Context-dependent response of endothelial cells to PIK3CA mutation

Context-dependent response of endothelial cells to PIK3CA mutation | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Context-dependent response of endothelial cells to PIK3CA mutation View ORCID Profile Helena Sabata , View ORCID Profile Ariadna Roca-Coll , View ORCID Profile Jose A. Dengra , View ORCID Profile Leonor Gouveia , View ORCID Profile Ane Martinez-Larrinaga , Alberto Collado-Remacha , View ORCID Profile Sandra D. Castillo , Elena Castillo , View ORCID Profile Nathalie Tisch , Macarena De Andrés-Laguillo , Svanhild Nornes , View ORCID Profile Judith Llena , View ORCID Profile Pilar Villacampa , View ORCID Profile Bart Vanhaesebroeck , View ORCID Profile María P. Alcolea , View ORCID Profile Sarah De Val , Susana Lopez-Fernandez , View ORCID Profile Katrien De Bock , View ORCID Profile Eulalia Baselga , View ORCID Profile Rui Benedito , View ORCID Profile Ana Angulo-Urarte , View ORCID Profile Mariona Graupera doi: https://doi.org/10.1101/2025.02.25.640041 Helena Sabata 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain 2 PhD Program in Biomedicine, Faculty of Medicine, University of Barcelona , Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Helena Sabata Ariadna Roca-Coll 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain 3 PhD Program in Biotechnology, Faculty of Pharmacy and Food Sciences, University of Barcelona , Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ariadna Roca-Coll Jose A. Dengra 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain 2 PhD Program in Biomedicine, Faculty of Medicine, University of Barcelona , Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Jose A. Dengra Leonor Gouveia 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Leonor Gouveia Ane Martinez-Larrinaga 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain 2 PhD Program in Biomedicine, Faculty of Medicine, University of Barcelona , Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ane Martinez-Larrinaga Alberto Collado-Remacha 4 Cell Signalling, UCL Cancer Institute, University College London , 72 Huntley Street, London WC1E 6BT Find this author on Google Scholar Find this author on PubMed Search for this author on this site Sandra D. Castillo 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain 5 Pediatric Cancer Center Barcelona, Hospital Sant Joan de Déu , 08950, Barcelona, Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Sandra D. Castillo Elena Castillo 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site Nathalie Tisch 6 Laboratory of Exercise and Health, Institute of Human Movement Sciences and Sport, Department of Health Sciences and Technology , ETH Zurich, Zurich, Switzerland Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Nathalie Tisch Macarena De Andrés-Laguillo 7 Molecular Genetics of Angiogenesis Group, Centro Nacional de Investigaciones Cardiovasculares (CNIC) , Madrid, Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site Svanhild Nornes 8 Department of Physiology, Anatomy and Genetics, Institute of Developmental and Regenerative Medicine, University of Oxford , Oxford, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site Judith Llena 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain 2 PhD Program in Biomedicine, Faculty of Medicine, University of Barcelona , Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Judith Llena Pilar Villacampa 9 Department of Physiological Sciences, Faculty of Medicine and Health Sciences, University of Barcelona, Feixa Llarga s/n, 08907 l’Hospitalet de Llobregat , Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Pilar Villacampa Bart Vanhaesebroeck 4 Cell Signalling, UCL Cancer Institute, University College London , 72 Huntley Street, London WC1E 6BT Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Bart Vanhaesebroeck María P. Alcolea 10 Cambridge Stem Cell Institute, Jeffrey Cheah Biomedical Centre, University of Cambridge, Puddicombe, Way , Cambridge, CB2 0AW, UK 11 Department of Physiology, Development and Neuroscience, University of Cambridge , Cambridge CB2 3EG, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for María P. Alcolea Sarah De Val 8 Department of Physiology, Anatomy and Genetics, Institute of Developmental and Regenerative Medicine, University of Oxford , Oxford, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Sarah De Val Susana Lopez-Fernandez 12 Department of Plastic Surgery, Hospital de la Santa Creu i de Sant Pau , Barcelona, Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site Katrien De Bock 6 Laboratory of Exercise and Health, Institute of Human Movement Sciences and Sport, Department of Health Sciences and Technology , ETH Zurich, Zurich, Switzerland Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Katrien De Bock Eulalia Baselga 13 Department of Dermatology, Hospital Sant Joan de Déu , 08950, Barcelona, Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Eulalia Baselga Rui Benedito 7 Molecular Genetics of Angiogenesis Group, Centro Nacional de Investigaciones Cardiovasculares (CNIC) , Madrid, Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Rui Benedito Ana Angulo-Urarte 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ana Angulo-Urarte For correspondence: aangulo{at}carrerasresearch.org mgraupera{at}carrerasresearch.org Mariona Graupera 1 Endothelial Pathobiology and Microenvironment, Josep Carreras Leukaemia Research Institute , 08916, Barcelona, Spain 14 CIBERONC, Instituto de Salud Carlos III, Av. de Monforte de Lemos 5 , 28029, Madrid, Spain 15 ICREA, Institució Catalana de Recerca i Estudis Avançats, Pg. Lluís Companys 23 , 08010, Barcelona, Spain Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Mariona Graupera For correspondence: aangulo{at}carrerasresearch.org mgraupera{at}carrerasresearch.org Abstract Full Text Info/History Metrics Preview PDF Abstract Cancer mutations in the PIK3CA gene cause congenital disorders. The endothelium is among the most frequently affected tissues in these disorders, displaying aberrant vascular overgrowth in the form of malformations. Pathological PIK3CA vascular phenotypes are found in veins and capillaries but rarely in arteries for reasons that are unclear at present. Here, using lineage tracing, we show that expression of mutated PIK3CA H1047R in endothelial cells leads to marked clonal expansions in capillary and venous endothelial cells. In contrast, mature arterial endothelial cells are refractory to PIK3CA mutation under these conditions and never display pathological phenotypes. Moreover, PIK3CA H1047R expression in arterial precursors interrupts arterial differentiation, thereby driving fate switch towards venous identity. This fate rewiring offers an additional layer of protection to prevent arterial damage in response to PIK3CA genetic perturbation. Molecularly, the PIK3CA H1047R -driven arterial-to-venous fate switch is orchestrated by upregulation of the vein-specifying transcription factor Nr2f2 /COUP-TFII. Our findings reveal that pathogenic responses to PIK3CA H1047R greatly depend on the diferentation stage and fate trajectory of the targeted cell. Arteries are thus shielded against PIK3CA mutation, solving the long-standing question on the rarity of PIK3CA -related arterial malformations observed in patients. Introduction Somatic and germline pathogenic variants in genes associated with cancer have been identified as a cause of a variety of congenital disorders ( Castel, Rauen, and