""Br&J&Cancer"113"(1):12G9."doi:" 10.1038/bjc.2015.196."

Grepin," R.," M." Guyot," M." Jacquin," J." Durivault," E." Chamorey," A." Sudaka," C." Serdjebi," B." Lacarelle,"J."Y."Scoazec,"S."Negrier,"H."Simonnet,"and"G."Pages."2012.""Acceleration"of"clear" cell"renal"cell"carcinoma"growth"in"mice"following"bevacizumab/Avastin"treatment:"the" role"of"CXCL"cytokines."""Oncogene"31"(13):1683G94."doi:"10.1038/onc.2011.360."

Gruenwald," V.," G." Beutel," S." SchuchGJantsch," C." Reuter," P." Ivanyi," A." Ganser," and" M." Haubitz." 2010." "Circulating" endothelial" cells" are" an" early" predictor" in" renal" cell" carcinoma" for" tumor" response" to" sunitinib."" " BMC&Cancer" 10:695." doi:" 10.1186/1471G 2407G10G695."

Grépin,"R.,"D."Ambrosetti,"A."Marsaud,"L."Gastaud,"J."Amiel,"F."Pedeutour,"and"G."Pagès." 2014." "The" relevance" of" testing" the" efficacy" of" antiGangiogenesis" treatments" on" cells" derived"from"primary"tumors:"a"new"method"for"the"personalized"treatment"of"renal"cell" carcinoma."""PLoS&One"9"(3):e89449."doi:"10.1371/journal.pone.0089449."

Grépin," R.," M." Guyot," S." Giuliano," M." Boncompagni," D." Ambrosetti," E." Chamorey," J." Y." Scoazec,"S."Negrier,"H."Simonnet,"and"G."Pagès."2014.""The"CXCL7/CXCR1/2"axis"is"a"key" driver"in"the"growth"of"clear"cell"renal"cell"carcinoma."""Cancer&Res"74"(3):873G83."doi:" 10.1158/0008G5472.CANG13G1267."

99" " Guan,"Y.,"K."R."Reddy,"Q."Zhu,"Y."Li,"K."Lee,"P."Weerasinghe,"J."Prchal,"G."L."Semenza,"and"N." Jing."2010.""GGrich"oligonucleotides"inhibit"HIFG1alpha"and"HIFG2alpha"and"block"tumor" growth."""Mol&Ther"18"(1):188G97."doi:"10.1038/mt.2009.219."

Guo,"F.,"Q."Ru,"J."Zhang,"S."He,"J."Yu,"S."Zheng,"and"J."Wang."2017.""Inflammation"factors"in" hepatoblastoma"and"their"clinical"significance"as"diagnostic"and"prognostic"biomarkers.""" J&Pediatr&Surg"52"(9):1496G1502."doi:"10.1016/j.jpedsurg.2017.01.059."

Guo,"Y."C.,"M."Zhang,"F."X."Wang,"G."C."Pei,"F."Sun,"Y."Zhang,"X."He,"Y."Wang,"J."Song,"F."M." Zhu,"N."S."Pandupuspitasari,"J."Liu,"K."Huang,"P."Yang,"F."Xiong,"S."Zhang,"Q."Yu,"Y."Yao,"and" C."Y."Wang."2017.""Macrophages"Regulate"Unilateral"Ureteral"ObstructionGInduced"Renal" Lymphangiogenesis" through" CGC" Motif" Chemokine" Receptor" 2GDependent" Phosphatidylinositol" 3GKinaseGAKTGMechanistic" Target" of"Rapamycin" Signaling" and" HypoxiaGInducible"FactorG1α/Vascular"Endothelial"Growth"FactorGC"Expression."""Am&J& Pathol"187"(8):1736G1749."doi:"10.1016/j.ajpath.2017.04.007."

Haas," N." B.," and" K." L." Nathanson." 2014." "Hereditary" kidney" cancer" syndromes."" " Adv& Chronic&Kidney&Dis"21"(1):81G90."doi:"10.1053/j.ackd.2013.10.001."

Hakimi,"A."A.,"C."G."Pham,"and"J."J."Hsieh."2013.""A"clear"picture"of"renal"cell"carcinoma.""" Nat&Genet"45"(8):849G50."doi:"10.1038/ng.2708."

Hanahan,"D.,"and"R."A."Weinberg."2000.""The"hallmarks"of"cancer."""Cell"100"(1):57G70."

Harmon,"C."S.,"S."E."DePrimo,"R."A."Figlin,"G."R."Hudes,"T."E."Hutson,"M."D."Michaelson,"S." Négrier,"S."T."Kim,"X."Huang,"J."A."Williams,"T."Eisen,"and"R."J."Motzer."2014.""Circulating" proteins" as" potential" biomarkers" of" sunitinib" and" interferonGα" efficacy" in" treatmentG naïve"patients"with"metastatic"renal"cell"carcinoma."""Cancer&Chemother&Pharmacol"73" (1):151G61."doi:"10.1007/s00280G013G2333G4."

Hay," E." D." 1995." "An" overview" of" epithelioGmesenchymal" transformation."" " Acta& Anat& (Basel)"154"(1):8G20."

Hay,"N.,"and"N."Sonenberg."2004.""Upstream"and"downstream"of"mTOR."""Genes&Dev"18" (16):1926G45."doi:"10.1101/gad.1212704."

He," Y.," K." Kozaki," T." Karpanen," K." Koshikawa," S." YlaGHerttuala," T." Takahashi," and" K." Alitalo."2002.""Suppression"of"tumor"lymphangiogenesis"and"lymph"node"metastasis"by" blocking"vascular"endothelial"growth"factor"receptor"3"signaling."""J&Natl&Cancer&Inst"94" (11):819G25."

He,"Y.,"I."Rajantie,"K."Pajusola,"M."Jeltsch,"T."Holopainen,"S."YlaGHerttuala,"T."Harding,"K." Jooss," T." Takahashi," and" K." Alitalo." 2005." "Vascular" endothelial" cell" growth" factor" receptor"3Gmediated"activation"of"lymphatic"endothelium"is"crucial"for"tumor"cell"entry" and" spread" via" lymphatic" vessels."" " Cancer& Res" 65" (11):4739G46." doi:" 10.1158/0008G 5472.CANG04G4576."

Hermann,"P."C.,"S."L."Huber,"T."Herrler,"A."Aicher,"J."W."Ellwart,"M."Guba,"C."J."Bruns,"and"C." Heeschen."2007.""Distinct"populations"of"cancer"stem"cells"determine"tumor"growth"and" metastatic" activity" in" human" pancreatic" cancer."" " Cell& Stem& Cell" 1" (3):313G23." doi:" 10.1016/j.stem.2007.06.002."

100" " Hess," P." R.," D." R." Rawnsley," Z." Jakus," Y." Yang," D." T." Sweet," J." Fu," B." Herzog," M." Lu," B." Nieswandt," G." Oliver," T." Makinen," L." Xia," and" M." L." Kahn." 2014." "Platelets" mediate" lymphovenous" hemostasis" to" maintain" bloodGlymphatic" separation" throughout" life."" " J& Clin&Invest"124"(1):273G84."doi:"10.1172/JCI70422."

Hirakawa," S.," S." Kodama," R." Kunstfeld," K." Kajiya," L." F." Brown," and" M." Detmar." 2005." "VEGFGA" induces" tumor" and" sentinel" lymph" node" lymphangiogenesis" and" promotes" lymphatic"metastasis."""J&Exp&Med"201"(7):1089G99."doi:"10.1084/jem.20041896."

Hong,"J.,"J."Zhou,"J."Fu,"T."He,"J."Qin,"L."Wang,"L."Liao,"and"J."Xu."2011.""Phosphorylation"of" serine"68"of"Twist1"by"MAPKs"stabilizes"Twist1"protein"and"promotes"breast"cancer"cell" invasiveness."""Cancer&Res"71"(11):3980G90."doi:"10.1158/0008G5472.CANG10G2914."

Hoot,"K."E.,"J."Lighthall,"G."Han,"S."L."Lu,"A."Li,"W."Ju,"M."KuleszGMartin,"E."Bottinger,"and"X."J." Wang." 2008." "KeratinocyteGspecific" Smad2" ablation" results" in" increased" epithelialG mesenchymal" transition" during" skin" cancer" formation" and" progression."" " J& Clin& Invest" 118"(8):2722G32."doi:"10.1172/JCI33713."

Hora," M.," O." Hes," T." Reischig," T." Urge," J." Klecka," J." Ferda," M." Michal," and" V." Eret." 2008." "Tumours" in" endGstage" kidney."" " Transplant& Proc" 40" (10):3354G8." doi:" 10.1016/j.transproceed.2008.08.135."

Hsieh,"J."J.,"M."P."Purdue,"S."Signoretti,"C."Swanton,"L."Albiges,"M."Schmidinger,"D."Y."Heng," J."Larkin,"and"V."Ficarra."2017.""Renal"cell"carcinoma."""Nat&Rev&Dis&Primers"3:17009."doi:" 10.1038/nrdp.2017.9."

Hu,"J.,"Y."Cheng,"Y."Li,"Z."Jin,"Y."Pan,"G."Liu,"S."Fu,"Y."Zhang,"K."Feng,"and"Y."Feng."2014." "microRNAG128"plays"a"critical"role"in"human"nonGsmall"cell"lung"cancer"tumourigenesis," angiogenesis"and"lymphangiogenesis"by"directly"targeting"vascular"endothelial"growth" factorGC."""Eur&J&Cancer"50"(13):2336G50."doi:"10.1016/j.ejca.2014.06.005."

Huang,"R."Y.,"P."Guilford,"and"J."P."Thiery."2012.""Early"events"in"cell"adhesion"and"polarity" during" epithelialGmesenchymal" transition."" " J& Cell& Sci" 125" (Pt" 19):4417G22." doi:" 10.1242/jcs.099697."

Hudes," G.," M." Carducci," P." Tomczak," J." Dutcher," R." Figlin,"A."Kapoor,"E."Staroslawska,"J." Sosman," D." McDermott," I." Bodrogi," Z." Kovacevic," V." Lesovoy," I." G." SchmidtGWolf," O." Barbarash,"E."Gokmen,"T."O'Toole,"S."Lustgarten,"L."Moore,"R."J."Motzer,"and"Global"ARCC" Trial."2007.""Temsirolimus,"interferon"alfa,"or"both"for"advanced"renalGcell"carcinoma.""" N&Engl&J&Med"356"(22):2271G81."doi:"10.1056/NEJMoa066838."

Huntington," T." W." 1912." "ADDRESS" OF" THE" PRESIDENT:" THE" MEDICAL" EDUCATION" PROBLEM."""Cal&State&J&Med"10"(6):233G6."

Imamura,"T.,"H."Kikuchi,"M."T."Herraiz,"D."Y."Park,"Y."Mizukami,"M."MinoGKenduson,"M."P." Lynch,"B."R."Rueda,"Y."Benita,"R."J."Xavier,"and"D."C."Chung."2009.""HIFG1alpha"and"HIFG 2alpha" have" divergent" roles" in" colon" cancer."" " Int& J& Cancer" 124" (4):763G71." doi:" 10.1002/ijc.24032."

Indraccolo," S." 2010." "InterferonGalpha" as" angiogenesis" inhibitor:" learning" from" tumor" models."""Autoimmunity"43"(3):244G7."doi:"10.3109/08916930903510963."

101" " Irrthum,"A.,"K."Devriendt,"D."Chitayat,"G."Matthijs,"C."Glade,"P."M."Steijlen,"J."P."Fryns,"M."A." Van"Steensel,"and"M."Vikkula."2003.""Mutations"in"the"transcription"factor"gene"SOX18" underlie" recessive" and" dominant" forms" of" hypotrichosisGlymphedemaGtelangiectasia.""" Am&J&Hum&Genet"72"(6):1470G8."doi:"10.1086/375614."

Issa," A.," T." X." Le," A." N." Shoushtari," J." D." Shields," and" M." A." Swartz." 2009." "Vascular" endothelial" growth" factorGC" and" CGC" chemokine" receptor" 7" in" tumor" cellGlymphatic" crossGtalk"promote"invasive"phenotype."""Cancer&Res"69"(1):349G57."doi:"10.1158/0008G 5472.CANG08G1875."

Ivan,"M.,"K."Kondo,"H."Yang,"W."Kim,"J."Valiando,"M."Ohh,"A."Salic,"J."M."Asara,"W."S."Lane," and" W." G." Kaelin." 2001." "HIFalpha" targeted" for" VHLGmediated" destruction" by" proline" hydroxylation:" implications" for" O2" sensing."" " Science" 292" (5516):464G8." doi:" 10.1126/science.1059817."

Iwasaki,"T.,"Y."Takeda,"K."Maruyama,"Y."Yokosaki,"K."Tsujino,"S."Tetsumoto,"H."Kuhara,"K." Nakanishi," Y." Otani," Y." Jin," S." Kohmo," H." Hirata," R." Takahashi," M." Suzuki," K." Inoue," I." Nagatomo,"S."Goya,"T."Kijima,"T."Kumagai,"I."Tachibana,"I."Kawase,"and"A."Kumanogoh." 2013.""Deletion"of"tetraspanin"CD9"diminishes"lymphangiogenesis"in"vivo"and"in"vitro.""" J&Biol&Chem"288"(4):2118G31."doi:"10.1074/jbc.M112.424291."

Jastrzebski,"K.,"K."M."Hannan,"E."B."Tchoubrieva,"R."D."Hannan,"and"R."B."Pearson."2007." "Coordinate"regulation"of"ribosome"biogenesis"and"function"by"the"ribosomal"protein"S6" kinase," a" key" mediator" of" mTOR" function."" " Growth& Factors" 25" (4):209G26." doi:" 10.1080/08977190701779101."

Joukov," V.," K." Pajusola," A." Kaipainen," D." Chilov," I." Lahtinen," E." Kukk," O." Saksela," N." Kalkkinen,"and"K."Alitalo."1996.""A"novel"vascular"endothelial"growth"factor,"VEGFGC,"is"a" ligand"for"the"Flt4"(VEGFRG3)"and"KDR"(VEGFRG2)"receptor"tyrosine"kinases."""EMBO&J"15" (7):1751."

Kaelin," W." G." 2007." "Von" HippelGLindau" disease."" " Annu& Rev& Pathol" 2:145G73." doi:" 10.1146/annurev.pathol.2.010506.092049."

Kaelin,"W."G.,"and"P."J."Ratcliffe."2008.""Oxygen"sensing"by"metazoans:"the"central"role"of" the" HIF" hydroxylase" pathway."" " Mol& Cell" 30" (4):393G402." doi:" 10.1016/j.molcel.2008.04.009."

Kang,"J.,"J."Yoo,"S."Lee,"W."Tang,"B."Aguilar,"S."Ramu,"I."Choi,"H."H."Otu,"J."W."Shin,"G."P." Dotto,"C."J."Koh,"M."Detmar,"and"Y."K."Hong."2010.""An"exquisite"crossGcontrol"mechanism" among" endothelial" cell" fate" regulators" directs" the" plasticity" and" heterogeneity" of" lymphatic" endothelial" cells."" " Blood" 116" (1):140G50." doi:" 10.1182/bloodG2009G11G 252270."

Kang,"Y.,"C."R."Chen,"and"J."Massagué."2003.""A"selfGenabling"TGFbeta"response"coupled"to" stress" signaling:" Smad" engages" stress" response" factor" ATF3" for" Id1" repression" in" epithelial"cells."""Mol&Cell"11"(4):915G26."

Karkkainen,"M."J.,"P."Haiko,"K."Sainio,"J."Partanen,"J."Taipale,"T."V."Petrova,"M."Jeltsch,"D."G." Jackson,"M."Talikka,"H."Rauvala,"C."Betsholtz,"and"K."Alitalo."2004.""Vascular"endothelial"

102" " growth"factor"C"is"required"for"sprouting"of"the"first"lymphatic"vessels"from"embryonic" veins."""Nat&Immunol"5"(1):74G80."doi:"10.1038/ni1013."

Karnezis,"T.,"R."Shayan,"C."Caesar,"S."Roufail,"N."C."Harris,"K."Ardipradja,"Y."F."Zhang,"S."P." Williams,"R."H."Farnsworth,"M."G."Chai,"T."W."Rupasinghe,"D."L."Tull,"M."E."Baldwin,"E."K." Sloan," S." B." Fox," M." G." Achen," and" S." A." Stacker." 2012." "VEGFGD" promotes" tumor" metastasis" by" regulating" prostaglandins" produced" by" the" collecting" lymphatic" endothelium."""Cancer&Cell"21"(2):181G95."doi:"10.1016/j.ccr.2011.12.026."

Katsuno," Y.," S." Lamouille," and" R." Derynck." 2013." "TGFGβ" signaling" and" epithelialG mesenchymal" transition" in" cancer" progression."" " Curr& Opin& Oncol" 25" (1):76G84." doi:" 10.1097/CCO.0b013e32835b6371."

Kearney," P." M.," M." Whelton," K." Reynolds," P." Muntner," P." K." Whelton," and" J." He." 2005." "Global"burden"of"hypertension:"analysis"of"worldwide"data."""Lancet"365"(9455):217G 23."doi:"10.1016/S0140G6736(05)17741G1."

Kerbel,"R."S."2001.""Molecular"and"physiologic"mechanisms"of"drug"resistance"in"cancer:" an"overview."""Cancer&Metastasis&Rev"20"(1G2):1G2."

Khromova,"N.,"P."Kopnin,"V."Rybko,"and"B."P."Kopnin."2012.""Downregulation"of"VEGFGC" expression" in" lung" and" colon" cancer" cells" decelerates" tumor" growth" and" inhibits" metastasis" via" multiple" mechanisms."" " Oncogene" 31" (11):1389G97." doi:" 10.1038/onc.2011.330."

Kodera,"Y.,"Y."Katanasaka,"Y."Kitamura,"H."Tsuda,"K."Nishio,"T."Tamura,"and"F."Koizumi." 2011.""Sunitinib"inhibits"lymphatic"endothelial"cell"functions"and"lymph"node"metastasis" in" a" breast" cancer" model" through" inhibition" of" vascular" endothelial" growth" factor" receptor"3."""Breast&Cancer&Res"13"(3):R66."doi:"10.1186/bcr2903."

Koltowska,"K.,"S."Paterson,"N."I."Bower,"G."J."Baillie,"A."K."Lagendijk,"J."W."Astin,"H."Chen,"M." Francois,"P."S."Crosier,"R."J."Taft,"C."Simons,"K."A."Smith,"and"B."M."Hogan."2015.""mafba"is"a" downstream" transcriptional" effector" of" Vegfc" signaling" essential" for" embryonic" lymphangiogenesis" in" zebrafish."" " Genes& Dev" 29" (15):1618G30." doi:" 10.1101/gad.263210.115."

Kondo,"K.,"W."Y."Kim,"M."Lechpammer,"and"W."G."Kaelin."2003.""Inhibition"of"HIF2alpha"is" sufficient" to" suppress" pVHLGdefective" tumor" growth."" " PLoS& Biol" 1" (3):E83." doi:" 10.1371/journal.pbio.0000083."

Kukk," E.," A." Lymboussaki," S." Taira," A." Kaipainen," M." Jeltsch," V." Joukov," and" K." Alitalo." 1996.""VEGFGC"receptor"binding"and"pattern"of"expression"with"VEGFRG3"suggests"a"role" in"lymphatic"vascular"development."""Development"122"(12):3829G37."

Kurosu," T.," N." Ohga," Y." Hida," N." Maishi," K." Akiyama," W." Kakuguchi," T." Kuroshima," M." Kondo,"T."Akino,"Y."Totsuka,"M."Shindoh,"F."Higashino,"and"K."Hida."2011.""HuR"keeps"an" angiogenic"switch"on"by"stabilising"mRNA"of"VEGF"and"COXG2"in"tumour"endothelium.""" Br&J&Cancer"104"(5):819G29."doi:"10.1038/bjc.2011.20."

103" " Lamouille," S.," E." Connolly," J." W." Smyth," R." J." Akhurst," and" R." Derynck." 2012." "TGFGβG induced"activation"of"mTOR"complex"2"drives"epithelialGmesenchymal"transition"and"cell" invasion."""J&Cell&Sci"125"(Pt"5):1259G73."doi:"10.1242/jcs.095299."

Lamouille," S.," and" R." Derynck." 2007." "Cell" size" and" invasion" in" TGFGbetaGinduced" epithelial"to"mesenchymal"transition"is"regulated"by"activation"of"the"mTOR"pathway."""J& Cell&Biol"178"(3):437G51."doi:"10.1083/jcb.200611146."

Lapidot,"T.,"C."Sirard,"J."Vormoor,"B."Murdoch,"T."Hoang,"J."CaceresGCortes,"M."Minden,"B." Paterson," M." A." Caligiuri," and" J." E." Dick." 1994." "A" cell" initiating" human" acute" myeloid" leukaemia" after" transplantation" into" SCID" mice."" " Nature" 367" (6464):645G8." doi:" 10.1038/367645a0."

Le"Bras,"B.,"M."J."Barallobre,"J."HommanGLudiye,"A."Ny,"S."Wyns,"T."Tammela,"P."Haiko,"M."J." Karkkainen,"L."Yuan,"M."P."Muriel,"E."Chatzopoulou,"C."Bréant,"B."Zalc,"P."Carmeliet,"K." Alitalo," A." Eichmann," and" J." L." Thomas." 2006." "VEGFGC" is" a" trophic" factor" for" neural" progenitors" in" the" vertebrate" embryonic" brain."" " Nat& Neurosci" 9" (3):340G8." doi:" 10.1038/nn1646."

Leclers,"D.,"K."Durand,"A."Dutour,"G."Barrière,"J."Monteil,"M."Rigaud,"and"F."Sturtz."2005." "[Lymphatic" vessels" and" cancer]."" " Med& Sci& (Paris)" 21" (10):839G47." doi:" 10.1051/medsci/20052110839."

Lee,"J."Y.,"S."Park,"D."C."Kim,"J."H."Yoon,"S."H."Shin,"W."S."Min,"and"H."J."Kim."2013.""A"VEGFRG 3" antagonist" increases" IFNGγ" expression" on" low" functioning" NK" cells" in" acute" myeloid" leukemia."""J&Clin&Immunol"33"(4):826G37."doi:"10.1007/s10875G013G9877G2."

Lee,"S.,"J."Kang,"J."Yoo,"S."K."Ganesan,"S."C."Cook,"B."Aguilar,"S."Ramu,"J."Lee,"and"Y."K."Hong." 2009.""Prox1"physically"and"functionally"interacts"with"COUPGTFII"to"specify"lymphatic" endothelial"cell"fate."""Blood"113"(8):1856G9."doi:"10.1182/bloodG2008G03G145789."

Leu," A." J.," D." A." Berk," A." Lymboussaki," K." Alitalo," and" R." K." Jain." 2000." "Absence" of" functional"lymphatics"within"a"murine"sarcoma:"a"molecular"and"functional"evaluation.""" Cancer&Res"60"(16):4324G7."

Li," X.," M." T." Lewis," J." Huang," C." Gutierrez," C." K." Osborne,"M."F."Wu,"S."G."Hilsenbeck,"A." Pavlick," X." Zhang," G." C." Chamness," H." Wong," J." Rosen," and" J." C." Chang." 2008." "Intrinsic" resistance"of"tumorigenic"breast"cancer"cells"to"chemotherapy."""J&Natl&Cancer&Inst"100" (9):672G9."doi:"10.1093/jnci/djn123."

Liedtke,"C.,"C."Mazouni,"K."R."Hess,"F."André,"A."Tordai,"J."A."Mejia,"W."F."Symmans,"A."M." GonzalezGAngulo,"B."Hennessy,"M."Green,"M."Cristofanilli,"G."N."Hortobagyi,"and"L."Pusztai." 2008.""Response"to"neoadjuvant"therapy"and"longGterm"survival"in"patients"with"tripleG negative"breast"cancer."""J&Clin&Oncol"26"(8):1275G81."doi:"10.1200/JCO.2007.14.4147."

Lin,"F."J.,"X."Chen,"J."Qin,"Y."K."Hong,"M."J."Tsai,"and"S."Y."Tsai."2010.""Direct"transcriptional" regulation"of"neuropilinG2"by"COUPGTFII"modulates"multiple"steps"in"murine"lymphatic" vessel"development."""J&Clin&Invest"120"(5):1694G707."doi:"10.1172/JCI40101."

104" " Lin," F.," P." L." Zhang," X." J." Yang," J." W." Prichard," M." Lun," and" R." E." Brown." 2006." "Morphoproteomic" and" molecular" concomitants" of" an" overexpressed" and" activated" mTOR"pathway"in"renal"cell"carcinomas."""Ann&Clin&Lab&Sci"36"(3):283G93."

Lin,"J.,"A."S."Lalani,"T."C."Harding,"M."Gonzalez,"W."W."Wu,"B."Luan,"G."H."Tu,"K."Koprivnikar," M." J." VanRoey," Y." He," K." Alitalo," and" K." Jooss." 2005." "Inhibition" of" lymphogenous" metastasis" using" adenoGassociated" virusGmediated" gene" transfer" of" a" soluble" VEGFRG3" decoy"receptor."""Cancer&Res"65"(15):6901G9."doi:"10.1158/0008G5472.CANG05G0408."

Lin," T.," A." Ponn," X." Hu," B." K." Law," and" J." Lu." 2010." "Requirement" of" the" histone" demethylase" LSD1" in" Snai1Gmediated" transcriptional" repression" during" epithelialG mesenchymal"transition."""Oncogene"29"(35):4896G904."doi:"10.1038/onc.2010.234."

Lin,"Y.,"Z."Yang,"A."Xu,"P."Dong,"Y."Huang,"H."Liu,"F."Li,"H."Wang,"Q."Xu,"Y."Wang,"D."Sun,"Y." Zou,"X."Zou,"D."Zhang,"X."Wu,"M."Zhang,"Y."Fu,"Z."Cai,"C."Liu,"and"S."Wu."2015.""PIK3R1" negatively"regulates"the"epithelialGmesenchymal"transition"and"stemGlike"phenotype"of" renal"cancer"cells"through"the"AKT/GSK3β/CTNNB1"signaling"pathway."""Sci&Rep"5:8997." doi:"10.1038/srep08997."

Liu,"C."C.,"J."Prior,"D."PiwnicaGWorms,"and"G."Bu."2010.""LRP6"overexpression"defines"a" class"of"breast"cancer"subtype"and"is"a"target"for"therapy."""Proc&Natl&Acad&Sci&U&S&A"107" (11):5136G41."doi:"10.1073/pnas.0911220107."

Liu,"G.,"X."Yuan,"Z."Zeng,"P."Tunici,"H."Ng,"I."R."Abdulkadir,"L."Lu,"D."Irvin,"K."L."Black,"and"J." S."Yu."2006.""Analysis"of"gene"expression"and"chemoresistance"of"CD133+"cancer"stem" cells"in"glioblastoma."""Mol&Cancer"5:67."doi:"10.1186/1476G4598G5G67."

Ljungberg,"B.,"K."Bensalah,"S."Canfield,"S."Dabestani,"F."Hofmann,"M."Hora,"M."A."Kuczyk,"T." Lam,"L."Marconi,"A."S."Merseburger,"P."Mulders,"T."Powles,"M."Staehler,"A."Volpe,"and"A." Bex."2015.""EAU"guidelines"on"renal"cell"carcinoma:"2014"update."""Eur&Urol"67"(5):913G 24."doi:"10.1016/j.eururo.2015.01.005."

Loges," S.," M." Mazzone," P." Hohensinner," and" P." Carmeliet." 2009." "Silencing" or" fueling" metastasis"with"VEGF"inhibitors:"antiangiogenesis"revisited."""Cancer&Cell"15"(3):167G70." doi:"10.1016/j.ccr.2009.02.007."

Low,"G.,"G."Huang,"W."Fu,"Z."Moloo,"and"S."Girgis."2016.""Review"of"renal"cell"carcinoma" and" its" common" subtypes" in" radiology."" " World& J& Radiol" 8" (5):484G500." doi:" 10.4329/wjr.v8.i5.484."

Luangdilok,"S.,"C."Box,"K."Harrington,"P."RhŷsGEvans,"and"S."Eccles."2011.""MAPK"and"PI3K" signalling"differentially"regulate"angiogenic"and"lymphangiogenic"cytokine"secretion"in" squamous" cell" carcinoma" of" the" head" and" neck."" " Eur& J& Cancer" 47" (4):520G9." doi:" 10.1016/j.ejca.2010.10.009."

Lund,"A."W.,"F."V."Duraes,"S."Hirosue,"V."R."Raghavan,"C."Nembrini,"S."N."Thomas,"A."Issa,"S." Hugues," and" M." A." Swartz." 2012." "VEGFGC" promotes" immune" tolerance" in" B16" melanomas"and"crossGpresentation"of"tumor"antigen"by"lymph"node"lymphatics."""Cell& Rep"1"(3):191G9."doi:"10.1016/j.celrep.2012.01.005."

105" " Luster,"A."D.,"and"P."Leder."1993.""IPG10,"a"GCGXGCG"chemokine,"elicits"a"potent"thymusG dependent"antitumor"response"in"vivo."""J&Exp&Med"178"(3):1057G65."

Mandriota," S." J.," L." Jussila," M." Jeltsch," A." Compagni," D." Baetens," R." Prevo," S." Banerji," J." Huarte,"R."Montesano,"D."G."Jackson,"L."Orci,"K."Alitalo,"G."Christofori,"and"M."S."Pepper." 2001." "Vascular" endothelial" growth" factorGCGmediated" lymphangiogenesis" promotes" tumour"metastasis."""EMBO&J"20"(4):672G82."doi:"10.1093/emboj/20.4.672."

Manning,"B."D.,"and"L."C."Cantley."2007.""AKT/PKB"signaling:"navigating"downstream.""" Cell"129"(7):1261G74."doi:"10.1016/j.cell.2007.06.009."

Marchetti,"A.,"M."Colletti,"A."M."Cozzolino,"C."Steindler,"M."Lunadei,"C."Mancone,"and"M." Tripodi."2008.""ERK5/MAPK"is"activated"by"TGFbeta"in"hepatocytes"and"required"for"the" GSKG3betaGmediated" Snail" protein" stabilization."" " Cell& Signal" 20" (11):2113G8." doi:" 10.1016/j.cellsig.2008.08.002."

Mashino,"K.,"N."Sadanaga,"H."Yamaguchi,"F."Tanaka,"M."Ohta,"K."Shibuta,"H."Inoue,"and"M." Mori." 2002." "Expression" of" chemokine" receptor" CCR7" is" associated" with" lymph" node" metastasis"of"gastric"carcinoma."""Cancer&Res"62"(10):2937G41."

Mendez,"M."G.,"S."Kojima,"and"R."D."Goldman."2010.""Vimentin" induces" changes" in" cell" shape,"motility,"and"adhesion"during"the"epithelial"to"mesenchymal"transition."""FASEB&J" 24"(6):1838G51."doi:"10.1096/fj.09G151639."

Merchant,"A."A.,"and"W."Matsui."2010.""Targeting"HedgehogGGa"cancer"stem"cell"pathway.""" Clin&Cancer&Res"16"(12):3130G40."doi:"10.1158/1078G0432.CCRG09G2846."

Mohammed," R." A.," A." Green," S." ElGShikh," E." C." Paish," I." O." Ellis," and" S." G." Martin." 2007." "Prognostic" significance" of" vascular" endothelial" cell" growth" factors" GA," GC" and" GD" in" breast"cancer"and"their"relationship"with"angioG"and"lymphangiogenesis."""Br&J&Cancer"96" (7):1092G100."doi:"10.1038/sj.bjc.6603678."

Mole," D." R.," C." Blancher," R." R." Copley," P." J." Pollard," J." M." Gleadle," J." Ragoussis," and" P." J." Ratcliffe."2009.""GenomeGwide"association"of"hypoxiaGinducible"factor"(HIF)G1alpha"and" HIFG2alpha"DNA"binding"with"expression"profiling"of"hypoxiaGinducible"transcripts."""J& Biol&Chem"284"(25):16767G75."doi:"10.1074/jbc.M901790200."

Morfoisse," F.," A." Kuchnio," C." Frainay," A." GomezGBrouchet," M." B." Delisle," S." Marzi," A." C." Helfer," F." Hantelys," F." Pujol," J." GuillermetGGuibert," C." Bousquet," M." Dewerchin," S." Pyronnet,"A."C."Prats,"P."Carmeliet,"and"B."GarmyGSusini."2014.""Hypoxia"induces"VEGFGC" expression" in" metastatic" tumor" cells" via" a" HIFG1αGindependent" translationGmediated" mechanism."""Cell&Rep"6"(1):155G67."doi:"10.1016/j.celrep.2013.12.011."

Motzer," R." J.," and" R." M." Bukowski." 2006." "Targeted" therapy" for" metastatic" renal" cell" carcinoma."""J&Clin&Oncol"24"(35):5601G8."doi:"10.1200/JCO.2006.08.5415."

Motzer," R." J.," B." Escudier," D." F." McDermott," S." George," H." J." Hammers,"S."Srinivas,"S."S." Tykodi,"J."A."Sosman,"G."Procopio,"E."R."Plimack,"D."Castellano,"T."K."Choueiri,"H."Gurney,"F." Donskov," P." Bono," J." Wagstaff," T." C." Gauler," T." Ueda," Y." Tomita," F." A." Schutz," C." Kollmannsberger,"J."Larkin,"A."Ravaud,"J."S."Simon,"L."A."Xu,"I."M."Waxman,"P."Sharma,"and"

106" " CheckMate"025"Investigators."2015.""Nivolumab"versus"Everolimus"in"Advanced"RenalG Cell"Carcinoma."""N&Engl&J&Med"373"(19):1803G13."doi:"10.1056/NEJMoa1510665."

Motzer,"R."J.,"B."Escudier,"S."Oudard,"T."E."Hutson,"C."Porta,"S."Bracarda,"V."Grünwald,"J."A." Thompson,"R."A."Figlin,"N."Hollaender,"A."Kay,"A."Ravaud,"and"RECORD1"Study"Group." 2010.""Phase"3"trial"of"everolimus"for"metastatic"renal"cell"carcinoma":"final"results"and" analysis"of"prognostic"factors."""Cancer"116"(18):4256G65."doi:"10.1002/cncr.25219."

Motzer," R." J.," T." E." Hutson," M." Ren," C." Dutcus," and" J." Larkin." 2016." "Independent" assessment" of" lenvatinib" plus" everolimus" in" patients" with" metastatic" renal" cell" carcinoma."""Lancet&Oncol"17"(1):e4G5."doi:"10.1016/S1470G2045(15)00543G4."

Motzer," R." J.," T." E." Hutson," P." Tomczak," M." D." Michaelson," R." M." Bukowski," O." Rixe," S." Oudard,"S."Negrier,"C."Szczylik,"S."T."Kim,"I."Chen,"P."W."Bycott,"C."M."Baum,"and"R."A."Figlin." 2007.""Sunitinib"versus"interferon"alfa"in"metastatic"renalGcell"carcinoma."""N&Engl&J&Med" 356"(2):115G24."doi:"10.1056/NEJMoa065044."

Motzer,"R."J.,"M."D."Michaelson,"J."Rosenberg,"R."M."Bukowski,"B."D."Curti,"D."J."George,"G."R." Hudes,"B."G."Redman,"K."A."Margolin,"and"G."Wilding." 2007." "Sunitinib" efficacy" against" advanced"renal"cell"carcinoma."""J&Urol"178"(5):1883G7."doi:"10.1016/j.juro.2007.07.030."

Muders,"M."H.,"H."Zhang,"E."Wang,"D."J."Tindall,"and"K."Datta."2009.""Vascular"endothelial" growth"factorGC"protects"prostate"cancer"cells"from"oxidative"stress"by"the"activation"of" mammalian"target"of"rapamycin"complexG2"and"AKTG1."""Cancer&Res"69"(15):6042G8."doi:" 10.1158/0008G5472.CANG09G0552."

Munson," J." M.," R." V." Bellamkonda," and" M." A." Swartz." 2013." "Interstitial" flow" in" a" 3D" microenvironment" increases" glioma" invasion" by" a" CXCR4Gdependent" mechanism.""" Cancer&Res"73"(5):1536G46."doi:"10.1158/0008G5472.CANG12G2838."

Murtomaki,"A.,"M."K."Uh,"Y."K."Choi,"C."Kitajewski,"V." Borisenko," J." Kitajewski," and" C." J." Shawber."2013.""Notch1"functions"as"a"negative"regulator"of"lymphatic"endothelial"cell" differentiation" in" the" venous" endothelium."" " Development" 140" (11):2365G76." doi:" 10.1242/dev.083865."