McCormick 2020 ; Angulo-Urarte and Graupera 2022 ; Nussinov, Tsai, and Jang 2022 ; Samuels et al. 2004 ). In these conditions, mutations in the germline can be either inherited or appear de novo . Instead, somatic mutations arise during embryonic development, resulting in mosaic distribution and a broad spectrum of phenotypes ( Castel, Rauen, and McCormick 2020 ). Recent findings have shown that cancer mutations also exist in phenotypically normal tissues ( Herms and Jones 2023 ; Martincorena et al. 2018 ). Understanding why and how these mutations result in context-dependent manifestations is crucial for identifying tissue susceptibility to pathogenic outcomes and explaining the diversity of phenotypes observed in patients. PIK3CA encodes the p110α phosphoinositide 3-kinase (PI3K) catalytic subunit (hereafter referred to PI3Kα) that regulates cell growth, proliferation, migration, and metabolism ( Vanhaesebroeck et al. 2010 ), mainly through activation of AKT-mTOR signaling ( Hoxhaj and Manning 2020 ; Manning and Toker 2017 ). PIK3CA pertains to the group of oncogenes frequently mutated in a mosaic fashion in congenital disorders and normal tissues. Activating mutations in PIK3CA span the entire gene, albeit there are common hotspots across conditions (E545K, E542K, H1047R, H1047L) ( Angulo-Urarte and Graupera 2022 ). Mesodermal derivatives such as the endothelium are among the tissues most affected in PIK3CA -related congenital disorders, manifesting in aberrant overgrowth and malformations ( Angulo-Urarte and Graupera 2022 ; Madsen, Vanhaesebroeck, and Semple 2018 ). While tissues carrying PIK3CA mutations in the epithelium have been extensively studied (Van Keymeulen et al. 2015 ; Koren et al. 2015 ; Herms et al. 2024 ; Yum et al. 2021 ), the pathogenic responses of the mesodermal lineage in general, and endothelial cells in particular, to these mutations are much less known. Vascular malformations are abnormal vascular channels that may occur anywhere in the body and strongly impact the patient’s quality of life. These vascular lesions appear during embryonic development and continue growing proportionally throughout the patient’s lifetime. They are classified as slow-flow lesions if they appear in capillaries, veins, and lymphatic vessels and fast-flow lesions when malformations affect arteries ( International Society for the Study of Vascular Anomalies 2018 ). PIK3CA mutations have been identified in the endothelium of all types of slow-flow vascular malformations ( i.e., capillary, venous, and lymphatic malformations). These lesions form through increased cell proliferation, although PIK3CA mutations alone do not confer a proliferative advantage to endothelial cells (ECs). Expansion of the mutant cells selectively occurs in the presence of mitogenic signals, which explains why PIK3CA mutations in ECs exhibit detrimental effects during organ growth and hormonal spurs ( Kobialka et al. 2022 ; Hassanein et al. 2012 ; Petkova et al. 2023 ; Martinez-Corral et al. 2020 ). Currently, there is no evidence that PIK3CA mutations cause fast-flow vascular malformations ( Angulo-Urarte and Graupera 2022 ). Thus far, it remains unknown whether arterial cells harbor PIK3CA mutations and are resistant to them or whether the lack of pathological traits in these vessels is because mutant clones are negatively selected or due to alterations in differentiation pathways. Retinal blood vessels in mice emerge postnatally in a process known as sprouting angiogenesis ( Selvam, Kumar, and Fruttiger 2018 ). During the first week of the angiogenesis, de novo vessels grow in a 2D fashion from the optic nerve and require the secretion of mitogenic endothelial factors. The sprouting process involves the selection of two molecularly distinct cell types: tip cells, which function as leaders of the sprout and primarily migrate, and stalk cells, which facilitate sprout elongation through proliferation ( Blanco and Gerhardt 2013 ; Potente, Gerhardt, and Carmeliet 2011 ). A subset of tip cells expresses arterial markers, migrates backward against the flow, and contributes to the growth of developing arteries ( Xu et al. 2014 ; Pitulescu et al. 2017 ). Retinal vessels are amenable to whole-mount analysis, which allows the tracing of cell behaviors in specific EC subsets ( i.e., tip, arterial, capillary, vein) at single-cell resolution. Based on these unique features, retinal vessels represent an ideal system to interrogate why the same mutation results in distinct phenotypic outcomes in different EC subsets. Here, to gain insight into why pathogenic phenotypes are restricted to some EC types, we map the behavior of cells bearing PIK3CA mutations at clonal resolution. We use a compendium of genetic approaches that allow the faithful tracing of PIK3CA H1047R mutant clones in different EC subsets. We further use single-cell RNA-sequencing analysis to characterize how PIK3CA H1047R impacts endothelial cell fate trajectories leading to context-dependent pathological phenotypes in the vasculature. Results Spatial mapping of Pik3ca H1047R -driven clonal expansion in the retinal vasculature Expression of the Pik3ca H1047R allele ( Pik3ca tm1.1Waph/+ ) in mouse retinal ECs using the Pdgfb-CreER T2 mouse line results in venous and capillary malformations while arteries remain unperturbed ( Extended Data Fig. 1a-c ) ( Kobialka et al. 2022 ). To trace how these phenotypes correlate with the presence or absence of mutant cells, we crossed the R26-mTmG reporter allele into the Pik3ca H1047R ; Pdgfb-CreER T2 mice ( Fig. 1a,b ). In these mice, tamoxifen treatment stochastically activates Cre in any EC subset, followed by co-expression of the Pik3ca H1047R allele from its endogenous promoter and a membrane-bound fluorescent GFP reporter that provides an approximation of mutant Pik3ca expressing cells. Pdgfb-CreER T2 ; R26-mTmG mice were used as controls (hereafter referred to as pEC-GFP) . We treated pups at postnatal day (P) 1 with a low dose of 4-hydroxytamoxifen (4-OHT) and isolated retinas at P6 ( Fig. 1c ). We obtained a mosaic recombination pattern in pEC-GFP and pEC-GFP-Pik3ca H1047R retinas with GFP+ cells scattered throughout the vascular plexus ( Fig. 1d ). First, we analyzed the incidence of vascular malformations in pEC-GFP-Pik3ca H1047R retinas by identifying aberrant overgrown vessels composed of Pik3ca H1047R -GFP+ ECs. We observed a very high incidence of malformed vessels in capillaries and veins (87% and 62%, respectively) but none in arteries ( Fig. 1e ). High magnification images confirmed that wild-type and mutant GFP+ ECs were present in all vessel types ( Fig. 1f ) but that Pik3ca H1047R -GFP+ ECs only expanded in capillaries and veins ( Fig. 1f,g ). Expansion of mutant clones led to aberrant vascular tubes with increased vessel width ( Fig. 1f,h ). We confirmed that this expansion was associated with increased EC proliferation at the onset of the malformation ( Extended Data Fig. 1d-f ). Despite Pik3ca H1047R -GFP+ cells in arteries, these mutant cells did not expand or alter arterial vessel width ( Fig. 1f-h ). We monitored PI3K/AKT/mTORC1 signaling activation by staining for phospho (p)-S6 (Ser235/236) and found that capillary and venous Pik3ca H1047R -GFP+ ECs exhibited higher activation of PI3K signaling compared to their control counterparts ( Fig. 1f,i ). In contrast, no p-S6 (Ser235/236) difference was found between wild-type and mutant GFP+ cells in arteries. These data show that Pik3ca H1047R ECs selectively expand in capillaries and veins and that arteries develop normally despite mutant cells. Download figure Open in new tab Figure 1. Spatial mapping of Pik3ca H1047R clone expansions in the retinal vasculature (a) Schematic illustration of retinal vasculature. Veins (V) are shown in blue, arteries (A) are shown in red, the capillary network is represented in grey and the sprouting front in green. (b) Left: Schematic illustration of the transgenic mouse strategy combining the pan-endothelial (pEC) tamoxifen-inducible Cre ( Pdgfb-CreER T2 ) allele with either (i) the R26-mTmG reporter allele alone ( pEC-GFP ), or (ii) the R26-mTmG reporter together with the Pik3ca H1047R allele ( pEC-GFP-Pik3ca H1047R ), both Cre-dependent. Right: Schematic of the growing vasculature in a petal of the retina showing in blue the population targeted by activating Pdgfb-CreER T2 in a stochastic fashion. (c) Experimental timeline showing postnatal administration of 0.05 mg/kg of 4-hydroxytamoxifen (4-OHT) at P1 littermates, followed by retina isolation and analysis at P6. (d) Representative confocal images of pEC- GFP and pEC-GFP-Pik3ca H1047R P6 mouse retinas stained for Isolectin B4 (IB4, red, blood vessels) and GFP (cyan, recombined cells). (e) Quantification of the incidence of vascular malformations in capillaries, veins, and arteries in pEC-GFP - Pik3ca H1047R P6 retinas, expressed as the percentage of capillary beds, veins or arteries exhibiting a pathological overgrowth. (f) High-magnification confocal images of IB4-(red), GFP-(cyan) and phospho (p)-S6 (Ser235/236)-(yellow) stained P6 pEC-GFP and pEC-GFP-Pik3ca H1047R retinas. White dotted rectangles mark regions of interest (ROIs), which are shown in the adjacent right panel. Note the expansion of GFP+ ECs and increased S6 phosphorylation in capillaries and veins but not in arteries in pEC-GFP-Pik3ca H1047R retinas. (g, h) Quantification of the percentage of GFP+ area (n ≥ 5 retinas per genotype) (g) and vessel width (μm, n ≥ 6 retinas per genotype) (h) in capillaries, veins and arteries from pEC-GFP and pEC-GFP-Pik3ca H1047R P6 retinas. (i) Quantification of p-S6 intensity within GFP+ cells in capillaries, veins, and arteries, expressed as fold change comparing pEC-GFP to pEC-GFP-Pik3ca H1047R vasculature (n ≥ 5 retinas per genotype). Data are presented as mean ± s.d. Statistical analysis was performed using nonparametric Mann-Whitney test. ns, non-statistical. Scale bars: 150 µm (d), 30 µm (f). Differentiated arterial cells do not respond to Pik3ca H1047R The lack of response to Pik3ca H1047R by arterial ECs led us to hypothesize that mutant cells could be ejected from arteries over time. To investigate this hypothesis, we used the Bmx(PAC)-CreER T2 mouse line, which targets mature arterial ECs in the retinal vasculature. As described above, wild-type and Pik3ca H1047R arterial ECs were traced by combining Pik3ca H1047R ; Bmx(PAC)-CreER T2 with R26-mTmG (hereafter referred to as aEC-GFP and aEC-GFP-Pik3ca H1047R , respectively; Fig. 2a ). First, we characterized Bmx promoter activity by treating aEC-GFP pups with a single saturating dose of 4-OHT at three different postnatal stages (P1, P4, and P7), followed by isolating retinas five days later (P6, P9, and P12, respectively) ( Extended Data Fig. 2a,b ). Inducing aEC-GFP mice at P1 did not label any arterial ECs at P6, indicating a lack of mature arteries at the induction stage. Targeting Bmx+ cells at P4 and P7 resulted in an overt presence of GFP+ cells in arteries ( Extended Data Fig. 2b ). In agreement with Bmx being a marker of mature arteries, the number of GFP+ cells was higher when 4-OHT was injected at P7 ( Extended Data Fig. 2b ). Based on these results, we analyzed aEC-GFP-Pik3ca H1047R retinas by injecting 4-OHT at P4 or P7 ( Fig. 2b-d ). In these retinas, the proportion of GFP+ cells and the width of arteries were similar to those in aEC-GFP littermate retinas ( Fig. 2c,d ,g,h ). In line with the data shown above ( Fig. 1f,i ), Pik3ca H1047R -GFP+ arteries did not exhibit an increase in PI3K signaling ( Fig. 2e-h ). Further supporting the lack of pathological phenotypes, arteries of aEC-GFP and aEC-GFP-Pik3ca H1047R retinas showed similar coverage of smooth muscle cells ( Extended Data Fig. 2c ). Next, we studied the long-term effects of Pik3ca H1047R mutant cells in the arteries by analyzing retinas at P21. We found no differences in the GFP+ area or distribution of wild-type and Pik3ca H1047R -GFP+ ECs in the main arteries and arteriolar branches ( Fig. 2I, j ). These results confirm that mutant cells are well-tolerated in arteries for long periods without triggering any pathological response. Download figure Open in new tab Figure 2. Differentiated arterial cells do not respond to Pik3ca H1047R ((a) Left: Schematic illustration of the transgenic mouse strategy combining the arterial EC-specific (aEC) tamoxifen-inducible Cre ( Bmx(PAC)-CreER T2 ) allele with either (i) the R26-mTmG reporter allele alone ( aEC-GFP ), or (ii) the R26-mTmG reporter together with the Pik3ca H1047R allele ( aEC-GFP - Pik3ca H1047R ), both Cre-dependent. Right: Schematic of the growing vasculature in a petal of the retina showing in red the population targeted by activating Bmx(PAC)-CreER T2 . (b) Experimental timeline showing postnatal administration of 10mg/kg of 4-OHT at P4 or P7, followed by retina isolation and analysis at P9, P12 or P21. (c,d) Representative confocal images of aEC-GFP and aEC-GFP-Pik3ca H1047R P9 (c) and P12 (d) retinas immunostained for IB4 (red, blood vessels) and GFP (cyan, recombined cells) showing the distribution of GFP+ cells along the arteries. (e,f) High-magnification images of arteries from aEC-GFP and aEC-GFP-Pik3ca H1047R P9 (e) and P12 (f) retinas showing immunodetection of GFP (cyan, recombined cells), phospho (p)-S6 (Ser235/236) (white) and IB4 (red). (g,h) Quantification of the percentage of GFP+ arterial area (left), artery width (middle) and p-S6 (Ser235/236) intensity within GFP+ area (right) comparing aEC-GFP and aEC-GFP - Pik3ca H1047R P9 (g) and P12 (h) retinas (n ≥ 4 retinas per genotype). (i) Representative confocal images of aEC-GFP and aEC-GFP - Pik3ca H1047R P21 retinas, induced with 10 mg/kg 4-OHT at P4. Retinas were stained for IB4 (red, blood vessels) and GFP (cyan, recombined cells). (j) Quantification of GFP+ area, presented as a percentage of GFP+/IB4+ area comparing aEC-GFP and aEC-GFP - Pik3ca H1047R P21 retinas within the total vasculature (n ≥ 8 retinas per genotype). Data are presented as mean ± s.d. Statistical analysis was performed by nonparametric Mann-Whitney test. ns, non-statistical. Scale bars: 150 µm (c,d,i) and 30 µm (e,f). Pik3ca H1047R expression reduces arterial mobilization To investigate whether the lack of pathological response in mature arteries was a matter of differentiation stage, we next targeted pre-arterial cells. Tip ECs express arterial markers and contribute to expanding arteries ( Xu et al. 2014 ; Pitulescu et al. 2017 ); thus, they are considered pre-arterial cells. To trace tip ECs and their progeny, we used the Esm1(BAC)-CreER T2 mouse line crossed with the R26-mTmG ( tEC-GFP ) alone or in combination with Pik3ca H1047R ( tEC-GFP-Pik3ca H1047R , Fig. 3a ). 