Mäkinen," T.," T." Veikkola," S." Mustjoki," T." Karpanen," B." Catimel," E." C." Nice," L." Wise," A." Mercer," H." Kowalski," D." Kerjaschki," S." A." Stacker," M." G." Achen," and" K." Alitalo." 2001." "Isolated"lymphatic"endothelial"cells"transduce"growth,"survival"and"migratory"signals" via" the" VEGFGC/D" receptor" VEGFRG3."" " EMBO& J" 20" (17):4762G73." doi:" 10.1093/emboj/20.17.4762."

Müller,"A.,"B."Homey,"H."Soto,"N."Ge,"D."Catron,"M."E."Buchanan,"T."McClanahan,"E."Murphy," W." Yuan," S." N." Wagner," J." L." Barrera," A." Mohar," E." Verástegui," and" A." Zlotnik." 2001." "Involvement" of" chemokine" receptors" in" breast" cancer" metastasis."" " Nature" 410" (6824):50G6."doi:"10.1038/35065016."

Nicenboim," J.," G." Malkinson," T." Lupo," L." Asaf," Y." Sela," O." Mayseless," L." GibbsGBar," N." Senderovich,"T."Hashimshony,"M."Shin,"A."JerafiGVider,"I."AvrahamGDavidi,"V."Krupalnik,"R." Hofi,"G."Almog,"J."W."Astin,"O."Golani,"S."BenGDor,"P."S."Crosier,"W."Herzog,"N."D."Lawson,"J." H." Hanna," I." Yanai," and" K." Yaniv." 2015." "Lymphatic" vessels" arise" from" specialized"

107" " angioblasts" within" a" venous" niche."" " Nature" 522" (7554):56G61." doi:" 10.1038/nature14425."

Niederleithner," H.," M." Heinz," S." Tauber," M." Bilban," H." Pehamberger," S." Sonderegger," M." Knöfler," A." Bracher," W." Berger," R." Loewe," and" P." Petzelbauer." 2012." "Wnt1" is" antiG lymphangiogenic"in"a"melanoma"mouse"model."""J&Invest&Dermatol"132"(9):2235G44."doi:" 10.1038/jid.2012.138."

Noman,"M."Z.,"G."Desantis,"B."Janji,"M."Hasmim,"S."Karray," P." Dessen," V." Bronte," and" S." Chouaib."2014.""PDGL1"is"a"novel"direct"target"of"HIFG1α,"and"its"blockade"under"hypoxia" enhanced" MDSCGmediated" T" cell" activation."" " J& Exp& Med" 211" (5):781G90." doi:" 10.1084/jem.20131916."

Nyhan," M." J.," G." C." O'Sullivan," and" S." L." McKenna." 2008." "Role" of" the" VHL" (von" HippelG Lindau)"gene"in"renal"cancer:"a"multifunctional"tumour"suppressor."""Biochem&Soc&Trans" 36"(Pt"3):472G8."doi:"10.1042/BST0360472."

O'Brien,"C."A.,"A."Pollett,"S."Gallinger,"and"J."E."Dick." 2007." "A" human" colon" cancer" cell" capable"of"initiating"tumour"growth"in"immunodeficient"mice."""Nature"445"(7123):106G 10."doi:"10.1038/nature05372."

Oka,"M.,"C."Iwata,"H."I."Suzuki,"K."Kiyono,"Y."Morishita,"T."Watabe,"A."Komuro,"M."R."Kano," and" K." Miyazono." 2008." "Inhibition" of" endogenous" TGFGbeta" signaling" enhances" lymphangiogenesis."""Blood"111"(9):4571G9."doi:"10.1182/bloodG2007G10G120337."

Okkenhaug," K." 2013." "Signaling" by" the" phosphoinositide" 3Gkinase" family" in" immune" cells."""Annu&Rev&Immunol"31:675G704."doi:"10.1146/annurevGimmunolG032712G095946."

Padera,"T."P.,"A."Kadambi,"E."di"Tomaso,"C."M."Carreira,"E."B."Brown,"Y."Boucher,"N."C."Choi," D."Mathisen,"J."Wain,"E."J."Mark,"L."L."Munn,"and"R."K."Jain."2002.""Lymphatic"metastasis"in" the" absence" of" functional" intratumor" lymphatics."" " Science" 296" (5574):1883G6." doi:" 10.1126/science.1071420."

Paik,"S.,"G."Tang,"S."Shak,"C."Kim,"J."Baker,"W."Kim,"M."Cronin,"F."L."Baehner,"D."Watson,"J." Bryant," J." P." Costantino," C." E." Geyer," D." L." Wickerham," and" N." Wolmark." 2006." "Gene" expression" and" benefit" of" chemotherapy" in" women" with" nodeGnegative," estrogen" receptorGpositive" breast" cancer."" " J& Clin& Oncol" 24" (23):3726G34." doi:" 10.1200/JCO.2005.04.7985."

Pan," Y.," W." D." Wang," and" T." Yago." 2014." "Transcriptional" regulation" of" podoplanin" expression" by" Prox1" in" lymphatic" endothelial" cells.""" Microvasc& Res" 94:96G102." doi:" 10.1016/j.mvr.2014.05.006."

Pan," Y.," and" L." Xia." 2015." "Emerging" roles" of" podoplanin" in" vascular" development" and" homeostasis."""Front&Med"9"(4):421G30."doi:"10.1007/s11684G015G0424G9."

Panka,"D."J.,"Q."Liu,"A."K."Geissler,"and"J."W."Mier."2013.""Effects"of"HDM2"antagonism"on" sunitinib" resistance," p53" activation," SDFG1" induction," and" tumor" infiltration" by" CD11b+/GrG1+" myeloid" derived" suppressor" cells."" " Mol& Cancer" 12:17." doi:" 10.1186/1476G4598G12G17."

108" " Pao,"W.,"V."A."Miller,"K."A."Politi,"G."J."Riely,"R."Somwar,"M."F."Zakowski,"M."G."Kris,"and"H." Varmus."2005.""Acquired"resistance"of"lung"adenocarcinomas"to"gefitinib"or"erlotinib"is" associated"with"a"second"mutation"in"the"EGFR"kinase"domain."""PLoS&Med"2"(3):e73."doi:" 10.1371/journal.pmed.0020073."

Pardal," R.," M." F." Clarke," and" S." J." Morrison." 2003." "Applying" the" principles" of" stemGcell" biology"to"cancer."""Nat&Rev&Cancer"3"(12):895G902."doi:"10.1038/nrc1232."

Park,"J."Y.,"K."H."Park,"S."Bang,"M."H."Kim,"J."E."Lee,"J."Gang,"S."S."Koh,"and"S."Y."Song."2007." "CXCL5" overexpression" is" associated" with" late" stage" gastric" cancer."" " J&Cancer&Res&Clin& Oncol"133"(11):835G40."doi:"10.1007/s00432G007G0225Gx."

Patel,"V.,"C."A."Marsh,"R."T."Dorsam,"C."M."Mikelis,"A."Masedunskas,"P."Amornphimoltham," C."A."Nathan,"B."Singh,"R."Weigert,"A."A."Molinolo," and" J." S." Gutkind." 2011." "Decreased" lymphangiogenesis" and" lymph" node" metastasis" by" mTOR" inhibition" in" head" and" neck" cancer."""Cancer&Res"71"(22):7103G12."doi:"10.1158/0008G5472.CANG10G3192."

Peinado," H.," D." Olmeda," and" A." Cano." 2007." "Snail," Zeb" and" bHLH" factors" in" tumour" progression:"an"alliance"against"the"epithelial"phenotype?"""Nat&Rev&Cancer"7"(6):415G28." doi:"10.1038/nrc2131."

Pennacchietti,"S.,"P."Michieli,"M."Galluzzo,"M."Mazzone,"S."Giordano,"and"P."M."Comoglio." 2003." "Hypoxia" promotes" invasive" growth" by" transcriptional" activation" of" the" met" protooncogene."""Cancer&Cell"3"(4):347G61."

PerezGGracia,"J."L.,"C."Prior,"F."GuillénGGrima,"V."Segura,"A."Gonzalez,"A."Panizo,"I."Melero," E."GrandeGPulido,"A."Gurpide,"I."GilGBazo,"and"A."Calvo."2009.""Identification"of"TNFGalpha" and" MMPG9" as" potential" baseline" predictive" serum" markers" of" sunitinib" activity" in" patients" with" renal" cell" carcinoma" using" a" human" cytokine" array."" " Br& J& Cancer" 101" (11):1876G83."doi:"10.1038/sj.bjc.6605409."

Pischon,"T.,"P."H."Lahmann,"H."Boeing,"A."Tjønneland,"J."Halkjaer,"K."Overvad,"K."KlipsteinG Grobusch,"J."Linseisen,"N."Becker,"A."Trichopoulou,"V."Benetou,"D."Trichopoulos,"S."Sieri," D." Palli," R." Tumino," P." Vineis," S." Panico," E." Monninkhof," P." H." Peeters," H." B." BuenoGdeG Mesquita," F." L." Büchner," B." Ljungberg," G." Hallmans," G." Berglund," C." A." Gonzalez," M." Dorronsoro," A." B." Gurrea," C." Navarro," C." Martinez," J." R." Quirós," A." Roddam," N." Allen," S." Bingham,"K."T."Khaw,"R."Kaaks,"T."Norat,"N."Slimani,"and"E."Riboli."2006.""Body"size"and" risk"of"renal"cell"carcinoma"in"the"European"Prospective"Investigation"into"Cancer"and" Nutrition"(EPIC)."""Int&J&Cancer"118"(3):728G38."doi:"10.1002/ijc.21398."

Postigo,"A."A.,"J."L."Depp,"J."J."Taylor,"and"K."L."Kroll."2003.""Regulation"of"Smad"signaling" through" a" differential" recruitment" of" coactivators" and" corepressors" by" ZEB" proteins.""" EMBO&J"22"(10):2453G62."doi:"10.1093/emboj/cdg226."

Proudfoot,"A."E.,"T."M."Handel,"Z."Johnson,"E."K."Lau,"P."LiWang,"I."ClarkGLewis,"F."Borlat,"T." N." Wells," and" M." H." KoscoGVilbois." 2003." "Glycosaminoglycan" binding" and" oligomerization"are"essential"for"the"in"vivo"activity"of"certain"chemokines."""Proc&Natl& Acad&Sci&U&S&A"100"(4):1885G90."doi:"10.1073/pnas.0334864100."

Raval,"R."R.,"K."W."Lau,"M."G."Tran,"H."M."Sowter,"S."J."Mandriota,"J."L."Li,"C."W."Pugh,"P."H." Maxwell," A." L." Harris," and" P." J." Ratcliffe." 2005." "Contrasting" properties" of" hypoxiaG

109" " inducible" factor" 1" (HIFG1)" and" HIFG2" in" von" HippelGLindauGassociated" renal" cell" carcinoma."""Mol&Cell&Biol"25"(13):5675G86."doi:"10.1128/MCB.25.13.5675G5686.2005."

Relf,"M.,"S."LeJeune,"P."A."Scott,"S."Fox,"K."Smith,"R."Leek,"A."Moghaddam,"R."Whitehouse,"R." Bicknell," and" A." L." Harris." 1997." "Expression" of" the" angiogenic" factors" vascular" endothelial"cell"growth"factor,"acidic"and"basic"fibroblast"growth"factor,"tumor"growth" factor"betaG1,"plateletGderived"endothelial"cell"growth"factor,"placenta"growth"factor,"and" pleiotrophin"in"human"primary"breast"cancer"and"its"relation"to"angiogenesis."""Cancer& Res"57"(5):963G9."

Ridley," A." J." 2011." "Life" at" the" leading" edge."" " Cell" 145" (7):1012G22." doi:" 10.1016/j.cell.2011.06.010."

Rinderknecht," M.," and" M." Detmar." 2008." "Tumor" lymphangiogenesis" and" melanoma" metastasis."""J&Cell&Physiol"216"(2):347G54."doi:"10.1002/jcp.21494."

Rini,"B."I.,"M."D."Michaelson,"J."E."Rosenberg,"R."M."Bukowski,"J."A."Sosman,"W."M."Stadler," T."E."Hutson,"K."Margolin,"C."S."Harmon,"S."E."DePrimo,"S."T."Kim,"I."Chen,"and"D."J."George." 2008." "Antitumor" activity" and" biomarker" analysis" of" sunitinib" in" patients" with" bevacizumabGrefractory" metastatic" renal" cell" carcinoma."" " J&Clin&Oncol" 26" (22):3743G8." doi:"10.1200/JCO.2007.15.5416."

Ristimäki," A.," K." Narko," B." Enholm," V." Joukov," and" K." Alitalo." 1998." "Proinflammatory" cytokines"regulate"expression"of"the"lymphatic"endothelial"mitogen"vascular"endothelial" growth"factorGC."""J&Biol&Chem"273"(14):8413G8."

Rizkalla," B.," J." M." Forbes," M." E." Cooper," and" Z." Cao." 2003." "Increased" renal" vascular" endothelial" growth" factor" and" angiopoietins" by" angiotensin" II" infusion" is" mediated" by" both"AT1"and"AT2"receptors."""J&Am&Soc&Nephrol"14"(12):3061G71."

Sabatini,"D."M."2006.""mTOR"and"cancer:"insights"into"a"complex"relationship."""Nat&Rev& Cancer"6"(9):729G34."doi:"10.1038/nrc1974."

Saharinen," P.," T." Tammela," M." J." Karkkainen," and" K." Alitalo." 2004." "Lymphatic" vasculature:" development," molecular" regulation" and" role" in" tumor" metastasis" and" inflammation."""Trends&Immunol"25"(7):387G95."doi:"10.1016/j.it.2004.05.003."

Sansal,"I.,"and"W."R."Sellers."2004.""The"biology"and"clinical"relevance"of"the"PTEN"tumor" suppressor"pathway."""J&Clin&Oncol"22"(14):2954G63."doi:"10.1200/JCO.2004.02.141."

Saran," U.," M." Foti," and" J." F." Dufour." 2015." "Cellular" and" molecular" effects" of" the" mTOR" inhibitor"everolimus."""Clin&Sci&(Lond)"129"(10):895G914."doi:"10.1042/CS20150149."

Sarbassov,"D."D.,"D."A."Guertin,"S."M."Ali,"and"D."M."Sabatini."2005.""Phosphorylation"and" regulation"of"Akt/PKB"by"the"rictorGmTOR"complex."""Science"307"(5712):1098G101."doi:" 10.1126/science.1106148."

SaukaGSpengler," T.," and" M." BronnerGFraser." 2008." "A" gene" regulatory" network" orchestrates" neural" crest" formation."" " Nat& Rev& Mol& Cell& Biol" 9" (7):557G68." doi:" 10.1038/nrm2428."

110" " Sayan,"A."E.,"T."R."Griffiths,"R."Pal,"G."J."Browne," A." Ruddick," T." Yagci," R." Edwards," N." J." Mayer," H." Qazi," S." Goyal," S." Fernandez," K." Straatman," G." D." Jones," K." J." Bowman," A." Colquhoun,"J."K."Mellon,"M."Kriajevska,"and"E."Tulchinsky."2009.""SIP1"protein"protects" cells" from" DNA" damageGinduced" apoptosis" and" has" independent" prognostic" value" in" bladder" cancer."" " Proc& Natl& Acad& Sci& U& S& A" 106" (35):14884G9." doi:" 10.1073/pnas.0902042106."

Scapini," P.," M." Morini," C." Tecchio," S." Minghelli," E." Di" Carlo," E." Tanghetti," A." Albini," C." Lowell," G." Berton," D." M." Noonan," and" M." A." Cassatella." 2004." "CXCL1/macrophage" inflammatory"proteinG2Ginduced"angiogenesis"in"vivo"is"mediated"by"neutrophilGderived" vascular"endothelial"growth"factorGA."""J&Immunol"172"(8):5034G40."

Schoppmann,"S."F.,"P."Birner,"J."Stöckl,"R."Kalt,"R."Ullrich,"C."Caucig,"E."Kriehuber,"K."Nagy," K."Alitalo,"and"D."Kerjaschki."2002.""TumorGassociated"macrophages"express"lymphatic" endothelial" growth" factors" and" are" related" to" peritumoral" lymphangiogenesis."" " Am& J& Pathol"161"(3):947G56."doi:"10.1016/S0002G9440(10)64255G1."

Semenza,"G."L."2013.""HIFG1"mediates"metabolic"responses"to"intratumoral"hypoxia"and" oncogenic"mutations."""J&Clin&Invest"123"(9):3664G71."doi:"10.1172/JCI67230."

Semenza," G." L." 2014." "Oxygen" sensing," hypoxiaGinducible" factors," and" disease" pathophysiology."" " Annu& Rev& Pathol" 9:47G71." doi:" 10.1146/annurevGpatholG012513G 104720."

Semenza,"G."L.,"and"G."L."Wang."1992.""A"nuclear"factor"induced"by"hypoxia"via"de"novo" protein"synthesis"binds"to"the"human"erythropoietin"gene"enhancer"at"a"site"required"for" transcriptional"activation."""Mol&Cell&Biol"12"(12):5447G54."

Sennino,"B.,"T."IshiguroGOonuma,"Y."Wei,"R."M."Naylor,"C."W."Williamson,"V."Bhagwandin," S." P." Tabruyn," W." K." You," H." A." Chapman," J." G." Christensen," D." T." Aftab," and" D." M." McDonald." 2012." "Suppression" of" tumor" invasion" and" metastasis" by" concurrent" inhibition"of"cGMet"and"VEGF"signaling"in"pancreatic"neuroendocrine"tumors."""Cancer& Discov"2"(3):270G87."doi:"10.1158/2159G8290.CDG11G0240."

Shayan,"R.,"R."Inder,"T."Karnezis,"C."Caesar,"K."Paavonen,"M."W."Ashton,"G."B."Mann,"G."I." Taylor,"M."G."Achen,"and"S."A."Stacker."2013.""Tumor"location"and"nature"of"lymphatic" vessels"are"key"determinants"of"cancer"metastasis."""Clin&Exp&Metastasis"30"(3):345G56." doi:"10.1007/s10585G012G9541Gx."

Shields,"J."D.,"M."S."Emmett,"D."B."Dunn,"K."D."Joory,"L."M."Sage,"H."Rigby,"P."S."Mortimer,"A." Orlando," J." R." Levick," and" D." O." Bates." 2007." "ChemokineGmediated" migration" of" melanoma" cells" towards" lymphaticsGGa" mechanism" contributing" to" metastasis.""" Oncogene"26"(21):2997G3005."doi:"10.1038/sj.onc.1210114."

Shields,"J."D.,"M."E."Fleury,"C."Yong,"A."A."Tomei,"G."J."Randolph,"and"M."A."Swartz."2007." "Autologous" chemotaxis" as" a" mechanism" of" tumor" cell" homing" to" lymphatics" via" interstitial" flow" and" autocrine" CCR7" signaling."" " Cancer& Cell" 11" (6):526G38." doi:" 10.1016/j.ccr.2007.04.020."

Shinriki,"S.,"H."Jono,"M."Ueda,"K."Ota,"T."Ota,"T."Sueyoshi,"Y."Oike,"M."Ibusuki,"A."Hiraki,"H." Nakayama,"M."Shinohara,"and"Y."Ando."2011.""InterleukinG6"signalling"regulates"vascular"

111" " endothelial"growth"factorGC"synthesis"and"lymphangiogenesis"in"human"oral"squamous" cell"carcinoma."""J&Pathol"225"(1):142G50."doi:"10.1002/path.2935."

Shipitsin,"M.,"L."L."Campbell,"P."Argani,"S."Weremowicz,"N."BloushtainGQimron,"J."Yao,"T." Nikolskaya," T." Serebryiskaya," R." Beroukhim," M." Hu," M."K."Halushka,"S."Sukumar," L." M." Parker,"K."S."Anderson,"L."N."Harris,"J."E."Garber,"A."L."Richardson,"S."J."Schnitt,"Y."Nikolsky," R."S."Gelman,"and"K."Polyak."2007.""Molecular"definition"of"breast"tumor"heterogeneity.""" Cancer&Cell"11"(3):259G73."doi:"10.1016/j.ccr.2007.01.013."

Shojaei,"F.,"J."H."Lee,"B."H."Simmons,"A."Wong,"C."O."Esparza,"P."A."Plumlee,"J."Feng,"A."E." Stewart,"D."D."HuGLowe,"and"J."G."Christensen."2010.""HGF/cGMet"acts"as"an"alternative" angiogenic"pathway"in"sunitinibGresistant"tumors."""Cancer&Res"70"(24):10090G100."doi:" 10.1158/0008G5472.CANG10G0489."

Shojaei,"F.,"B."H."Simmons,"J."H."Lee,"P."B."Lappin,"and"J."G."Christensen."2012.""HGF/cGMet" pathway" is" one" of" the" mediators" of" sunitinibGinduced" tumor" cell" typeGdependent" metastasis."""Cancer&Lett"320"(1):48G55."doi:"10.1016/j.canlet.2012.01.026."

Shojaei,"F.,"X."Wu,"A."K."Malik,"C."Zhong,"M."E."Baldwin,"S."Schanz,"G."Fuh,"H."P."Gerber,"and" N." Ferrara." 2007." "Tumor" refractoriness" to" antiGVEGF" treatment" is" mediated" by" CD11b+Gr1+"myeloid"cells."""Nat&Biotechnol"25"(8):911G20."doi:"10.1038/nbt1323."

Shojaei,"F.,"X."Wu,"C."Zhong,"L."Yu,"X."H."Liang,"J."Yao,"D."Blanchard,"C."Bais,"F."V."Peale,"N." van"Bruggen,"C."Ho,"J."Ross,"M."Tan,"R."A."Carano,"Y."G."Meng,"and"N."Ferrara."2007.""Bv8" regulates" myeloidGcellGdependent" tumour" angiogenesis."" " Nature" 450" (7171):825G31." doi:"10.1038/nature06348."

Siegfried," G.," A." Basak," J." A." Cromlish," S." Benjannet," J." Marcinkiewicz," M." Chrétien," N." G." Seidah,"and"A."M."Khatib."2003.""The"secretory"proprotein"convertases"furin,"PC5,"and" PC7" activate" VEGFGC" to" induce" tumorigenesis."" " J& Clin& Invest" 111" (11):1723G32." doi:" 10.1172/JCI17220."

Siekmann," A." F.," and" N." D." Lawson." 2007." "Notch" signalling" limits" angiogenic" cell" behaviour" in" developing" zebrafish" arteries."" " Nature" 445" (7129):781G4." doi:" 10.1038/nature05577."

Singh," A.," and" J." Settleman." 2010." "EMT," cancer" stem" cells" and" drug" resistance:" an" emerging" axis" of" evil" in" the" war" on" cancer."" " Oncogene" 29" (34):4741G51." doi:" 10.1038/onc.2010.215."

Singh,"S."K.,"C."Hawkins,"I."D."Clarke,"J."A."Squire,"J."Bayani,"T."Hide,"R."M."Henkelman,"M."D." Cusimano,"and"P."B."Dirks."2004.""Identification"of"human"brain"tumour"initiating"cells.""" Nature"432"(7015):396G401."doi:"10.1038/nature03128."

Siva,"S.,"G."Kothari,"A."Muacevic,"A."V."Louie,"B."J."Slotman,"B."S."Teh,"and"S."S."Lo."2017." "Radiotherapy" for" renal" cell" carcinoma:" renaissance" of" an" overlooked" approach."" " Nat& Rev&Urol"14"(9):549G563."doi:"10.1038/nrurol.2017.87."

Skobe," M.," T." Hawighorst," D." G." Jackson," R." Prevo," L." Janes," P." Velasco," L." Riccardi," K." Alitalo," K." Claffey," and" M." Detmar." 2001." "Induction" of" tumor" lymphangiogenesis" by" VEGFGC"promotes"breast"cancer"metastasis."""Nat&Med"7"(2):192G8."doi:"10.1038/84643."

112" " Sobin," L." H.," and" C." C." Compton." 2010." "TNM" seventh" edition:" what's" new," what's" changed:"communication"from"the"International"Union"Against"Cancer"and"the"American" Joint"Committee"on"Cancer."""Cancer"116"(22):5336G9."doi:"10.1002/cncr.25537."

Soumaoro,"L."T.,"H."Uetake,"Y."Takagi,"S."Iida,"T."Higuchi,"M."Yasuno,"M."Enomoto,"and"K." Sugihara."2006.""Coexpression"of"VEGFGC"and"CoxG2"in"human"colorectal"cancer"and"its" association" with" lymph" node" metastasis."" " Dis& Colon& Rectum" 49" (3):392G8." doi:" 10.1007/s10350G005G0247Gx."

Sowter,"H."M.,"R."R."Raval,"J."W."Moore,"P."J."Ratcliffe,"and"A."L."Harris."2003.""Predominant" role" of" hypoxiaGinducible" transcription" factor" (Hif)G1alpha" versus" HifG2alpha" in" regulation"of"the"transcriptional"response"to"hypoxia."""Cancer&Res"63"(19):6130G4."

Srinivasan,"R."S.,"M."E."Dillard,"O."V."Lagutin,"F."J."Lin,"S."Tsai,"M."J."Tsai,"I."M."Samokhvalov," and"G."Oliver."2007.""Lineage"tracing"demonstrates"the"venous"origin"of"the"mammalian" lymphatic"vasculature."""Genes&Dev"21"(19):2422G32."doi:"10.1101/gad.1588407."

Srinivasan," R." S.," N." Escobedo," Y." Yang," A." Interiano," M." E." Dillard," D." Finkelstein," S." Mukatira,"H."J."Gil,"H."Nurmi,"K."Alitalo,"and"G."Oliver."2014.""The"Prox1GVegfr3"feedback" loop"maintains"the"identity"and"the"number"of"lymphatic"endothelial"cell"progenitors.""" Genes&Dev"28"(19):2175G87."doi:"10.1101/gad.216226.113."

Srinivasan,"R."S.,"X."Geng,"Y."Yang,"Y."Wang,"S."Mukatira,"M."Studer,"M."P."Porto,"O."Lagutin," and" G." Oliver." 2010." "The" nuclear" hormone" receptor" CoupGTFII" is" required" for" the" initiation" and" early" maintenance" of" Prox1" expression" in" lymphatic" endothelial" cells.""" Genes&Dev"24"(7):696G707."doi:"10.1101/gad.1859310."

Srinivasan,"R."S.,"and"G."Oliver."2011.""Prox1"dosage"controls"the"number"of"lymphatic" endothelial"cell"progenitors"and"the"formation"of"the"lymphovenous"valves."""Genes&Dev" 25"(20):2187G97."doi:"10.1101/gad.16974811."

Stacker,"S."A.,"C."Caesar,"M."E."Baldwin,"G."E."Thornton,"R."A."Williams,"R."Prevo,"D."G." Jackson,"S."Nishikawa,"H."Kubo,"and"M."G."Achen."2001.""VEGFGD"promotes"the"metastatic" spread"of"tumor"cells"via"the"lymphatics."""Nat&Med"7"(2):186G91."doi:"10.1038/84635."

Stanton,"M."J.,"S."Dutta,"H."Zhang,"N."S."Polavaram,"A."A."Leontovich,"P."Hönscheid,"F."A." Sinicrope," D." J." Tindall," M." H." Muders," and" K." Datta." 2013." "Autophagy" control" by" the" VEGFGC/NRPG2"axis"in"cancer"and"its"implication"for"treatment"resistance."""Cancer&Res" 73"(1):160G71."doi:"10.1158/0008G5472.CANG11G3635."

Steffens,"S.,"M."Janssen,"F."C."Roos,"F."Becker,"S."Schumacher,"C."Seidel,"G."Wegener,"J."W." Thüroff,"R."Hofmann,"M."Stöckle,"S."Siemer,"M."Schrader,"A."Hartmann,"M."A."Kuczyk,"K." Junker," and" A." J." Schrader." 2012." "Incidence" and" longGterm" prognosis" of" papillary" compared" to" clear" cell" renal" cell" carcinomaGGa" multicentre" study."" " Eur& J& Cancer" 48" (15):2347G52."doi:"10.1016/j.ejca.2012.05.002."

Strieter," R." M.," P." J." Polverini," S."L."Kunkel,"D."A."Arenberg,"M."D."Burdick,"J."Kasper," J." Dzuiba,"J."Van"Damme,"A."Walz,"and"D."Marriott."1995.""The"functional"role"of"the"ELR" motif"in"CXC"chemokineGmediated"angiogenesis."""J&Biol&Chem"270"(45):27348G57."

113" " Su,"J."L.,"J."Y."Shih,"M."L."Yen,"Y."M."Jeng,"C."C."Chang,"C."Y."Hsieh,"L."H."Wei,"P."C."Yang,"and"M." L." Kuo." 2004." "CyclooxygenaseG2" induces" EP1G" and" HERG2/NeuGdependent" vascular" endothelial"growth"factorGC"upGregulation:"a"novel"mechanism"of"lymphangiogenesis"in" lung"adenocarcinoma."""Cancer&Res"64"(2):554G64."

Su,"J."L.,"P."C."Yang,"J."Y."Shih,"C."Y."Yang,"L."H."Wei,"C."Y."Hsieh,"C."H."Chou,"Y."M."Jeng,"M."Y." Wang," K." J." Chang," M." C." Hung," and" M." L." Kuo." 2006." "The" VEGFGC/FltG4" axis" promotes" invasion" and" metastasis" of" cancer" cells."" " Cancer& Cell" 9" (3):209G23." doi:" 10.1016/j.ccr.2006.02.018."

Sun," S." Y.," L." M." Rosenberg," X." Wang," Z." Zhou," P." Yue," H." Fu," and" F." R." Khuri." 2005." "Activation" of" Akt" and" eIF4E" survival" pathways" by" rapamycinGmediated" mammalian" target" of" rapamycin" inhibition."" " Cancer& Res" 65" (16):7052G8." doi:" 10.1158/0008G 5472.CANG05G0917."

SánchezGTilló,"E.,"A."Lázaro,"R."Torrent,"M."Cuatrecasas,"E."C."Vaquero,"A."Castells,"P."Engel," and"A."Postigo."2010.""ZEB1"represses"EGcadherin"and"induces"an"EMT"by"recruiting"the" SWI/SNF" chromatinGremodeling" protein" BRG1."" " Oncogene" 29" (24):3490G500." doi:" 10.1038/onc.2010.102."

Takahashi,"A.,"K."Kono,"J."Itakura,"H."Amemiya,"R."Feng"Tang,"H."Iizuka,"H."Fujii,"and"Y." Matsumoto."2002.""Correlation"of"vascular"endothelial"growth"factorGC"expression"with" tumorGinfiltrating"dendritic"cells"in"gastric"cancer."""Oncology"62"(2):121G7."doi:"48257."

Tammela," T.," G." Zarkada," E." Wallgard," A." Murtomäki," S." Suchting," M." Wirzenius," M." Waltari," M." Hellström," T." Schomber," R." Peltonen," C." Freitas," A." Duarte," H." Isoniemi," P." Laakkonen," G." Christofori," S." YläGHerttuala," M." Shibuya," B." Pytowski," A." Eichmann," C." Betsholtz,"and"K."Alitalo."2008.""Blocking"VEGFRG3"suppresses"angiogenic"sprouting"and" vascular"network"formation."""Nature"454"(7204):656G60."doi:"10.1038/nature07083."

Tannir,"N."M.,"R."A."Figlin,"M."E."Gore,"M."D."Michaelson,"R."J."Motzer,"C."Porta,"B."I."Rini,"C." Hoang," X." Lin," and" B." Escudier." 2017." "LongGTerm" Response" to" Sunitinib" Treatment" in" Metastatic" Renal" Cell" Carcinoma:" A" Pooled" Analysis" of" Clinical" Trials."" " Clin&Genitourin& Cancer."doi:"10.1016/j.clgc.2017.06.005."

Tewalt,"E."F.,"J."N."Cohen,"S."J."Rouhani,"C."J."Guidi,"H."Qiao,"S."P."Fahl,"M."R."Conaway,"T."P." Bender,"K."S."Tung,"A."T."Vella,"A."J."Adler,"L."Chen,"and"V."H."Engelhard."2012.""Lymphatic" endothelial"cells"induce"tolerance"via"PDGL1"and"lack"of"costimulation"leading"to"highG level" PDG1" expression" on" CD8" T" cells."" " Blood" 120" (24):4772G82." doi:" 10.1182/bloodG 2012G04G427013."

Thiery," J." P.," H." Acloque," R." Y." Huang," and" M." A." Nieto." 2009." "EpithelialGmesenchymal" transitions" in" development" and" disease."" " Cell" 139" (5):871G90." doi:" 10.1016/j.cell.2009.11.007."

Timoshenko,"A."V.,"S."Rastogi,"and"P."K."Lala."2007.""MigrationGpromoting"role"of"VEGFGC" and"VEGFGC"binding"receptors"in"human"breast"cancer"cells."""Br&J&Cancer"97"(8):1090G8." doi:"10.1038/sj.bjc.6603993."

114" " Toivola,"D."M.,"G."Z."Tao,"A."Habtezion,"J."Liao,"and"M."B."Omary."2005.""Cellular"integrity" plus:" organelleGrelated" and" proteinGtargeting" functions" of" intermediate" filaments.""" Trends&Cell&Biol"15"(11):608G17."doi:"10.1016/j.tcb.2005.09.004."

Toschi,"A.,"E."Lee,"N."Gadir,"M."Ohh,"and"D."A."Foster."2008.""Differential"dependence"of" hypoxiaGinducible"factors"1"alpha"and"2"alpha"on"mTORC1"and"mTORC2."""J&Biol&Chem" 283"(50):34495G9."doi:"10.1074/jbc.C800170200."

Tsai," P." W.," S." G." Shiah," M." T." Lin," C." W." Wu," and" M." L." Kuo." 2003." "UpGregulation" of" vascular"endothelial"growth"factor"C"in"breast"cancer"cells"by"heregulinGbeta"1."A"critical" role"of"p38/nuclear"factorGkappa"B"signaling"pathway."""J&Biol&Chem"278"(8):5750G9."doi:" 10.1074/jbc.M204863200."

Tsui,"K."H.,"O."Shvarts,"R."B."Smith,"R."A."Figlin,"J."B."deKernion,"and"A."Belldegrun."2000." "Prognostic"indicators"for"renal"cell"carcinoma:"a"multivariate"analysis"of"643"patients" using"the"revised"1997"TNM"staging"criteria."""J&Urol"163"(4):1090G5;"quiz"1295."

Vallés," A." M.," B." Boyer," G." Tarone," and" J." P." Thiery." 1996." "Alpha" 2" beta" 1" integrin" is" required"for"the"collagen"and"FGFG1"induced"cell"dispersion"in"a"rat"bladder"carcinoma" cell"line."""Cell&Adhes&Commun"4"(3):187G99."

Van"Poppel,"H.,"L."Da"Pozzo,"W."Albrecht,"V."Matveev,"A."Bono,"A."Borkowski,"M."Colombel," L." Klotz," E." Skinner," T." Keane," S." Marreaud," S." Collette," and" R." Sylvester." 2011." "A" prospective," randomised" EORTC" intergroup" phase" 3" study" comparing" the" oncologic" outcome" of" elective" nephronGsparing" surgery" and" radical" nephrectomy" for" lowGstage" renal"cell"carcinoma."""Eur&Urol"59"(4):543G52."doi:"10.1016/j.eururo.2010.12.013."

Vandercappellen,"J.,"J."Van"Damme,"and"S."Struyf."2008.""The"role"of"CXC"chemokines"and" their" receptors" in" cancer."" " Cancer& Lett" 267" (2):226G44." doi:" 10.1016/j.canlet.2008.04.050."

Volpe," A.," G." Novara," A." Antonelli," R." Bertini," M." Billia," G." Carmignani," S." C." Cunico," N." Longo," G." Martignoni," A." Minervini," V." Mirone," A." Simonato," C." Terrone," F." Zattoni," V." Ficarra,"Surveillance"and"Treatment"Update"on"Renal"Neoplasms"(SATURN)"Project,"and" Leading" Urological" NoGProfit" Foundation" for" Advanced" Research" (LUNA)" Foundation." 2012.""Chromophobe"renal"cell"carcinoma"(RCC):"oncological"outcomes"and"prognostic" factors" in" a" large" multicentre" series."" " BJU& Int" 110" (1):76G83." doi:" 10.1111/j.1464G 410X.2011.10690.x." von"Hundelshausen,"P.,"F."Petersen,"and"E."Brandt."2007.""PlateletGderived"chemokines"in" vascular"biology."""Thromb&Haemost"97"(5):704G13." von"Rahden,"B."H.,"H."J."Stein,"F."Pühringer,"I."Koch,"R."Langer,"G."Piontek,"J."R."Siewert,"H." Höfler," and" M." Sarbia." 2005." "Coexpression" of" cyclooxygenases" (COXG1," COXG2)" and" vascular"endothelial"growth"factors"(VEGFGA,"VEGFGC)"in"esophageal"adenocarcinoma.""" Cancer&Res"65"(12):5038G44."doi:"10.1158/0008G5472.CANG04G1107."

Voulgari," A.," and" A." Pintzas." 2009." "EpithelialGmesenchymal" transition" in" cancer" metastasis:" mechanisms," markers" and" strategies" to" overcome" drug" resistance" in" the" clinic."""Biochim&Biophys&Acta"1796"(2):75G90."doi:"10.1016/j.bbcan.2009.03.002."

115" " Wang,"C."A.,"P."Jedlicka,"A."N."Patrick,"D."S."Micalizzi,"K."C."Lemmer,"E."Deitsch,"M."CasásG Selves," J." C." Harrell," and" H." L." Ford." 2012." "SIX1" induces" lymphangiogenesis" and" metastasis"via"upregulation"of"VEGFGC"in"mouse"models"of"breast"cancer."""J&Clin&Invest" 122"(5):1895G906."doi:"10.1172/JCI59858."

Wang,"H.,"H."S."Wang,"B."H."Zhou,"C."L."Li,"F."Zhang,"X."F."Wang,"G."Zhang,"X."Z."Bu,"S."H."Cai," and"J."Du."2013.""EpithelialGmesenchymal"transition"(EMT)"induced"by"TNFGα"requires" AKT/GSKG3βGmediated" stabilization" of" snail" in" colorectal" cancer."" " PLoS& One" 8" (2):e56664."doi:"10.1371/journal.pone.0056664."

Wang,"J.,"W."Zhao,"Y."Guo,"B."Zhang,"Q."Xie,"D."Xiang,"J."Gao,"B."Wang,"and"Z."Chen."2009." "The" expression" of" RNAGbinding" protein" HuR" in" nonGsmall" cell" lung" cancer" correlates" with" vascular" endothelial" growth" factorGC" expression" and" lymph" node" metastasis.""" Oncology"76"(6):420G9."doi:"10.1159/000216837."

Weikert," S.," H." Boeing," T." Pischon," C." Weikert," A." Olsen," A." Tjonneland," K." Overvad," N." Becker,"J."Linseisen,"A."Trichopoulou,"T."Mountokalakis,"D."Trichopoulos,"S."Sieri,"D."Palli," P." Vineis," S." Panico," P." H." Peeters," H." B." BuenoGdeGMesquita," W." M." Verschuren," B." Ljungberg,"G."Hallmans,"G."Berglund,"C."A."González,"M."Dorronsoro,"A."Barricarte,"M."J." Tormo,"N."Allen,"A."Roddam,"S."Bingham,"K."T."Khaw,"S."Rinaldi,"P."Ferrari,"T."Norat,"and"E." Riboli." 2008." "Blood" pressure" and" risk" of" renal" cell" carcinoma" in" the" European" prospective" investigation" into" cancer" and" nutrition."" " Am& J& Epidemiol" 167" (4):438G46." doi:"10.1093/aje/kwm321."

Welti,"J.,"S."Loges,"S."Dimmeler,"and"P."Carmeliet."2013.""Recent"molecular"discoveries"in" angiogenesis" and" antiangiogenic" therapies" in" cancer."" " J&Clin&Invest" 123" (8):3190G200." doi:"10.1172/JCI70212."

Wenzel," J.," B." Bekisch," M." Uerlich," O." Haller," T." Bieber," and" T." Tüting." 2005." "Type" I" interferonGassociated" recruitment" of" cytotoxic" lymphocytes:" a" common" mechanism" in" regressive" melanocytic" lesions."" " Am& J& Clin& Pathol" 124" (1):37G48." doi:" 10.1309/4EJ9KL7CGDENVVLE."

Wigle,"J."T.,"N."Harvey,"M."Detmar,"I."Lagutina,"G."Grosveld,"M."D."Gunn,"D."G."Jackson,"and" G."Oliver."2002.""An"essential"role"for"Prox1"in"the"induction"of"the"lymphatic"endothelial" cell"phenotype."""EMBO&J"21"(7):1505G13."doi:"10.1093/emboj/21.7.1505."

Wigle,"J."T.,"and"G."Oliver."1999.""Prox1"function"is"required"for"the"development"of"the" murine"lymphatic"system."""Cell"98"(6):769G78."

Willers," H.," C." G." Azzoli," W." L." Santivasi," and" F." Xia." 2013." "Basic" mechanisms" of" therapeutic" resistance" to" radiation" and" chemotherapy" in" lung" cancer."" " Cancer& J" 19" (3):200G7."doi:"10.1097/PPO.0b013e318292e4e3."

Wilson," W." R.," and" M." P." Hay." 2011." "Targeting" hypoxia" in" cancer" therapy."" " Nat& Rev& Cancer"11"(6):393G410."doi:"10.1038/nrc3064."

Witta,"S."E.,"R."M."Gemmill,"F."R."Hirsch,"C."D."Coldren,"K."Hedman,"L."Ravdel,"B."Helfrich,"R." Dziadziuszko," D." C." Chan," M." Sugita," Z." Chan," A." Baron," W." Franklin," H." A." Drabkin," L." Girard,"A."F."Gazdar,"J."D."Minna,"and"P."A."Bunn."2006.""Restoring"EGcadherin"expression"

116" " increases"sensitivity"to"epidermal"growth"factor"receptor"inhibitors"in"lung"cancer"cell" lines."""Cancer&Res"66"(2):944G50."doi:"10.1158/0008G5472.CANG05G1988."

Xian,"X.,"J."Håkansson,"A."Ståhlberg,"P."Lindblom,"C."Betsholtz,"H."Gerhardt,"and"H."Semb." 2006." "Pericytes" limit" tumor" cell" metastasis."" " J& Clin& Invest" 116" (3):642G51." doi:" 10.1172/JCI25705."

Xiang," L.," G." Xie," J." Ou," X." Wei," F." Pan," and" H." Liang." 2012." "The" extra" domain" A" of" fibronectin" increases" VEGFGC" expression" in" colorectal" carcinoma" involving" the" PI3K/AKT" signaling" pathway."" " PLoS& One" 7" (4):e35378." doi:" 10.1371/journal.pone.0035378."

Xu,"J.,"S."Lamouille,"and"R."Derynck."2009.""TGFGbetaGinduced"epithelial"to"mesenchymal" transition."""Cell&Res"19"(2):156G72."doi:"10.1038/cr.2009.5."

Xu,"Y.,"L."Yuan,"J."Mak,"L."Pardanaud,"M."Caunt,"I." Kasman," B." Larrivée," R." Del" Toro," S." Suchting,"A."Medvinsky,"J."Silva,"J."Yang,"J."L."Thomas,"A."W."Koch,"K."Alitalo,"A."Eichmann," and" A." Bagri." 2010." "NeuropilinG2" mediates" VEGFGCGinduced" lymphatic" sprouting" together"with"VEGFR3."""J&Cell&Biol"188"(1):115G30."doi:"10.1083/jcb.200903137."

Yamashita,"T.,"J."Ji,"A."Budhu,"M."Forgues,"W."Yang,"H."Y."Wang,"H."Jia,"Q."Ye,"L."X."Qin,"E." Wauthier,"L."M."Reid,"H."Minato,"M."Honda,"S."Kaneko,"Z."Y."Tang,"and"X."W."Wang."2009." "EpCAMGpositive" hepatocellular" carcinoma" cells" are" tumorGinitiating" cells" with" stem/progenitor" cell" features."" " Gastroenterology" 136" (3):1012G24." doi:" 10.1053/j.gastro.2008.12.004."

Yamazaki,"T.,"Y."Yoshimatsu,"Y."Morishita,"K."Miyazono,"and"T."Watabe."2009.""COUPGTFII" regulates" the" functions" of" Prox1" in" lymphatic" endothelial" cells" through" direct" interaction."""Genes&Cells"14"(3):425G34."doi:"10.1111/j.1365G2443.2008.01279.x."

Yang," F.," L." Sun," Q." Li," X." Han," L." Lei," H." Zhang," and" Y." Shang." 2012." "SET8" promotes" epithelialGmesenchymal" transition" and" confers" TWIST" dual" transcriptional" activities.""" EMBO&J"31"(1):110G23."doi:"10.1038/emboj.2011.364."

Yang,"L.,"L."M."DeBusk,"K."Fukuda,"B."Fingleton,"B."GreenGJarvis,"Y."Shyr,"L."M."Matrisian,"D." P."Carbone,"and"P."C."Lin."2004.""Expansion"of"myeloid"immune"suppressor"Gr+CD11b+" cells" in" tumorGbearing" host" directly" promotes" tumor" angiogenesis."" " Cancer& Cell" 6" (4):409G21."doi:"10.1016/j.ccr.2004.08.031."

Yang,"M."H.,"and"K."J."Wu."2008.""TWIST"activation"by"hypoxia"inducible"factorG1"(HIFG1):" implications" in" metastasis" and" development."" " Cell& Cycle" 7" (14):2090G6." doi:" 10.4161/cc.7.14.6324."

Yang," Y.," J." M." GarcíaGVerdugo," M." SorianoGNavarro," R." S." Srinivasan," J." P." Scallan," M." K." Singh,"J."A."Epstein,"and"G."Oliver."2012.""Lymphatic"endothelial"progenitors"bud"from"the" cardinal"vein"and"intersomitic"vessels"in"mammalian"embryos."""Blood"120"(11):2340G8." doi:"10.1182/bloodG2012G05G428607."

Yaniv," K.," S." Isogai," D." Castranova," L." Dye," J." Hitomi," and" B." M." Weinstein." 2006." "Live" imaging" of" lymphatic" development" in" the" zebrafish."" " Nat& Med" 12" (6):711G6." doi:" 10.1038/nm1427."

117" " Ye," J.," X." Wu," D." Wu," P." Wu," C." Ni," Z." Zhang," Z." Chen," F." Qiu," J." Xu," and" J." Huang." 2013." "miRNAG27b"targets"vascular"endothelial"growth"factor"C"to"inhibit"tumor"progression" and" angiogenesis" in" colorectal" cancer."" " PLoS& One" 8" (4):e60687." doi:" 10.1371/journal.pone.0060687."

Yilmaz,"M.,"and"G."Christofori."2009.""EMT,"the"cytoskeleton,"and"cancer"cell"invasion.""" Cancer&Metastasis&Rev"28"(1G2):15G33."doi:"10.1007/s10555G008G9169G0."

Yoshimatsu,"Y.,"Y."G."Lee,"Y."Akatsu,"L."Taguchi,"H."I."Suzuki,"S."I."Cunha,"K."Maruyama,"Y." Suzuki,"T."Yamazaki,"A."Katsura,"S."P."Oh,"T."A."Zimmers,"S."J."Lee,"K."Pietras,"G."Y."Koh,"K." Miyazono," and" T." Watabe." 2013." "Bone" morphogenetic" proteinG9" inhibits" lymphatic" vessel" formation" via" activin" receptorGlike" kinase" 1" during" development" and" cancer" progression."" " Proc& Natl& Acad& Sci& U& S& A" 110" (47):18940G5." doi:" 10.1073/pnas.1310479110."

You,"L."R.,"F."J."Lin,"C."T."Lee,"F."J."DeMayo,"M."J."Tsai,"and"S."Y."Tsai."2005.""Suppression"of" Notch"signalling"by"the"COUPGTFII"transcription"factor"regulates"vein"identity."""Nature" 435"(7038):98G104."doi:"10.1038/nature03511."

Yu,"F.,"S."B."White,"Q."Zhao,"and"F."S."Lee."2001.""HIFG1alpha"binding"to"VHL"is"regulated" by"stimulusGsensitive"proline"hydroxylation."""Proc&Natl&Acad&Sci&U&S&A"98"(17):9630G5." doi:"10.1073/pnas.181341498."

Zavadil,"J.,"M."Bitzer,"D."Liang,"Y."C."Yang,"A."Massimi,"S."Kneitz,"E."Piek,"and"E."P."Bottinger." 2001." "Genetic" programs" of" epithelial" cell" plasticity" directed" by" transforming" growth" factorGbeta."""Proc&Natl&Acad&Sci&U&S&A"98"(12):6686G91."doi:"10.1073/pnas.111614398."

Zavadil," J.," and" E." P." Böttinger." 2005." "TGFGbeta" and" epithelialGtoGmesenchymal" transitions."""Oncogene"24"(37):5764G74."doi:"10.1038/sj.onc.1208927."

Zhang,"H.,"M."H."Muders,"J."Li,"F."Rinaldo,"D."J."Tindall,"and"K."Datta."2008.""Loss"of"NKX3.1" favors"vascular"endothelial"growth"factorGC"expression"in"prostate"cancer."""Cancer&Res" 68"(21):8770G8."doi:"10.1158/0008G5472.CANG08G1912."

Zhang," L.," M." Bhasin," R." SchorGBardach," X." Wang," M." P." Collins," D." Panka," P." Putheti," S." Signoretti,"D."C."Alsop,"T."Libermann,"M."B."Atkins,"J."W."Mier,"S."N."Goldberg,"and"R."S." Bhatt."2011.""Resistance"of"renal"cell"carcinoma"to"sorafenib"is"mediated"by"potentially" reversible" gene" expression."" " PLoS& One" 6" (4):e19144." doi:" 10.1371/journal.pone.0019144."

Zhang,"X."H.,"D."P."Huang,"G."L."Guo,"G."R."Chen,"H."X."Zhang,"L."Wan,"and"S."Y."Chen."2008." "Coexpression" of" VEGFGC" and" COXG2" and" its" association" with" lymphangiogenesis" in" human"breast"cancer."""BMC&Cancer"8:4."doi:"10.1186/1471G2407G8G4."

Zhu,"X.,"L."Sun,"J."Lan,"L."Xu,"M."Zhang,"X."Luo,"J."Gong,"G."Wang,"X."Yuan,"J."Hu,"and"J."Wang." 2016." "BRG1" targeting" STAT3/VEGFC" signaling" regulates" lymphangiogenesis" in" colorectal"cancer."""Oncotarget"7"(24):36501G36509."doi:"10.18632/oncotarget.9038."

Zini,"L.,"P."Perrotte,"C."Jeldres,"U."Capitanio,"D."Pharand,"P."Arjane,"S."Lapointe,"F."Montorsi," J."J."Patard,"and"P."I."Karakiewicz."2008.""Nephrectomy"improves"the"survival"of"patients"

118" " with" locally" advanced" renal" cell" carcinoma."" " BJU& Int" 102" (11):1610G4." doi:" 10.1111/j.1464G410X.2008.07917.x."

Zlotnik,"A.,"and"O."Yoshie."2000.""Chemokines:"a"new"classification"system"and"their"role" in"immunity."""Immunity"12"(2):121G7."

Şenbabaoğlu,"Y.,"R."S."Gejman,"A."G."Winer,"M."Liu,"E."M."Van"Allen,"G."de"Velasco,"D."Miao," I." Ostrovnaya," E." Drill," A." Luna," N." Weinhold," W." Lee," B." J." Manley," D." N." Khalil," S." D." Kaffenberger,"Y."Chen,"L."Danilova,"M."H."Voss,"J."A."Coleman,"P."Russo,"V."E."Reuter,"T."A." Chan,"E."H."Cheng,"D."A."Scheinberg,"M."O."Li,"T."K."Choueiri,"J."J."Hsieh,"C."Sander,"and"A."A." Hakimi." 2016." "Tumor" immune" microenvironment" characterization" in" clear" cell" renal" cell" carcinoma" identifies" prognostic" and" immunotherapeutically" relevant" messenger" RNA"signatures."""Genome&Biol"17"(1):231."doi:"10.1186/s13059G016G1092Gz." " " " " " " " " " " " " " " " " " " " " " " " " " 119" " ANNEXES!!

" " " " " " " " " " " " " " " " " " " " " " " " " " !

120" " Annexe!1!: Effects of proton versus photon irradiation on (lymph) angiogenic, inflammatory, proliferative and anti-tumor immune responses in head and neck squamous cell carcinoma

! Durant"ces"trois"années"de"thèse"j’ai"également"participé"à"un"projet"sur"la"comparaison" de" la" radiothérapie" par" protons" et" " photons" dans" les" cancers" de" la" tête" et" du" cou" (HNSCC)."La"radiothérapie"induit"la"dérégulation"des"voies"de"signalisation"cellulaire"qui" entrainent" des" résistances" (Willers" et" al." 2013)." Nous" avons" montré" que" des" cellules" résistantes"aux"photons"ou"aux"protons"pouvaient"être"obtenues."Ces"cellules"prolifèrent" moins"vite"in"vitro"par"rapport"aux"cellules"contrôles"et"génère"des"tumeurs"dont"la"taille" est" plus" importante" que" celle" obtenue" avec" des" cellules" non" irradiées." Cependant," les" tumeurs" générées" avec" les" cellules" insensibles" aux" protons" ne" développent" pas" de" vaisseaux" lymphatiques" alors" que" celles" générées" avec" les" cellules" résistantes" aux" photons" contiennent" en" leur" centre" des" vaisseaux" lymphatiques." " Ce" résultat" dépend" d’un" profil" sécrétoire" différent" entre" les" deux" types" cellulaires" notamment" de" la" présence"de"VEGFGC"dans"les"cellules"résistantes"aux"photons."La"présence"de"vaisseaux" lymphatiques" dans" des" tumeurs" de" patients" en" récidive" locale" après" radiothérapie" donne"un"rationnel"clinique"à"nos"résultats"sur"des"tumeurs"expérimentales." Nos"résultats"mettent"en"évidence"un"mécanisme"moléculaire"nouveau"lié"aux"rechutes" observées"postGradiothérapie"de"patients"avec"des"tumeurs"HNSCC."Ce"travail"met"donc" en" évidence" un" nouveau" marqueur" prédictif" d’évolution" péjorative" de" ces" cancers" et" potentiellement"une"cible"thérapeutique"pertinente."" " " " " " " ! ! ! ! !

121" " OPEN Citation: Oncogenesis (2017) 6,e354;doi:10.1038/oncsis.2017.56 www.nature.com/oncsis

ORIGINAL ARTICLE Effects of proton versus photon irradiation on (lymph) angiogenic, inﬂammatory, proliferative and anti-tumor immune responses in head and neck squamous cell carcinoma

M Lupu-Plesu1,2,7, A Claren1,2,3,4,7, S Martial1,2, P-D N’Diaye1,2, K Lebrigand1,5, N Pons1,5, D Ambrosetti1,6, I Peyrottes4, J Feuillade4, J Hérault4, M Duﬁes1,2, J Doyen1,2,4 and G Pagès1,2

The proximity of organs at risk makes the treatment of head and neck squamous cell carcinoma (HNSCC) challenging by standard radiotherapy. The higher precision in tumor targeting of proton (P) therapy could promote it as the treatment of choice for HNSCC. Besides the physical advantage in dose deposition, few is known about the biological impact of P versus photons (X) in this setting. To investigate the comparative biological effects of P versus X radiation in HNSCC cells, we assessed the relative biological effectiveness (RBE), viability, proliferation and mRNA levels for genes involved in (lymph)angiogenesis, inflammation, proliferation and anti-tumor immunity. These parameters, particularly VEGF-C protein levels and regulations, were documented in freshly irradiated and/or long-term surviving cells receiving low/high-dose, single (SI)/multiple (MI) irradiations with P/X. The RBE was found to be 1.1 Key (lymph)angiogenesis and inflammation genes were downregulated (except for vegf-c) after P and upregulated after X irradiation in MI surviving cells, demonstrating a more favorable profile after P irradiation. Both irradiation types stimulated vegf-c promoter activity in a NF-κB-dependent transcriptional regulation manner, but at a lesser extent after P, as compared to X irradiation, which correlated with mRNA and protein levels. The cells surviving to MI by P or X generated tumors with higher volume, anarchic architecture and increased density of blood vessels. Increased lymphangiogenesis and a transcriptomic analysis in favor of a more aggressive phenotype were observed in tumors generated with X-irradiated cells. Increased detection of lymphatic vessels in relapsed tumors from patients receiving X radiotherapy was consistent with these findings. This study provides new data about the biological advantage of P, as compared to X irradiation. In addition to its physical advantage in dose deposition, P irradiation may help to improve treatment approaches for HNSCC.

Oncogenesis (2017) 6, e354; doi:10.1038/oncsis.2017.56; published online 3 July 2017

INTRODUCTION tumors based on the advantage to deliver a much lower integral Approximately 50% of all cancer patients are subject to dose, which significantly reduces the risk of radiation induced 2 radiotherapy during the course of their illness with an estimation cancers in a long-life expectancy setting. Several retrospective that radiotherapy contributes to approximately 40% towards and dosimetry studies have suggested an advantage of P 1 radiotherapy in other tumors located near organs at risk, such as curative treatment. The goal of radiotherapy is to deliver 2 loco-regionally a specific dose of radioactivity that will allow the the head and neck squamous cell carcinoma (HNSCC). destruction of cancer cells, while limiting the exposure of The head and neck cancers are among the 10 most common types of cancer and the seventh cause of mortality from cancer surrounding healthy tissues. Among the ionizing radiation worldwide. Depending on disease stage, the treatment of HNSCC treatments, the large majority consists of photons (X) of 3 – consists of either chemoradiotherapy and/or surgical excision. high energy (5 20 MeV). However, the main disadvantage of X However, conventional radiotherapy with X in HNSCC remains radiotherapy is represented by the deposition of radiation also at difficult, due to the proximity of numerous organs at risk (that is, the level of surrounding healthy tissues, leading to side effects. salivary glands, esophagus and larynx). Recent studies have shown Although the ionizing radiation by proton beams (P) is currently an advantage of P, over X radiotherapy, in inducing lower more expensive and more difficult to produce, it has the physical toxicities4 and lower dose delivery to organs of risk5 in HNSCC advantage of delivering no radiation outside of the intended patients. targeted area, thanks to the so-called Bragg peak.2 Despite of the currently available therapeutic strategies, the P radiotherapy is mainly proposed for the treatment of uveal five-year overall survival rate of HNSCC patients is only 53%6 melanoma, skull base and paraspinal tumors due to its high because of a high percentage of a poor response to therapy and precision in tumor targeting with a very high irradiation dose next high recurrence rates. Sentinel lymph node metastasis, the first to radiosensitive structures.2 It is also proposed for the pediatric sign of tumor progression, was directly correlated to prognosis in

1University of Nice Sophia Antipolis, Nice, France; 2Institute for Research on Cancer and Aging, Nice, CNRS UMR 7284, INSERM U1081, Nice, France; 3University of Aix Marseille, Marseille, France; 4Centre Antoine Lacassagne, Nice, France; 5University of the Côte d'Azur, CNRS, Institute of Molecular and Cellular Pharmacology, Sophia Antipolis, France and 6Nice University Hospital, Nice, France. Correspondence: Dr G Pagès, CNRS UMR7284, INSERM U 1081, Institute for Research on Cancer and Aging, Nice UNS 28 Avenue de Valombrose, 06107 Nice, France. E-mail: [email protected] 7These authors contributed equally to this work. Received 27 February 2017; revised 7 May 2017; accepted 10 May 2017 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 2 7 HNSCC patients. Vascular endothelial growth factor C (VEGF-C) is 15 * ** a major pro-lymphangiogenic factor responsible for the metastatic dissemination of cancer cells.8 A significant correlation has been ### observed between intra-tumor lymphatic vessel density and ### 9 10 ### lymph node metastasis in patients with HNSCC. Moreover, ** VEGF-C expression levels correlated with lymphatic vessel density ### and lymph node metastasis in these patients.10,11 VEGF-C- # dependent development of the lymphatic network might also ### Cell number 5 be the major route of spread of tumor cells when the patients increase) (Fold ** * become resistant to therapy.8 # Beside the physical advantage of P versus X irradiation and the RBE, few comparative preclinical studies have been conducted 0 that contrast cellular/biological response to P versus X CT X2 X8 P2 P8 CT X2 X8 P2 P8 CT X2 X8 P2 P8 radiations.12–15 48 h 72 h 96 h P irradiation led to distinct gene and protein expression Figure 1. CAL33 proliferative ability following multiple X or P profiles.12 Mice receiving total-body irradiation with either P or X irradiations. Counts of CAL33 cells following multiple low (2 Gy) or had enhanced plasma levels of transforming growth factor-β, high (8 Gy) dose(s) of P or X irradiation and cell expansion after only after X irradiation.13 Moreover, X irradiation promoted the third irradiation (CR-MI). The values correspond to fold increase, angiogenesis, thus enhancing metastasis by upregulation of as compared to the viable cell number at 24 h after cell seeding. 14 Significantly decreased viable cell counts, as compared to CT: #, various pro-angiogenic factors. By contrast, low dose P Po0.05; ###, Po0.001. Significantly increased viable cell counts for irradiation did not induce the pro-angiogenic and comparisons between X and P groups: *, Po0.05; **, Po0.01. CT, pro-inflammatory genes, impaired tumor cell invasion in vitro control (non-irradiated cells). and attenuated tumor growth rate in mice.14 By downregulating integrins and matrix-metalloproteinases (MMP), P irradiation also the same biological parameters on cells that have survived to 15 reduced invasive and migratory properties of tumor cells. multiple irradiations (MI) and that have been expanded as new Therefore, beside the physical advantage in dose deposition, P populations were qualified as the ‘chronic response (CR)’. may have different biological properties, as compared to X In order to calibrate our experiments, we first determined a radiation at a similar dose. The purpose of the present study was relative biological effectiveness (RBE) of photons and protons on thus to analyze the different biological behaviors of HNSCC cells our model cell lines following SI. According to the literature, P when exposed to P versus X radiation. The study focused on therapy treatments are based on a RBE of 1.1, relative to molecules with key roles in the progression and prognosis of high-energy X therapy.25 The surviving curve of CAL33 cells HNSCC, such as the inflammatory cytokines: Interleukin 6 (IL6),16 following administration of escalating doses of either P or Interleukin 8 (IL8);17 (lymph)angiogenic factors: VEGF A, C and D X irradiation confirmed a RBE of 1.1 for P, as compared to X and their receptors: vascular endothelial growth factor receptor irradiation (Supplementary Figure S1). This experiment confirms (VEGFR) 1, 2 and 3, Neuropilin (NRP) 1 and 2;18,19 factors involved the literature data showing that P kills tumor cells more efficiently in lymphatic vessels development: lymphatic vessel endothelial than X irradiation.25 hyaluronan receptor 1 (LYVE1), prospero homeobox 1 (PROX1) However, patients are irradiated several times to reach a transcription factor, and podoplanin (PDPN), a mucin-type maximal therapeutic efficacy. Therefore, our next purpose was transmembrane protein20; pro-inflammatory chemokine C-C Motif to compare the relative aggressiveness of cells that were resistant Chemokine Ligand 2 (CCL2) involved in cell migration21; cell cycle to MI by X or P. Hence, we performed our experiments on two regulators: polo-like kinase 1 (PLK1)22 and telomeric independent cell lines (CAL33 and CAL27). The proliferative ability repeat-binding factor 2 (TRF2) transcription factor23; immune along a time course of CAL33 (Figure 1) or CAL27 (Supplementary checkpoint molecule programmed death-ligand 1 (PD-L1) Figure S2) that have survived to MI was determined. As compared involved in anergy and tumor progression.24 to non-irradiated cells, the proliferation of X or P irradiated cells The role of the above-mentioned molecules in the was reduced in both models and the difference was striking 96 h post-irradiation progression of HNSCC has not been elucidated. following cell seeding (Po0.001 for CAL33; P = 0.049 for CAL27). Our working hypothesis was that different radiation types would However, the difference in proliferation became statistically lead to different intrinsic and extrinsic biological responses, significant earlier for X irradiated cells in the CAL33 model allowing the adaptation of tumor cells. Therefore, we studied (P = 0.02 for X8 at 48 h; P = 0.014 and 0.009 for X2 and X8, the impacts of P versus X irradiation on human HNSCC cells respectively, at 72 h; Po0.001 for all conditions at 96 h). Whereas viability; proliferation; whole transcriptome profile and expression the difference in proliferation did not reach statistical significance of key genes/proteins implicated in (lymph)angiogenesis/metas- between X2 and X8 irradiated cells, X8 cells proliferated to a lesser tasis, inflammation, tumor cell proliferation and anti-tumor extent, as compared to P2 and P8, at 48 h post seeding (P = 0.006 immunity; tumorigenic potential, and depicted the molecular and P = 0.035, respectively), to P2 at 72 h post seeding (P = 0.018), mechanisms of post-irradiation VEGF-C regulation, to set the basis and P2 and P8 irradiated cells at 96 h post seeding (P = 0.012 and for improved therapeutic approaches for HNSCC. P = 0.008, respectively). Therefore, the overall therapeutic advantage, attested by reduced cell viability and proliferation capacity following SI RESULTS switched in favor of X post MI for CAL33 cells. For CAL27, no Cell survival/proliferation is in favor of P following single difference in the proliferative ability of MI X and P cells was irradiation, and X following multiple irradiations observed, suggesting that X and P exert different outcomes, Our hypothesis was that irradiation would lead to different cell depending on the HNSCC type. viability and proliferation profiles depending on the radiation type and dose, number of irradiations and time of assessment. PirradiationleadstooveralllowerinductionofmRNAcoding We qualified as the ‘acute response (AR)’ the modifications of pro-inflammatory, pro-(lymph)angiogenic and pro-proliferative genes biological parameters (proliferation, survival, gene expression) a The gene expression levels for CAL33 cells following SI or MI, few hours following a single irradiation (SI). The modifications of represented as percentage of control, and the gene expression

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 3 (Po0.0001) and IL-8 (Po0.0001). By contrast, among these Table 1. Quantitative gene expression, as percentage of control genes, X irradiation led to downregulation of IL-8 only, within the (0 Gy), in either P or X irradiated CAL33 cells belonging to (a) AR-SI and low dose CR-MI settings. Notably, VEGF-C mRNA levels were (b) CR-MI groups systematically increased after both P and X irradiation, but they fi (a) mRNA levels (% of control) − AR − SI X P were signi cantly lower after P, as compared to X irradiation, after high dose within the CR-MI setting (po0.001). Among all investigated genes, IL-8 was the gene whose mRNA was induced Role Gene 2 Gy 8 Gy 2 Gy 8 Gy at the highest level after X (79-fold, as compared to control), (Lymph)angiogenesis and VEGF-A 132 a191 143 160 but not after P irradiation, within the high dose CR-MI setting metastasis (Po0.0001). Moreover, both P and X irradiation augmented PD-L1 VEGF-C 130 a227 159 a209 mRNA expression in a dose-dependent manner within the AR-SI VEGF-D 103 96 105 98 and CR-MI settings, and in an irradiation number-dependent manner within the AR-SI setting. The generated gene expression ab Inflammation IL6 129 136 119 72 scores showed that P irradiation is associated with a more IL8 147 a436 162 a401 fi b a favorable pro le (reduced proliferation, (lymph)angiogenesis, CCL2 104 123 79 136 inflammation)). A similar gene score, in favor of P irradiation, Proliferation TRF-2 88 a112 93 102 was also obtained for CAL27 cells, within the CR-MI setting, Plk-1 85 a108 89 a99 despite an increase in VEGF-C, VEGF-D, NRP1, NRP2, IL-8 and PD-L1 mRNA expression (Supplementary Table S3). Anti-tumor immunity PD-L1 108 103 91 112 Induction of VEGF-C protein is reduced in P irradiated cells Gene score 4603 Because lymph node metastasis is frequent at diagnosis in HNSCC and in patients who relapse locally after radiotherapy, we focused (b) mRNA levels XP our research on VEGF-C, the major growth factor for lymphatic (% of control)–CR–MI endothelial cells. Although the mRNA levels of VEGF-C were increased after both low and high dose(s) of P or X irradiation, Role Gene 2 Gy 8 Gy 2 Gy 8 Gy they were lower after high dose(s) of P irradiation. To confirm the results obtained at mRNA level, we next assessed VEGF-C protein a b ab (Lymph)angiogenesis VEGF-A 118 208 93 107 levels in CAL33 and CAL27 cells. and metastasis a b In CAL33 cells, VEGF-C protein levels increased in a VEGF-C 281 626 226 215 dose-dependent manner following both P and X irradiation. VEGF-D 127 134 b95 b79 a b Furthermore, they were significantly lower after P irradiation. Within VEGFR-1 96 73 72 73 fi VEGFR-2 81 88 76 77 the AR-SI setting (Figure 2a), VEGF-C protein levels were signi cantly VEGFR-3 116 91 100 80 increased after a low and high dose of irradiation with either P NRP1 105 90 b92 85 (P =0.038and P =0.046,respectively)orX (P = 0.0002 for both dose NRP2 133 a169 b102 b129 types). A significantly lower expression was observed after a high dose of P, as compared to X irradiation (by 59%, P = 0.018). However, b ab Inflammation IL6 120 126 74 100 significantly increased levels were observed after a high versus low IL8 87 a7916 b75 b90 a b b dose of X irradiation (3-fold increase, P =0.002). CCL2 112 751 83 94 The VEGF-C protein induction was also maintained at signifi- Proliferation TRF-2 101 95 95 106 cantly increased levels in CAL33 cells of the CR-MI group Plk-1 91 97 89 a117 (Figure 2b), after both low and high doses of P and X irradiation (Po0.001), with significantly decreased levels after high doses of Anti-tumor immunity PD-L1 119 140 b92 a116 P versus X irradiation (by 50%, Po0.001). In addition, there were significantly increased levels after high, as compared to low doses Gene score 4 6 − 8 − 4 of X irradiation (P = 0.002). These observations were confirmed in — fi o CAL27 cells within the CR-MI setting (Supplementary Figure S2B), Highlighted values signi cantly different (P 0.05) expression levels, as fi compared to control, for genes associated to favorable (dark gray) and where VEGF-C protein levels were signi cantly increased after non-favorable (black) outcomes. asignificantly different expression levels both P and X irradiations (Po0.001), with lower levels after high after low, as compared to high dose(s) of either P or X irradiation. doses of P versus X irradiation (P = 0.001). bsignificantly different expression levels after either low or high dose(s) of P, as compared to X irradiation. X and P irradiations stimulate the VEGF-C promoter activity Irradiation by either X or P stimulated the activity of the vegf-c promoter especially in CAL33 cells surviving to multiple X scores are listed in Table 1a and b. The mRNA levels of the irradiations (6- and 18-fold increase, respectively, Po0.001, different tested genes overall increased in a dose-dependent Figure 2c). This result is consistent with the induction of the manner and with the irradiation number after both P and X VEGF-C mRNA within the CR-MI setting (Table 1) and suggests a irradiation. Genes involved in (lymph)angiogenesis, inflammation chronic induction of vegf-c gene transcription, an increase in vegf-c and immune tolerance were overall less expressed after high dose mRNA half-life or a combination of both mechanisms. Mutation of (s) of P, as compared to X, irradiation in all investigated groups; the NF-κB binding site (MUT) had no effect on the basal vegf-c the genes involved in (lymph)angiogenesis, inflammation and promoter activity in non-irradiated cells. However, in cells immune tolerance were downregulated after P irradiation, surviving to MI by P and X, the activity of the MUT, as compared showing significantly lower mRNA levels, as compared to X to WT, promoter was significantly decreased (by 33%, P = 0.004 irradiation, within the following settings: (i) AR-SI after low dose: and by 30%, P = 0.027, respectively, Figure 2c) suggesting that the CCL2 (P = 0.035) and high dose: IL-6 (P = 0.0001); (ii) CR-MI after increase in the transcriptional activation of the vegf-c promoter low dose: VEGF-A (Po0.0001), IL-6 (Po0.0001), IL-8 (P = 0.046), depends in part on a constitutive activation of NF-κB. In the CAL27 CCL2 (P = 0.041), PD-L1 (P = 0.002) and high dose: VEGF-D cell line, the irradiation by either P or X did not stimulate the

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 4

Figure 2. VEGF-C protein expression levels and regulation in CAL33 cells following P or X irradiation. (a) VEGF-C protein levels at 48 h post- single irradiation (AR-SI): * and *, significantly (Po0.05) increased levels after a low (2 Gy) or high (8 Gy) dose of P and X irradiation, respectively, as compared to CT; #, significantly decreased levels after a high dose of P, as compared to X irradiation; §, significantly increased levels after a high, as compared to a low X irradiation dose; (b) VEGF-C protein levels after cell expansion following the third irradiation (CR-MI): * and *, significantly increased levels after low and high doses of P and X irradiation, respectively, as compared to CT; Concentration in ng/ml, normalized to 1 × 106 cells, and represented as percentage of CT. #, significantly decreased levels after high doses of P, as compared to X irradiation; §, significantly increased levels after high, as compared to low doses of X irradiation; (c) Activity of a short vegf-c promoter (CR-MI); (d) Activity of an artificial promoter having three binding sites for NF-kB (CR-MI); (e) Activity of a VEGF-C 3′UTR reporter gene (CR-MI). * and *, significantly (Po0.05) increased promoter activity after P and X irradiation, respectively, as compared to CT; # and #, significantly decreased activity of MUT, as compared to WT vegf-c promoter after P and X irradiation, respectively; §, significantly decreased promoter activity after P, as compared to X irradiation. CT, control (non-irradiated cells); MUT, mutated, WT, wild type.