4-OHT was administrated at P1, followed by lineage tracing of the Esm1 -GFP+ progeny at P2, P6 and P8 ( Fig. 3b,c ). Twenty-four hours after 4-OHT administration, GFP+ cells were mainly detected at the sprouting front of tEC-GFP and tEC-GFP-Pik3ca H1047R retinas ( Fig. 3c ). At P6, arteries of tEC-GFP retinas were extensively composed of wild-type-GFP+ cells, and the GFP+ area in arteries increased by P8 ( Fig. 3c ). Consistent with a proportion of Esm1+ cells contributing to the capillary bed ( Pitulescu et al. 2017 ), we also detected wild-type-GFP+ cells distributed along capillaries. Surprisingly, we found that the area of GFP+ ECs that covered arteries at P6 and P8 was reduced by half in tEC-GFP-Pik3ca H1047R mice ( Fig. 3c,e ,f ). Remarkably, none of the few Pik3ca H1047R -GFP+ cells that reached an artery triggered arterial overgrowth ( Fig. 3d ). Most Esm1 -traced Pik3ca H1047R -GFP+ cells remained in the capillary bed, substantially expanding ( Fig. 3c,e,f ) and forming multiple vascular malformations ( Fig. 3d ). The expansion of the mutant clones in capillaries was associated with a significant increase in PI3K pathway signaling ( Fig. 3g,h ) and enhanced cell proliferation ( Extended Data Fig. 3a-c ). Intriguingly, many veins in tEC-GFP-Pik3ca H1047R retinas also contained GFP+ cells, a phenotype rarely seen in control retinas. We also observed that the proportion of GFP+ area in the veins of tEC-GFP-Pik3ca H1047R mice grew over time (P8 vs. P6) ( Fig.3f ). The loss of Pik3ca H1047R -GFP+ cells in arteries and their gain in veins suggest that Pik3ca H1047R induces a fate switch. We also analyzed P21 retinas and found that Pik3ca H1047R -GFP+ clones remained in veins and capillaries ( Extended Data Fig. 3d-f ). Consistent with the expansion of mutant clones in the presence of mitogenic signals( Kobialka et al. 2022 ), there was no further expansion of Pik3ca H1047R -GFP+ cells at later time points (P21), in contrast to their active expansion at P6 and P8 ( Extended Data Fig. 3g ). These data indicate that Pik3ca H1047R expression in arterial precursors prevents their mobilization to arteries, thereby remaining in capillaries and veins where they clonally expand and form vascular malformations. Download figure Open in new tab Figure 3. Pik3ca H1047R expression interferes with arterial mobilization (a) Left: Schematic illustration of the transgenic mouse strategy combining the tip EC-specific ( tEC ) tamoxifen-inducible Cre ( Esm1(BAC)-CreER T2 ) allele with either (i) the R26-mTmG reporter allele alone ( tEC-GFP ), or (ii) the R26-mTmG reporter together with the Pik3ca H1047R allele ( tEC-GFP - Pik3ca H1047R ), both Cre-dependent. Right: Schematic of the growing vasculature in a petal of the retina showing in green the population targeted by activating Esm1(BAC)-CreER T2 . (b) Experimental timeline showing postnatal administration of 10 mg/kg of 4-OHT at P1 littermates, followed by retina isolation and analysis at P2, P6, and P8. (c) Representative confocal images of tEC-GFP and tEC-GFP-Pik3ca H1047R of P2, P6 and P8 retinas stained with IB4 (red, blood vessels) and anti-GFP (cyan, recombined cells). (d) Quantification of the incidence of vascular malformations in the sprouting front, capillaries, veins, and arteries of P6 tEC-GFP - Pik3ca H1047R retinal vasculature, expressed as the percentage of vessels or capillary beds exhibiting a pathological phenotype. (e) High-magnification confocal images of tEC-GFP and tEC-GFP-Pik3ca H1047R P6 (left panel) and P8 (right panel) retinas stained with IB4 (red, blood vessels) and anti-GFP (cyan, recombined ECs) showing the expansion of GFP+ ECs (P1 Esm1 -derived progeny) in capillaries, veins, and arteries. (f) Quantification of GFP+ EC expansion (P1 Esm1 -derived progeny) in tEC-GFP and tEC-GFP-Pik3ca H1047R P6 (left) and P8 (right) retinas, comparing capillaries, veins, and arteries (n ≥ 8 retinas per genotype). (g) High-magnification confocal images of capillary regions from tEC-GFP and tEC-GFP-Pik3ca H1047R P6 (left panel) and P8 (right panel) retinas, stained for IB4 (red, blood vessels), GFP (cyan, recombined cells), and phospho (p)-S6 (Ser235/236) (yellow). (h) Quantification of p-S6 (Ser235/236) intensity within the GFP+ capillary area comparing tEC-GFP and tEC-GFP - Pik3ca H1047R P6 (left) and P8 (right) retinas (n ≥ 5 retinas per genotype). Data are presented as mean ± s.d. Statistical analysis was performed using nonparametric Mann-Whitney test. Scale bars: 150 µm (c) and 30 µm (e, g). Multispectral mosaic tracing of wild-type and Pik3ca H1047R endothelial clones To unequivocally trace both wild-type and mutant cells in the same tissue microenvironment, we generated a new genetic mouse model, hereafter abbreviated as iFluMosaic-PIK3CA H1047R ( Rosa26 CAG-LSL-H2B-Cherry/H2B-GFP-2A-HAHuPIK3CAH1047R ). This new allele contains two pairs of distinct and mutually exclusive Lox sites, which Cre can stochastically recombine to induce the expression of either His-H2B-Cherry or equimolar co-expression of H2B-EGFP-V5 along with HA - PIK3CA H1047R , due to the viral 2A peptide separating them ( Extended Data Fig. 4a ). First, we assessed the functionality of the construct by breeding iFluMosaic-PIK3CA H1047R with Cdh5(PAC)-CreER T2 mice ( Extended Data Fig. 5a ). Lung ECs from these mice were isolated, cultured, and treated with 4-OHT to induce stochastic recombination of the mutually exclusive LoxP sites ( Extended Data Fig. 5b ). Upon early recombination, we observed that the proportion of Cherry+ ECs (wild type, ≈60%) was higher than EGFP- PIK3CA H1047R ECs (≈40%) ( Extended Data Fig. 5c,e ). These data are consistent with the notion that the LoxN sites in front of the Cherry reporter are closer together than the Lox2272 sites and follow the ratiometric recombination pattern shown in similar constructs ( Pontes-Quero et al. 2017 ). As expected, EGFP- PIK3CA H1047R ECs showed higher proliferative rates ( Extended Data Fig. 5c,f ) and higher PI3K signaling than Cherry-Wild-type ECs ( Extended Data Fig. 5d,g ). This higher proliferation led to a higher fraction of EGFP- PIK3CA H1047R ECs over time, offsetting the initial Cherry bias ( Extended Data Fig. 5c-e ). Next, we fate-mapped EGFP- PIK3CA H1047R and Cherry-Wild-type pre-arterial tip ECs in vivo by crossing iFluMosaic-PIK3CA H1047R with Esm1(BAC)-CreER T2 mice ( tEC-iFluMosaic-PIK3CA H1047R ). Pups were treated with a high dose of 4-OHT at P1 and P2, followed by assessing the behavior of EGFP- PIK3CA H1047R and Cherry-Wild-type targeted cells 6 days after ( Fig. 4a,b ). Cherry+ cells were distributed between capillaries and arteries ( Fig. 4c,d ). In contrast, Esm1 -targeted EGFP- PIK3CA H1047R cells and their progeny were found in veins and capillaries of the sprouting front, the central plexus, and the immediate vicinity of arteries ( Fig. 4c,d ). In these locations, EGFP- PIK3CA H1047R cells clonally expanded and formed vascular malformations ( Fig. 4d ), which overall resulted in a more significant number of EGFP+ cells (i.e., 7680) compared to Cherry+ cells (i.e., 511) ( Fig. 4e ). Vascular lesions were of different sizes, including big (> 30 cells), medium (between 10-20 cells), and small (<10 cells) ( Fig. 4f ). EGFP- PIK3CA H1047R cells showed increased Ki67 positivity ( Fig. 4g and Extended Data Fig. 5h ) and p-S6 immunostaining compared to Cherry-Wild-type cells ( Fig. 4h and Extended Data Fig. 5i ). We noticed that few EGFP- PIK3CA H1047R cells (0,47% of total EGFP+ cells) settled in the artery, albeit at an almost fifty-fold lower proportion than Cherry+ cells (23% of total Cherry+ cells) ( Fig. 4d, e ). Overall, among arteries that contain recombined cells only 16% presented EGFP- PIK3CA H1047R