activity of the WT vegf-c promoter but the activity of the MUT The average tumor volume was significantly increased (Po0.05) promoter was completely inhibited in both non-irradiated and for P and X tumors, but no differences were observed between irradiated cells (Po0.001, Supplementary Figure S2C). To further the irradiation types (Figure 3a and b). These results were assess the role of NF-κB on vegf-c promoter, the activity of an inconsistent with the in vitro proliferative abilities of the cells artificial promoter containing three binding sites for human NF-κB surviving after MI with either P or X. To determine whether P and was determined in control and irradiated cells. In CAL33, the X irradiated cells ‘educated’ the microenvironment to favor tumor NF-κB-dependent promoter activity was lower in P irradiated cells, growth, we performed a whole transcriptomic screening of the which is consistent with the activity of the vegf-c promoter tumors. Indeed, distinct profiles for both the mouse (Figure 3c) having a WT NF-κBbindingsite(Figure2d).ForCAL27,the and human (Figure 3d) 10 most up- and downregulated genes NF-κB-dependent promoter activity is almost equivalent in control were detected. Among the 10 most up- and downregulated mouse genes, some (Figure 3c) such as collagen type XVII alpha 1 and either X or P irradiated cells (Supplementary Figure S2D). 26 This result indicates that the vegf-c promoter activity exclusively and carbonic anhydrase 2 (Car2) had a shared pattern of fi relies on an NF-κB-dependent transcriptional mechanism in CAL27 expression in P and X tumors (Figure 3c). In addition, we identi ed fi cells, whereas the dependency to NF-κB is partial in CAL33 cells. distinct pro les for the 10 most up- and downregulated mouse Moreover, a reporter gene used to assess VEGF-C mRNA half-life (Supplementary Figure S3) and human (Supplementary Figure S4) genes involved in angiogenesis, inflammation, metastasis, M1/M2 was not affected by either P or X irradiation in CAL33 cells macrophage transition. Some of these genes had a shared pattern (Figure 2e), suggesting that the increase in vegf-c mRNA levels of expression in P and X tumors. does not depend on modifications in mRNA half-life. Furthermore, we identified 70 (26%) common upregulated and 3(5.8%)commondownregulatedgenes(Figure3e)betweenX Cells surviving multiple irradiations by P and X generate tumors and P tumors, with roles in angiogenesis/metastasis, inflammation, with distinct characteristics M1/M2 macrophage transition and proliferation (Table 2). The cells resistant to MI by either P or X served to generate Tumors induced by irradiated cells presented less necrosis and experimental tumors in mice to test their relative aggressiveness. increased intra-tumor vessels density (P = 0.031 for P and P = 0.002

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 5

1000 Control0 Gy * 800 XX 8 Gy PP 8 Gy ) 600 3 * (mm 400 * 200 * Average tumor volume 0 * 024681012141618 Days post-injection

Mouse Mouse Mouse 0 vs X 0 vs P P vs X

Human Human Human 0 vs X 0 vs P P vs X

UP DOWN X vs 0 P vs 0 X vs 0 P vs 0

109 70 90 48 3 1 (40.5%) (26%) (33.5%) (92.3%) (5.8%) (1.9%)

for X group, Figure 4a and Supplementary Figure S5A). In addition, P and X groups, Supplementary Figure S5B). Lymphatic vessels irradiation by either P or X led to generation of tumors with were detected in the tumor-skin border of the control and P destabilized vessel architecture (Figure 4b), attested by a decrease groups; however, they were also present in the core of the X in vessels with co-staining for CD31 and αSMA (Po0.001 for both tumors (Figure 4c), ﬁnding consistent with the over-expression of

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 6 VEGF-C observed in vitro. Since VEGF-C was particularly discrimi- the cells surviving after three irradiations with P. These observations native between the two experimental irradiation conditions, we suggest that P radiotherapy would lead to less pronounced tested whether it had induced the development of lymphatic lymphangiogenesis/metastasis, as compared to X radiotherapy. vessels. LYVE1, PDPN and PROX1 markers of lymphatic vessels Therefore, we postulated that over-expression of VEGF-C may were then tested. Density of LYVE1-positive lymphatic vessels was represent an extrinsic mechanism responsible for the significantly increased in tumors generated with X-, as compared post-irradiation tumor dissemination/metastasis in HNSCC. to non- and P-irradiated cells (P = 0.006 and P = 0.009, respectively, VEGF-C expression was associated with lymph node metastasis, Supplementary Figure S5C). LYVE1 and PDPN mRNA were recurrence and a poorer five-year survival rate in patients with upregulated (P = 0.015 and 0.044, respectively) in X and HNSCC,11 being an independent prognostic factor.11,29 Moreover, downregulated (Po0.001 for both markers) in P tumors. Lower the online available database cBioPortal (http://www.cbioportal. mRNA levels of LYVE1 and PDPN were detected in P, as compared org) shows that over-expression of VEGF-C correlated to o to X tumors (P = 0.003 and P 0.001, respectively). PROX1 mRNA significantly lower disease-free (P = 0.0022, Supplementary level was downregulated (P = 0.02) and unchanged in P and X Figure S7A) and overall (P = 0.015, Supplementary Figure S7B) tumors, respectively (Figure 4d). survival rates in patients with HNSCC (n = 517). It has been reported that gamma rays irradiation induced VEGF-C expression Conventional radiotherapy by X increases tumor and endothelial cell proliferation in lung cancer.30 These lymphangiogenesis in patients with HNSCC observations, corroborated with ours, suggest that VEGF-C may To further correlate the relationshipbetweenirradiation-dependent be an important therapeutic target for HNSCC patients who VEGF-C expression and lymphatic vessels development, we tested relapse after radiotherapy with either P or X. the presence of lymphatic markers in biopsies from primary and Because VEGF-C might be a major factor responsible for locally relapsed human HNSCC, after conventional radiotherapy. post-irradiation disease progression in HNSCC patients, via Recent reports described that the expression of PDNP, one of the promotion of lymphangiogenesis, we further started investigating major makers of lymphatic vessels,wasnotrestrictedtolymphatic the mechanisms involved in its induction, which may serve to its vessels but it was also expressed in HNSCC cells.27 Expression of therapeutic targeting. Regulation of VEGF-C expression has been PDPN was indeed detected in tumors from patients with oral and poorly addressed.27,31,32 Irradiation-mediated induction of VEGF-C pharyngeal SCC (Figures 5a–1.a). However, we observed a significant mRNA suggested stimulation of transcription, stabilization of increase of PDNP labeling, in both tumor and lymphatic cells, in mRNA or a combination of these mechanisms.31 Our data indicate sections from relapsed tumors after treatment with conventional X that both P and X irradiation stimulated the activity of a short form radiotherapy, as compared to the initial tumors (P =0.048,Figures5a– of vegf-c promoter in CAL33 cells. The vegf-c promoter contained a 1.b, Supplementary Figure S6A). In the same patients, the vascular binding site for NF-κB. The dependency of this site is variable fi network, attested by CD31 labeling, was not modi ed (P =0.059, considering the two cell lines we tested, but nevertheless NF-κB – Supplementary Figure S6B) intherelapsed(Figures5b1.b), as plays a key role in VEGF-C regulation, as suggested in another – compared to the initial tumors (Figures 5b 1.a). In addition, a cancer type.32 As these cell lines came from two different patients, tendency for increased mRNA expression of PDPN (P =0.088),along fi our results highlight the inter-patient variability in VEGF-C with signi cantly increased mRNA expression of VEGF-C (P =0.005), expression and regulation, stressing out the importance of LYVE1 (P =0.025) and PROX1 (P =0.003) were detected in relapsed implementing personalized diagnosis and treatment strategies. patient tumors after conventional X radiotherapy (Figure 5c). In the cells surviving after three irradiations, the VEGF-A and VEGF-D genes were downregulated by P and upregulated by X DISCUSSION irradiation. VEGF-A expression significantly correlated with lymph node metastasis in patients with HNSCC.11 High VEGF-A Our in vitro results indicate that P irradiation led to lower expression of factors involved in (lymph)angiogenesis, inflamma- expression was also associated with higher clinical stages and worse overall survival, being a significant predictor of poor tion and immune tolerance. This suggests the acquisition of less 33 aggressive phenotypes after P therapy. The selection of surviving prognosis in patients with HNSCC. Furthermore, VEGF-D expression correlated with lymphatic vessel density and lymph cells was still possible after MI, indicating a mechanism of acquired 10 28 node metastasis in these patients. In addition, VEGFR-2, VEGFR-3 resistance secondary to irradiation. However, the molecular 34 profiling of the surviving cells suggests a more aggressive in vivo and NRP1, highly expressed by HNSCC cells, were phenotype after MI with X. Therefore, due to its physical and downregulated in the surviving cells selected after three biological properties, P irradiation may be more efficient in tumor irradiations with P, but not with X. High NRP1 and NRP2 levels size control through dose escalation. correlated with poor prognosis in HNSCC patients, NRP2 being an 35 The long-term surviving cells after three irradiations with P independent prognostic markers for overall survival. showed a downregulation of the investigated pro-angiogenic/pro- Therefore, our study sets the basis for clinical assays investigat- inflammatory genes, except for vegf-c,whilemostofthesegenes ing more efficient treatments, combining P radiotherapy with were upregulated after X irradiation. The implication of VEGF-C in anti-angiogenic-targeted therapies. Such combinations would the metastatic dissemination process after irradiation has not been eventually lead to decreased selection of post-irradiation surviving elucidated. To our knowledge, this is the first report showing P or X cells and lower relapse rates in patients with HNSCC, for which the radiation-induced VEGF-C over-expression at both gene and protein current treatments include X irradiation.3 A case report describing levels in HNSCC cells. The VEGF-C mRNA levels increased in a the successful treatment of a patient with chondrosarcoma by dose-dependent manner and with the irradiation number, except in combining P radiotherapy with sunitinib, an inhibitor of VEGFRs

Figure 3. Evaluation of tumors generated following xenografting of either non-irradiated, P or X irradiated CAL33 cells in immunodeﬁcient mice. (a) Average tumor volume (mm3); (b) Representative images of tumor xenografts; (c) Heatmap of 10 most up- and downregulated mouse genes in tumors generated by non-irradiated cells versus P or X tumors, and in P versus X tumors; (d) Heatmap of 10 most up- and downregulated human genes in tumors generated by non-irradiated cells versus P or X tumors, and in P versus X tumors; (e) Venn diagrams showing common upregulated and downregulated human genes between P and X tumors. Framed genes are commonly expressed in P and X tumors. Selection is adjusted P-valueo0.05 and lofFC41.

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 7

Table 2. Common upregulated and downregulated human genes in tumors generated with either X or P irradiated cells

Common up-regulated genes between X versus 0 and P versus 0

Role Gene abbreviation Gene full name

Metastasis/Angiogenesis KRT16 Keratin 16 SERPINB3 Serpin family B member 3 CAPNS2 Calpain small subunit 2 GRHL3 Grainyhead like transcription factor 3 CSTB Cystatin B PRSS27 Protease, serine 27 TLE4 Transducin like enhancer of split 4 TMPRSS11D Transmembrane protease, serine 11D

Inﬂammation PGLYRP3 Peptidoglycan recognition protein 3 RASGRP1 RAS guanyl releasing protein 1 ENDOU Endonuclease, poly(U) speciﬁc METRNL Meteorin like, glial cell differentiation regulator S100A8 S100 calcium binding protein A8 S100A9 S100 calcium binding protein A9 A2ML1 Alpha-2-macroglobulin like 1 HCN2 Hyperpolarization activated cyclic nucleotide gated potassium channel 2 CHST2 Carbohydrate sulfotransferase 2

M1/M2 ABCG1 ATP binding cassette subfamily G member 1

Proliferation HPGD Hydroxyprostaglandin dehydrogenase 15-(NAD) BNIPL BCL2 interacting protein like PPP2R2C Protein phosphatase 2 regulatory subunit Bgamma KLK8 Kallikrein related peptidase 8 GJB6 Gap junction protein beta 6 EEF1A2 Eukaryotic translation elongation factor 1 alpha 2 EPHA4 EPH receptor A4 GAS7 Growth arrest speciﬁc7 DSG1 Desmoglein 1 PDZK1IP1 PDZK1 interacting protein 1 TMPRSS11A Transmembrane protease, serine 11A FLRT2 Fibronectin leucine rich transmembrane protein 2

Other C12orf36 Putative uncharacterized protein C12orf36 FRMPD1 FERM and PDZ domain containing 1 TMEM45A Transmembrane protein 45A LIPK Lipase family member K CTC-490G23.2 CTC-490G23.2 HOPX HOP homeobox PLIN2 Perilipin 2 SDR9C7 Short chain dehydrogenase/reductase family 9C, member 7 STXBP5-AS1 STXBP5 antisense RNA 1 ARRDC4 Arrestin domain containing 4 FRY FRY microtubule binding protein FAM25A Family with sequence similarity 25 member A [ SCEL Sciellin GJB2 Gap junction protein beta 2 UNC5B-AS1 UNC5B antisense RNA 1 RP11-21B23.2 Pre-mRNA processing factor SPRR1B Small proline rich protein 1B NAV3 Neuron navigator 3 SLC10A6 Solute carrier family 10 member 6 RP11-275I14.4 Pre-mRNA processing factor RP11-356I2.4 Pre-mRNA processing factor C9orf169 Cysteine rich tail 1 RP11-321G12.1 Pre-mRNA processing factor LINC01094 Long intergenic non-protein coding RNA 1094 OR7E62P Olfactory receptor family 7 subfamily E member 62 pseudogene FAM3D Family with sequence similarity 3 member D SMIM5 Small integral membrane protein 5 FBXL16 F-box and leucine rich repeat protein 16 RP11-783K16.5 Pre-mRNA processing factor KCNK7 Potassium two pore domain channel subfamily K member 7 FAM25HP Family with sequence similarity 25, member H pseudogene WI2-85898F10.1 Uncharacterized LOC107985535 IVL Involucrin

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 8

Table 2. (Continued )

Common up-regulated genes between X versus 0 and P versus 0

Role Gene abbreviation Gene full name

TCN1 Transcobalamin 1 KLHL4 Kelch like family member 4 LRRC7 Leucine rich repeat containing 7 RP11-557H15.3 Pre-mRNA processing factor TSHZ2 Teashirt zinc ﬁnger homeobox 2 OLFM2 Olfactomedin 2 ADH7 Alcohol dehydrogenase 7 (class IV), mu or sigma polypeptide

Common down-regulated genes between X versus 0 and P versus 0

Role Gene abbreviation Gene full name

Proliferation LIF leukemia inhibitory factor Other RNA5-8SP2 RNA, 5.8S ribosomal pseudogene 2 P2RX5 purinergic receptor P2X 5 In bold are shown genes upregulated in either P or X tumors, but down-regulated in tumors generated with non-irradiated cells. Selection is adjusted P- valueo0.05 and lofFC41.

and platelet-derived growth factor receptor, underlines the response was increased in TRF2 knocked-down cells and that TRF2 effectiveness of such approach.36 over-expression had a negative impact on patients’ survival.23 We also showed that P and X radiations differently modulated Irradiation leads to adaptive changes in the tumor microenvir- the pro-inflammatory gene expression in HNSCC cells. Among the onment that may limit the generation of an anti-tumor immune assessed genes, the highest determined mRNA level was for IL-8. response.24 Indeed, we showed a significant increase of PD-L1 Stress and drug-induced IL-8 signaling conferred chemotherapeu- expression after P, and confirmed the X radiation-induced PD-L1 tic resistance to cancer cells.37 Serum and tumor IL-8 significantly expression in other cancers.24,45 In patients with HNSCC, high affected the disease-free survival in patients with early stage PD-L1 expression in primary tumors correlated with metastasis HNSCC.38 Therefore, inhibiting the effects of IL-8 signaling in and poor prognosis, being an independent prognostic factor.46 combination to chemoradiotherapy may be of significant PD-L1 was also a significant predictor for poor treatment response therapeutic value. and shorter survival in X radiotherapy-treated patients with P but not X irradiation downregulate IL-6 expression at HNSCC.45 A phase II, multi-center, single-arm, global study of the mRNA level. IL-6 expression predicted a poor response to monotherapy with durvalumab, a Fc optimized monoclonal radio-chemotherapy and a non-favorable prognosis in HNSCC antibody directed against PD-L1, is ongoing in our institution in patients.39 It was also linked to radiation resistance and patients with recurrent/metastatic HNSCC and PD-L1 positive development of chronic toxicities after irradiation.40 Depending status. Therefore, our data, associated to the progress in the field, on tumor location, the most common side effects after set the basis for the investigation of novel therapeutic strategies conventional radiotherapy of HNSCC include mucositis, for HNSCC, based on the PD-L1–PD-1 interaction, in combination xerostomia, dysphagia requiring short-term or permanent with radiotherapy. gastrostomy, soft tissue/bone necrosis, neck fibrosis, and thyroid We also demonstrated that the aggressiveness of the irradiated dysfunction.41 Although the primary goal in radiotherapy is tumor cells was augmented in vivo through increased tumor volume, control, a parallel essential goal is to spare normal tissues from density of tumor vessels and blood vessels with destabilized radiation toxicity. Therefore, our data bring further pre-clinical architecture. These observations suggest that the irradiation- evidence that the use of P irradiation in the treatment of HNSCC adapted cells have acquired different transcriptome and may lead to less inflammatory side effects. secretome profiles. Indeed, among the common human genes We also showed that, in the cells surviving long-term after three upregulated in either X or P tumors, but downregulated in irradiations, another major pro-inflammatory cytokine, CCL2, was tumors generated with non-irradiated cells, we identified PDZK1 downregulated after P, while being highly upregulated after X interacting protein 1 (PDZK1IP1, known also as MAP17)47 and irradiation. As serum CCL2 levels were associated with HNSCC fibronectin leucine rich transmembrane protein 2,48 known for progression,42 our data suggest that P therapy might be more promoting cell proliferation. In addition, mouse Car2 expression beneficial for these patients. was downregulated in P and X tumors, while upregulated in Our results also showed that PLK1 and TRF2 genes were tumors generated with non-irradiated cells. Interestingly, low differently regulated after P or X irradiation and correlated to the CAR2 protein expression has been associated with increased proliferation patterns. By inhibiting apoptosis, PLK1 tumor size.26 In addition, the X tumors showed upregulation of over-expression was associated with poor survival in patients human genes involved in metastasis, angiogenesis and epithelial with HNSCC, being an independent prognostic factor.43 Its mesenchymal transition, such as MMP2, MMP9, MMP13, MMP16, targeting with a multi-kinase inhibitor led to encouraging MMP28 and vimentin,15 while P tumors showed upregulation of anti-tumor activity in patients with SCC.44 These data suggest human C–C Motif Chemokine Ligand 5 chemokine gene involved that PLK1 might be a potential therapeutic target for HNSCC in CD8+ T lymphocytes recruitment associated with better clinical patients undergoing radiotherapy. TRF2 may also become an outcomes.49 established predictive marker for treatment efficacy and a marker To get further insights whether tumor cell adaptation following of survival in HNSCC. We previously showed that the treatment radiotherapy may contribute to clinical disease progression, in

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 9 0 Gy X8 Gy P8 Gy

HES

10x 10x 10x

CD31 40x 40x 40x αSMA DAPI

40x 40x 40x

INT

LYVE-1 10x 10x 10x DAPI

40x 40x 40x

Figure 4. Histology, immunoﬂuorescence and quantitative gene expression of vascular and lymphatic markers in murine xenografts. (a) Representative images of HES staining, indicating increased necrosis (black arrowhead, delimited by dashed black lines) in CT and increased blood vessels density (white arrowhead showing collagen surrounding the vessels) in the irradiated cells-derived tumors; (b) Representative images of CD31 (endothelial cells, green)/αSMA (pericytes, red)/Hoechst (nuclei, blue) staining, showing anarchic blood vessels structures and lack of pericyte coverage of blood vessels in the irradiated cells-derived tumors; (c) Representative images of LYVE1 (lymphatic endothelial cells, red)/Hoechst (nuclei, blue) staining, showing different patterns of lymphatic vessels development in X (both periphery and interior of the tumor), P and CT (periphery of the tumor) groups; dashed white lines delimit the tumor edge; CT, control (tumors generated by non-irradiated cells); (d) Murine LYVE1, PDPN and PROX1 mRNA quantitative mRNA expression, as percentage of control (0 Gy). HES, Hematoxylin Eosin Saffron. part through lymphangiogenesis, we investigated lymphatic radiotherapy may promote lymphangiogenesis. It has also been markers expression in patients with relapsed HNSCC after X reported by others that high PDPN expression is associated with radiotherapy. Biopsies at relapse are very rarely sampled in aggressive tumor behavior, poor prognosis and metastatic radiotherapy-treated patients. However, in this small cohort, all regulation through interaction with VEGF-C, suggesting that PDPN patients presented increased protein and/or mRNA levels of PDPN, may be used as a potential prognostic biomarker for HNSCC.27 VEGF-C, LYVE1 and PROX1, bringing evidence that conventional However, our in vitro studies did not reveal increased PDPN

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 10 Primary tumor Relapsed tumor

1.a 10x 1.b 10x PDPN

2.a 10x 2.b 10x

Primary tumor Relapsed tumor

1.a 10x 1.b 10x CD31

2.a 10x 2.b 10x

PDPN mRNA VEGF-C mRNA LYVE1 mRNA PROX1 mRNA 1000 1000 ** 1000 * 1000 500 500 500 500 ** mRNA 0 0 0 0 Pre-X Post-X Pre-X Post-X Pre-X Post-X Pre-X Post-X (% of Control) Figure 5. Evaluation of vascular and lymphatic markers in biopsies from patients diagnosed with HNSCC. Representative images of immunohistochemistry for (a) PDPN and (b) CD31 expression: (1) oral and (2) hypopharyngeal localization; Left panels (1.a, 2.a)—primary tumor; Right panels (1.b, 2.b)—relapsed tumor in the same patient after surgery and chemo-X radiotherapy (brown, PDPN/CD31; blue, hematoxylin - nuclei); (c) quantitative PDPN, VEGF-C, LYVE1 and PROX1 mRNA expression, as percentage of control (0 Gy); * and **, signiﬁcantly increased values (Po0.05 and Po0.01, respectively) post-, as compared to pre-X radiotherapy.

expression in HNSCC cells that resisted to MI (Supplementary needed to refine the strategies for blocking VEGF-C activity and its Figure S8). effects on the vascular/lymphatic endothelial or tumor cells with In conclusion, our study highlighted the differential anti-angiogenic therapies. The implementation of P therapy in gene/protein expression profile after P versus X irradiation in combination with anti-angiogenic or anti-immune checkpoint HNSCC and potential candidate markers for prognosis, efficacy of drugs for HNSCC will therefore require prospective randomized anti-tumor treatments and new anti-tumor targets, such as clinical trials to measure the toxicity and disease control. VEGF-C. Beside the physical advantage of P irradiation in dose deposition, our observations provide preclinical evidence that beam therapy with P might be superior to conventional X therapy MATERIALS AND METHODS in HNSCC patients, due to its biological advantages. P irradiation Cell lines and culture could therefore permit dose escalation without increasing the side Two human HNSCC cell lines, CAL33 and CAL27, were provided through a effects, while increasing the tumor control. Further work is also Material Transfer Agreement with the Oncopharmacology Laboratory,

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 11 Centre Antoine Lacassagne (CAL), where they had initially been isolated.50 Luciferase assays The cells were cultured in Dulbecco’s modified Eagle's medium CAL33 cells belonging to the CR-MI group were transfected by using 50 μl supplemented with 7% fetal bovine serum (Thermo Fisher Scientific, NaCl buffer, 1.25 μl of polyethylenimine transfection reagent (Sigma Waltham, MA, USA). Aldrich) and 0.5 μg of total test plasmid DNA-renilla luciferase. The plasmids encoded either (i) a human vegf-c promoter fragment with either Cell irradiations a non-mutated (wild type, WT) or a mutated (MUT) binding site for the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB),32 2 fl Five million cells were seeded onto 12 cm tissue culture asks, 48 h prior (ii) an artificial promoter containing three binding sites for human NF-κB or to the irradiations, which were carried out at CAL (four independent (iii) a human VEGF-C 3′UTR reporter (LightSwitch, S803537, Active Motif, experiments) with either P (63 MeV Cyclotron MEDICYC, CAL, Nice, France) Carlsbad, CA, USA), all cloned downstream of the luciferase reporter gene. or X (6 MeV Dual energy Clinac 21EX Linear Accelerator, Varian Inc., Palo A CMV plasmid was used to control the variability of transfection efficiency Alto, CA, USA). For clonogenicity assays, the cells were irradiated once in the reporter assays. (single irradiation, SI) with 1, 2, 4, 6 or 8 Grays (Gy; physical dose) and processed immediately after irradiation. To the purpose of all other Tumor xenografts experiments, the cells were irradiated either once or three times, 1 week apart (multiple irradiations, MI) with either 2 Gy (low dose) or 8 Gy (high The study was carried out in strict accordance with the recommendations dose), and processed 6 h after irradiation. In the MI setting, cells were of the United Kingdom Coordinating Committee on Cancer Prevention Research’s Guidelines for the Welfare of Animals in Experimental re-seeded after each irradiation and kept in culture until the next Neoplasia. Our experiments were approved by the ‘Comité National irradiation to reproduce the clinical situation where patients are usually Institutionnel d'Éthique Pour l'Animal de Laboratoire’ (CIEPAL, reference: given several irradiations. The CR was evaluated to determine if the NCE/2013-97). One million non-irradiated, P or X irradiated CAL33 cells changes associated with the AR persist late (3 weeks) after irradiation. (CR-MI group) were injected subcutaneously into the flank of 6-week-old Two cell groups were thus generated from each independent irradiation NMRI-Foxn1nu/Foxn1nu female mice (Janvier Labs, Le Genest-Saint-Isle, experiment. They consisted of cells subjected to: (1) SI and analysis 48 h France, n = 10/group). The tumor volume (v = L × l2 × 0.52) was determined thereafter (AR-SI); (2) MI and culture expansion (3 weeks) after the third following measurement with a caliper. When the tumors reached 1 cm3, irradiation (CR-MI). All cell experiments were performed in triplicate wells the mice were killed and the tumors collected. for each condition and repeated at least three times. Whole transcriptomic screening of tumor xenografts Clonogenicity assays For the sequencing and secondary analysis, 1 μg of total RNA was They were performed to quantify the radio-induced cell mortality, to extracted from tumor xenografts, generated with either non-irradiated, P generate the cell surviving curves and to determine the RBE. Owing to or X irradiated cells (n = 3/group), by using the AllPrep DNA/RNA/Protein radiation dose-induced differences in plating efficiency, the cells were Mini Kit (Qiagen). Lack of RNA degradation (ratio 28S/18S ⩾ 1.6 and RIN47) seeded at different densities: 3000 cells/dish for 0, 1, 2 and 4 Gy; 6000 was documented (Bioanalyzer 2100, Agilent Technologies, Santa Clara, CA, cells/dish for 6 Gy and 9000 cells/dish for 8 Gy. On day 10 of culture, cells USA). The libraries were generated by using Truseq Stranded mRNA kit were stained for 20 min with Giemsa (Sigma Aldrich, St. Louis, MO, USA). (Illumina, San Diego, CA, USA). Libraries were then quantified with KAPA Stained plates were scanned and the number of cell colonies was library quantification kit (Kapa Biosystems, Inc., Wilmington, MA, USA) and determined with the ImageJ processing software (National Institutes of pooled; 4 nM of this pool were loaded on a Nextseq 500 high output fl × Health, Bethesda, MD, USA). The RBE was calculated as ratio of the owcell and sequenced with a 2 75 bp paired-end chemistry. STAR biological effectiveness of P versus X irradiation, given the same (2.4.0i) was used to map reads versus a STAR database containing: Ensembl dose/amount of absorbed energy.25 hg19 build (GRCh37.75), Ensembl mm10 build (GRCm38) and the ERCC spikes-in set, formatted with splice junctions information described from Ensembl release GRCh37.75 and GRCm38.83. STAR options were set to Cell counting for viability and proliferation assessment the recommended Encode RNA-seq options ‘--outFilterType BySJout The cell counting for the CR-MI group was done every day, for 4 days --outFilterMultimapNmax 20 --alignSJoverhangMin 8 --alignSJDBoverhang- post-seeding, with an automatic cell counter (Advanced Detection Min 1 --outFilterMismatchNmax 999 --outFilterMismatchNoverLmax 0.04 Accurate Measurement system, LabTech, Tampa, FL, USA), according to --alignIntronMin 20 --alignIntronMax 1000000 --alignMatesGapMax ’ the manufacturer’s instructions. 1000000 . Gene counts were obtained with featureCounts (subread-1.5.0- p3-Linux-x86_64) and ‘--primary -p -s 1 -C’ options, by using the same GTF files used for STAR splice junctions training. Data were deposited in Gene Quantification of gene expression Expression Omnibus (accession code GSE90761, https://www.ncbi.nlm.nih. Molecular characterization of the irradiated cells was done by using the gov/geo/query/acc.cgi?token = opybisygbzotvkh&acc = GSE90761). quantitative real-time–polymerase chain reaction. Total RNA was extracted For the heatmaps gene lists selection, genes involved in angiogenesis, with the RNeasy Mini Kit; first-strand cDNA synthesis was performed by inflammation, metastasis and cell proliferation were selected by using the using the QuantiTect Reverse Transcription Kit (all from Qiagen, Hilden, Ingenuity Pathway Analysis (Qiagen) database. To define M1/M2 Germany). cDNA samples were amplified by using the StepOnePlus macrophages-related genes, the GEO data set GSE69607 has been RT–PCR System (Thermo Fisher Scientific) for 40 cycles with the Takyon Rox reanalyzed by using geo2R online resource. Genes up- and downregulated fi (Abs (logFC)42) in both M1 versus M0 and M2 versus M0 comparisons SYBR Master Mix, dTTP Blue (Eurogentec, Liege, Belgium) and speci c ‘ ’ oligonucleotides (Sigma Aldrich, Supplementary Table S1), to assess mRNA were selected as the M1/M2 macrophages -related gene list. expression for VEGF-A, VEGF-C, VEGF-D, VEGFR-1, VEGFR-2, VEGFR-3, NRP1, NRP2, IL-6, IL-8, CCL2, TRF2, PLK1, PD-L1, LYVE1, PDPN and PROX1. mRNA Histochemistry and immunofluorescence levels were normalized to a housekeeping mRNA coding for either the Murine tumor sections were handled as previously described.8 To assess human or murine ribosomal protein, large, P0 (RPLP0). The gene tumor architecture, the sections were subjected to hematoxylin eosin expression levels were given the individual scores of − 1, 0 and 1 when saffron staining. For immunofluorescence, the frozen sections were they were significantly decreased, not significantly changed and incubated overnight, at 4 °C, with the following primary antibodies: significantly increased, respectively, as compared to control. For each polyclonal rabbit anti-mouse/human LYVE1 (1:200; Abcam, Cambridge, irradiation setting, a global gene expression score was then calculated by UK), monoclonal mouse anti-mouse/human alpha smooth muscle actin cumulating the individual scores allocated to each gene expression level. (αSMA, 1:400, Sigma Aldrich) and monoclonal rat anti-mouse CD31 (1:50, clone MEC 13.3, BD Pharmigen, Heidelberg, Germany) primary antibodies, fi then incubated for 2 h at room temperature, in the dark, with the Protein quanti cation secondary antibodies: anti-rabbit FP594, anti-mouse FP547 (1:1000, VEGF-C protein was quantified by using an enzyme-linked immunosorbent FluoroProbes, Interchim, Montluçon, France) and anti-rat AF488 (1:1000, assay (human DuoSet ELISA kit, R&D Systems, Minneapolis, MN, USA). AlexaFluor, Thermo Fisher Scientific); cell nuclei were stained with Hoechst Protein concentration was normalized to the viable cell number. (1:1000, Thermo Fisher Scientific). Cell and tissue preparations were

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 12 examined under an inverted epifluorescence microscope (Axio 6 Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin Observer Z1) with an incorporated digital camera system for imaging 2005; 55: 74–108. (AxioCam Icc1); images acquisition and stitching, as well as the assessment 7 Mamelle G, Pampurik J, Luboinski B, Lancar R, Lusinchi A, Bosq J. Lymph node of tumor vessels density, were performed with ZEN 2.3 software (all from prognostic factors in head and neck squamous cell carcinomas. Am J Surg 1994; Carl Zeiss MicroImaging GmbH, Weinheim, Germany). 168: 494–498. 8GrepinR,GuyotM,JacquinM,DurivaultJ,ChamoreyE,SudakaAet al. Immunohistochemistry Acceleration of clear cell renal cell carcinoma growth in mice following bevacizumab/Avastin treatment: the role of CXCL cytokines. Oncogene 2012; 31: Patient biopsy samples were collected with the approval of the local Ethics 1683–1694. Committee, and their use in research was in accordance with the 9 Frech S, Hormann K, Riedel F, Gotte K. Lymphatic vessel density in correlation to Declaration of Helsinki. The patient, disease and treatment characteristics lymph node metastasis in head and neck squamous cell carcinoma. Anticancer Res fi were described in Supplementary Table S2. Sections from formalin- xed 2009; 29: 1675–1679. fi and paraf n-embedded biopsies from initial and relapsed tumors were 10 Sugiura T, Inoue Y, Matsuki R, Ishii K, Takahashi M, Abe M et al. VEGF-C and incubated at room temperature with monoclonal, primary mouse VEGF-D expression is correlated with lymphatic vessel density and lymph node anti-human PDPN and CD31 antibodies, as well as biotinylated secondary metastasis in oral squamous cell carcinoma: implications for use as a antibodies, by using an automated slide stainer (Ventana Medical Systems, prognostic marker. Int J Oncol 2009; 34: 673–680. Inc., Basel, Switzerland). Binding was detected with the diaminobenzidine 11 Yanase M, Kato K, Yoshizawa K, Noguchi N, Kitahara H, Nakamura H. Prognostic substrate against a hematoxylin counterstain. Evaluation of marker value of vascular endothelial growth factors A and C in oral squamous cell expression was performed by an accredited clinical pathologist (IP). carcinoma. J Oral Pathol Med 2014; 43:514–520. 12 Tian J, Zhao W, Tian S, Slater JM, Deng Z, Gridley DS. Expression of genes involved Statistical analyses in mouse lung cell differentiation/regulation after acute exposure to photons and 176 – Statistical analysis for all test, excepting whole transcriptomic screening, protons with or without low-dose preirradiation. Radiat Res 2011; : 553 564. was performed by two-tailed unpaired t test on at least three independent 13 Kajioka EH, Andres ML, Mao XW, Moyers MF, Nelson GA, Gridley DS. experiments; the results were considered statistically significant when Hematological and TGF-beta variations after whole-body proton irradiation. In 14 – P-valueo0.05. The error bars were defined as standard error of the mean. Vivo 2000; : 703 708. For the whole transcriptomic screening, statistical analyses were 14 Girdhani S, Lamont C, Hahnfeldt P, Abdollahi A, Hlatky L. Proton irradiation conducted separately for human and mouse gene expression counts. suppresses angiogenic genes and impairs cell invasion and tumor growth. Radiat 178 – Quality of libraries was assessed based on the Pearson correlation between Res 2012; :33 45. observed versus expected ERCC counts (R240.90 for all samples). 15 Ogata T, Teshima T, Kagawa K, Hishikawa Y, Takahashi Y, Kawaguchi A et al. Normalization and differential analysis were conducted within R/Biocon- Particle irradiation suppresses metastatic potential of cancer cells. Cancer Res 65 – ductor environment, by using DESeq2. P-values were corrected for multiple 2005; :113 120. testing, by using the Benjamini and Hochberg method. Heatmaps were 16 Shinriki S, Jono H, Ueda M, Ota K, Ota T, Sueyoshi T et al. Interleukin-6 signalling generated with TMeV software. Heatmaps used the top 10 most up- and regulates vascular endothelial growth factor-C synthesis and lymphangiogenesis 225 – downregulated genes, based on logFC and adjusted P-valueo0.05 for in human oral squamous cell carcinoma. JPathol2011; : 142 150. human genes, and logFC only for mouse genes. 17 Li M, Zhang Y, Feurino LW, Wang H, Fisher WE, Brunicardi FC et al. Interleukin-8 increases vascular endothelial growth factor and neuropilin expression and stimulates ERK activation in human pancreatic cancer. Cancer Sci 2008; 99:733–737. CONFLICT OF INTEREST 18 Cursiefen C, Chen L, Borges LP, Jackson D, Cao J, Radziejewski C et al. VEGF-A stimulates lymphangiogenesis and hemangiogenesis in inflammatory neovascu- fl The authors declare no con ict of interest. The views expressed are purely those of larization via macrophage recruitment. J Clin Invest 2004; 113: 1040–1050. fi the authors and may not in any circumstances be regarded as stating an of cial 19 Zachary I. Neuropilins: role in signalling, angiogenesis and disease. Chem Immunol position of the European Commission. Allergy 2014; 99: 37–70. 20 Pan Y, Wang WD, Yago T. Transcriptional regulation of podoplanin expression by 94 – ACKNOWLEDGEMENTS Prox1 in lymphatic endothelial cells. Microvasc Res 2014; :96 102. 21 Loberg RD, Day LL, Harwood J, Ying C St, John LN, Giles R et al. CCL2 is a potent We thank Drs Heide L. Ford and Kari Alitalo for the kind gift of human VEGF-C regulator of prostate cancer cell migration and proliferation. Neoplasia 2006; 8: promoter DNA, as well as to Julien Parola, Catherine Pons, Christian Baudoin, and 578–586. ‘UCAGenomiX’ Functional Genomics Platform of the University of Nice Sophia 22 Kim SA, Kwon SM, Yoon JH, Ahn SG. The antitumor effect of PLK1 and HSF1 Antipolis, a partner of the National Infrastructure ‘France Génomique’, for their double knockdown on human oral carcinoma cells. Int J Oncol 2010; 36:867–872. technical support. We are grateful for the research funding received from the 23 Benhamou Y, Picco V, Raybaud H, Sudaka A, Chamorey E, Brolih S et al. Telomeric European Commission Marie Curie actions—FP7 IEF grant ‘VELYMPH’ No. 626449- repeat-binding factor 2: a marker for survival and anti-EGFR efficacy in oral car- /2015-2016 (ML-P), University of Nice Sophia Antipolis/Aix Marseille University Master cinoma. Oncotarget 2016; 7:44236–44251. II program (AC), ‘Fondation ARC pour la recherche sur le cancer’, France (GP), 24 Dovedi SJ, Illidge TM. The antitumor immune response generated by fractionated ‘Fondation de France’ (MD), Ph.D. grant Region PACA, France (PND), ‘Cordons de Vie’, radiation therapy may be limited by tumor cell adaptive resistance and can be Monaco (GP), ‘Electricité de France (EDF)’ (AC, JD, GP) and National Institute of Cancer circumvented by PD-L1 blockade. Oncoimmunology 2015; 4:e1016709. (INCA), France (GP). 25 Zuofeng Li. Prescribing, recording, and reporting proton-beam therapy: International Commission on Radiation Units and Measurements Report 78. JICRU2007; 7: 210. 26 Hu X, Huang Z, Liao Z, He C, Fang X. Low CA II expression is associated with tumor REFERENCES aggressiveness and poor prognosis in gastric cancer patients. Int J Clin Exp Pathol 1 Begg AC, Stewart FA, Vens C. Strategies to improve radiotherapy with 2014; 7:6716–6724. targeted drugs. Nat Rev Cancer 2011; 11: 239–253. 27 Kim HY, Rha KS, Shim GA, Kim JH, Kim JM, Huang SM et al. Podoplanin is involved 2 Doyen J, Falk AT, Floquet V, Herault J, Hannoun-Levi JM. Proton beams in cancer in the prognosis of head and neck squamous cell carcinoma through interaction treatments: clinical outcomes and dosimetric comparisons with photon therapy. with VEGF-C. Oncol Rep 2015; 34:833–842. Cancer Treat Rev 2016; 43: 104–112. 28 Willers H, Azzoli CG, Santivasi WL, Xia F. Basic mechanisms of therapeutic resis- 3 Seiwert TY, Salama JK, Vokes EE. The chemoradiation paradigm in head and tance to radiation and chemotherapy in lung cancer. Cancer J 2013; 19:200–207. neck cancer. Nat Clin Pract Oncol 2007; 4: 156–171. 29 Matsui T, Shigeta T, Umeda M, Komori T. Vascular endothelial growth factor C 4 Sio TT, Lin HK, Shi Q, Gunn GB, Cleeland CS, Lee JJ et al. Intensity modulated (VEGF-C) expression predicts metastasis in tongue cancer. Oral Surg Oral Med Oral proton therapy versus intensity modulated photon radiation therapy for Pathol Oral Radiol 2015; 120: 436–442. oropharyngeal cancer: first comparative results of patient-reported outcomes. 30 Chen YH, Pan SL, Wang JC, Kuo SH, Cheng JC, Teng CM. Radiation-induced Int J Radiat Oncol Biol Phys 2016; 95: 1107–1114. VEGF-C expression and endothelial cell proliferation in lung cancer. Strahlenther 5BlanchardP,WongAJ,GunnGB,GardenAS,MohamedAS,RosenthalDIet al. Onkol 2014; 190:1154–1162. Toward a model-based patient selection strategy for proton therapy: external 31 Enholm B, Paavonen K, Ristimaki A, Kumar V, Gunji Y, Klefstrom J et al. validation of photon-derived normal tissue complication probability models in a Comparison of VEGF, VEGF-B, VEGF-C and Ang-1 mRNA regulation by serum, head and neck proton therapy cohort. Radiother Oncol 2016; 121: 381–386. growth factors, oncoproteins and hypoxia. Oncogene 1997; 14: 2475–2483.

Oncogenesis (2017), 1 – 13 Proton versus photon radiation impact in squamous cell carcinoma M Lupu-Plesu et al 13 32 Du Q, Jiang L, Wang X, Wang M, She F, Chen Y. Tumor necrosis factor-alpha survivin level in esophageal squamous cell carcinoma. Int J Cancer 2009; 124: promotes the lymphangiogenesis of gallbladder carcinoma through nuclear 578–588. factor-kappaB-mediated upregulation of vascular endothelial growth factor-C. 44 Bowles DW, Diamond JR, Lam ET, Weekes CD, Astling DP, Anderson RT et al. Phase Cancer Sci 2014; 105:1261–1271. I study of oral rigosertib (ON 01910.Na), a dual inhibitor of the PI3K and Plk1 33 Smith BD, Smith GL, Carter D, Sasaki CT, Haffty BG. Prognostic significance of pathways, in adult patients with advanced solid malignancies. Clin Cancer Res vascular endothelial growth factor protein levels in oral and oropharyngeal 2014; 20: 1656–1665. squamous cell carcinoma. J Clin Oncol 2000; 18: 2046–2052. 45 Chen MF, Chen PT, Chen WC, Lu MS, Lin PY, Lee KD. The role of PD-L1 in 34 Chu W, Song X, Yang X, Ma L, Zhu J, He M et al. Neuropilin-1 promotes the radiation response and prognosis for esophageal squamous cell epithelial-to-mesenchymal transition by stimulating nuclear factor-kappa B and is carcinoma related to IL-6 and T-cell immunosuppression. Oncotarget 2016; 7: – associated with poor prognosis in human oral squamous cell carcinoma. 7913 7924. PLoS ONE 2014; 9: e101931. 46 Oliveira-Costa JP, de Carvalho AF, da Silveira da GG, Amaya P, Wu Y, Park KJ et al. 35 Zhang B, Gao Z, Sun M, Li H, Fan H, Chen D et al. Prognostic significance of Gene expression patterns through oral squamous cell carcinoma development: VEGF-C, semaphorin 3F, and neuropilin-2 expression in oral squamous cell PD-L1 expression in primary tumor and circulating tumor cells. Oncotarget 2015; 6 – carcinomas and their relationship with lymphangiogenesis. J Surg Oncol 2015; :20902 20920. 111:382–388. 47 de Miguel-Luken MJ, Chaves-Conde M, de Miguel-Luken V, Munoz-Galvan S, Lopez-Guerra JL, Mateos JC et al. MAP17 (PDZKIP1) as a novel prognostic 36 Dallas J, Imanirad I, Rajani R, Dagan R, Subbiah S, Gaa R et al. Response to sunitinib biomarker for laryngeal cancer. Oncotarget 2015; 6: 12625–12636. in combination with proton beam radiation in a patient with chondrosarcoma: a 48 Xu Y, Wei K, Kulyk W, Gong SG. FLRT2 promotes cellular proliferation and case report. J Med Case Rep 2012; 6:41. inhibits cell adhesion during chondrogenesis. JCellBiochem2011; 112: 37 Huang D, Ding Y, Zhou M, Rini BI, Petillo D, Qian CN et al. Interleukin-8 mediates 3440–3448. resistance to antiangiogenic agent sunitinib in renal cell carcinoma. Cancer Res 49 Liu J, Li F, Ping Y, Wang L, Chen X, Wang D et al. Local production of the 2010; 70:1063–1071. chemokines CCL5 and CXCL10 attracts CD8+ T lymphocytes into esophageal 38 Fujita Y, Okamoto M, Goda H, Tano T, Nakashiro K, Sugita A. Prognostic squamous cell carcinoma. Oncotarget 2015; 6: 24978–24989. significance of interleukin-8 and CD163-positive cell-infiltration in tumor tissues in 50 Gioanni J, Fischel JL, Lambert JC, Demard F, Mazeau C, Zanghellini E et al. patients with oral squamous cell carcinoma. PLoS ONE 2014; 9: e110378. Two new human tumor cell lines derived from squamous cell carcinomas of 39 Jinno T, Kawano S, Maruse Y, Matsubara R, Goto Y, Sakamoto T et al. Increased the tongue: establishment, characterization and response to cytotoxic treatment. expression of interleukin-6 predicts poor response to chemoradiotherapy and Eur J Cancer Clin Oncol 1988; 24: 1445–1455. unfavorable prognosis in oral squamous cell carcinoma. Oncol Rep 2015; 33: 2161–2168. 40 Wu CT, Chen MF, Chen WC, Hsieh CC. The role of IL-6 in the radiation response of Oncogenesis is an open-access journal published by Nature Publishing 8 prostate cancer. Radiat Oncol 2013; : 159. Group. This work is licensed under a Creative Commons Attribution 4.0 41 Holliday EB, Frank SJ. Proton radiation therapy for head and neck cancer: International License. The images or other third party material in this article are included a review of the clinical experience to date. Int J Radiat Oncol Biol Phys 2014; 89: in the article’s Creative Commons license, unless indicated otherwise in the credit line; if 292–302. the material is not included under the Creative Commons license, users will need to 42 Ding L, Li B, Zhao Y,Fu YF, Hu EL, Hu QG et al. Serum CCL2 and CCL3 as potential obtain permission from the license holder to reproduce the material. To view a copy of biomarkers for the diagnosis of oral squamous cell carcinoma. Tumour Biol 2014; this license, visit http://creativecommons.org/licenses/by/4.0/ 35:10539–10546. 43 Feng YB, Lin DC, Shi ZZ, Wang XC, Shen XM, Zhang Y et al. Overexpression of PLK1 is associated with poor survival by inhibiting apoptosis via enhancement of © The Author(s) 2017

Supplementary Information accompanies this paper on the Oncogenesis website (http://www.nature.com/oncsis)

Oncogenesis (2017), 1 – 13 Annexe!2!:"Role"of"Autophagy"in"lung"cancer" Après" l’obtention" de" mon" master," je" n’ai" pas" aussitôt" commencé" ma" thèse." J’ai" eu" la" chance"d’abord"d’intégrer"une"équipe"formidable":"celle"du"Pr"Hofman"à"l’IRCAN."Durant" ces"trois"années,"j’ai"participé"à"des"projets"qui"ont"généré"des"publications"que"je"vous" présente"ciGdessous."L’équipe"est"spécialisée"dans"le"cancer"du"poumon"et"l’autophagie." Nous"avons"démontré"le"rôle"à"double"tranchant"de"l’autophagie"dans"le"développement" des"cancers"du"poumon."Ce"projet"a"permis"de"décrire"la"fonction"de"la"petite"protéine" RHOA""dans"la"progression"tumorale"et"sa"régulation"par"l’autophagie"par"un"mécanisme" dépendant"de"la"protéine"SQSTM1"qui"interagit"directement"avec"RHOA."" " " " " " " " " " " " "

122" " Autophagy

ISSN: 1554-8627 (Print) 1554-8635 (Online) Journal homepage: http://www.tandfonline.com/loi/kaup20

Autophagy and SQSTM1 on the RHOA(d) again

Amine Belaid, Papa Diogop Ndiaye, Michaël Cerezo, Laurence Cailleteau, Patrick Brest, Daniel J Klionsky, Georges F Carle, Paul Hofman & Baharia Mograbi

To cite this article: Amine Belaid, Papa Diogop Ndiaye, Michaël Cerezo, Laurence Cailleteau, Patrick Brest, Daniel J Klionsky, Georges F Carle, Paul Hofman & Baharia Mograbi (2014) Autophagy and SQSTM1 on the RHOA(d) again, Autophagy, 10:2, 201-208, DOI: 10.4161/ auto.27198 To link to this article: http://dx.doi.org/10.4161/auto.27198

View supplementary material

Published online: 26 Nov 2013.

Submit your article to this journal

Article views: 923

View related articles

View Crossmark data

Citing articles: 10 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=kaup20

Download by: [Inserm Disc Ist] Date: 16 October 2017, At: 19:52 Downloaded by [Inserm Disc Ist] at 19:52 16 October 2017 a trigger of tumorigenesis. of trigger a Moreover, prevent transformation. can autophagy inflammatio c inhibiting thus tosignaling, oncogenic response in cence senes- promotecellular also may Autophagy mutations. DNA in resu genotoxic stress cause would otherwise which chondria, mit as such organelles, damaged and old of dispose can tions) nutrien physiological under (i.e., autophagy constitutive t proposed been it has cancer, against mechanism safeguard a initiatio tumor prevent to acts initially process autophagic the that is consensus current The pathway. tumor-promoting a tumor-suppress still a as either act cancer may it in controversial; remain autophagy of roles the work, of amount huge a www.landesbioscience.com Autophagy 10:2,201–208;February 2014;©2014LandesBioscience ent limitation in the inner area of the tumor. the of area inner the in limitation ent nutri and hypoxia to response in dormancy tumor and cell survival sustain may autophagy promoter, tumor a As cells. cancer of needs the meet to induced dramatically is autophagy lished, monly mutated or downregulated in human cancers. human in downregulated or mutated monly Abbreviations: Abbreviations: http://dx.doi.org/10.4161/auto.27198 05/21/2013;Submitted: 10/12/2013; Revised: Accepted: 11/13/2 [email protected] Email: Mograbi; to: Baharia *Correspondence Nice, France; France; Nice, MAP1LC3 (LC3), microtubule-associated protein 1 light chain 3; shRNA 3; chain protein 1light microtubule-associated (LC3), MAP1LC3 Amine Belaid, Nice, France; France; Nice, oe opnns f h atpai mciey r com- are machinery autophagic the of components Some 1 Institute of Research on Cancer and Ageing of Nice (I Nice of Ageing and Cancer on Research of Institute TCIRG1/a3, T-cell immune regulator 1, H TCIRG1/a3, ATPase, regulator T-cell immune Autophagy and SQSTM1 on the RHOA(d) again B A SIC BRIE cell proliferation, migration, and genome stability. genome and migration, cell proliferation, progression. We therefore propose that autophagy acts as a deg a as acts autophagy that propose We therefore progression. ulated, failure, thus and aneuploidy driving motili cytokinesis spindle midbody or spindle midbody to the cell as surface, we herei demonstrate RHOA. The active RHOA is sequestered via SQSTM1/p62 within autol SQSTM1/p62within via sequestered is RHOA active The RHOA. degradatio autophagosome of inhibition Specifically, pathway. stressful conditions. Our recent studies revealed that const revealed studies recent Our conditions. stressful t and components, downstream kinases, degrading by pathways e Emerging motility. its and survival, its growth, own its control Degradation of signaling proteins is one of the most powerful tu powerful most the of one is proteins signaling of Degradation 3 F R F Equipe Labellisée par l’ARC; Villejuif, France; France; Villejuif, l’ARC; par Labellisée Equipe 6 Emerging roles of autophagy in the degradation University of Michigan; Life Sciences Institute; Ann Arbor, MI USA; USA; MI Arbor, Ann Institute; Sciences Life Michigan; of University EPORT Atg, autophagy-related; GEF, guanine nucleotide exchange factor; LA factor; exchange nucleotide GEF, guanine autophagy-related; Atg, 1,2,3,4 8 Centre Hospitalier Universitaire de Nice; Pasteur Hospital; Labo Hospital; Pasteur Nice; de Universitaire Hospitalier Centre Papa Diogop Ndiaye, Papa Diogop Keywords: 1 Conversely, once the cancer is estab- is cancer the Conversely, once Georges FCa Georges autophagy, RHOA, tumor suppression, cytokinesis, aneuploidy, migr suppression, cytokinesis, tumor autophagy, RHOA, 1,2,3,4 of signaling proteins RC 4 rle, 1 AN); INS AN); Centre Antoine Lacassagne; Nice, France; France; Nice, Lacassagne; Antoine Centre Later, autophagy autophagy Later, Michaël C Michaël 2,4,7 013 Paul Hofman, ER 1,2 M U1081; CNRS UMR7284; Nice, France; France; M U1081; Nice, UMR7284; CNRS Despite Despite t condi- t + transporting, lysosomal V0 subunit A3; v-ATPase, vacuolar-ATPas v-ATPase, A3; V0 subunit lysosomal transporting, ellular ellular WT, wild type WT, wild ive or ive erezo, n. As As n. lting lting Autophagy hat hat o- n, 7 - Laboratoire Laboratoire itutive autophagy temporally and spatially controls and spatially the RHOA temporally autophagy itutive 1,2,5 ty, three processes that havedirectly an upon impact cancer 1,2,3,4,8 suggests that the levels of inactive RHOA are also controlled controlled also are RHOA inactive of levels the that suggests interest, recent tein) accumulating factors. Of particular pr activating GTPase (a inhibiting nucleoti and GEF) factor, exchange guanine (a activating by controlled tightly be must RHOA that concept the to led have GTPases small studying of Decade RHOA. GTPase small the is genome stability influence promoting promoting or preventing ofthe effect cytotoxic anticancer sword,” “double-edged a be can therapy cancer in autophagy of modulation result, a As chemotherapy. during death cell evade with tumorigenesis. with integ genomic in defects homeo- between cell correlation the in and stasis, integrity genomic of importance the given cal issue is This c unknown. remain still mechanisms underlying enables cancer cells to survive anoikis during metastasis during anoikis to survive cells cancer enables was first described in 2003, in described first was 5 years ago the role of autophagy in genomic stability, genomic in autophagy of role the ago years 5 not ofKey understood. observations Eileen White et al. descr ratory of Clinical and Experimental Pathology; Nice, France Nice, Pathology; Experimental and Clinical of ratory Laurence C n. n. As a are responses result, all dereg- RHOA-downstream vidence suggests that autophagy limits several signaling signaling several limits autophagy that suggests vidence n induces the accumulation of the GTP-bound form of of form GTP-bound the of accumulation the induces n radative brake for RHOA signaling and thereby controls controls thereby and signaling RHOA for brake radative Even though the tumor-suppression function of autophagy autophagy of function tumor-suppression the though Even ranscription factors; however, this often occurs under under occurs often this however, factors; ranscription and Baharia Mograbi TIRO mor-suppressive mechanisms by which a cell can can cell a which by mechanisms mor-suppressive ysosomes, and accordingly fails to localize to the the to localize to fails accordingly and ysosomes, , short hairpin RNA; SQSTM1/p62, sequestosome 1; sequestosome SQSTM1/p62, RNA; hairpin , short - MA 5 INS TOs UMR E4320; Commissariat à l’Energie Atomique; Nice, France; France; Nice, Atomique; àl’Energie E4320; Commissariat UMR TOs ail ER M U895/C3M: Centre Méditerranéen de Médecine Moléculaire; Moléculaire; Médecine de Méditerranéen Centre M U895/C3M: leteau, 2 Université de Nice-Sophia Antipolis; Faculté de Médecine; Médecine; de Faculté Antipolis; Nice-Sophia de Université MP, lysosomal-associated membrane protein; protein; membrane MP, lysosomal-associated 10 Foremost among the signaling players that that players signaling the among Foremost 1,2 Patrick Brest, 1,2,3,4, 6,7 its precise role in tumorigenesis is is tumorigenesis in role precise its * ation 1,2,3,4 Daniel JKlionsky, BA SIC BRIE 8,9 F R F 3,4 evidence evidence but the the but and to and drugs. e; e; EPORT ibed ibed riti- rity rity 6 201

by de de o- s 5 ©2014 Landes Bioscience. Do not distribute. Downloaded by [Inserm Disc Ist] at 19:52 16 October 2017 tion. migra cell in essential is inactivation RHOA where region one 202 the lamella of crawling crawling of under just lamella vesicles, the autophagic within RHOA of staining tate shRNA released active RHOA released shRNA at the plasma membrane of and impaired formation of cell protrusions ( protrusions cell of formation impaired and filamen into polymerization ACTIN allowed which cells, null niiin f uohgsm frain epeso of expression formation: autophagosome of inhibition shRNA-mediated the by demonstrated then was RHOA-GTP S1 Vid. work ( a3 pathway (such as phagy GEF ECT2, and ARHGDIA/RHOGDI are not affected. are ARHGDIA/RHOGDI and ECT2, GEF the RHOA vation of RHOA KIF23/MKLP1, such as the kinesin act proper for required regulators upstream the as well as RAC as the closely related GTP specific by is remarkably autophagy cancer. Importantly, of the control aggressive ofhallmark and aneupl failure, cytokinesis driving responses, stream RHOA all dow tion, deregulates or sequestration, degradation) fo a defects: studied result, the of a (irrespective autophagy in As failure autophagosomes. to RHOA ubiquitinated active the and targets that adaptor molecular the as SQSTM1 fied id we level, molecular the At cytokinesis. undergoing cells of midbody the to close autolysosomes, within RHOA-GTP of accumulation the demonstrate to us allowed subunit) TCIRG1 t of loss the (by degradation autophagosome of Inhibition sis. cytokine- during midbody at the activation RHOA for restricted mechanism important most the is autophagy that established we ing, another striking hallmark of hallmark striking another ing, cells by time-lapse microscopy and observed that the the that observed and microscopy time-lapse by cells cells migrated 7 times as fast as the WT cells ( cells WT the as fast as times 7 migrated cells autophagy. by levels low at maintained constitutively is RHOA active that far, elusive. so remains, degraded larly regulatory light chain (P-MLC, (P-MLC, chain light regulatory myo of phosphorylation downstream of levels higher panels), the lamellipodia to allow cell motility ( motility cell to allow lamellipodia the at RHOA of amount appropriate the maintaining for necessary be absolutemight we autophagy propose that offindings, these ity ( ity accumulation of RHOA at the cell surface ( surface cell the at RHOA of accumulation proteasomal degradation, proteasomal lamellipodia ( lamellipodia ACTIN-ric of formation the and fibers, stress of loss the with cytoskeleton ACTIN the of remodeling increase dramatic a an size, cell by in characterized was TCIRG1- defect v-ATPase autophagy the dependent Remarkably, progression. cancer for relev response cell RHOA-controlled another migration, cell affect might autophagy in howis defects issue a key RHOA, and that depletion of depletion that cells cells would be to sufficient impair cell ( motility A549 in autophagy-competent tumor sequestration of autophagy ] knockdowns, gene deletion, and use of chemical inhibitors chemical of use and deletion, gene knockdowns, ] Instead of Instead the proteasome, however, we demonstrate recently Considering the apparent connection between autophagy autophagy between connection apparent the Considering If our model is correct, one would expect that the inhibition inhibition the that expect would one correct, is model our If Fig. 18,19 Fig. ). A role for autophagy in controlling the localization of localization the controlling in autophagy for role A ). We thus followed the wild-type (WT) and and (WT) wild-type the followed thus We 2E and F and 2E 2B and C 17 Indeed, through targeted manipulation of the auto- of the manipulation targeted through Indeed, Fig. ATG5

; 1A Fig. S1DFig. ; Fig. S1C ). Correlated with the mesenchymal ). spread- with Correlated the mesenchymal or Atg5, Atg5, Atg7, Sqstm1 Tcirg1 11-16 SQSTM1 ; but whether RHOA-GTP is simi- is RHOA-GTP whether but ) and consistently suppressed motil- Vids. S2–S6 Vids. nl cls ( cells -null Fig. S1BFig. Tcirg1 ( Fig. S1AFig. Fig. ), a denser ACTIN net- ACTIN denser a ), -null cells was a punc- a was cells -null , and ), in comparison with with comparison in ), 2G Fig. Fig. Fig. ) recapitulated the the recapitulated ) ). Fig.

Tcirg1 1A Fig. 2B and C and 2B

1D 2 , left panel), panel), left , 1B and C and 1B ). ). We found ). [v-ATPase Tcirg1 Tcirg1 17 oidy, one In light light In RHOA RHOA Tcirg1 17 ATG5 rma- -null -null -null -null , left left , enti- Autophagy ant ant ase ase sin sin ts, ts, he n- ly i- ), ), h d - - ;

control shRNA-transduced cells ( cells shRNA-transduced control the inhibition of autophagosome formation by formation autophagosome of inhibition the produced similar effects ( effects similar produced atively regulates several signaling pathways by degrading degrading by pathways signaling kinases, several regulates atively its cell signaling involve both ubiquitination and SQSTM1 SQSTM1 and ubiquitination both ( involve signaling cell its infections, and treatment with infections, chemotherapy drugs. inhibitio proteasome inhibition, HSP90 m detachment, (ECM) extracellular starvation, nutrient as such conditions ful RHOB, within lysosomes within RHOB, and RHOH proteins, RHO active constitutively two of dation nated and recognized by the autophagy receptor SQSTM1. receptor autophagy the by recognized and nated ubiquiti- is RHOA that identified we treatment, chloroquine by pcfcly erds h mmrn-soitd n atv p RHOA. of active and membrane-associated the degrades specifically the and degrades forms, cytosolic inactive the path autophagy RHOA routes uses distinct for while the degradation: proteas evidence of lines first the provide we state, activation its on along with the proteasome and autophagy ( autophagy and proteasome the with along proteins GTPase-activating and GEFs itinvolves that in unique is RHOA of regulation The RHOA-GTP. of regulator master a established. to be remains functions pressive tumor-s to thereby ensure and conditions growth basal under pro signaling degrade to autophagy of ability the Therefore, ride (LPS). lipopolysacch ride inducer, autophagy an with activation after or atr ( factors of RHOA signaling as reflected by the loss of F-ACTIN, and and F-ACTIN, ( of promotionlamellipodia of the loss the by reflected as signaling RHOA of SQSTM1 migration of migration c and lamellipodia, broad of formation spreading, cell moted pro Y27632 by treatment kinase RHO inhibiting scenario, this the phenotype ( phenotype the fo responsible directly was overactivation RHOA that showing for fine-tuning the RHOA pathway to ensure cell motility. cell to ensure pathway RHOA the for fine-tuning required is SQSTM1-dependentautophagy that support provide autophagic autophagic pathway in which LC3 RHOA dampens byactivation the and RHOA between loops feedback of existence the suggest connec recent 2 concept, this with auto- line In targeting. selective phagic for LC3 marker autophagosome the to RHOA bridg that SQSTM1 protein scaffold the recruiting uitination, ubiq- undergoes activated, once RHOA, that propose therefore RHOA to undefined intracellular puncta during cell migratio cell during puncta RHOA intracellular to undefined where RHOA should be inactivated. be should RHOA where localizat temporal and subcellular same the cells, migrating edge ofleading at the autolysosomes targeting the we reported Accordi substrates. its downstream with able to interact and is active, where place GTPase a and c RHO time the dictate might extracellular of integration its with together autophagy of ( activation RHOA off turns constitutively that pathway suppressor tumor a Fig. Our data raise the importance of constitutive autophagy as as autophagy constitutive of importance the raise data Our eet oplig vdne ugss ht uohg neg- autophagy that suggests evidence compelling Recent So far, the emerging mechanisms by which autophagy lim- autophagy which by mechanisms emerging the far, So

2H 20-27 Fig. ). (rescue) back into these cells cancelled the stabilization the stabilization cancelled intocells back these (rescue) 27-30,32,33,35 17 38 e dwsra components downstream key Fig. This is consistent with the recently reported degra- reported recently the with consistent is This

We therefore propose that basal autophagy acts as as acts autophagy basal that propose therefore We 2H ATG5 2G Fig. ). 30-35 Importantly, upon inhibition of autophagy autophagy of inhibition upon Importantly, -depleted cells and and cells -depleted ). In this model, the remarkable dynamics dynamics remarkable the model, this In ).

2D–F oee, hs fe ocr udr stress- under occurs often this However, 36,37 Fig. S1 Fig. ; Vids. S3–S6 Vids. and the redistribution of the active active of the redistribution the and Fig. ). Reintroduction of Reintroduction ). Fig. 2B and C and 2B 2A and D and 2A SQSTM1 ). These data together together data These ). 28,29 Fig. and transcription transcription and ). In keeping with with keeping In ). 2G ATG7 Volume 10Issue2 -depleted cells, cells, -depleted ). As expected, expected, As ). ). Depending Depending ). 20,23-27,29,31-33,35 depletion depletion ATG5 17 tions tions ngly, teins teins atrix atrix ome ions that that We way way up- n, ues ues ool ool or ell ell of n, es es a- 19 r - ,

©2014 Landes Bioscience. Do not distribute. Downloaded by [Inserm Disc Ist] at 19:52 16 October 2017 www.landesbioscience.com normal small morphology and the formation of actin fibers (arrowhead) fibers actin of formation the and morphology small normal and XY migration tracks (left, cell positions were recorded e were recorded cell positions (left, tracks and migration XY phagosome formation by by formation phagosome trusion at the leading edge (arrowhead). ( (arrowhead). edge leading the at trusion (inset, arrowhead), and the formation of F-AC of formation the and arrowhead), (inset, AC video microscopy (right) and XY migration tracks (left) sho (left) tracks migration XY and (right) microscopy video at at 15.2 ± 3.9 of the apparent RHOA sequestration, sequestration, RHOA apparent the of phagosomes (LC3-positive, arrowhead) and autolysosomes (L autolysosomes and arrowhead) (LC3-positive, phagosomes Figure TI N (green) and microtubule (red) staining showing that that showing staining (red) microtubule and (green) N

1. Cell motility is induced by v-ATPase by induced is motility Cell μ m/h over long distances (still-images from m/h over (still-images long distances Atg5 sh RNA dramatically increases localization of active RHOA at t at RHOA active of localization increases dramatically RNA Tcirg1 C ) Left. Representative confocal images showing in WT cells th cells WT in showing images confocal Representative Left. ) -null cells lacked AC lacked cells -null TI Tcirg1 N (arrowhead). Time-lapse video microscopy (middle, image (middle, microscopy video Time-lapse (arrowhead). N loss. ( loss. Vid. S1Vid. A wing that the the that wing very 60 min very for 7 h) cells at that moved slowly showing WT 2 ± 1 Tcirg1 ) Confocal images of v-ATPase of images Confocal ) C3- and LAMP1-positive, arrows), close to highly dynamic lam dynamic highly to close arrows), LAMP1-positive, and C3- ; intervals in ; h:min).of intervals cell the mov Arrows indicate direction TI -null cells displayed a classical front-rear polarized morph polarized front-rear classical a displayed cells -null N stress fibers and developed aberrant lamellipodia (arrow lamellipodia aberrant developed and fibers stress N Autophagy . Tcirg1 -null cells established short-lived cellular contacts, and m and contacts, cellular short-lived established cells -null Tcirg1 he plasma membrane of of membrane plasma he -null cells showing colocalization of RHOA with auto- with RHOA of colocalization showing cells -null e presence of RHOA at the plasma membrane membrane plasma the at RHOA of presence e s selected from from selected s Tcirg1 μ ement. Inset: representative representative Inset: ement. ology, with membrane pro- membrane with ology, -null cells, which restored a restored which cells, -null ellipodia (right). As a result result a As (right). ellipodia m/h. ( m/h. Vid. S1 Vid. head, left). ( left). head, D ) Inhibition of ) auto- Inhibition ; intervals in h:min) in intervals ; igrated rapidly rapidly igrated B ) T ) im e-lapse e-lapse 203

©2014 Landes Bioscience. Do not distribute. Downloaded by [Inserm Disc Ist] at 19:52 16 October 2017 www.landesbioscience.com tial control of RHOA, of control tial sp and temporal the light a new under to revisit helps us phagy brane. mem- plasma the with ATG16L1) associate least autophagy (at the machinery of components some that and phagophores, of in the formati membrane participates plasma that endocytosis, increa stimuli autophagy that observations the are evidence of nals that target connexins for autophagy. selective connexins target that nals sig- also are SQSTM1 and ubiquitination RHOA, to Similarly eitr f autophagy. of mediator of the BECN1,degradation RHOA active while triggers a critic GABRA/GABAA receptors, and connexins. and receptors, GABRA/GABAA protein membrane plasma internalized 2 of degradation the in Remarkably, was recently involvautophagy somal degradation. lyso- by followed endocytosis involves which of common most th degradation, for targeted be can proteins membrane plasma which in exclusive, non-mutually ways, of number a are There degradation autophagy to targeted is protein membrane plasma migration. directed and exit cycle ible cell irrevers- to cells commit might RHOA and autophagy between h pam mmrn of membrane plasma at the RHOA of “retention” higher the is study our of finding ing to rapid trafficking and fusion with the lysosomes, as recen as suggested. lysosomes, the with fusion and trafficking rapid to leading endosomes, RHOA-containing the to ATG7) (ATG5 and LC3, machinery autophagy of recruitment the is possibility Anothe autophagosomes. with fuse that into endosomes nalized inter- are membrane plasma the at present RHOA active blethat internalization (3-methyladenine, (3-methyladenine, internalization its preventing by membrane plasma the at 43) (connexin GJA1 of accumulation induces autophagy of blockage the that dence already is there accordance, In accumulation. intracellular of instead cells), A549 and cells PCT in (both cells depleted inactivating the RHO-GEF activity of AKAP13/AKAP-Lbc, activity RHO-GEF the inactivating rapid migration, by enhancing the degradation of RHOA. Of Of RHOA. of degradation the enhancing by thei to migration, contribute rapid can cells cancer in autophagy of activation inapprop that possibility the raise also findings Our phagy. be our next challenge. next our be to coordinated ofprotein allow timely degradation signaling a endocytosis and signal how autophagy Understanding proteins. ing of degradation and endocytosis orchestrates autophagy f SQSTM1, of opposite effects ( effects opposite cell migration through the increased RHOA sequestration ( sequestration RHOA increased the through migration cell have a can profoun pathway autophagy in the alterations result, the plasma membrane ( membrane plasma the for for the control of the RHOA pathway cell and mi the subsequent and and [ microscopy video expressing proteins encoded by shRNA-resistant murine autopha murine shRNA-resistant by encoded proteins expressing tumor cell migration and only compromised rear retraction ( rear retraction and cell only tumor compromised migration and the following degradation would limit the level of activ of level the limit would degradation following the and phagy defects on the RHOA pathway in A549 Depleti cells. tumor defects phagy Figure Future work is required to understand Future how,work to is required understand mechanistically In conclusion, this unexpected link between In RHOA conclusion, between link and unexpected this auto- SQSTM1 46