mutant ECs ( Fig. 4i ). EGFP- PIK3CA H1047R cells that arrived at the artery formed small lesions (composed of an average of 7 cells/lesion) ( Fig. 4d,j ). Given that malformations in the artery do not occur when expression of Pik3ca H1047R is regulated through the endogenous locus ( Fig. 3 ), these data suggest that pre-arterial cells downregulate the levels of PI3Kα before differentiating into mature arterial cells. Next, we wondered whether similar phenotypes were observed upon forced overexpression of PIK3CA H1047R in mature arterial cells. We combined iFluMosaicPIK3CA H1047R with Bmx(PAC)-CreER T2 and injected 4-OHT between P7 and P9 ( Fig. 4k,l ). P12 arteries contained the expected ratio of Cherry+ (65%) and EGFP- PIK3CA H1047R (35%) cells ( Fig. 4m,n ). Yet, mutant EGFP+ cells in mature arterial ECs did not trigger any pathological phenotype ( Fig. 4o ), indicating that forced overexpression of PIK3CA H1047R in mature arterial cells is insufficient to overcome their resistance to expand upon expression of these mutations. Download figure Open in new tab Figure 4. Multispectral mosaic tracing of wild-type and Pik3ca H1047R endothelial clones (a) Left: Schematic illustration of the transgenic mouse strategy combining the tip EC-specific ( tEC ) tamoxifen-inducible Cre ( Esm1(BAC) -CreER T2 ) allele with the iFluMosaic-PIK3CA H1047R allele. The iFluMosaic-PIK3CA H1047R knock-in model contains two mutually exclusive Lox sites that, upon recombination by Cre activity, either result in the expression of (i) the nuclear His-H2B-Cherry (recombined cell remains wild type for Pik3ca ) or (ii) H2B-EGFP-V5 and HA- PIK3CA H1047R separated by a 2A peptide (recombined cell overexpresses human PIK3CA H1047R cDNA tagged with HA at the C-terminal along with the nuclear EGFP). This system allows for the simultaneous tracing of both wild-type and mutant cells within the same sample using coupled fluorescent proteins. For detailed information, refer to Figure S4 and M&M section. Right: Schematic of the growing vasculature in a petal of the retina showing in dark grey the population targeted by activating Esm1(BAC)-CreER T2 . Recombined cells are distinguished by fluorescent nuclear markers: Cherry-Wild-type EC for Pik3ca are represented in cyan, while ECs expressing EGFP and PIK3CA H1047R are represented in yellow. (b) Experimental timeline showing postnatal administration of 50 mg/kg of 4-OHT at P1 and P2 mouse littermates, followed by retina isolation and analysis at P8. (c) Representative confocal images of tEC-iFluMosaic-PIK3CA H1047R P8 mouse retinas immunostained for IB4 (magenta, blood vessels), Cherry (cyan, wild-type ECs for Pik3ca ) and EGFP (yellow, PIK3CA H1047R expressing ECs). “A” refers to the artery, “V” refers to the vein. Cyan arrows point to wild-type ECs for Pik3ca . (d) High-magnification confocal images of tEC-iFluMosaic-PIK3CA H1047R P8 retinas. From left to right: healthy artery populated with Cherry-Wild-type ECs, malformed capillaries, sprouting front, and vein, showing pathological expansion of EGFP- PIK3CA H1047R clones, and a malformed artery (indicated with a white asterisk) with EGFP- PIK3CA H1047R clones, as an example of an occasional lesion. (e) Stacked bar graph comparing the percentage of P1 and P2 recombined Esm1 -Cherry-Wild-type and EGFP- PIK3CA H1047R -derived progeny located within arteries vs capillaries and veins in P8 tEC-iFluMosaic-PIK3CA H1047R retinas. Numbers within the bar show the total number of analyzed cells and the resulting percentage for each category (n = 6 retinas). (f) Pie chart categorizing lesions based on their extent, defined by the number of EGFP- PIK3CA H1047R ECs in tEC-iFluMosaic-PIK3CA H1047R P8 retinas (n = 14 retinas). (g) Quantification of cell proliferation assessed as a percentage of Ki67+ staining within Cherry-Wild-type and EGFP- PIK3CA H1047R ECs in tEC-iFluMosaic-PIK3CA H1047R P8 retinas (n = 6 retinas). (h) Quantification of phospho (p)-S6 (Ser235/236) intensity in Cherry-Wild-type or EGFP- PIK3CA H1047R ECs of tEC-iFluMosaic-PIK3CA H1047R P8 retinas (n = 4 retinas). (i) Quantification of the percentage of arteries containing Cherry-Wild-type and EGFP- PIK3CA H1047R ECs among total number of arteries with any recombined EC. (j) Quantification of artery width in regions containing Cherry-Wild-type and EGFP- PIK3CA H1047R ECs compared to areas with non-labeled wild-type ECs (n = 7 arteries from 6 retinas). This parameter is used to detect arterial overgrowth. (k) Left: Schematic illustration of the transgenic mouse strategy combining the artery EC-specific (aEC) tamoxifen-inducible Cre ( Bmx(PAC) -CreER T2 ) allele with iFluMosaic-PIK3CA H1047R allele. Right: Schematic of the growing vasculature in a petal of the retina showing in dark grey the population targeted by activating Bmx(PAC) -CreER T2 . Recombined cells are distinguished by fluorescent nuclear markers: Cherry-Wild-type EC for Pik3ca are represented in cyan, while ECs expressing EGFP and PIK3CA H1047R are represented in yellow. (l) Experimental timeline showing postnatal administration of 50mg/kg of 4-OHT at P7, P8 and P9 mouse littermates, followed by retina isolation and analysis at P12. (m) A representative confocal image of aEC-iFluMosaic-PIK3CA H1047R P12 mouse retina immunostained for IB4 (magenta, blood vessels), Cherry (cyan, wild-type ECs for Pik3ca ) and EGFP (yellow, PIK3CA H1047R expressing ECs). The white dotted square marks the ROI (R’), which is magnified to show an artery in greater detail in the right panel. (n) Quantification of the percentage of recombined Cherry-Wild-type and EGFP- PIK3CA H1047R ECs within aEC - iFluMosaic-PIK3CA H1047R P12 arteries (n = 14 retinas). (o) Quantification of artery width in regions containing Cherry-Wild-type and EGFP- PIK3CA H1047R ECs compared to regions with non-labeled wild-type ECs (n= 14 retinas). Data are presented as mean ± s.d. Statistical analysis was performed using nonparametric Mann-Whitney test (g,h,j, and o). ns, non-statistical. Scale bars: 150 µm (c, m) and 30 µm (d). Our data demonstrate that arterial differentiation is interrupted upon acquiring PIK3CA H1047R mutations in pre-arterial cells. Pik3ca H1047R expression triggers a transcriptional switch toward a venous phenotype To study how Pik3ca mutations induce a cell fate switch, we analyzed the transcriptomic profile of ECs by single-cell RNA sequencing (scRNAseq). Given the low number of Esm1 + cells in the retina, particularly in wild-type mice, we chose an in vitro strategy. We aimed to capture the presence of various EC subsets and identify whether acute expression of Pik3ca H1047R interfered with cell fate and state transitions. To study these questions, we used ECs derived from Pdgfb-CreER T2 ; Pik3ca H1047R ( pEC-Pik3ca H1047R ) mice. Parental ECs were treated with vehicle (wild type for Pik3ca ) or 4-OHT (expressing the Pik3ca H1047R allele) and analyzed at 48 h post-induction. scRNAseq analysis of control and Pik3ca H1047R ECs yielded 5 clusters (C1-C5) with different signatures of marker enrichment in the two-dimensional (2D) Uniform Manifold Approximation and Projection (UMAP) ( Fig. 5a and Extended Data Fig. 6a,b ). Specifically, the C1 cluster exhibited expression of genes characteristic of tip and arterial ECs, such as Cxcr4, Esm1, Apln, and Sox17 ( Fig. 5d and Extended Data Fig. 6b ). The C2 cluster contained cells with high expression of known markers of venous identity, including Nrp2, Ephb4, and Nr2f2 . Some cells