Altogether this might suggest that SQSTM1-dependent that suggest might this Altogether 2 (See opposite page). opposite (See 2 45 t ds ih noyoi hptei, n intrigu- an hypothesis, endocytosis with odds At -depleted A549 cells inhi (right) kinase RHO upon were rescued -depleted 48-50 ATG5 n te uo-upesv fnto o auto- of function tumor-suppressive the and D–F , ATG7 ] and wound closure assay [see also also [see assay closure wound and ] A–C 18,19,47 41 Together, this sophisticated crosstalk crosstalk sophisticated this Together, , ; left), a denser F-AC denser a left), ; SQSTM1 ATG5 the signaling and tumorigenic roles roles tumorigenic and signaling the Autophagy is required for tumor A549 cell migration through t through migration cell A549 tumor for required is Autophagy shRNA). ( shRNA). -depleted cells and and cells -depleted Atg7 - and and - H ) Growing list of signaling proteins degraded by autophagy. by degraded proteins signaling of list ) Growing TI N reticulation ( reticulation N 42-44 Atg5 It is thus possi- thus is It 43,44 nl cells). -null TCIRG1 e RHOA present at the plasma membrane, thereby relieving inh relieving thereby membrane, plasma at the present e RHOA Other lines Other lines D Fig. S1 Fig. , lower (See panel). also SQSTM1 gration; once activated, the sequestration of RHO the active the sequestration once gration; activated, d impact upon cell motility: inhibition of autophagy degrada autophagy of inhibition motility: cell upon d impact on on of A–C gy genes (red). ( (red). genes gy loss), whereas blocking autophagosome formation or seques or formation autophagosome blocking loss),whereas ]). Shown are confocal images of A549 cells cotransfected with h with cotransfected cells A549 of images confocal are Shown ]). s will s will riate riate Autophagy ; right) and a strong suppression of tumor cell migration (ex migration cell tumor of suppression strong a and right) ; evi- 39,40 ATG5 tly tly on an an ed ed , a re se se a- al al 44 s, s, e r r - - . ,

bition with a of bition Y27632low concentration (2.5 ( B and E tubules (PCT) of wild-type (WT, of wild-type (PCT) tubules convo proximal from derived cells renal autolysosomes, tive ing membrane. ing by line cell A549 cancer lung human the in formation phagosome n te erimn o atpay usrts ihn h auto- the by vesicles within phagic substrates autophagy of recruitment the transduction, and SHCLNV-NM_NM_006395) (Sigma, shRNA v-ATPase (mutations in PS1), in (mutations v-ATPase of v-ATPase, of n dwsra targets. downstream and regulator upstream their GTPases, Rho several degrade which that 86) te hrpui bnft f rg sc a chloroquin as such investigation. careful warrants drugs of benefit therapeutic the refs. 58–60), and study (our migration acidification cell enhance autolysosomal might degradation of inhibition that Given cancer. metastati of treatment for useful be might initiation autophagy v-ATPase ln 51 ad QT1 B Tasuto Laboratories™, 610833). Transduction BD SQSTM1, and Nanotools 5F10 clone LC3-II, Inc.; Technology Signaling Cell D12B11, clone ATG7, Nanotools; 7C6, clone (ATG5, immunoblotting by monitored were impairment autophagy and downregulation u study. our with agreement in LC3, of staining punctate and SQSTM1 overexpress of i.e. autophagy, “enhanced” an by characterized (melanoma, cancers all prone to metastas cancers) interestingly, colorectal and pancreatic, lung, deaths; cancer of 90% importan mo for responsible is Metastasis suggests implications. therapeutic that observation an migrate, to failed cells tum A549 the autophagy, silencing when interest, particular SHCLNV-NM_NM_003900). n RHOA-GTP, SQSTM1 and LC3-II, substrates autophagy and/or the of autolysosomes accumulation the by reflected as fusion, of downstream degradatio autolysosomal blocks thereby and pH lysosomal the ously. prev described as vector, neo pSV3 the with immortalized and To inhibit the maturation of autophagosomes into degrada- into autophagosomes of maturation the inhibit To treatments and culture Cell ATG5 E and F and E Tcirg1 17 ) and Vids. S2–S6 Vids. PCT cells were chosen, as they express the highest level level highest the express they as chosen, were cells PCT hN (im, HLVN_089 or SHCLNV-NM_004849) (Sigma, shRNA 51-57 a3 SQSTM1 ) The migratory defects of of defects migratory The ) depletion does not affect the activity of proteasomes, proteasomes, of activity the affect not does depletion 61 -deficient, Jackson Laboratory) mice were isolated isolated were mice Laboratory) Jackson -deficient, Based on our result, we propose that inhibiting inhibiting that propose we result, our on Based the a3 subunit is localized in the lysosomal limit- lysosomal the in localized is subunit a3 the 62 he degradation of RHOA. ( RHOA. of degradation he In agreement with the other reported defect of defect reported other the with agreement In ). ( ( Materials and Methods C and F 17 SQSTM1 G codn to. according ) Proposed model by is model ) which essential autophagy Proposed ) in RHOA A549 cells levels at to led increased 17 , through shRNA treatment (Sigma, (Sigma, treatment shRNA through , lentvl, e rvne auto- prevented we Alternatively, 63 V-ATPase 17 hN-eitd protein ShRNA-mediated ATG5 Tcirg1 64 Likewise, we have checked checked have we Likewise, μ -depleted A549 cells (middle) (middle) cells A549 -depleted A–F M) that A549 did not impede A within autophagic vesicles A vesicles autophagic within ibitory RHOA signaling. As a As signaling. RHOA ibitory +/+ tion would promote rapid rapid promote would tion ) or ) Consequences of auto- of Consequences ) Tcirg1 tration would have the the have would tration amined by time-lapse time-lapse by amined tcirg1 uman shRNA and and shRNA uman depletion raises raises depletion − / − (lysosomal (lysosomal re than than re ize are are ize ATG7 luted ion 205 or s, s, i- n e c / t

©2014 Landes Bioscience. Do not distribute. Downloaded by [Inserm Disc Ist] at 19:52 16 October 2017 6. Qu X, Yu J, Bhagat G, Furuya N, Hibshoosh H, H, Hibshoosh N, Furuya G, Bhagat J, Yu X, Qu 6. 206 5. Livesey KM, Tang D, Zeh HJ, Lotze MT. Autophagy MT. Autophagy Lotze HJ, Tang D, Zeh KM, Livesey 5. Fung C, R, J. Lock E, Gao Debnath S, Induction Salas 4. 3. Lock R, Debnath J. Extracellular matrix regulation regulation matrix Extracellular J. Debnath R, Lock 3. 2. Levine B, Kroemer G. Autophagy in the pathogenesis pathogenesis the in Autophagy G. Kroemer B, Levine 2. 1. Mathew R, Karantza-Wadsworth V, White E. Role of Role V,E. White Karantza-Wadsworth R, Mathew 1. with a CO equippe a with Zeiss) Carl Observer.Z1; (Axio microscope an inverted on imaged were medium growth complete in cultured cells (Zeiss microscope confocal Zeiss). a Carl LSM5; using examined and PBS, with antifade washes 3 Gold after DAPI P36931)with Long Probes, Pro (Molecular reagent with mounted Sigma, were 1:1,000; Slides temperature; P1951). room h; with (1 cytoskeleton TRITC-phalloidin ACTIN the of staining before temperature room at min 15 for paraformaldehyde 3.7% with PBS in fixed anti-LC3 BD 553792) PharMingen, Alternatively, cell (1:500, labeling. antibody Biotechnology), anti-LAMP1 and Cruz Nanotools) 5F10, clone Santa (1:500, 1:100; 26C4; (clone (LC3 for 5 min. To of detect the translocation RHOA to autolysosomes Triton 0.3% in X-100 permeabilized PBS in and PBS, in washed associated, but not cytoplasmic, RHOA. cytoplasmic, not but associated, membrane- preserves that method a 15 for min, TCA 10% cold seeded oncells thus coverslips activated, w were glass fixed (2.5 Y27632 inhibitor Kinase RHO selective the of dose low a cated, ind Where migration. iv)and cell cytoskeleton; tion of ACTIN re the iii) 3671); blotting; Technology,western Signaling Cel (P-MLC, chain light regulatory myosin of phosphorylation Ser19- downstream the ii) membranes; to ofRHOA recruitment dent experiments. Images were captured every 5 min with a with 5 min every were captured Images dent experiments. photographed w and a of total 18 were condition followed in fields 3 indep per fields different Three stage. X-Y motorized For monitoring single-cell migration, exponentially growi exponentially migration, single-cell monitoring For microscopy video Time-lapse To determine whether RHOA is membrane-associated and and membrane-associated is RHOA whether determine To staining Immunofluorescence the i) by monitored was pathway RHOA the of activity The pathway RHOA of the Analysis J Clin Invest 2003; 112:1809-20; PMID:14638851 112:1809-20; 2003; Invest J Clin gene. autophagy 1 beclin the of N, disruption erozygous Mizushima het- by EL, tumorigenesis of Promotion al. et Y, Eskelinen Ohsumi J, Rosen A, Troxel Curr Opin Investig Drugs 2009; 10:1269-79; 10:1269-79; treatment. 2009; Drugs cancer Investig PMID:19943199 combination Opin Curr in inhibition mbc.E07-10-1092 19:797- 806; 2008; Cell Biol Mol survival. cell promotes detachment matrix extracellular during autophagy of ceb.2008.05.002 f uohg. ur pn el il 08 20:583- 2008; Biol 8; Cell Opin Curr autophagy. of http://dx.doi.org/10.1016/j.cell.2007.12.018 of disease. Cell 2008; 132:27-42; 2008; Cell disease. of nrc2254 uohg i cne. a Rv acr 07 7:961- 2007; 7; Cancer Rev Nat cancer. in autophagy μ PMID:17972889 PMID:18573652 + and LAMP1 M; Calbiochem, 688002) was added. was 688002) Calbiochem, M; MD1043; http://dx.doi.org/10.1091/ PMID:18094039; 2 -equilibrated chamber (37 °C and 5% CO 5% and °C (37 chamber -equilibrated References + ; ), cells ), were cells then incubated with anti-RHOA http://dx.doi.org/10.1038/ ; http://dx.doi.org/10.1016/j. PMID:18191218 ; 65 The cells were then then were cells The 10 11 9. Mathew R, Kongara S, Beaudoin B, Karp CM, CM, Karp B, Beaudoin S, Kongara R, Mathew 9. . aataWdwrh , ae S Kacu O, Kravchuk S, Patel V, Karantza-Wadsworth 8. 7. Yue Z, Jin S, Yang C, Levine AJ, Heintz N. Beclin Beclin N. Heintz AJ, Levine C, Yang S, Jin Z, Yue 7. . Doye A, Mettouchi A, Bossis G, Clément R, Buisson- R, G, Clément Bossis A, Mettouchi A, Doye . . Fujiwara T, Bandi M, Nitta M, Ivanova EV, Bronson EV, Bronson Ivanova M, Nitta M, T, Bandi Fujiwara . activation for bacterial host cell invasion. Cell Cell invasion. cell host 111:553-64; 2002; bacterial for GTPase Rho activation restrict to ubiqui- machinery M, the exploits tin-proteasome Piechaczyk CNF1 E. L, Lemichez P, Gagnoux Boquet G, Flatau C, Touati raploids promotes tumorigenesis in p53-null cells. cells. p53-null 437:1043-7; 2005; in Nature tumorigenesis tet- promotes generating raploids failure Cytokinesis D. Pellman RT, gad.1545107 dx.doi.org/10.1038/nature04217 org/10.1016/S0092-8674(02)01132-7 81; 21:1367- 2007; Dev Genes instability. E. chromosomal White limiting by S, progression Jin tumor G, suppresses Autophagy Chen K, Degenhardt K, Bray gad.1565707 35; in 21:1621- 2007; damage Dev Genes genome tumorigenesis. Autophagy and mammary E. stress White metabolic S, Jin mitigates R, Mathew G, Chen o. rc al cd c U A 03 100:15077- 2003; A S U 82; Sci Acad Natl suppres- Proc tumor sor. haploinsufficient a embryonic is early for development, essential gene autophagy an 1, pnas.2436255100 PMID:17510285 PMID:17606641 PMID:14657337 2 ) and a and ) ith ith ice- ticula- s were × Autophagy 20 20 en- ere ere ng ng PMID:12437928 i- d l ; http://dx.doi.org/10.1101/ ; http://dx.doi.org/10.1101/ ; www.landesbioscience.com/journals/autophagy/article/ “STEATOX” and GM053396). (DJK: NIH and # ANR-13-CESA-0009-01) ANR-11-LABX-0028-01, # SIGNALIFE LABE Future” the for (“Investments agency research national French SL220110603478), n° Grants (ARC Cancer” le Recherche contre la pour “Association 6.3.3.4), and n°6.3.3.3 PACA Alpes Côte d’Azur (A B: plan régional santé environnement PR logementProvencdu et l’aménagement de l’Environnement, de régiona Direction and d’Azur Côte Alpes Provence santé nale régio- Agence 0044), n°C 62 08 ADEME convention C: A and Médicale,” Recherche la B (A de l’Energie” de Maîtrise la de et l’Environnement “Agence et de Santé support la Grant de discussions. National helpful “Institut for Albrengues Jean and 6) ±s.d. 6) = (n clones independent 2 from scratches 3 of means are values the wound area that was closed using Visilog 5.3software Visilog using closed was that area wound the o percentage the as quantified was later,h 48 migration and ing wound- after immediately captured were ofwounds domimages remove ra to Three applied. then was complete medium fresh and debris, washed were Cells area. wound uniform a create to ImageJ. with images acquired the from were generated Videos during 18 Devices h (Molecular using 2.0 MetaMorph software ph2 ( ph2 http://dx.doi.org/10.1073/ PMID:16222300 Supplemental materials may be found here: here: found be may materials Supplemental Boulter, Singer, Etienne Nathalie Rochet, Nathalie Wethank were disclosed. of interest conflicts No potential Confluent cell monolayers were scratched with a pipet tip tip pipet a with scratched were monolayers cell Confluent assay wound Scratch Tcirg1 http://dx.doi. ; -null) and and -null) Disclosure of Potential Conflicts of Interest Conflicts ofPotential Disclosure ; http:// ; Supplemental Materials Supplemental 15 13 14 12 × Acknowledgments 32 ph1 (WT) phase-contrast objective objective phase-contrast (WT) ph1 32 . Chen Y, Yang Z, Meng M, Zhao Y, Dong N, Yan H, H, Yan N, Y,Dong Zhao M, Meng Z, Y,Yang Chen . . Asanuma K, Yanagida-Asanuma E, Faul C, Tomino C, Faul E, Yanagida-Asanuma K, Asanuma . . Ozdamar B, Bose R, Barrios-Rodiles M, Wang Wang M, Barrios-Rodiles R, Bose B, Ozdamar . . Wang HR, Zhang Y, Ozdamar B, Ogunjimi AA, AA, Ogunjimi B, Ozdamar Y, Zhang HR, Wang . f hA inlig Nt el il 06 8:485- 2006; Biol Cell Nat 91; regulation signalling. via RhoA motility of cell orchestrates and Synaptopodin organization P. actin Mundel K, Kim Y, 55; cytoskeleton 35:841- 2009; Mol Cell actin movement. cell and structure control con- to adaptors evolutionarily BTB through served RhoA of mediates Cullin degradation F. Shao HB, Peng M, Ding L, Liu ncb1400 molcel.2009.09.004 9; 302:1775- 2003; Science degradation. target- for by RhoA ing formation protrusion and polarity cell of Regulation JL. Wrana GH, Thomsen E, Alexandrova pteil el lsiiy Sine 05 307:1603- 2005; controls Science receptors 9; plasticity. TGFbeta cell by epithelial Par6 polar- protein the of ity Regulation JL. Wrana Y, Zhang HR, science.1090772 science.1105718 PMID:14657501 PMID:15761148 PMID:16622418 PMID:19782033; PMID:19782033; ; ; http://dx.doi.org/10.1038/ ; http://dx.doi.org/10.1016/j. http://dx.doi.org/10.1126/ http://dx.doi.org/10.1126/ Volume 10Issue2 27198 ® . The The . SE n- X le ). ). e f : ©2014 Landes Bioscience. Do not distribute. Downloaded by [Inserm Disc Ist] at 19:52 16 October 2017 27 28 26 25 24 23 22 21 20 18 19 17 www.landesbioscience.com 16 . Belaid A, Cerezo M, Chargui A, Corcelle-Termeau Corcelle-Termeau A, Chargui M, Cerezo A, Belaid . . Goussetis DJ, Gounaris E, Wu EJ, Vakana E, Sharma Sharma E, Wu E, EJ, Vakana DJ, Gounaris Goussetis . . Heasman SJ, Carlin LM, Cox S, Ng T, Ridley AJ. AJ. Ridley T, Ng S, Cox LM, Carlin SJ, Heasman . . Niida M, Tanaka M, Kamitani T. Downregulation Downregulation T. Kamitani M, Tanaka M, Niida . . Sandilands E, Serrels B, Wilkinson S, Frame MC. MC. Frame S, Wilkinson B, Serrels E, Sandilands . . Machacek M, Hodgson L, Welch C, Elliott H, Pertz Pertz H, Elliott C, Welch L, Hodgson M, Machacek . . Boulter E, Garcia-Mata R, Guilluy C, Dubash A, A, Dubash C, Guilluy R, Garcia-Mata E, Boulter . . Fliss PM, Jowers TP, Brinkmann MM, Holstermann Holstermann MM, TP,Brinkmann Jowers PM, Fliss . Sniad E Sres , cwn G Mro JP, Morton DG, McEwan B, Serrels E, Sandilands . . Qing G, Yan P, Xiao G. Hsp90 inhibition results results inhibition Hsp90 G. Xiao P, Yan G, Qing . . Paul S, Kashyap AK, Jia W, He YW, Schaefer BC. BC. Schaefer YW, He W, Jia AK, Kashyap S, Paul . Huh S Yn C Si N, hag J L CF, Li NJ, Chiang NY, Shih CC, Yen YS, Hsueh . . Kim JE, You DJ, Lee C, Ahn C, Seong JY, Hwang Hwang JY, Seong C, Ahn C, Lee DJ, You JE, Kim . http://dx.doi.org/10.1016/j.immuni.2012.04.008 muiy 02 36:947-58; 2012; Immunity Blood 2012; 120:3555-62; 2012; Blood trioxide. arsenic by responses antileukemic of generation and oncoprotein BCR-ABL the of Autophagic degradation LC. Platanias JK, Altman M, Bogyo B, auto.22802 netnl toa tmr. uohg 21; 9:220- 33; 2013; Autophagy tumors. stromal gastrointestinal in degradation KIT NVP-AUY922-induced and endogenous in involved is LT.Autophagy Chen ouae T el eetr ciain f NF- of activation receptor Bcl10 cell protein T adaptor the modulates of autophagy Selective dx.doi.org/10.1182/blood-2012-01-402578 reduced FAK signalling. Nat Cell Biol 2012; 14:51- 60; 2012; Biol following Cell Nat survival signalling. cell FAK reduced target- cancer promotes Autophagic Src al. of et ing M, Vidal VG, WY, Brunton C, Langdon Stevens K, McLeod JP, Macagno org/10.1038/embor.2012.92 Rep EMBO 13:733-40; in 2012; cells. Ret cancer of FAK-signalling-defective degradation autophagic Src-dependent ncb2386 nal.ppat.1002517 oy acd. LS ahg 02 8:e1002517; 2012; Pathog PLoS PMID:22319449 cascade. tory rn W Vrl eitd eieto o NEMO/ of redirection IKK mediated Viral W. P, Ghazal Brune H, Hohenberg P, Dickinson C, Mack B, dx.doi.org/10.1016/j.cellsig.2010.06.004http:// PMID:20600852; 22:1645-54; 2010; Signal Cell phosphorylation. of autophagic inhibition and through degradation activity IKKbeta of KEAP1 by regulation signaling NF-kappaB of Suppression JI. http://dx.doi.org/10.1016/j.molimm.2010.05.004 of active IKK beta by Ro52-mediated autophagy. autophagy. Ro52-mediated 47:2378-87; 2010; by Immunol Mol beta IKK active of org/10.1038/sj.cr.7310109 degradation of IkappaB kinase (IKK). Cell Res Res Cell (IKK). kinase 16:895-901; 2006; IkappaB proteasome-independent of degradation autophagy-mediated in http://dx.doi.org/10.1083/jcb.201002067 tion. J tion. Biol 2010; Cell 190:553-63; migra- transendothelial cell T for required is uropod and edge leading the at signaling RhoA Coordinated is uig el rtuin Ntr 20; 461:99- 2009; Nature 103; protrusion. activi- cell GTPase during Rho KM, ties of Hahn Coordination GL, G. Johnson Danuser A, Abell P, Nalbant O, org/10.1158/0008-5472.CAN-12-4142 nature08242 cell cytokinesis, and genomic stability. Cancer Res Res Cancer 73:4311-22; stability. 2013; genomic and of control cytokinesis, the role cell RHOA, critical active of a plays degradation the Autophagy in al. et S, Tauc Barale I, M, Rubera M, Ilie S, Giuliano F, Pedeutour E, ncb2049 12:477-83; 2010; Biol activ- PMID:20400958 Cell and Nat RhoGDI1. degradation by Regulation ity crosstalk, K. GTPase Burridge Rho PJ, of Brennwald G, Rossi PMID:23196876 PMID:22138575 γ PMID:19693013 o uohgsms utis h inflamma- the curtails autophagosomes to http://dx.doi.org/10.1371/jour- ; http://dx.doi.org/10.1038/ ; PMID:22732841 PMID:23704209 PMID:17088896 http://dx.doi.org/10.1038/ ; http://dx.doi.org/10.4161/ ; http://dx.doi.org/10.1038/ ; PMID:22898604 PMID:22658522 PMID:20627395 PMID:20733052 http://dx.doi. ; http://dx.doi. ; ; http://dx.doi. ; ; http:// κ B. B. ; ; ; 41 38 40 39 36 37 35 33 34 32 31 29 30 . Schmidt-Mende J, Geering B, Yousefi S, Simon HU. HU. Simon S, Yousefi B, Geering J, Schmidt-Mende . . Sukhdeo K, Mani M, Hideshima T, Takada K, Pena- K, T, Takada Hideshima M, Mani K, Sukhdeo . Yoo BH, Wu T, X, Li Y,M, Shirasawa Sasazuki Haniff . . Baisamy L, Cavin S, Jurisch N, Diviani D. The ubiq- D. The Diviani N, Jurisch S, Cavin L, Baisamy . . Jia L, Gopinathan G, Sukumar JT, Gribben JG. JG. Gribben JT, Sukumar G, Gopinathan L, Jia . . Gao C, Cao W, Bao L, Zuo W, Xie G, Cai T, Fu Fu T, Cai G, Xie W, Zuo L, Bao W, Cao C, Gao . . Wang Z, Cao L, Kang R, Yang M, Liu L, Zhao Y, Yu Zhao L, Liu M, Yang R, Kang L, Cao Z, Wang . . Isakson P, M, Isakson Bjørås Bøe A. SO, Autophagy Simonsen . . Cetin S, Ford HR, Sysko LR, Agarwal C, Wang Wang C, Agarwal LR, Sysko HR, Ford S, Cetin . . Bauer PO, Wong HK, Oyama F, Goswami A, Okuno Okuno A, F, Goswami PO, Oyama Wong HK, Bauer . . Chang CP, Chang Su YC, Hu CW, Lei HY. TLR2-dependent . . Pérez-Sala D, Boya Pérez-Sala P, I, Ramos M, Herrera Stamatakis . Lv Q, Wang W, J, Xue F,Hua H, Yan Mu J, Lin R, Lv . epithelial cells. J Biol Chem 2010; 285:5438-49; M109.046789 285:5438-49; 2010; Chem Biol PMID:19778902 J cells. intestinal of epithelial transformation malignant for is Beclin-1 required mediator autophagy of down-regulation ras-induced Oncogenic KV. Rosen EL, Eskelinen S, huntingtin. J Biol Chem 2009; 284:13153-64; 284:13153-64; M809229200 2009; Chem Biol mutant PMID:19278999 of J degradation the huntingtin. enhances of Inhibition kinases N. Rho Nukina H, Miyazaki Y, Kino M, PMID:15169791 increased 279:24592-600; and 2004; Chem Biol J adhesions. Rho-GTPase focal of DJ. restitution activation Hackam epithelial through G, intestinal Apodaca inhibits C, Endotoxin Baty MD, Neal J, dx.doi.org/10.1002/eji.200939556 J Eur cells. 40:525-9; B 2010; not Immunol but T in activation receptor gen anti- upon protein RhoH of degradation Lysosomal M109.054668 PMID:19696020 284:28232-42; 2009; Chem Biol J AKAP-Lbc. of ity activ- Rho-GEF the regulates LC3 protein uitin-like M313620200 degradation through an endo-lysosomal pathway. pathway. 4:e8117; endo-lysosomal 2009; One PLoS an through degradation protein directs RhoB of sequence C-terminal The K. nal.pone.0032584 dx.doi.org/10.1371/journal.pone.0008117 Blocking autophagy prevents bortezomib-induced bortezomib-induced NF- prevents autophagy Blocking dx.doi.org/10.1158/0008-5472.CAN-11-3832 72:3238-50; 2012; Res Cancer cancer. breast human in transition epithelial-mesenchymal attenuate and autophagy activate to PI3KC3 with interacts DEDD ZW. Hu X, Chen X, in lymphoma cells. PLoS One 2012; 7:e32584; 7:e32584; 2012; One PLoS PMID:22393418 cells. lymphoma in eitd erdto o PML-RAR of degradation p62/SQSTM1- mediated by differentiation cell myeloid lates regu- Autophagy al. et KM, Livesey X, Yin M, Y,Xie blood-2010-01-261040 Autophagy 2011; 7:401-11; Autophagy dx.doi.org/10.4161/auto.7.4.14397 M/AA norti. lo 21; 116:2324- 2010; 31; Blood the oncoprotein. of PML/RARA degradation therapy-induced to contributes autophagosome-lysosome pathway. Leukemia Leukemia org/10.1038/leu.2011.303 pathway. 26:1116-9; 2012; proteasome-aggresome- the autophagosome-lysosome by myeloma multiple Cruz V, Mendez G, Ito S, Anderson KC, Carrasco Carrasco KC, DR. Anderson S, Ito G, Mendez V, Cruz cdd.2012.146 ifrnito. el et Dfe 21; 20:515- 2013; Differ 23; Death Cell macrophage M2 differentiation. hepatoma-derived in degradation 90; promoting by 12:781- 2010; Biol Cell Nat degradation. Autophagy signalling Dishevelled al. Wnt et regulates X, Zhang negatively W, Wu J, Zhang W, eetv atpay euae NF- regulates autophagy selective ncb2082 PMID:20574048 PMID:23175187 PMID:20639871 κ β atvto b rdcn I- reducing by activation B -catenin is dynamically stored and cleared in in cleared and stored dynamically is -catenin Autophagy ; ; http://dx.doi.org/10.1371/jour- ; ; ; PMID:22051532 http://dx.doi.org/10.1074/jbc. http://dx.doi.org/10.1074/jbc. http://dx.doi.org/10.1074/jbc. http://dx.doi.org/10.1074/jbc. http://dx.doi.org/10.1038/ ; ; http://dx.doi.org/10.1038/ ; PMID:22719072 http://dx.doi.org/10.1182/ PMID:19950172 PMID:19956591 PMID:21187718 κ B http://dx.doi. ; α κ α oncoprotein. oncoprotein. B lysosomal lysosomal B degradation degradation http:// ; ; http:// ; ; http:// ; ; http:// ; 53 52 51 50 49 42 44 48 43 46 47 45 55 54 . Miller AL, Bement WM. Regulation of cytokinesis cytokinesis of Regulation WM. Bement AL, Miller . Lzv R Cm R, lm V Sdiu SF, Siddiqui V, Klump RL, Camp R, Lazova . . Sanjuan MA, Dillon MA, CP, Sanjuan Tait SW, S, Moshiach Dorsey . . Mathew R, Karp CM, Beaudoin B, Vuong N, Chen Chen N, Vuong B, Beaudoin CM, Karp R, Mathew . . Fujii S, Mitsunaga S, Yamazaki M, Hasebe T, G, M, Hasebe Ishii S, Yamazaki Fujii S, Mitsunaga . . Sato K, Tsuchihara K, Fujii S, Sugiyama M, Goya T, Goya M, Sugiyama S, Fujii K, Tsuchihara K, Sato . Hn , u B Wn W Ci , o D Sn Y, Sun D, Lou W, Cai W, Wang B, Sun C, Han . . Kadandale P, Stender JD, Glass CK, Kiger AA. AA. Kiger CK, Glass JD, Stender P, Kadandale . . Fong JT, Kells RM, Gumpert AM, Marzillier JY, JY, Marzillier AM, Gumpert RM, Kells JT, Fong . . Duran A, Linares JF, Galvez AS, Wikenheiser K, K, Wikenheiser AS, Galvez JF, Linares A, Duran . . Yoshioka A, Miyata H, Doki Y, Yamasaki M, Sohma Sohma M, Y,Yamasaki Doki H, Miyata A, Yoshioka . . Ravikumar B, Moreau K, Jahreiss L, Puri C, C, Puri L, Jahreiss K, Moreau B, Ravikumar . . Bejarano E, Girao H, Yuste A, Patel B, Marques C, C, Marques B, Patel A, Yuste H, Girao E, Bejarano . . Rowland AM, Richmond JE, Olsen JG, Hall DH, DH, Hall JG, Olsen JE, Richmond AM, Rowland . activated in colorectal cancer cells and contributes contributes and cells cancer colorectal in activated is Autophagy H. Esumi A, T,Ochiai Ueno Y, Atomi dx.doi.org/10.1620/tjem.223.243 e 21; 223:243-51; 2011; Med Exp J Tohoku mimicry. vasculogenic and melanoma metastasis with associated is 3 chain microtubule-associated light protein-1 of Overexpression X. Zhao Clin Cancer Res 2012; 18:370-9; 2012; Res Cancer Clin and poor outcome. metastasis, proliferation, ated with associ- and tumors solid of feature common a is sion expres- LC3B Punctate JM. Pawelek RK, Amaravadi http://dx.doi.org/10.1158/1078-0432.CCR-11-1282 Acad Sci U Acad S A 2010; 107:10502-7; Proc Natl and recruitment. blood cell remodeling tical cor- Rho1-mediated in autophagy for role Conserved http://dx.doi.org/10.1073/pnas.0914168107 cell.2009.03.048 C, et al. Autophagy suppresses tumorigenesis tumorigenesis 137:1062- suppresses 2009; 75; Cell p62. Autophagy of elimination al. through et Gelinas C, G, Bhanot A, Reddy K, Bray HY, Chen G, ccr.2008.02.001 811; 8:794- 2012; Autophagy autophagy. by degraded are junctions gap Internalized MM. MW, Falk Davidson receptors in Caenorhabditis elegans. J Neurosci Neurosci J elegans. 26:1711-20; 2006; Caenorhabditis in GABAA of receptors autophagy and clustering synaptic ulate reg- independently terminals Presynaptic BA. Bamber in a ubiquitin-dependent manner. Mol Biol Cell Cell Biol Mol 23:2156-69; manner. 2012; ubiquitin-dependent a in membrane plasma the at connexins of dynamics lates modu- Autophagy AM. P, Cuervo Pereira DC, Spray y h GPs fu. a Cl Bo 20; 11:71- 2009; Biol Cell Nat 7; flux. GTPase Rho by auto.19390 tor in tumorigenesis. Cancer Cell 2008; 13:343-54; PMID:18394557 13:343-54; 2008; Cell Cancer tumorigenesis. in tor media- NF-kappaB signal- important an The is p62 J. adaptor ing Moscat MT, Diaz-Meco JM, Flores ncb1814 Cell Biol 2010; 12:747-57; 2010; Biol Cell Nat structures. to pre-autophagosomal of contributes formation the membrane Plasma DC. Rubinsztein org/10.1523/JNEUROSCI.2279-05.2006 http://dx.doi.org/10.1038/nature06421 dx.doi.org/10.1038/ncb2078 oi. aue 07 450:1253-7; 2007; phagocy- Nature to tosis. pathway autophagy the links rophages mac- in signalling receptor Toll-like al. et S, Withoff JL, Cleveland K, Tanaka M, Komatsu S, Connell F, org/10.1091/mbc.E11-10-0844 2008; 99:1813-9; PMID:18616529 99:1813-9; 2008; Sci Cancer outcome. patient cells poor with cancer correlates and pancreatic in activated is Autophagy Ochiai A. H, Esumi T, Ueno T, Kinoshita M, Kojima ihy xrse i gsritsia cnes It J Int cancers. PMID:18695874 is 33:461-8; 2008; Oncol gastrointestinal marker, in autophagosome expressed an highly LC3, M. Y, Uchiyama Y,Monden Fujiwara S, Takiguchi K, Gotoh I, org/10.1158/0008-5472.CAN-07-1462 07 67:9677-84; Res 2007; Cancer deprivation. nutrient to tolerance the to PMID:19060892 PMID:19524509 PMID:22635056 ; PMID:22496425 PMID:17942897 PMID:16467519 ; http://dx.doi.org/10.1038/ ; http://dx.doi.org/10.4161/ ; http://dx.doi.org/10.1016/j. http://dx.doi.org/10.1016/j. PMID:21415575 PMID:20639872 PMID:22080440 PMID:18097414 PMID:20498061 http://dx.doi. ; http://dx.doi. ; ; http://dx.doi. ; http:// ; ; http:// ; 207 ; ; ; ©2014 Landes Bioscience. Do not distribute. Downloaded by [Inserm Disc Ist] at 19:52 16 October 2017 57 58 59 208 56 . Jo YK, Kim SC, Park IJ, Park SJ, Jin DH, Hong SW, Hong DH, Jin SJ, Park IJ, Park SC, Kim YK, Jo . . Tu C, Ortega-Cava CF, Chen G, Fernandes ND, ND, Fernandes G, Chen CF, Ortega-Cava C, Tu . Ci , h H Hn , hag , u , u T. Xu X, Wu Y, Zhuang M, Han H, Zhu C, Chi . . Rolland P, Madjd Z, Durrant L, Ellis IO, Layfield Layfield IO, Ellis L, Durrant Z, Madjd P, Rolland . 02 7:e52705; One PLoS 2012; metastasis. node lymph and invasion lar lymphovascu- with associated is ATG10cancer of colorectal in expression Increased JC. Kim DH, Cho erc.1.01312 is expressed in breast cancers showing features of of 80; features 14:73- 2007; showing Cancer Relat Endocr disease. cancers aggressive breast in expressed p62 is protein ubiquitin-binding The I. Spendlove R, 00 285:21817-23; Chem Biol J 2010; Drosophila. in tumor metastasis promotes and growth function lysosome of Disruption org/10.1371/journal.pone.0052705 acr e 20; 68:9147-56; http://dx.doi.org/10.1158/0008-5472.CAN-07-5127 2008; Res fibroblasts. Cancer v-Src in lysosomes secreted podo- via invasion and the degradation in matrix extracellular participates some-mediated B H. Band cathepsin V, Band Lysosomal BF, Sloane D, Cavallo-Medved dx.doi.org/10.1074/jbc.M110.131714 PMID:17395976 PMID:23285162 PMID:20418542 ; http://dx.doi.org/10.1677/ PMID:19010886 http://dx.doi. ; http:// ; ; 61 60 62 . Hurtado-Lorenzo A, Skinner M, J, A, El Skinner M, Futai Annan Hurtado-Lorenzo . . Taddei TH, Kacena KA, Yang M, Yang R, Malhotra Malhotra R, Yang M, Yang KA, Kacena TH, Taddei . . Toyomura T, Murata Y, Yamamoto A, Oka T, Sun- T, Oka A, Yamamoto Y, Murata T, Toyomura . 06 8:124-36; Biol 2006; Cell Nat pathway. degradative protein the lates regu- and endosomes early in Arf6 and ARNO with interacts V-ATPase al. et DA, Ausiello S, Bechoua A, Wildeman J, Casanova S, Bourgoin GH, Sun-Wada ajh.21362 ik n 0 ptet. m Hmtl 09 84:208- 2009; 14; Hematol J Am patients. 403 in risk cancer of assessment and disease Gaucher N370S of nature progressive GM, Pastores underrecognized The G, PK. Rennert Mistry KA, Aleck M, Boxer A, differentiation. J Biol Chem 2003; 278:22023-30; 278:22023-30; 2003; PMID:12672822 Chem Biol J osteoclast during differentiation. isoform a3 the with -ATPase H+ vacuolar-type of to localization lysosomes membrane: From plasma M. the Futai Y, Wada GH, Wada org/10.1038/ncb1348 M302436200 PMID:19260119 Autophagy ; PMID:16415858 http://dx.doi.org/10.1074/jbc. ; http://dx.doi.org/10.1002/ http://dx.doi. ; 65 63 64 Yc O Pen A Gozr . n ECT2- An M. Glotzer A, Piekny O, Yüce . . Lee JH, Yu WH, Kumar A, Lee S, Mohan PS, PS, Mohan S, Lee A, Kumar WH, Yu JH, Lee . . Fass E, Shvets E, Degani I, Hirschberg K, Elazar Elazar K, Hirschberg I, Degani E, Shvets E, Fass . n fnto o Ro. Cl Bo 20; 170:571- 2005; Biol 82; Cell J localization RhoA. of the function and regulates complex centralspindlin jcb.200501097 16; J 281:36303- Biol lysosomes. with Chem fusion 2006; and targeting their not but starvation- autophagosomes of induced production support Microtubules Z. M607031200 rupted by Alzheimer-related PS1 mutations. Cell Cell mutations. dis- PS1 are 141:1146-58; 2010; and Alzheimer-related 1 by presenilin rupted require proteolysis Lysosomal autophagy al. and et M, G, Sovak Martinez-Vicente AC, DM, Massey Wolfe CM, Peterhoff org/10.1016/j.cell.2010.05.008 PMID:16963441 PMID:16103226; PMID:16103226; PMID:20541250 ; http://dx.doi.org/10.1074/jbc. http://dx.doi.org/10.1083/ Volume 10Issue2 ; http://dx.doi. ;

Amine Belaid1, 2, 3, Papa Diogop Ndiaye1, 2, 3, Jérémie Roux1, 2, 3, Éric Röttinger1, 2, Yacine Graba4, Patrick Brest1, 2, 3, Paul Hofman1, 2, 3, 5, 6, and Baharia Mograbi1, 2, 3*

1. Institute of Research on Cancer and Ageing of Nice (IRCAN), INSERM U1081, CNRS UMR7284, Nice, F-06107, France.

2. Université de Nice-Sophia Antipolis, Faculté de Médecine, Nice, F-06107, France.

3. Euipe Laellisée pa lA‘C ue Guy Môuet F- 94803 Villejuif, France. 4. CNRS, Aix Marseille Université, IBDML, UMR 7288, Campus de Luminy, Marseille, cedex 09, 13288, France.

5. Centre Hospitalier Universitaire de Nice, Pasteur Hospital, Laboratory of Clinical and Experimental Pathology, Nice, F-06002, France.

6. Centre Hospitalier Universitaire de Nice, Pasteur Hospital, Human Biobank, Nice, F- 06002, France.

Corresponding Author

Baharia Mograbi; Institute of Research on Cancer and Ageing of Nice (IRCAN); Université de Nice- Sophia Antipolis; Centre Antoine Lacassagne, Avenue de Valombrose; 06107 Nice Cedex 02, France; Tel: +33.4.92.03.12.45. Fax: +33.4.92.03.12.41; E. mail: [email protected].

Running title: Targeting autophagy in cancer

Page 1 of 27

Abstract

Survival rates for patients with metastatic or relapsed cancers has remained virtually unchanged during the past 30 years, and new therapeutic options are urgently needed. An attractive option would be to target autophagy, an essential quality control process that degrades toxic aggregates, damaged organelles, and signaling proteins, and acts as a tumor suppressor pathway of tumor initiation. Conversely, other fascinating observations suggest that autophagy supports cancer progression, relapse, metastasis, dormancy and resistance to therapy. This review provides an overview of the Janus-faced roles that autophagy has in cancer initiation and progression and discusses the promises and the challenges of current strategies that target autophagy for cancer therapy.

Graphical Abstract

Keywords Antineoplastic Agents/drug effects/ autophagy targeted therapy; Autophagy addiction/KRAS/BRAF-driven cancers, Tumor metabolism, Tumor resistance, Tumor relapse, Tumor metastasis, Tumor dormancy. Page 2 of 27

O Deee , , a PuMed seah of autophagy ad ae yielded eties, Autophagy and cancer hih ostitutes % of all , atiles pulished o the topi of autophagy. This amount of literature illustrates how autophagy has paradoxical roles in tumorigenesis: Promoting tumor growth or tumor suppression. Not surprisingly, all the signaling pathways that control cancer development, either tumor suppressors or oncogenes, regulate autophagy. Likewise, the results of fifteen years of research do not clarify whether cancer therapies can suppress or upregulate autophagy, and whether upregulation of autophagy can favor tumor cell survival or death. The exact role that autophagy plays in cancer is therefore complex and warrants a further unifying model; consequently, we discuss in the following sections the ways in which autophagy can be both tumorigenic and tumor suppressive.

The beginning of the research on autophagy can be tracked back to 1963 when Christian de Duve Autophagy at a glance observed the occasional appearance of autophagosomes and established the nomenclature for different lysosomal pathways (Nobel Prize in physiology or medicine 1974).

Autophagy, as suggested y its Geek aoy self-eatig, tagets itacellular organelles and constituents to the lysosomes for degradation. So far, three different types of autophagy have been described; namely chaperone-mediated autophagy (CMA), microautophagy, and macroautophagy. These three types essentially differ in the mechanism through which they deliver substrates (cargo) to the lysosomal lumen. Chaperone-mediated autophagy ensures the translocation of a KFERQ-motif bearing protein into the lysosome by the chaperone heat shock protein cognate 70 (HSC70) [1]. Microautophagy involves the direct internalization of small cytoplasmic material into the lysosome through invagination of its membrane [2]. Macroautophagy delivers proteins and organelles to lysosomes for degradation upon sequestration in a double-membrane vesicle – termed autophagosome.

These three autophagic pathways are part of a complex poga that eets the ells needs. The complementary, rapidity, specificity, and most importantly the reversibility of these pathways allow the complete adaptation of cell to its environment.

– Complementarity. All three of these pathways are constitutively active at basal levels to maintain cell homeostasis. In response to environmental stresses, such as those encountered

Page 3 of 27 in cancer, including nutrient starvation, hypoxia, and other forms of metabolic stress, both macroautophagy and CMA are upregulated, but their activation does not occur simultaneously. Instead, within as quickly as within 30 minutes, cells overcome the nutrient depletion by dramatically up-regulating autophagy. Even under persistent starvation, cells do not cannibalize themselves: The upregulation of autophagy is transient, reaches maximal activity around 4–6 first hours, and then gradually declines to basal levels. The decrease in macroautophagy is concomitant with a progressive switch to CMA that may allow the cell to degrade only unneeded proteins for cell survival [1]. In this review, we will focus the discussion on the role of macroautophagy (hereafter referred to as autophagy) in cancer.

– Reversibility. The autophagosomes are formed and degraded within only 8 minutes. As a result, the autophagy pathway is highly dynamic, rapidly increased and suppressed once the stress is removed, providing evidence that, unlike the other stress fates, namely apoptosis, necrosis, or senescence, autophagy is a reversible phenomenon.

– Cargo specificity. Since its discovery in the 1950s, macroautophagy was first thought to be a bulk, nonseletie self-eatig poess. Emerging evidence now suggests that autophagy is a quality control process, which selectively degrades damaged and unneeded proteins/organelles that would otherwise unnecessarily auulate duig a ells life. The autophagy substrates include organelles such as mitochondria (mitophagy), peroxisomes (pexophagy), large protein aggregates (aggrephagy), and even portions of the nucleus and micronuclei (nucleophagy, chromatophagy). Far from being simply a housecleaner, we and others recently provide the evidence that autophagy also negatively regulates signaling pathway by degrading kinases, cell-cycle regulators, G protein and transcription factors [3].

As an adaptive process, the destiny of autophagic cargo, even following degradation, can also be modulated. Under most circumstances, nutrients of digested cytoplasmic material are recycled into biosynthetic pathways. However, under metabolic stress, the products of autophagy can be further catabolized to fuel ATP synthesis required for survival.

By all of these features (adaptation, rapidity, reversion), autophagy is not only a housekeeping process that suppresses tumor initiation by removing harmful components from the cells, but also supplies all intracellular nutrients absolutely required for the survival, the proliferation, the metastasis, and the dormancy of cancer cells.

Page 4 of 27

The Inside autophagy a utophagy: pathway begins A withthreesome the formation of vesicles of a double-membrane with 3 servants compartment, teed a phagophoe that seuestes a potio of the ytosol. The phagophore expands into a completed vesile, a autophagosoe. During the maturation step, the autophagosome acquires an acidic pH and hydrolases by fusing with a lysosome to generate a autolysosoe hee the otet is degaded. The products generated by degradation are then transferred back to the cytosol by permeases in the autolysosomal membrane and recycled into different metabolic pathways (Figure 1).

At the molecular level, a family of 36 autophagy-related (ATG) genes controls the execution of autophagy — from the initiation, maturation to the degradation of the autophagosomes [4].

– The nucleation of the phagophore is critically dependent on the class III PI3K (phosphatidylinositol 3-kinase) complex containing an enzyme VPS34, together with Vps15/p150, BECN1 (Beclin-1/Atg6), and ATG14. PI3K complex catalyzes the production of phosphatidylinositol 3-phosphate [PI3P], thereby generating a signal that recruits several WIPI (WD-repeat domain phosphoinositide-interacting) proteins, which together with ATG2, regulates the trafficking of ATG9 vesicles, the only core ATG protein with a transmembrane domain.

– The elongation of the isolation membrane and subsequent closure of the autophagosome require two ubiquitin-like conjugates. ATG12 is conjugated to ATG5 by the sequential activity of ATG7 and ATG10. The resulting ATG5-ATG12 conjugate then associates with ATG16L1, to form a ~800-kDa multimeric complex (referred to as the ATG16L complex). A fraction of the ATG16L complex localizes to the phagophore and mediates the binding of the LC3/Atg8- phosphatidylethanolamine conjugate (microtubule-associated protein 1 light chain 3-II, LC3- II) via the activity of ATG4, ATG7 and ATG3. Upon recruitment, the incorporation of LC3-II to the isolation membrane governs its elongation, its curvature, and its closure as well as the substrate recruitment into the autophagosome. While the unprocessed form of LC3 (LC3-I) is diffusely distributed throughout the cytoplasm, the lipidated form of LC3 (LC3-II) specifically accumulates on nascent autophagosome until its degradation and thus represents a marker to monitor autophagy.

Page 5 of 27

– The autophagosome eventually seals off and fuses with lysosomes through mechanisms that remain poorly characterized. Upon completion of the autophagosome, the ATG16L complex and most components of autophagy machinery are released from the membrane. Some regulators of the autophagosome-lysosome fusion include LC3, the lysosomal proteins LAMP-1 and LAMP-2 (lysosomal-associated membrane protein), the small GTP-binding protein RAB7, the SNARE protein Syntaxin 17 and the AAA-type ATPase SKD1. Autophagosome- lysosome fusion then results in the activation of the hydrolases which completely degrade the autophagosomal cargo.

Figure 1. Autophagy (macroautophagy) pathway. Shown are the vesicular and molecular steps of autophagy, enabling cell to digest its own cytosol. Inset: At ultrastructural level, the double membrane–enclosed autophagosomes sequestrating morphologically intact cytoplasm and entire organelles (mitochondria) is a hallmark of autophagy.

Page 6 of 27

Regulation of autophagy by oncogenes and tumor To cope continuously with the environment, autophagy flux (from the formation to the suppressors degradation of autophagosomes) must be highly dynamic, rapidly increased and suppressed by signaling pathways. Within minutes, reactions of phosphorylation, acetylation, ubiquitination, lipidation, and proteolytic cleavage, increase the activity of the autophagic machinery [5,6]. This is followed by an general increase in the expression of autophagy and lysosomal proteins (ATP6VB, ATG, BECN, LC… y the transcription factors Tp53, NFκB, FOXO3a, ATF4 and TFEB [7].

Not surprisingly, the mammalian target of rapamycin (mTOR) kinase, a sensor of the nutritional status of the cell, controls the initiation of autophagy. When nutrients and growth factors are available, mTORC1 inhibits autophagy by phosphorylating and maintaining in an inactive state ULK1, which is required for the formation of the phagophore. mTOR and its direct regulators are among the most frequently mutated oncogenes and tumor suppressors in cancer. Specifically, tumor suppressors that negatively regulate mTOR, such as TSC1/2, LKB1, PTEN, and AMPK stimulate autophagy while, conversely, oncogenes that activate mTOR, such as the tyrosine kinase EGFR, RAS, the class I PI3K, and AKT, inhibit autophagy, suggesting that inhibition of autophagy likely contributes to the onset to tumor development [8].

The discovery,Autophagy over a as decade a tumor ago, that suppressor the essential autophagy pathway gene BECN1, suppresses tumor development, has been enthusiastically greeted because of the potential idea that this could lead to new therapeutic strategies for cancer. The first evidence came from the monoallelic deletion of BECN1 in 40% to 75% of breast, ovarian, colon, and prostate cancers [9,10]. Consistently, allelic loss of Becn1 was demonstrated in mice to predispose to lymphomas, hepatocellular carcinomas, and lung carcinomas [11,12]. Similarly, several partners of BECN1 that positively regulate autophagy, such as AMBRA1 [13], BIF-1/SH3GLB1 [14], and UVRAG [15], have been shown to display tumor suppressive effects. Soon afterward, this tumor suppressor function was extended to Atg5, Atg7, and Atg4C [16-18]. It is thus widely accepted that the entire autophagy pathway could suppress tumorigenesis and a growing list of underlying mechanisms has been proposed.

Page 7 of 27

4.1. Mechanisms by which autophagy suppresses tumor development

4.1.1. To switch off oncogenic signaling. The degradation of key signaling proteins is one of the most powerful tumor-suppressive mechanisms by which a cell can control its own growth, its survival, and its motility. We and others have provided the unexpected evidence that autophagy limits several key signaling pathways (mTOR, Wnt, NFNB, RHOA) by degrading kinases, G proteins, downstream components, and transcription factors. We therefore propose the term "SIGNALphagy" to indicate a dedicated type of macroautophagy that degrades and thereby maintains the appropriate level of active signaling proteins to achieve tumor suppression [3,19,20].

4.1.2. To degrade the SQSTM1 oncoprotein. Since 1996 [21], SQSTM1 (sequestosome- 1; also known as p62) has emerged as a critical oncoprotein involved in a myriad of cellular functions. This multifunctional role of SQSTM1 is explained by its ability to recruit and activate key sigalig poteis that otol ell suial NFκB [22]), nutrient sensing (mTOR [23]), oxidative detoxifying stress (NRF2 [24]), and migration (Twist1 [25]); all crucial events that have a direct impact upon cancer development. Not surprisingly, SQSTM1 is absolutely required for KRASG12D-induced tumorigenesis of lung and pancreas in mice [22,26]. In humans, SQSTM1 overexpression was associated with worse cancer-specific survival in lung, gastrointestinal, prostate, liver, kidney and breast cancers, suggesting a broader role for SQSTM1 overexpression in cancer progression [27].

Besides these signaling functions, SQSTM1 was the first identified autophagic substrate and the first characterized autophagic receptor [28]. Indeed, the presence of the LIR (LC3- interacting region) and UBA (ubiquitin-associated) domains enables SQSTM1 to serve as an adaptor for selective autophagy of ubiquitinated substrates (misfolded proteins, signaling proteins and damaged organelles). Autophagy therefore mediates the clearance of SQSTM1 and impairment of autophagy results in SQSTM1 accumulation and oncogenesis in multiple setting [16,29,30]. Overall, the emerging concept is that autophagy is required to degrade SQSTM1 and thereby suppress the inappropriate activation of oncogenic signaling pathway, which can promote survival and tumorigenesis [30].

4.1.3. To limit oxidative stress and DNA damage. Given their highly active metabolism, cancer cells have higher levels of Reactive Oxygen Species (ROS) than normal cells. A high ROS level can damage the DNA, and induce the activation of signaling pathways, thus stimulating Page 8 of 27 carcinogenesis, cancer initiation, and progression. Mitochondria are the main source of ROS and their ROS production increases as these organelles age or become damaged. As a quality control mechanism, mitophagy is a form of selective autophagy, which degrades compromised mitochondria that would otherwise induce genotoxic stress and DNA mutation [31,32]. Autophagy is also an integral part of the DNA damage response that favors DNA repair not only by recycling key proteins involved in this process [33], but also by maintaining the pools of ATP and dNTPs [34,35]; and ultimately by degrading the injured DNA (chromatophagy) [36].

4.1.4. To maintain genomic stability. Genomic instability is present in 90% of solid human tumors, and associated with tumor progression, aggressiveness, drug resistance∼ and poor patient outcome. Evidence to date suggests that autophagy supports genomic stability by impeding retrotransposon insertions [37] and by controlling cytokinesis [20,31,32]. Each time a cell divides, it must duplicate its entire genome, distribute one copy of each chromosome to each pole, and then split, during cytokinesis, the cytoplasm into two identical daughter cells. Recent elegant studies reveal that the autophagy pathway also functions as a guadia of ellula geoe. Defets of the etie autophagy pathay [irrespective of the defect: i.e., either at the step of autophagosome formation (Atg5−/−, Becn1−/+, ATG7 shRNA), sequestration (SQSTM1 shRNA) or degradation (V-ATPase a3−/−] drive cytokinesis failure, multi-nucleation, and losses/gains of entire chromosomes [20,31,32]. In support, activation of autophagy was shown to maintain genome stability in yeast [38] and reduce genomic instability in hepatocellular carcinoma [39]. For faithful inheritance of a diploid genome, we have shown that autophagy is essential to degrade and thereby maintain the appropriate level of the active RHOA GTPase at midbody, a key regulator of the contractile ring that separates the two daughter cells during cytokinesis [20].

4.1.5. To promote autophagic cell death and senescence. Although autophagy first serves as a survival mechanism, a massive autophagy can act as an accomplice that accelerates the mitotic catastrophe or behaves as an actual killer that commits the cell to an autophagic suicide. This non-apoptotic cell death, the type II (autophagic cell death) is defined by the enlarged vesicles, and the demonstrated dependence on the autophagy machinery [40]. It is noteworthy that the forced expression of the HRASV12 oncogene induces the demise of several cell types through a type II pathway that depends on the autophagy genes BECN1 and

Page 9 of 27

ATG5 as well as the BH3-only protein NOXA, which ultimately limits the oncogenic potential of RAS [41,42]. This is in line with the observation that autophagy can be induced during the oncogene-induced senescence (OIS) [43], and depending on its level of induction the RAS- induced autophagy can dictate the cellular response to RAS: to senesce or proliferate [44] (see below). In response to chemotherapy, autophagy has also been connected to senescence, but its role remains controversial. Depending on the cancer cell type and the drug, autophagy activation seems to suppress cellular senescence (imatinib, leukemia cells [45]), favor it (microtubule poison, colon cancer cells and melanoma cells [46]; temozolomide in glioma cells [47]; lymphoma cells [48]) or even occur in parallel but in an independent manner [49].

4.1.6. To limit inflammation. Chronic inflammation has long been recognized as important factor for tumorigenesis. A tumor microenvironment rich in inflammatory cells and pro-inflammatory cytokines supports all stages of tumor development from the growth, the stimulation of angiogenesis and metastasis, to the reduced response to therapy. In different models, autophagy impairment has been shown to lead to exacerbated inflammation, necrotic cell death, and tumor growth [50]. It turns out that autophagy may contribute to tumor suppression by controlling the intensity and duration of the inflammatory responses not only by the rapid removal of apoptotic corpses, but also by blocking cell necrosis, protecting cells from oxidative stress and limiting the production of inflammatory cytokines [51,52].

4.2. Polemic on the tumor-suppressive function of autophagy. Before finishing up this discussion, it is important to note that skeptics have correctly pointed out that all the above mechanisms that explain the tumor suppressive function of autophagy have been mainly reported in vitro and there is little evidence that they actually occur in vivo. Perhaps the most puzzling observation is that the BECN1 gene is located on chromosome 17q21 next to BRCA1, a known tumor-suppressor gene. Thanks to the availability of the Cancer Genome Atlas (https://tcga-data.nci.nih.gov/tcga/), Laddha et al. have evidenced that large deletions encompassing both BRCA1 and BECN1, and deletions of only BRCA1 but not BECN1, are found in breast and ovarian cancers, consistent with BRCA1 loss being a primary driver mutation in these cancers [53].

In response to these concerns, it might be answered that the deletion of the essential autophagy genes, Atg5, and Atg7 in mice, similarly to Becn1, does produce multiple benign Page 10 of 27 tumors (adenomas, PanIn), but only in the liver [16], in the lung [54-56] and in the pancreas [17] and not in any other tissues. In light of these finding, the role of autophagy in cancer needs to be revisited: rather than a general mechanism, these studies support the notion that autophagy may play an important role in tumor suppression but primarily at early stages and specifically to certain cancer types. As only benign lesions did emerge and fail to progress, these findings also reveal the Janus-faced nature of autophagy in tumor progression (see below).

.5.1. Autophagy Autophagy Addiction as a tumor of RA“ -drivenpromoter: tumors. ToThe fuelRAS GTPases tumor (HRAS progression, KRAS and NRAS ) are the most commonly mutated oncogenes in human cancers, associated with worse prognosis, early metastasis, and resistance to therapy. Activating RAS mutations, mainly in codons 12 and 13 are highly prevalent in colorectal, pancreatic, and lung cancers ( 35% of all human cancers). There is currently no effective therapy to treat the RAS-driven cancers.∼ Thus, beyond RAS, there is clearly an urgent need to identify downstream vulnerabilities that are directly involved in tumor progression.

Classically, the rapidly proliferating tumor cells are viewed as being critically dependent on a single oncogene that reprograms their metabolism towards aerobic glycolysis ad glutaiolysis to geeate the iks that ae eeded to podue a e ell. Hoee, it is apparent that the autophagy degradation also allows the cells to acquire these essential metabolites, particularly under stressful conditions, as found in cancer progression. Supporting this idea, the White [54,57,58], Debnath [59], and Kimmelman [60] laboratories elegantly demonstrated that i) autophagy is constitutively activated by the RAS and BRAF oncogenes and ii) the genetic inhibition of autophagy is sufficient to block their tumorigenicity in lung and pancreas, giving rise to benign oncocytomas [54,56,57,60] (Table 1). This has led to the notio that ae ells ae addited to autophagy for their progression: The tumor cells that activate autophagy and thereby successfully adapt to their hostile microenvironment, survive and proliferate whereas the losers that do not activate autophagy, die. This autophagy addiction is not limited to the RAS pathway, as inactivation of Fip200 (also known as Rb1cc1) in the polyoma middle T (PyMT) mammary cancer model impairs tumor

Page 11 of 27 growth [61], and deletion of Atg5 or Atg7 in the liver causes hepatoma formation without progression to hepatocellular carcinoma [16].

5.2. To fuel mitochondrial metabolism. To support their rapid cell growth, cancer cells require an increased generation of ATP and metabolic intermediates to maintain energy status and a higher biosynthesis of macromolecules. Both metabolic requirements are satisfied by increasing autophagy.

5.2.1. To maintain a pool of functional mitochondria. Contrary to the conventional wisdom, mitochondria are not defective in tumor cells, and still have an important role in tumorigenesis: Cancer cells are highly dependent on mitochondria to produce the ATP and the tricarboxylic cycle (TCA) cycle intermediates required for biosynthesis of lipids, proteins, and nucleic acids. Remarkably, only autophagy is capable of degrading damaged mitochondria, in a selective process termed mitophagy.

5.2.2. To supply metabolic substrates. Mitochondria can generate cellular energy from different fuel sources, including glucose, fatty acids, and amino acids. Through the degradation of proteins and lipids, the upregulation of autophagy supplies the nutrient required to maintain aerobic glycolysis [59,62], fatty acid oxidation [54,57] and glutaminolysis [56]. Of note, neoplastic tissues produce high levels of ammonia as a result of an intense flux through glutaminolysis [63]. Ammonia then diffuses to the outside of the cell and act as a signaling molecule that activates autophagy in both the tumor cells and the adjacent stromal cells to optimize via autophagy the supply of nutrients [64,65]. This intricate relationship between metabolism and autophagy is critical for cancer cell growth and survival (as discussed below).

As a result, deletion of Atg5 or Atg7 results in the accumulation of damaged mitochondria, which, in turn, leads to multiple defects in mitochondrial metabolism, including decreased production of TCA cycle intermediates, reduced mitochondrial respiration, and diminished ATP production. This, ultimately, impedes the growth and the tumorigenesis of cancer cells expressing RAS/BRAF, and the PyMT oncogenes [54-61].

5.3. To increase resistance to apoptosis. Apoptotic cell death is a crucial defense mechanism against malignancy that is initiated by a diverse range of signals, many of which are autophagy activators, such as nutrient deprivation, matrix detachment, and DNA damage. Page 12 of 27

Oncogene Autophagy function Models Ref [54,57,58] PRO-TUMORAL HRAS V12 Ç Proliferation, tumor progression Lung Cancer NSCLC KRAS V12 Ç Mitochondrial respiration (fatty acid Mouse model KRAS G12D oxidation) Atg7– , Tp53 – Ç Glutamine-dependent [55] Lung Cancer NSCLC PRO-TUMORAL KRAS G12D Mouse model Atg5–

Cells HRAS V12 TUMOR SUPPRESSOR [41,42] Ç Autophagic cell death fibroblast

TUMOR SUPPRESSOR – Early stages È Oxidative stress (NRF2) [56] Lung Cancer NSCLC BRAF V600E PRO-TUMORAL – Advanced stages Mouse model

Ç Proliferation, tumor progression Atg7– Ç Mitochondrial respiration Ç Glutamine-dependent

PRO-TUMORAL HRAS V12 Ç Glycolysis Cells [59] KRAS Ç Transformation (soft agar)

PRO-TUMORAL Ç ATG5 ATG7 Liver cancer (HCC) KRAS V12 [62] Ç Transformation (soft agar, xenograft) Cells rat2 Ç Mitophagy Pancreas cancer (PDAC) [17,60] PRO-TUMORAL Mouse model KRAS G12D independent of p53 status patient-derived xenografts, CQ DUAL ROLES dependent of p53 status Lung ADK Atg5– [55] KRAS G12D TUMOR SUPPRESSOR – TP53– Pancreas cancer (PDAC) [66] Atg5–, Atg7–, HCQ PRO-TUMORAL – TP53+

TUMOR SUPPRESSOR PRO-TUMORAL

Table 1: Summary of the studies showing the two facets of autophagy in RAS/BRAF-driven cancers.

Page 13 of 27

5.3.1. To enable survival during starvation. During tumorigenesis, cancer cells frequently endure limited supply of growth factors, nutrients and oxygen, in the inner area of the tumor. Autophagy is then activated and inhibition of autophagy by monoallelic loss of Becn1 promotes cell death (necrosis and apoptosis) specifically in the hypoxic regions of tumors [50]. These observations suggest a role for autophagy in promoting tumor cell survival under conditions of metabolic stress.

5.3.2. To enable survival during metastasis. Metastasis is responsible for more than 90% of cancer deaths. Disseminated tumor cells that survive in the circulation, and metastasize to a distant organ site, need to overcome anoikis, a form of apoptosis that takes place when cells are detached from the extracellular matrix (ECM). Following ECM detachment, autophagy was found to be robustly activated and essential to protect against anoikis a large array of cancer cell-lines with RAS-oncogenic mutations [67,68]. Besides survival, we have shown that autophagy promotes the migration of lung cancer cells with KRAS mutation by degrading RHOA at the lamellipodia [19]. Consistently, there are clinical associations between hyperactive autophagy (BECN1 overexpression, punctate LC3B expression) and early metastases in melanoma and colorectal cancers [69-71].

5.3.3. To enable chemoresistance. Radiation and chemotherapeutic drugs, such as imatinib, paclitaxel, cisplatin, 5-fluorouracil (5-FU), arsenic trioxide (As2O3), and TRAIL, all induce autophagy along with apoptosis. A key contributor to drug and/or radiation resistance in several cancer types (melanoma, brain cancer, gastric cancer, prostate cancer or non-small- cell lung cancer) is undoubtedly autophagy: As autophagy precedes apoptosis and it helps cancer cells to evade apoptosis by degrading the drug molecules, the mitochondria, and the activated caspase-8 [72,73]. Inversely once apoptosis is initiated, the caspases 3 and 8 can cleave BECN1, thus inhibiting autophagy initiation [74,75]. As a result, all chemo-sensitive cell lines turn out to exhibit apoptosis, whereas cell populations that respond to drugs by inducing autophagy are more drug-resistant and will recover after the withdrawal of the chemotherapeutic agents [76].

5.4. To fuel cancer stem cell dormancy. Tumor dormancy is the leading factor in treatment failure, metastasis, and tumor recurrence even decades after resection of the primary tumor. When their microenvironments is not sufficiently protective (following metastasis or therapy), only a small population of tumor cells, likely the cancer stem cells, survive, stop dividing and

Page 14 of 27 enter a dormant state. Given their dormancy, these tumor cells became resistant to the chemotherapies that exploit the rapid cell cycling of cancer cells. Moreover, as the dormancy is reversible, cancer cells can resume proliferation when the stressful environment has improved, being the source of tumor relapse.

The regulation of cancer dormancy is poorly understood. Recently, transcriptomics analyses reveal that the dormancy of cancer stem cells is characterized by major metabolic changes with an increased autophagy flux [77,78]. During the dormancy period, ARHI/DIRAS3 is the molecular switch that promotes initiation of autophagosome formation in cancer cells [79,80]. Importantly, inhibition of autophagy using chloroquine or silencing (Atg7 and Atg12 shRNA) suppresses the increase in ATP levels, impedes cancer cell survival, and reduces tumor regrowth [79,81], all of which could be partially rescued by the addition of glutamine as an energy source [82]. Once again this supports the notion that autophagy provides nutrients necessary to meet the metabolic needs absolutely required for cancer stem cell dormancy; thus providing the rationale for targeting this pathway to eradicate these aggressive cancer cells.

FromAutophagy a therapeutic as perspective, a therapeutic the ever-expanding target roles of autophagy in tumorigenesis provides the rationale to prioritize this pathway as a new potential target for drug development. More than 500 patent applications for autophagy regulators have been filed. This massive investment has been fuelled by the realization that all classes of anticancer insults including DNA damaging agents, microtubule-targeted drugs, glycolytic inhibitors, death receptor agonists, hormonal agents, anti-angiogenic agents, proteasome inhibitors, histone deacetylase inhibitors, and targeted kinase inhibitors all affect autophagy [83] (Table 2). It was originally proposed that autophagic cell death may be part of their cytotoxicity, particularly in apoptotic-defective cancer cells. However, similar to its two-sided tumor- suppressive and tumor-promoting effects, autophagy is also exploited by dying cancer cells to deal with the cytotoxicity of anticancer agents rather than a cause of cell death [84].

6.1. Activators of autophagy. Clinically available drugs that upregulate autophagy include the inhibitors of receptor tyrosine kinases (for example, EGFR, ERBB2, PDGFR and VEGFR2: Imatinib, gefitinib and Erlotinib), the inhibitors of PI3K pathway (PIK3CA, Akt) and the

Page 15 of 27 inhibitors of mTOR (Rapamycin, and its analogues Temsirolimus and Everolimus). These three groups of oncogenic targets are mechanistically linked because receptor tyrosine kinases activate the PI3K/Akt pathway and then mTORC1, which is a negative regulator of autophagosome formation. Moreover, treatment of cancer cells with the proteasome inhibitor Bortezomib can induce the expression of endogenous mTOR inhibitor sestrin-2, while Perifosine can induce the degradation of mTORC1 complex components. Likewise, the activation of the PI3K/AKT/mTOR pathway can be suppressed indirectly by the sirtuin activator resveratrol, the glucocorticoid Dexamethasone and the anti-diabetic drug metformin. Whatever the underlying mechanism, these kinases inhibitors and drugs which are currently used in oncology can exert their antitumor effect by inducing autophagy (table 2).

Other autophagy activators directly target BECN1 itself. A good example comes from BH3 mimetics (ABT-737), which disrupt the interaction between BCL2 proteins and BECN1 to induce autophagy [85-87]. Direct regulation of BECN1 is also responsible for autophagy induction by tyrosine kinase inhibitors, such as Erlotinib, that inhibits the phosphorylation of BECN1 by EGFR [88]. Tamoxifen, a well-recognized antitumor drug for breast cancer treatment, can increase the level of BECN1 to stimulate autophagy [89,90] As a chemotherapeutic vitamin D analog, EB1089 may trigger and induce BECN1-dependent autophagy in MCF-7 cells [91].

Because BECN1 and other ATG are haploinsufficient tumor suppressor genes that undergo heterozygous loss in a substantial proportion of human tumors, it might be expected that drugs that may re-activate the remaining copy and thereby restore autophagy back to its physiological levels of activation may suppress tumor growth. Despite the excitement surrounding these promises, clinical progress has been uneven. Autophagy activators have been least ineffective in treating the cancer types that have the highest mortality rates, such as lung, colorectal, pancreatic, skin, and brain cancer.

Why has the clinical application of autophagy activators been so challenging? One reason is that many of these drugs (for instance DNA-damaging drugs) confer resistance by a cytoprotective autophagy [92-96]. Intuitively, overcoming these resistance will require targeting tumor cells through a cocktail of autophagy activators that efficiently upregulate the autophagy flux to a cytotoxic level when applied as concurrent treatment.

Page 16 of 27

6.2. Inhibitors of autophagy. A general hallmark of KRAS/BRAF-driven cancer is an upregulation of basal autophagy, when compared to their normal counterparts. Evidence has accumulated that this upregulation of autophagy intimately accompanies and allows for all different facets of malignant progression (growth, survival, invasion, and dormancy; section 5). Therefore, autophagy is a central aspect of tumor biology that might be turned into cancer's Achilles heel.

Various autophagy inhibitors have been developed such as the inhibitors of PI3KCIII (3- methyladenine, wortmannin, and LY294002 [97]), of microtubule (vinblastine, colchicine [83]), of Vacuolar-ATPase (bafilomycin A1 [98]), of lysosomal proteases (Pepstatin A, [99]) and weak bases (chloroquine, hydroxychloroquine, and Monessen) to block the formation of autophagosome, its maturation/fusion with lysosome and then its degradation, respectively.

Currently, there are 40 clinical trials targeting autophagy addiction using primarily chloroquine (CQ) and hydroxychloroquine (HCQ) to kill cancer cells (http://www.clinicaltrial.gov. Table 2). Indeed these antimalarial drugs are already approved for human use, relatively safe, and inexpensive. Both are weak bases that accumulate in the lysosomes causing a rise in lysosomal pH and thus preventing the activity of autolysosomes or the fusion of autophagosomes with lysosomes.

The rationale of combining a chemotherapy and a lysosomal autophagy inhibitor is that the former induces massive autophagic flux and the latter prevents autophagic contents from being degraded, leading to an accumulation of ineffective autophagic vesicles and a burst of ROS, presumably from the damaged mitochondria within autolysosomes. This ROS in turn produces permeabilization of lysosomal membrane, release of cathepsin and activation of apoptotic cell death. A large body of preclinical results claims the successes of CQ to trigger senescence, apoptosis, and autophagic cell death of cancer cells/cancer stem cells.

Future directions: translational challenges towards Despiteindividualized the great promises health of autophagy-based care… therapies, there remain, undoubtedly, many burning issues to address. In particular, it will be important to identify the cancer-type that might respond to these therapies and evaluate the efficacy, safety and long-term outcomes of modulating autophagy.

Page 17 of 27

– What are the best patient populations for autophagy-based therapies? The concept of autophagy additio as fist suggested fo KRAS-, and BRAF-driven tumors [56,57]. In transgenic mice, the inhibition of autophagy induces the regression of adenocarcinomas to a more benign oncocytoma, providing a rationale for targeting autophagy to treat these refractory KRAS-mutated patients. Importantly, it was demonstrated that autophagy ablation demonstrate efficacy to advanced tumors that have grown to an appreciable size in mice [58], such as would be the case in patients receiving therapy. However, at present it is unclear whether all cancer-bearing patients with a mutated RAS or activation of the RAS pathway will respond similarly to autophagy inhibition. Two recent papers evidenced that inhibition of autophagy (by CQ, ATG5 or ATG7 knockout) inhibits the growth of KRAS–driven tumors in pancreas and lung when Tp53 is wild type but stimulated tumor growth in RAS mutant, Tp53 null cells [55,66] (Table1). In the near future, these puzzling findings may revolutionize the stratification of patients that will receive autophagy inhibition treatment. However, in the ongoing clinical trials, no strategy for patient selection is being pursued.

– What are the main safety concerns? The ultimate goal of any cancer therapy is to robustly target cancerous cells while sparing normal cells. Currently, the major concern of using autophagy modulators is their lack of selectivity, targeting both the cancer and the healthy tissue. A study by Karsli-Uzunbas and colleagues recently indicates that the systemic inhibition of autophagy for 5 weeks in mice is not only efficacious in regressing established lung adenocarcinomas to benign oncocytoma but also safe, in line with a greater reliance of tumor cells on autophagy. However, if the inhibition is sustained for 6 to 12 weeks, this strategy is extremely toxic causing severe liver, muscle and brain degenerations [58]. This suggests that, with proper control of the extent and/or timing of autophagy inhibition, there is likely to be a therapeutic window to suppress tumorigenesis while mostly sparing normal tissue.

– What are the best autophagy-based therapies? The regression or the promotion of cancer by either upregulating or silencing autophagy makes autophagy an attractive therapeutic target, but its unpredictable potential might also make such targeted therapy challenging and risky. For instance, CQ displays dramatic effects for some drugs/tumor models and modest or no effects in others, even in a panel of KRAS-mutated cancer cell lines [129,130]. Of particular concern, CQ might also exacerbate the progression of established cancers, as suggested by certain studies [19,66,131,132]. Furthermore, the possibility that CQ fulfills an autophagic- Page 19 of 27 independent role cannot be excluded yet [133]. All of the above allows us to presume that autophagy-based therapies have to be improved by a specific modulation. Among the druggable targets are the autophagic kinases ULK1/2, and the cysteine protease Atg4. In addition to pharmacological inhibitors, specific gene interference using siRNA or CRISPR/Cas technology against various ATG genes may improve the effectiveness of autophagy inhibition.

– What are the best clinical markers for monitoring autophagy? Actually, autophagy markers in tumor samples are the detection of autophagic vesicles, the expression of autophagy- related genes, and the degradation of the autophagy substrate SQSTM1. However, it is challenging to predict from these markers whether the drug activates or blocks autophagy and whether autophagy upregulation would be beneficial or detrimental for patients. We have developed methods and guidelines to measure this dynamic process in vitro [134], but these approaches are not feasible in a patient fixed biopsy. For instance, a major caveat is the difficulty to distinguish by immunohistochemistry the LC3-I positive aggregates from LC3-II positie eal autophagosoes. Moeoe, eause autophagosomes represent a mid-point in the dynamic autophagic flux, i.e. rapidly formed and degraded within 8 minutes, accumulation of autophagosomes can occur through induction of autophagy, but can also arise through inhibition of autophagic degradation. As a result, only a few autophagosomes and autolysosomes are observed in normal tissues even under nutrient starvation. Thus, it might be interpreted that the robust accumulation of LC3 and SQSTM1 spots most likely reflect an impaired autophagy flux in cancer [20,135], while recognizing that any delay in the processing/fixation of biospecimens would artifactually upregulate autophagy.