within the C2 cluster also showed enrichment for canonical markers of lymphatic ECs ( Prox1, Lyve-1, Itga9 ) ( Extended Data Fig. 6c ). The C4 and C5 clusters showed enrichment of genes involved in cell proliferation, while the C3 cluster contained cells with elevated expression of inflammation-related genes ( Cd44, Ptgs2, Serpine1, Cxcl12, Fig. 5a,d and Extended Data Fig. 6b ). Download figure Open in new tab Figure 5. Pik3ca H1047R expression promotes a transcriptional shift towards a venous phenotype with an increase in Nr2f2 levels (a) UMAP representation of Louvain clustering of merged Pdgfb-CreER T2 (pEC)- Pik3ca H1047R mouse lung endothelial cells (ECs) 48 h after treatment with ethanol (vehicle, wild type for Pik3ca ) and 4-OHT (expressing Pik3ca H1047R ) derived from scRNA-seq data. (b) Bar plot showing the distribution of pEC -Wild-type (ethanol-treated) and pEC-Pik3ca H1047R (4-OHT-treated) ECs clusters. (c) UMAP representation of pEC- Wild-type (ethanol-treated, grey) and pEC-Pik3ca H1047R (4-OHT-treated, blue) ECs shown independently. (d) Dot plot showing the average expression of key cell type markers across the 5 clusters. (e) Tip, arterial and venous markers selected from differentially expressed genes (DEG, FDR adjusted p-value < 0.05) analysis in bulk RNA-seq data ( Kobialka et al. 2022 ) comparing pEC-Pik3ca H1047R (4-OHT-treated) with pEC -Wild-type (ethanol-treated) ECs 24 h post-induction. (f) RT-qPCR for Nr2f2 gene expression 24 h and 72 h after ethanol or 4-OHT treatment in pEC-Pik3ca H1047R ECs (n = 6 biological replicates). (g) Representative immunoblot showing the activation of PI3K/AKT pathway (by assessing the levels of phopsho(p)-AKT (Ser473)) and COUP-TFII protein levels in pEC-Pik3ca H1047R ECs 24 h and 72 h after ethanol or 4-OHT treatment. (h) Quantification of COUP-TFII protein levels normalized to vinculin levels (n = 6 biological replicates). (i) RT-qPCR for Nr2f2 gene expression in pEC-Pik3ca H1047R ECs treated with the PI3Kα specific inhibitor, BYL719. ECs were first treated with ethanol or 4-OHT for 6 h, followed by treatment with vehicle or BYL719 (1 µM and 5 µM) for 66 h. (n = 6 biological replicates). (j) Representative immunoblot showing the impact BYL719 inhibitor on PI3K/AKT signaling and COUP-TFII levels in pEC-Pik3ca H1047R ECs. ECs were first treated with ethanol or 4-OHT for 6 h, followed by treatment with vehicle or BYL719 (1 µM and 5 µM) for 66 h. (k) Quantification of COUP-TFII protein levels normalized to vinculin levels in pEC-Pik3ca H1047R ECs treated with BYL719 (vehicle DMSO, 1 µM and 5 µM) (n = 6 biological replicates). (l) Representative immunoblot showing the impact of BYL719, on PI3K/AKT signaling and COUP-TFII levels in PIK3CA H1047R and PIK3CA E545G patient-derived venous ECs. These ECs were treated with vehicle DMSO and BYL719 (1 µM and 5 µM) for 72 h. (m) RT-qPCR for NR2F2 gene expression in PIK3CA H1047R and PIK3CA E545G patient-derived venous ECs treated with BYL719 (vehicle DMSO and 5 µM) (n = 3 technical replicates). (N) Representative confocal images of control and tEC-Pik3ca H1047R P6 mouse retinas after 25 mg/kg 4-OHT induction at P1 and P2. Immunostaining for IB4 (white, blood vessels), ERG (blue, EC nuclei) and COUP-TFII (yellow) showing upregulation of COUP-TFII in the capillary area of tEC-Pik3ca H1047R P6 retinas. (o) Quantification of COUP-TFII intensity within ERG+ areas (EC nuclei) in capillary vessels from control and tEC-Pik3ca H1047R P6 retinas (n = 4 retinas per genotype). Data are presented as mean ± s.d. Statistical analysis was performed using an unpaired t-test (f, h, m), a 2-way ANOVA (i, k), and a nonparametric Mann-Whitney test (o). ns, non-statistical. Scale bars: 30 µm (n). Consistent with published data ( Kobialka et al. 2022 ), the C4 cluster, composed of proliferative cells, increased by about 10% following Pik3ca H1047R expression ( Fig. 5b ). We also noticed that there was a marked loss of cells in the C1 cluster enriched in the tip and arterial markers and an overt increase of the C2 venous-like cluster in the 4-OHT treated population ( Fig. 5b,c ). The Principal component analysis of the first two components revaled a high degree of similarity between the C1 and C2 culsters. Hence, we proceed to analyze the differentialy expressed genes (DEG) in these group of cells between Pik3ca WT and Pik3ca H1047R -expressing cells ( Extended Data Fig. 6d, e ). This comparison identified that 4141 genes were differentially expressed (FDR adjusted p-value < 0.05), with tip ECs markers strongly decreased (i.e., Esm1 , Angpt2, Cxcr4 , Mecom , Adm, and Apln ) and venous markers (i.e., Nrp2 , Nr2f2, and Ephb4) increased upon expression of Pik3ca H1047R ( Extended Data Fig. 6f,g ). To validate these molecular changes in an independent data set, we used published bulk RNAseq data ( Kobialka et al. 2022 ). We specifically focused on tip, arterial, and venous genes and found a significantly decreased expression of tip and arterial markers in Pik3ca H1047R ECs and a prominent increase in expression of venous markers ( Fig. 5e ). Collectively, these data indicate that in addition to adopting proliferative behavior ( Kobialka et al. 2022 ), expression of Pik3ca H1047R in ECs induces a molecular transition from tip/arterial cell-like to venous identity phenotype, which aligns with the phenotypes observed in the retinal vasculature. Pik3ca H0147R mutation increases NR2F2 and COUP-TFII expression Our scRNAseq and bulk RNAseq data showed that Pik3ca H1047R expression significantly increased Nr2f2 mRNA levels ( Extended Data Fig. 6f,g and Fig. 5e ). NR2F2 encodes for COUP-TFII, a key transcription factor (TF) that induces EC proliferation and suppresses arterial identity ( Chen et al. 2012 ; Aranguren et al. 2013 ; You et al. 2005 ; Su et al. 2018 ). Building on these results, we hypothesized that COUP-TFII could act downstream of Pik3ca H1047R to rewire arterial identity and promote clone expansion. First, we confirmed by quantitative (q)PCR that Nr2f2 levels rose rapidly 24 h after 4-OHT treatment of mutant ECs in vitro and remained elevated over time ( Fig. 5f ). Higher Nr2f2 mRNA levels in Pik3ca H1047R ECs led to a significant increase in COUP-TFII protein levels at 72 h post-recombination ( Fig. 5g,h ). PI3K signaling was also overactivated at 24 h and 72 h after 4-OHT treatment, as shown by increased phospho (p)-AKT (Ser473) levels ( Fig. 5g and Extended Data Fig. 6h ). Next, we treated these cells with Alpelisib (BYL-719). This PI3Kα-selective inhibitor led to a significant reduction in the levels of p-AKT (Ser473) ( Fig. 5j and Extended Data Fig. 6i ) and Nr2f2 mRNA and COUP-TFII protein in a dose-dependent manner ( Fig. 5i-k ). PI3Kα overactivation also regulated Nr2f2 and COUP-TFII levels in patient-derived ECs with an activating PIK3CA mutation ( Fig. 5l,m ). Finally, COUP-TFII immunostaining of retinas from tEC-Pik3ca H1047R and control P6 littermates revealed a significant increase in COUP-TFII expression in the progeny of Esm1 -targeted Pik3ca H1047R cells in vivo ( Fig. 5n,o ). These results establish COUP-TFII as a downstream target of PI3K signaling and that this regulation primarily occurs at the mRNA level. Nr2f2 deletion allows arterialization of Pik3ca H1047R pre-arterial cells To elucidate whether COUP-TFII mediates the Pik3ca H1047R -related fate switch and pathogenic proliferative response in vivo , we examined the behavior of PIK3CA H1047R pre-arterial tip cells in the absence of COUP-TFII expression. For this, we used both genetic strategies of Pik3ca H1047R expression: endogenous expression, which provides a pathophysiological setting ( Fig. 6a ), and iFluMosaic-PIK3CA H1047R to faithfully trace wild-type and mutant clones, both in combination with the Nr2f2 flox/flox allele ( Fig. 6f ). To maximize the recombination of all alleles in their endogenous locus, two consecutive doses of 4-OHT were injected ( Fig. 6a,b ). Consequently, more recombined cells were observed in the tEC-GFP control retinas ( Fig. 6c ) than those treated with one 4-OHT dose ( Fig. 3c ). First, we confirmed that COUP-TFII expression was reduced in tEC-Pik3ca H1047R -Nr2f2 ΔEC/ΔEC retinas ( Extended Data Fig. 7a,b ). Deletion of Nr2f2 in a wild-type background did not interfere with the mobilization of Esm1- targeted Nr2f2 ΔEC/ΔEC -GFP+ ECs, nor with the overall vascular density ( Fig. 6c,d and Extended Data Fig. 7c,d ). Pik3ca H1047R - Nr2f2 ΔEC/ΔEC -GFP+ mutant cells kept expanding regardless of Nr2f2 deletion ( Fig. 6c,e ). In contrast, we noticed that Esm1- targeted Pik3ca H1047R -Nr2f2 ΔEC/ΔEC -GFP+ cells regained the ability to form arteries compared to Pik3ca H1047R - GFP+ cells ( Fig. 6c,d ). The tEC - iFluMosaic-PIK3CA H1047R model ( Fig. 6f,g ) also showed that EGFP- PIK3CA H1047R -Nr2f2 ΔEC/ΔEC mutant cells overexpanded in the capillary bed ( Extended Data Fig. 7e,f ). Yet, while Cherry and Cherry- Nr2f2 ΔEC/ΔEC ECs colonized arteries to a similar extend, EGFP- PIK3CA H1047R -Nr2f2 ΔEC/ΔEC cells reached arteries significantly more frequently than EGFP- PIK3CA H1047R cells (35% vs 10% of arteries respectively) ( Fig. 6h,i ). Notably, the majority of EGFP- PIK3CA H1047R -Nr2f2 ΔEC/ΔEC cells did not form a lesion when settled in the artery in contrast to EGFP- PIK3CA H1047R cells, which consistently formed small lesions ( Fig. 6h,j ). These data suggest that tip EGFP- PIK3CA H1047R -Nr2f2 ΔEC/ΔEC cells recover the capacity to colonize arteries and differentiate into mature arterial cells despite expressing PIK3CA H1047R . Collectively, these findings indicate that COUP-TFII orchestrates PIK3CA -driven fate switch toward vein identity. Download figure Open in new tab Figure 6. Nr2f2 deletion allows arterialization of Pik3ca H1047R pre-committed arterial cells (a) Left: Schematic illustration of the transgenic mouse strategy combining the tip EC-specific (tEC) tamoxifen-inducible Cre ( Esm1(BAC)-CreER T2 ) with the R26-mTmG reporter allele (i) alone ( tEC-GFP ), (ii) combined with the Pik3ca H1047R allele ( tEC-GFP - Pik3ca H1047R ), (iii) combined with the Nr2f2 flox/flox allele ( tEC-GFP-Nr2f2 flox/flox ) or (iv) with both of them ( tEC-GFP-Pik3ca H1047R - Nr2f2 flox/flox ). Right: Schematic of the growing vasculature in a petal of the retina showing in green the population targeted by activating Esm1(BAC)-CreER T2 . (b) Experimental timeline showing postnatal administration of 25 mg/kg of 4-OHT at P1 and P2, followed by retina isolation and analysis at P6. (c) Representative confocal images of tEC-GFP , tEC-GFP-Nr2f2 ΔEC/ΔEC , tEC-GFP-Pik3ca H1047R , and tEC-GFP-Pik3ca H1047R - Nr2f2 ΔEC/ΔEC of P6 mouse retinas, stained for IB4 (red, blood vessels) and GFP (cyan, P1/P2 Esm1 -derived progeny). “A” denotes for artery. (d, e) Quantification of the GFP+ area specifically within arteries (d) and across all retina vasculature (e), presented as a percentage of the GFP+/IB4+ area, used to assess the expansion of the Esm1 -derived progeny at P6 (n ≥ 7 retinas per genotype). (f) Left: Schematic illustration of the transgenic mouse strategy combining the tip EC-specific (tEC) tamoxifen-inducible Cre ( Esm1(BAC)-CreER T2 ) allele with iFluMosaic-PIK3CA H1047R , either alone or in combination with the Nr2f2 flox/flox allele. Right: Schematic of the vasculature in a retinal petal showing in dark grey the population targeted by activating Esm1(BAC)-CreER T2 . Recombined cells are distinguished by fluorescent nuclear markers: Cherry-Wild-type EC for Pik3ca are represented in cyan, while ECs expressing EGFP and PIK3CA H1047R are represented in yellow. (g) Experimental timeline showing postnatal administration of 50 mg/kg of 4-OHT at P1, followed by retina isolation and analysis at P8. (h) Representative confocal images of tEC-iFluMosaic-PIK3CA H1047R and tEC-iFluMosaic-PIK3CA H1047R -Nr2f2 ΔEC/ΔEC P8 retinas stained for IB4 (magenta, blood vessels), Cherry (cyan, wild-type ECs for Pik3ca ) and EGFP (yellow, PIK3CA H1047R expressing ECs) in arteries with no lesions (left) and in malformed arteries (right). (i) Quantification of the number of arteries with Cherry+ or EGFP+ ECs, expressed as the percentage of positive arteries out of all arteries in tEC-iFluMosaic-PIK3CA H1047R and tEC-iFluMosaic-PIK3CA H1047R -Nr2f2 ΔEC/ΔEC P8 retinas (n = 16 retinas per genotype). (j) Quantification of EGFP+ arteries in tEC-iFluMosaic-PIK3CA H1047R and tEC-iFluMosaic-PIK3CA H1047R -Nr2f2 ΔEC/ΔEC P8 retinas, categorizing them as healthy or malformed (n ≥ 7 retinas per genotype). Data are presented as mean ± s.d (d, e) or ± s.e.m (i, j). Statistical analysis was performed using one-way ANOVA, followed by Tukey test for multiple comparisons (d, e), and a nonparametric Mann-Whitney test (i). Scale bars: 200 µm (c) and 50 µm (h). Discussion Mutations traditionally associated with cancer have been identified in a variety of diseases, including congenital syndromes, clonal hematopoiesis, and also in healthy tissues. This has sparked the interest of the scientific community in understanding the role of these mutations beyond cancer. In this regard, the impact of these mutations in the epithelium has been extensively studied, providing light on the mechanisms of mutant clone survival ( Ciwinska et al. 2024 ; Herms et al. 2024 ; Colom et al. 2021 ). Yet, a question remains: why pathogenic phenotypes are restricted to some tissues and cell types? In this study, we take advantage of the vasculature, a mesoderm-derived tissue commonly affected by cancer mutations, which exhibits restricted pathogenic patterns to these mutations. We specifically focus on PIK3CA H1047R , a cancer mutation frequently found in capillary, vein, and lymphatic malformations but not in arteries. Our findings identify a wide range of context-dependent effects of PIK3CA H1047R mutations in ECs, including (i) generation of large clone expansions (i.e., in venous and capillary ECs) causing vascular malformations, (ii) silent presence in refractory cell types (i.e., arterial ECs) without pathological manifestations; and (iii) rewiring of cell fate (i.e., in arterial precursors) favoring pathogenic expansion ( Fig. 7 ). Download figure Open in new tab Figure 7. Schematic illustration of the context-dependent effects of PIK3CA H1047R mutation in ECs Stochastic expression of PIK3CA H1047R in any ECs results in the generation of large clones specifically in venous and capillary ECs, causing vascular malformations while arterial ECs remain unperturbed. Expression of PIK3CA H1047R in mature arterial ECs does not cause any pathological response. Instead, the expression of PIK3CA H1047R in pre-arterial tip cells interrupts arterial differentiation, and induces fate switch towards venous identity. The arterial-to-venous fate switch induced by PIK3CA H1047R is orchestrated through the upregulation of Nr2f2 /COUP-TFII expression, a major venous fate determinant. By combining genetic approaches that faithfully trace PIK3CA mutant ECs, our data demonstrate that