In conclusion, we are embarking on an exciting journey. All the remarkable studies that have been performed over the past 14 years on the role of autophagy in cancer are culminating in clinical trials. Still in its infancy, the challenge for the cancer research community will be now to identify those patients who are more likely to respond and the combination of autophagy modulators that would be most effective.

Page 20 of 27

The authors declare no conflict of interest. Conflict of Interest

This work was supported by the Istitut Natioal de la “até et de la ‘ehehe Médiale, Acknowledgements Region Poee Alpes Côte d'Azu, Agee égioale saté Poee Alpes Côte d'Azu, Dietio égioale de lEioeet, de laéageet et du logement Provence Alpes Côte d'Azu A B: pla égioal saté eioeet P‘“E PACA °... ad ..., Assoiatio pou la ‘ehehe ote le Cae A‘C Gats ° “L, ad Feh atioal eseah agey Iestets fo the Futue LABEX “IGNALIFE : program reference # ANR-11-LABX-0028-01).

ATG, autophagy-related; BECN1, beclin 1; CQ, chloroquine; HCQ, Hydroxychloroquine; CMA, Abbreviations chaperon-mediated autophagy; LAMP, lysosomal-associated membrane protein; LC3 (MAP1LC3), microtubule-associated protein 1 light chain 3; mTOR, mammalian target of rapamycin; shRNA, short hairpin RNA; SQSTM1, sequestosome 1/p62; ROS, reactive oxygen species ; TCA , tricarboxylic acid; ATPase, H+ transporting, lysosomal V0 subunit A3; v-ATPase; vacuolar-ATPase.

1. Arias, E., et al. Chaperone-mediated autophagy in protein quality control. Curr Opin Cell Biol References2011, 23, 184-9. 2. Li, W.W., et al. Microautophagy: lesser-known self-eating. Cell Mol Life Sci 2012, 69, 1125-36. 3. Belaid, A., et al. Signalphagy Scheduled signal termination by macroautophagy. Autophagy 2013, 9, 1629-1630. 4. Parzych, K.R., et al. An Overview of Autophagy: Morphology, Mechanism, and Regulation. Antioxid Redox Signal 2014, 20, 460-473. 5. McEwan, D.G., et al. The Three Musketeers of Autophagy: phosphorylation, ubiquitylation and acetylation. Trends Cell Biol 2011, 21, 195-201. 6. Morselli, E., et al. Spermidine and resveratrol induce autophagy by distinct pathways converging on the acetylproteome. J Cell Biol 2011, 192, 615-29. 7. Pietrocola, F., et al. Regulation of autophagy by stress-responsive transcription factors. Semin Cancer Biol 2013, 23, 310-22. 8. Choi, A.M., et al. Autophagy in human health and disease. N Engl J Med 2013, 368, 1845-6. 9. Aita, V.M., et al. Cloning and genomic organization of beclin 1, a candidate tumor suppressor gene on chromosome 17q21. Genomics 1999, 59, 59-65. 10. Liang, X.H., et al. Induction of autophagy and inhibition of tumorigenesis by beclin 1. Nature 1999, 402, 672-6.

Page 21 of 27

11. Qu, X., et al. Promotion of tumorigenesis by heterozygous disruption of the beclin 1 autophagy gene. J. Clin. Invest. 2003, 112, 1809-20. 12. Yue, Z., et al. Beclin 1, an autophagy gene essential for early embryonic development, is a haploinsufficient tumor suppressor. Proc Natl Acad Sci USA 2003, 100, 15077-82. 13. Fimia, G.M., et al. Ambra1 regulates autophagy and development of the nervous system. Nature 2007, 447, 1121-5. 14. Takahashi, Y., et al. Bif-1 interacts with Beclin 1 through UVRAG and regulates autophagy and tumorigenesis. Nat Cell Biol. 2007, 9, 1142-1151. 15. Liang, C., et al. Beclin1-binding UVRAG targets the class C Vps complex to coordinate autophagosome maturation and endocytic trafficking. Nat Cell Biol 2008, 10, 776-87. 16. Takamura, A., et al. Autophagy-deficient mice develop multiple liver tumors. Genes Dev 2011, 25, 795-800. 17. Yang, A., et al. Autophagy Is Critical for Pancreatic Tumor Growth and Progression in Tumors with p53 Alterations. Cancer Discov 2014, 4, 905-913. 18. Mariño, G., et al. Tissue-specific autophagy alterations and increased tumorigenesis in mice deficient in Atg4C/autophagin-3. J. Biol. Chem. 2007, 282, 18573-18583. 19. Belaid, A., et al. Autophagy and SQSTM1 on the RHOA(d) again Emerging roles of autophagy in the degradation of signaling proteins. Autophagy 2014, 10, 201-208. 20. Belaid, A., et al. Autophagy Plays a Critical Role in the Degradation of Active RHOA, the Control of Cell Cytokinesis, and Genomic Stability. Cancer Res 2013, 73, 4311-4322. 21. Joung, I., et al. Molecular cloning of a phosphotyrosine-independent ligand of the p56lck SH2 domain. Proc Natl Acad Sci USA 1996, 93, 5991-5. 22. Duran, A., et al. The signaling adaptor p62 is an important NF-kappaB mediator in tumorigenesis. Cancer Cell 2008, 13, 343-54. 23. Duran, A., et al. p62 is a key regulator of nutrient sensing in the mTORC1 pathway. Mol Cell 2011, 44, 134-46. 24. Komatsu, M., et al. The selective autophagy substrate p62 activates the stress responsive transcription factor Nrf2 through inactivation of Keap1. Nat Cell Biol 2010, 12, 213-23. 25. Qiang, L., et al. Regulation of cell proliferation and migration by p62 through stabilization of Twist1. Proc Natl Acad Sci USA 2014, 111, 9241-9246. 26. Ling, J., et al. KrasG12D-induced IKK2/beta/NF-kappaB activation by IL-1alpha and p62 feedforward loops is required for development of pancreatic ductal adenocarcinoma. Cancer Cell 2012, 21, 105-20. 27. Moscat, J., et al. p62 at the crossroads of autophagy, apoptosis, and cancer. Cell 2009, 137, 1001-4. 28. Bjorkoy, G., et al. p62/SQSTM1 forms protein aggregates degraded by autophagy and has a protective effect on huntingtin-induced cell death. J Cell Biol 2005, 171, 603-14. 29. Inami, Y., et al. Persistent activation of Nrf2 through p62 in hepatocellular carcinoma cells. J Cell Biol 2011, 193, 275-84. 30. Mathew, R., et al. Autophagy suppresses tumorigenesis through elimination of p62. Cell 2009, 137, 1062-75. 31. Karantza-Wadsworth, V., et al. Autophagy mitigates metabolic stress and genome damage in mammary tumorigenesis. Genes Dev 2007, 21, 1621-35. 32. Mathew, R., et al. Autophagy suppresses tumor progression by limiting chromosomal instability. Genes Dev 2007, 21, 1367-81. 33. Robert, T., et al. HDACs link the DNA damage response, processing of double-strand breaks and autophagy. Nature 2011, 471, 74-79. 34. Katayama, M., et al. DNA damaging agent-induced autophagy produces a cytoprotective adenosine triphosphate surge in malignant glioma cells. Cell Death Differ 2007, 14, 548-558. 35. Dyavaiah, M., et al. Autophagy-Dependent Regulation of the DNA Damage Response Protein Ribonucleotide Reductase 1. Mol Cancer Res 2011, 9, 462-475.

Page 22 of 27

36. Changou, C.A., et al. Arginine starvation-associated atypical cellular death involves mitochondrial dysfunction, nuclear DNA leakage, and chromatin autophagy. Proc Natl Acad Sci USA 2014, 111, 14147-52. 37. Guo, H., et al. Autophagy supports genomic stability by degrading retrotransposon RNA. Nat Commun 2014, 5, 5276. 38. Matsui, A., et al. The Role of Autophagy in Genome Stability through Suppression of Abnormal Mitosis under Starvation. PLoS Genet 2013, 9, e1003245. 39. Xie, R., et al. Autophagy Enhanced by Microtubule- and Mitochondrion-Associated MAP1S Suppresses Genome Instability and Hepatocarcinogenesis. Cancer Res 2011, 71, 7537-7546. 40. Liu, Y., et al. Autosis and autophagic cell death: the dark side of autophagy. Cell Death Differ 2014. 41. Byun, J.Y., et al. The Rac1/MKK7/JNK pathway signals upregulation of Atg5 and subsequent autophagic cell death in response to oncogenic Ras. Carcinogenesis 2009, 30, 1880-1888. 42. Elgendy, M., et al. Oncogenic Ras-Induced Expression of Noxa and Beclin-1 Promotes Autophagic Cell Death and Limits Clonogenic Survival. Mol Cell 2011, 42, 23-35. 43. Young, A.R., et al. Autophagy mediates the mitotic senescence transition. Genes Dev 2009, 23, 798-803. 44. Wang, Y.H., et al. Autophagic activity dictates the cellular response to oncogenic RAS. Proc Natl Acad Sci USA 2012, 109, 13325-13330. 45. Drullion, C., et al. Apoptosis and autophagy have opposite roles on imatinib-induced K562 leukemia cell senescence. Cell Death Dis 2012, 3, e373. 46. Biggers, J.W., et al. Autophagy, cell death and sustained senescence arrest in B16/F10 melanoma cells and HCT-116 colon carcinoma cells in response to the novel microtubule poison, JG-03-14. Cancer Chemother Pharmacol 2013, 71, 441-455. 47. Knizhnik, A.V., et al. Survival and Death Strategies in Glioma Cells: Autophagy, Senescence and Apoptosis Triggered by a Single Type of Temozolomide-Induced DNA Damage. PLoS One 2013, 8, e55665. 48. Dorr, J.R., et al. Synthetic lethal metabolic targeting of cellular senescence in cancer therapy. Nature 2013, 501, 421-5. 49. Goehe, R.W., et al. The Autophagy-Senescence Connection in Chemotherapy: Must Tumor Cells (Self) Eat Before They Sleep? J Pharmacol Exp Ther 2012, 343, 763-778. 50. Degenhardt, K., et al. Autophagy promotes tumor cell survival and restricts necrosis, inflammation, and tumorigenesis. Cancer Cell 2006, 10, 51-64. 51. Brest, P., et al. Autophagy and Crohn's disease: at the crossroads of infection, inflammation, immunity, and cancer. Curr Mol Med 2010, 10, 486-502. 52. Levine, B., et al. Autophagy in immunity and inflammation. Nature 2011, 469, 323-35. 53. Laddha, S.V., et al. Mutational Landscape of the Essential Autophagy Gene BECN1 in Human Cancers. Mol Cancer Res 2014, 12, 485-490. 54. Guo, J.Y., et al. Autophagy suppresses progression of K-ras-induced lung tumors to oncocytomas and maintains lipid homeostasis. Genes Dev 2013, 27, 1447-61. 55. Rao, S., et al. A dual role for autophagy in a murine model of lung cancer. Nat Commun 2014, 5. 56. Strohecker, A.M., et al. Autophagy Sustains Mitochondrial Glutamine Metabolism and Growth of BrafV600E-Driven Lung Tumors. Cancer Discov 2013, 3, 1272-85. 57. Guo, J.Y., et al. Activated Ras requires autophagy to maintain oxidative metabolism and tumorigenesis. Genes Dev 2011, 25, 460-70. 58. Karsli-Uzunbas, G., et al. Autophagy Is Required for Glucose Homeostasis and Lung Tumor Maintenance. Cancer Discov 2014, 4, 914-927. 59. Lock, R., et al. Autophagy facilitates glycolysis during Ras-mediated oncogenic transformation. Mol Biol Cell 2011, 22, 165-178. 60. Yang, S., et al. Pancreatic cancers require autophagy for tumor growth. Genes Dev 2011, 25, 717-29.

Page 23 of 27

61. Wei, H., et al. Suppression of autophagy by FIP200 deletion inhibits mammary tumorigenesis. Genes Dev 2011, 25, 1510-27. 62. Kim, M.J., et al. Involvement of autophagy in oncogenic K-Ras-induced malignant cell transformation. J Biol Chem 2011, 286, 12924-32. 63. Galluzzi, L., et al. Metabolic targets for cancer therapy. Nat Rev Drug Discov 2013, 12, 829-46. 64. Ko, Y.H., et al. Glutamine fuels a vicious cycle of autophagy in the tumor stroma and oxidative mitochondrial metabolism in epithelial cancer cells Implications for preventing chemotherapy resistance. Cancer Biol Ther 2011, 12, 1085-1097. 65. Eng, C.H., et al. Ammonia Derived from Glutaminolysis Is a Diffusible Regulator of Autophagy. Sci. Signal 2010, 3. 66. Rosenfeldt, M.T., et al. p53 status determines the role of autophagy in pancreatic tumour development. Nature 2013, 504, 296-300. 67. Fung, C., et al. Induction of autophagy during extracellular matrix detachment promotes cell survival. Mol Biol Cell 2008, 19, 797-806. 68. Peng, Y.F., et al. Autophagy inhibition suppresses pulmonary metastasis of HCC in mice via impairing anoikis resistance and colonization of HCC cells. Autophagy 2013, 9, 2056-2068. 69. Giatromanolaki, A.N., et al. Autophagy patterns and prognosis in uveal melanomas. Mod. Pathol 2011, 24, 1036-1045. 70. Lazova, R., et al. Punctate LC3B Expression Is a Common Feature of Solid Tumors and Associated with Proliferation, Metastasis, and Poor Outcome. Clin Cancer Res 2012, 18, 370- 379. 71. Jo, Y.K., et al. Increased Expression of ATG10 in Colorectal Cancer Is Associated with Lymphovascular Invasion and Lymph Node Metastasis. PLoS One 2012, 7, e52705. 72. Hou, W., et al. Autophagic degradation of active caspase-8 A crosstalk mechanism between autophagy and apoptosis. Autophagy 2010, 6, 891-900. 73. Cojoc, M., et al. A role for cancer stem cells in therapy resistance: Cellular and molecular mechanisms. Semin Cancer Biol 2014, [Epub ahead of print]. 74. Wirawan, E., et al. Caspase-mediated cleavage of Beclin-1 inactivates Beclin-1-induced autophagy and enhances apoptosis by promoting the release of proapoptotic factors from mitochondria. Cell Death Dis 2010, 1, e18. 75. Li, H., et al. Following Cytochrome c Release, Autophagy Is Inhibited during Chemotherapy- Induced Apoptosis by Caspase 8-Mediated Cleavage of Beclin 1. Cancer Res 2011, 71, 3625- 3634. 76. O'Donovan, T.R., et al. Induction of autophagy by drug-resistant esophageal cancer cells promotes their survival and recovery following treatment with chemotherapeutics. Autophagy 2011, 7, 509-524. 77. Gong, C., et al. Beclin 1 and autophagy are required for the tumorigenicity of breast cancer stem-like/progenitor cells. Oncogene 2013, 32, 2261-2272. 78. Viale, A., et al. Oncogene ablation-resistant pancreatic cancer cells depend on mitochondrial function. Nature 2014, 514, 628-632. 79. Lu, Z., et al. The tumor suppressor gene ARHI regulates autophagy and tumor dormancy in human ovarian cancer cells. J Clin Invest 2008, 118, 3917-3929. 80. Lu, Z., et al. DIRAS3 regulates the autophagosome initiation complex in dormant ovarian cancer cells. Autophagy 2014, 10, 1071-1092. 81. Gupta, A., et al. Autophagy inhibition and antimalarials promote cell death in gastrointestinal stromal tumor (GIST). Proc. Natl Acad. Sci. USA 2010, 107, 14333-14338. 82. Shimizu, T., et al. IGF2 Preserves Osteosarcoma Cell Survival by Creating an Autophagic State of Dormancy That Protects Cells against Chemotherapeutic Stress. Cancer Res 2014, 74, 6531-41. 83. Shen, S., et al. Association and dissociation of autophagy, apoptosis and necrosis by systematic chemical study. Oncogene 2011, 30, 4544-4556.

Page 24 of 27

84. Kroemer, G., et al. Autophagic cell death: the story of a misnomer. Nat Rev Mol Cell Biol 2008, 9, 1004-1010. 85. Martin, A.P., et al. BCL-2 Family Inhibitors Enhance Histone Deacetylase Inhibitor and Sorafenib Lethality via Autophagy and Overcome Blockade of the Extrinsic Pathway to Facilitate Killing. Mol Pharmacol 2009, 76, 327-341. 86. Wei, Y., et al. The Combination of a Histone Deacetylase Inhibitor with the Bcl-2 Homology Domain-3 Mimetic GX15-070 Has Synergistic Antileukemia Activity by Activating Both Apoptosis and Autophagy. Clin Cancer Res 2010, 16, 3923-3932. 87. Oltersdorf, T., et al. An inhibitor of Bcl-2 family proteins induces regression of solid tumours. Nature 2005, 435, 677-681. 88. Wei, Y.J., et al. EGFR-Mediated Beclin 1 Phosphorylation in Autophagy Suppression, Tumor Progression, and Tumor Chemoresistance. Cell 2013, 154, 1269-1284. 89. John, S., et al. Regulation of estrogenic effects by Beclin 1 in breast cancer cells. Cancer Res 2008, 68, 7855-7863. 90. Schoenlein, P.V., et al. Autophagy facilitates the progression of ERalpha-positive breast cancer cells to antiestrogen resistance. Autophagy 2009, 5, 400-3. 91. Hoyer-Hansen, M., et al. Vitamin D analog EB1089 triggers dramatic lysosomal changes and Beclin 1-mediated autophagic cell death. Cell Death Differ 2005, 12, 1297-1309. 92. Abedin, M.J., et al. Autophagy delays apoptotic death in breast cancer cells following DNA damage. Cell Death Differ 2007, 14, 500-510. 93. Li, J., et al. Inhibition of autophagy augments 5-fluorouracil chemotherapy in human colon cancer in vitro and in vivo model. Eur. J. Cancer 2010, 46, 1900-1909. 94. Vazquez-Martin, A., et al. Autophagy Facilitates the Development of Breast Cancer Resistance to the Anti-HER2 Monoclonal Antibody Trastuzumab. PLoS One 2009, 4, e6251. 95. Gorzalczany, Y., et al. Combining an EGFR directed tyrosine kinase inhibitor with autophagy- inducing drugs: A beneficial strategy to combat non-small cell lung cancer. Cancer Lett 2011, 310, 207-215. 96. Chaachouay, H., et al. Autophagy contributes to resistance of tumor cells to ionizing radiation. Radiother Oncol 2011, 99, 287-92. 97. Boya, P., et al. Inhibition of macroautophagy triggers apoptosis. Mol Cell Biol 2005, 25, 1025- 40. 98. Kanematsu, S., et al. Autophagy Inhibition Enhances Sulforaphane-induced Apoptosis in Human Breast Cancer Cells. Anticancer Research 2010, 30, 3381-3390. 99. Hsu, K.F., et al. Cathepsin L mediates resveratrol-induced autophagy and apoptotic cell death in cervical cancer cells. Autophagy 2009, 5, 451-460. 100. Takeuchi, H., et al. Synergistic augmentation of rapamycin-induced autophagy in malignant glioma cells by phosphatidylinositol 3-kinase/protein kinase B inhibitors. Cancer Res 2005, 65, 3336-3346. 101. Carayol, N., et al. Critical roles for mTORC2-and rapamycin-insensitive mTORC1-complexes in growth and survival of BCR-ABL-expressing leukemic cells. Proc Natl Acad Sci USA 2010, 107, 12469-12474. 102. Yazbeck, V.Y., et al. Temsirolimus downregulates p21 without altering cyclin D1 expression and induces autophagy and synergizes with vorinostat in mantle cell lymphoma. Exp Hematol 2008, 36, 443-450. 103. Crazzolara, R., et al. Potentiating effects of RAD001 (Everolimus) on vincristine therapy in childhood acute lymphoblastic leukemia. Blood 2009, 113, 3297-3306. 104. Fu, L., et al. Perifosine Inhibits Mammalian Target of Rapamycin Signaling through Facilitating Degradation of Major Components in the mTOR Axis and Induces Autophagy. Cancer Res 2009, 69, 8967-8976. 105. Bruning, A., et al. Nelfinavir and bortezomib inhibit mTOR activity via ATF4-mediated sestrin- 2 regulation. Mol Oncol 2013, 7, 1012-1018.

Page 25 of 27

106. Zhu, K., et al. Proteasome inhibitors activate autophagy as a cytoprotective response in human prostate cancer cells. Oncogene 2010, 29, 451-462. 107. Ding, W.X., et al. Oncogenic transformation confers a selective susceptibility to the combined suppression of the proteasome and autophagy. Mol Cancer Ther 2009, 8, 2036-2045. 108. Liu, T.J., et al. NVP-BEZ235, a novel dual phosphatidylinositol 3-kinase/mammalian target of rapamycin inhibitor, elicits multifaceted antitumor activities in human gliomas. Mol Cancer Ther 2009, 8, 2204-2210. 109. Fan, Q.W., et al. Inhibition of PI3K-Akt-mTOR signaling in glioblastoma by mTORC1/2 inhibitors. Methods Mol Biol 2012, 821, 349-59. 110. Wang, K., et al. Quercetin induces protective autophagy in gastric cancer cells: involvement of Akt-mTOR- and hypoxia-induced factor 1alpha-mediated signaling. Autophagy 2011, 7, 966-78. 111. Scarlatti, F., et al. Role of non-canonical Beclin 1-independent autophagy in cell death induced by resveratrol in human breast cancer cells. Cell Death Differ 2008, 15, 1318-1329. 112. Zhang, X., et al. Plant natural compounds: targeting pathways of autophagy as anti-cancer therapeutic agents. Cell Prolif. 2012, 45, 466-476. 113. Laane, E., et al. Cell death induced by dexamethasone in lymphoid leukemia is mediated through initiation of autophagy. Cell Death Differ 2009, 16, 1018-1029. 114. Ertmer, A., et al. The anticancer drug imatinib induces cellular autophagy. Leukemia 2007, 21, 936-942. 115. Pan, H.M., et al. Autophagy inhibition sensitizes hepatocellular carcinoma to the multikinase inhibitor linifanib. Sci Rep 2014, 4, 6683. 116. Milano, V., et al. Dasatinib-induced autophagy is enhanced in combination with temozolomide in glioma. Mol Cancer Ther 2009, 8, 394-406. 117. Chang, C.Y., et al. Autophagy contributes to gefitinib-induced glioma cell growth inhibition. Exp Cell Res 2014, 327, 102-112. 118. D'Amato, V., et al. The dual PI3K/mTOR inhibitor PKI-587 enhances sensitivity to cetuximab in EGFR-resistant human head and neck cancer models. Br J Cancer 2014, 110, 2887-2895. 119. Qian, W., et al. Arsenic trioxide induces not only apoptosis but also autophagic cell death in leukemia cell lines via up-regulation of Beclin-1. Leuk Res 2007, 31, 329-39. 120. Kanzawa, T., et al. Induction of autophagic cell death in malignant glioma cells by arsenic trioxide. Cancer Res 2003, 63, 2103-2108. 121. Zhang, W., et al. Knockdown of autophagy-related protein 6, Beclin-1, decreases cell growth, invasion, and metastasis and has a positive effect on chemotherapy-induced cytotoxicity in osteosarcoma cells. Tumour Biol 2014, [Epub ahead of print]. 122. Kanzawa, T., et al. Role of autophagy in temozolomide-induced cytotoxicity for malignant glioma cells. Cell Death Differ 2004, 11, 448-457. 123. Amaravadi, R.K., et al. Autophagy inhibition enhances therapy-induced apoptosis in a Myc- induced model of lymphoma. J Clin Invest 2007, 117, 326-336. 124. Maclean, K.H., et al. Targeting lysosomal degradation induces p53-dependent cell death and prevents cancer in mouse models of lymphomagenesis. J Clin Invest 2008, 118, 79-88. 125. Shao, Y.F., et al. Apoptotic and autophagic cell death induced by histone deacetylase inhibitors. Proc Natl Acad Sci USA 2004, 101, 18030-18035. 126. Carew, J.S., et al. Targeting autophagy augments the anticancer activity of the histone deacetylase inhibitor SAHA to overcome Bcr-Abl-mediated drug resistance. Blood 2007, 110, 313-322. 127. Ellis, L., et al. The histone deacetylase inhibitors LAQ824 and LBH589 do not require death receptor signaling or a functional apoptosome to mediate tumor cell death or therapeutic efficacy. Blood 2009, 114, 380-93. 128. Ben Sahra, I., et al. Targeting Cancer Cell Metabolism: The Combination of Metformin and 2- Deoxyglucose Induces p53-Dependent Apoptosis in Prostate Cancer Cells. Cancer Res 2010, 70, 2465-2475.

Page 26 of 27

129. Bristol, M.L., et al. Autophagy inhibition for chemosensitization and radiosensitization in cancer: do the preclinical data support this therapeutic strategy? J Pharmacol Exp Ther 2013, 344, 544-52. 130. Morgan, M.J., et al. Regulation of autophagy and chloroquine sensitivity by oncogenic RAS in vitro is context-dependent. Autophagy 2014, 10, 1814-26. 131. Chi, C.W., et al. Disruption of Lysosome Function Promotes Tumor Growth and Metastasis in Drosophila. J Biol Chem 2010, 285, 21817-21823. 132. Tu, C., et al. Lysosomal Cathepsin B Participates in the Podosome-Mediated Extracellular Matrix Degradation and Invasion via Secreted Lysosomes in v-Src Fibroblasts. Cancer Res 2008, 68, 9147-9156. 133. Maes, H., et al. Tumor Vessel Normalization by Chloroquine Independent of Autophagy. Cancer Cell 2014, 26, 190-206. 134. Klionsky, D.J., et al. Guidelines for the use and interpretation of assays for monitoring autophagy. Autophagy 2012, 8, 445-544. 135. Chargui, A., et al. Cadmium-Induced Autophagy in Rat Kidney: An Early Biomarker of Subtoxic Exposure. Toxicol Sci 2011, 121, 31-42.

Page 27 of 27

Autophagy

ISSN: 1554-8627 (Print) 1554-8635 (Online) Journal homepage: http://www.tandfonline.com/loi/kaup20

Signalphagy

Amine Belaid, Papa Diogop Ndiaye, Daniel J Klionsky, Paul Hofman & Baharia Mograbi

To cite this article: Amine Belaid, Papa Diogop Ndiaye, Daniel J Klionsky, Paul Hofman & Baharia Mograbi (2013) Signalphagy, Autophagy, 9:10, 1629-1630, DOI: 10.4161/auto.25880 To link to this article: http://dx.doi.org/10.4161/auto.25880

Published online: 13 Aug 2013.

Submit your article to this journal

Article views: 390

View related articles

Citing articles: 6 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=kaup20

Download by: [77.140.255.75] Date: 16 October 2017, At: 19:55 Downloaded by [77.140.255.75] at 19:55 16 October 2017 www.landesbioscience.com Autophagy 2013;©2013LandesBioscience 9:10,1629–1630;October RHOA, cytokinesis, migration cytokinesis, RHOA, active suppression, signaling, Keywords: Faculté de Médecine; Nice, France; France; Nice, Médecine; de Faculté Email: Email: Mograbi; to: Baharia *Correspondence http://dx.doi.org/10.4161/auto.25880 Accepted: 07/24/2013 07/18/2013Revised: 07/04/2013Submitted: Nice, France Nice, 1 Belaid, Amine terminationScheduled signal by macroautophagy Signalphagy University of Michigan; Ann Arbor, MI USA; USA; MI Arbor, Ann ofMichigan; University Institute of Research on Cancer and Ageing of Nice (IRCAN); INSERM U1081; CNRS U U1081; CNRS INSERM (IRCAN); ofNice Ageing and on Cancer ofResearch Institute of signaling proteins. Autophagy 2013; Autophagy proteins. signaling of press In degradation in the autophagy of roles emerging RHOA(d) SQSTM1 the and on again: Autophagy P, DJ, GF, Carle P, Klionsky Hofman B. Mograbi Brest L, PD, Ndiaye Cailleteau M, A, Cerezo Belaid and org/10.1158/0008-5472.CAN-12-4142 73:4311–22; PMID:23704209; http://dx.doi. 2013; Res Cancer stability. genomic and cell cytokinesis of control the RHOA, active of degradation inthe role acritical plays Autophagy al. et S, Tauc I, Barale Rubera M, M, Corcelle-Termeau F, Pedeutour E, Ilie S, Giuliano A, Chargui M, Cerezo A, to: Belaid Punctum [email protected]

AUTOPH A autophagy, tumor autophagy, tumor GIC P 1,2,3,4 UNCTUM Papa Diogop Ndiaye, 3 Equipe Labellisée par l’ARC; Villejuif, France; France; Villejuif, l’ARC; par Labellisée Equipe 6 Centre Hospitalier Universitaire de Nice; Pasteur Hospital; Labora Hospital; Pasteur Nice; de Universitaire Hospitalier Centre 1,2,3,4 Daniel JKlionsky, Daniel u rcn suis in studies recent signaling. Our off turn can strategies cell a the which by among are com- interactions protein–protein that and transcription loops bine feedback controlled. Negative carefully signaling be key must of proteins localization the and duration the abundance, the potential, oncogenic their of Because response. cell appropriate the to lead should pathway f nry edd o cl sria. s a As survival. cell for needed supply energy of the provide to upregulated cally emer- dramati- is these macroautophagy states, gency to response In challenges. environmental other and starvation with cope to process “self-eating” nonselective bulk, a be to thought ﬁrst was autophagy achieve to suppression. tumor proteins signaling active of level appropriate the and maintains thereby degrades that macroautophagy of type dedicated “sig- a indicate term to the nalphagy” progression. propose cancer therefore We upon impact have an directly that processes motility—three and aneuploidy, failure, kinesis cyto- drive compromises autophagy all result, a As migration. directed and sis cytokine- faithful for manner temporally restricted and spatially limits a in that signaling mechanism represents safeguard autophagy one constitutive by RHOA-GTP as such proteins active key and A Since its discovery Since in its the discovery 1950s, macro- Autophagy

is how the activation of a signaling signaling a of activation the how is biology cell in issue fundamental Autophagy show that degradation of degradation that show 5 Paul Hofman, Paul 4 Centre Antoine Lacassagne; Nice, France; France; Nice, Lacassagne; Antoine Centre acr Research Cancer MR7284; Nice, France; France; Nice, MR7284; 1,2,3,4,6 and Baharia Mograbi Baharia and

conditions such as extracellular matrix matrix extracellular as such conditions stressful under simi- occurs this DVL, However,to larly substrates. autophagy the among factors transcription and tors, regula- cell-cycle signal- kinases, molecules, report ing observations fascinating Other starvation. nutrient under occurs this however, (disheveled protein); polarity DVL segment degrading by naling sig- family) site integration MMTV type (wingless- WNT regulates negatively also et al. provides the that autophagy evidence Gao by study pioneering a housecleaner, a simply being from Far (nucleophagy). the of portions even and (aggrephagy), (xenophagy), bacteria invading aggregates protein large (pexophagy), peroxisomes (mitophagy), mitochondria as such organelles expected, as include, autophagy by degraded tively is phagy open: the that substrates are selec- life. acell’s during accumulate otherwise and would that proteins/organelles damaged unneeded selectively degrades and conditions rich in levels basal at tinuously con- evidence occurs autophagy that now Emerging suggests cancers. myopa- and thies, neurodegeneration, including diseases human list devastating common growing of a in involved be to shown recently were defects its until studied was less autophagy constitutive contrast, In starvation. to adaptation is autophagy of function primary the that led notion the to has This conditions. nutrient-poor under die rapidly formation phagosome auto- for deﬁcient mice and yeast result, tory of Clinical and Experimental Pathology; Pathology; Experimental and ofClinical tory The The Pandora’s box auto- of constitutive 2 Université de Nice-Sophia Antipolis; Antipolis; Nice-Sophia de Université 5 Life Sciences Institute; Institute; Sciences Life 1,2,3,4, nucleus AUTOPH * n micronuclei and A GIC P UNCTUM 1629

©2013 Landes Bioscience. Do not distribute. Downloaded by [77.140.255.75] at 19:55 16 October 2017 1630 the of disruption some. For this purpose, we made a targeted autolyso- the within step, degradation the at pathway, autophagy the block to strategy was our of feature a protein, naling sig- a To protein. such identify signaling a degrading by growth cell check in keeps autophagy constitutive that assume We its substrates. be to characterize will better autophagy of function suppressive tumor treatment drugs. chemotherapy with and infections, detachment, for degradation. The substrates that are are that substrates The degradation. for of the type same cargo and target function same the fulfill starvation-induced not do which autophagy, and constitutive is, of autophagy, there types that are that two propose we emerging: is paradigm new a Mechanistically, and ubiquitination. conditions involves basal under occurs that process degrada- selective tion of RHOA a is highly the macroautophagy, induced starvation- to contrast By autophagy. by degraded is form active the whereas some, protea- the by degraded is ofRHOA form inactive the that fact the from stems bility sta- in difference the Strikingly, half-life. shorter substantially a has form inactive RHOA is a long-lived protein, the whereas of form active the substrates, autophagic other criti- with As progression. pathway cancer for a cal pathway, RHOA controls the autophagy that demonstration Tcirg1 autolysosomes. the within substrates of the to accumulation the leading dation, degra- autophagic and proteases acidic the to pHthe thereby completely and block raise order in pump proton v-ATPase the One key step in understanding the the understanding in step key One By analyzing the phenotype of the the of phenotype the analyzing By -null cells we made the surprising surprising the made we cells -null TCIRG1/A3 subunit of subunit activation of RHOA allowing irreversible irreversible restricted allowing RHOA of activation is, efficiency—that ensure to signaling autophagy by degraded be to appears RHOA cases, during both In migration. lamella the and cytokinesis, ing dur- midbody the zones: dynamic highly in activated is RHOA of fraction small a only Instead, simultaneously. on turned are molecules RHOA the all not vation, acti- upon that mind in keep should We activity? RHOA modulate to events phosphorylation and or GAP GEF, transcription, reversible of instead selective using autophagy for reasons the are what Therefore, it. phosphorylate that kinases that several with GDIs concert in 3 RHOA sequester it, inactivate that GAPs) 60 (RHO- proteins factors GTPase-activating RHO RHOA, activate exchange that (RHO-GEFs) nucleotide guanine RHO 70 over are there date, to instance for remain: questions Yet,ronment. many envi- its to appropriately respond to cell a allows signaling and autophagy between turn in interplay Such intricate controls signaling. autophagy how and pathways, signaling by regulated is autophagy how substrates. targeted of the degradation and sequestration selective the ensuring thereby autophagosome, nascent the on LC3/GABARAP to simultaneously binds the ubiquitinated. Ubiquitination helps nature—aggregates, recruit all proteins—are their signaling and organelles of spective irre- autophagy, by degraded selectively e r rpdy ann isgt into insight gaining rapidly are We autophagy adaptor SQSTM1/p62 autophagy that Perspectives, Promises, Promises, Perspectives, and Challenges and Autophagy in regulating cellular physiology. cellular regulating in autophagy of role the into insight further gaining our to regard with previous the as exciting as be to going are years next the that guess We modulation. dynamic and compartmentalization specificity; naling sig- achieves “signalphagy,” call we which autophagy, by termination signal that late postu- therefore We etc.). of GAPs, GEFs, action the phosphorylation, as events (such enzymatic reversible and gradual with contrasts autophagy by degradation irreversible migration. and timely this directed Undoubtedly, and mitosis of exit NIH (DJK: GM053396). (DJK: NIH the and Cancer” INCa), POUMON du PNES (PH: National “Institut II), axe 2007, (PROCAN PACA” “Cancéropole PHRC 2003), (PH: CHU Nice Clinique” Recherche De Hospitalier “Programme Can- SL220110603478),no le Grants (ARC contre cer” Recherche la “Associa- pour 6.3.3.4), tion et n°6.3.3.3 PACA (AB: PRSE environnement santé Logement régional plan du de et l’Environnement, l’Aménagement de régionale tion Direc- and d’Azur Côte Alpes Provence santé régionale Agence 0862C0044), n° ADEME (AB: de de l’Energie” la Maîtrise Recherche Médicale”, “Agence la de et l’Environnement de et Santé la de “Institut National from grants by supported work was This discussions. helpful for gues Albren- Jean and Boulter, Etienne Singer, disclosed. Disclosure of Potential Conflicts ofInterest Conflicts ofPotential Disclosure We thank Nathalie Rochet, Nathalie Nathalie Rochet, Nathalie thank We were interest of conflicts potential No Acknowledgments Volume 9Issue10