expression of PIK3CA mutations is compatible with their existence in mature arteries ( Bmx traced), albeit they do not induce any response. This is surprising, given that PI3Kα is the most active class IA PI3K isoform in ECs ( Graupera et al. 2008 ) and is ubiquitously expressed ( Vanhaesebroeck et al. 2010 ). Yet, we show that PI3K signaling is relatively low in arteries and is not activated in response to PIK3CA H1047R . In addition, we demonstrate that not even forced overexpression of PIK3CA H1047R can break arterial refractoriness in mature arterial cells. Building on these observations, we speculate that mature arterial ECs display cell-autonomous mechanisms to silence PI3Kα expression. Notably, the lack of arterial response to different mutations is not unique to PIK3CA, as other mutations, such as the loss-of-function mutation in PDCD10/CCM3 provoke similar impassive responses ( Orsenigo et al. 2020 ), thus suggesting that arteries are also equipped with broad protective mechanisms against genetic insult. Despite mature arterial ECs not responding to PIK3CA H1047R , this mutation interrupts the differentiation of pre-arterial cells ( i.e., tip cells). As soon as they emerge, tip cells are marked as pre-committed arterial ECs by expressing key arterial markers such as CXCR4, DATCH1, HEY1, DLL4, and SOX17 ( Pitulescu et al. 2017 ; Stewen et al. 2024 ; Park et al. 2021 ). Later, these cells migrate backward against the flow from the sprouting front to the inner plexus, settling in the artery and contributing to the growth of this vessel type. In this transition, a reduction of cell-cycle activity is required ( Luo et al. 2020 ). We showed that upon expression of Pik3ca H1047R , tip cells are rewired to express high levels of cell cycle genes and block arterial identity. Consequently, these cells are retained in the venocapillary bed, overexpanding and forming lesions. These results align with PI3K signaling promoting venous differentiation during vascular development ( Chu et al. 2016 ) while blocking arterial differentiation ( Ang et al. 2022 ) and favoring stemness states ( Madsen et al. 2019 ). Indeed, in the aging human esophagus, PIK3CA H1047R mutations in progenitor cells favor progenitor cell fate, whereas wild-type cells exhibit equal numbers of progenitor and differentiated cells per average division ( Herms et al. 2024 ). Likewise, PIK3CA H1047R mutations evoke cell dedifferentiation during tumorigenesis in the mammary gland ( Koren et al. 2015 ; Van Keymeulen et al. 2015 ). As new mechanism findings, we discovered that PIK3CA mutant tip cells circumvent arterial differentiation by upregulating COUP-TFII. These data align with discoveries showing that COUP-TFII can block pre-arterial, but not arterial, specification ( Su et al. 2018 ) and that COUP-TFII promotes stemness under proliferative pressure ( Mauri et al. 2021 ). Despite some Pik3ca H1047R ; Nr2f2 iΔEC/iΔEC tip cells regaining the ability to become arterial ECs, many others remain in the capillary bed where they keep expanding. This incomplete rescue suggests that venous and proliferative programs are uncoupled downstream of PIK3CA mutations. This is surprising given that COUP-TFII is a master regulator of both venous identity and cell cycle progression ( You et al. 2005 ; Wu et al. 2016 ) and that COUP-TFI blocks arterial differentiation by upregulating cell cycle genes ( Su et al. 2018 ). Recent evidence has suggested that the cell cycle state determines cell fate ( Chavkin et al. 2022 ). For instance, an early G1 cycle state is necessary to embrace a venous identity program. In contrast, arterial specification is only compatible with being in late G1. In this regard, not all tip cells exhibit a similar cell cycle state: approximately 50% of these cells are in early G1, and the remaining 50% tend to be in late G1 ( Chavkin et al. 2022 ). Hence, it is possible that COUP-TFII depletion selectively affects tip cells in late G1, which are those more permissive to arterial signals. Instead, tip cells in the early G1 phase, and thus prone to proliferate, may already exhibit high PI3K signaling and, in turn, be less sensitive to Nr2f2 deletion or COUP-TFII depletion. Of note, one should bear in mind that it is also conceivable that the expression of Pik3ca H1047R and deletion of Nr2f2 occur in an asynchronous fashion with Pik3ca H1047R , thus resulting in a much faster expression than COUP-TFII depletion. This would imply that when COUP-TFII was fully depleted, many tip cells expressing Pik3ca H1047R had already clonally amplified and undergone an irreversible fate switch. We found that both Nr2f2 mRNA and COUP-TFII protein expression levels are increased in Pik3ca H1047R mutant ECs. This fits with previous work showing that activation of the TIE2-PI3K axis stabilizes COUP-TFII protein levels during venous specification ( Chu et al. 2016 ). Yet, our data suggest that overactivation of PI3K signaling primarily regulates COUP-TFII at the transcriptional level. Currently, it is not understood which TFs regulate Nr2f2 mRNA expression. Indirect evidence has suggested that SOX7, SOX18, and R-SMADs control Nr2f2 expression in ECs ( McCracken et al. 2023 ; Neal et al. 2019 ; Swift et al. 2014 ), yet how this connects with PI3K signaling is uncertain. Various enhancers have also been described as regulating Nr2f2 expression in ECs. For instance, ERG is the highest expressed ETS factor in mature ECs ( Shah, Birdsey, and Randi 2016 ) and is a crucial cognate binding TF for Nr2f2 ( Payne, Neal, and De Val 2024 ). However, ERG exhibits genome-wide enhancer occupancy in ECs, which has impaired understanding of how and when it promotes arterial or venous fate identity programs. Intriguingly, PI3K signaling via AKT, has been shown to modify the ERG cistrome and promote the expression of specific cell fate genes in prostate cancer ( Strittmatter, Jerde, and Hollenhorst 2021 ). In an independent line of research, overexpression of COUP-TFII has also been found in PI3K-driven prostate tumors, with high COUP-TFII levels cooperating with PI3K signaling to sustain cancer progression ( Qin et al. 2013 ). Based on these similarities, it is tempting to speculate that Pik3ca H1047R regulates Nr2f2 expression via ERG in ECs. Overall, our findings reveal that arterial ECs have a double layer of protection against PIK3CA . On the one hand, we identify that arterial cells are refractory to these mutations. On the other hand, we demonstrate that when PIK3CA mutations are expressed at early fate stages, a fate tilt towards venous and capillaries occurs, reducing the likelihood that mutant cells reach an artery. Our model shows that cell fate and differentiation stage determine the context-dependent pathological manifestations of PIK3CA mutations and opens the groundwork for understanding why PIK3CA -related tissue overgrowth is only observed in certain tissues. Author contribution M.G., A.A-U., H.S. and A.R. were the main contributors to the conception, design, acquisition, and interpretation of the data, and in writing the article. J.D., A. C-R., E.C., N.T., M. A-L., S.N., J. L., P. V., performed experiments and data analysis with inputs from S.D.C., B.V., M.A.P, S.D.V., K.D.B, and R.B. L.G., and A.M-L. performed bioinformatic analysis. S.L., and E.B. liaised with human subjects and provided access to human tissue samples and clinical input for the study. Declaration of interests M.G. has a research agreement with Relay Therapeutics, Inc., A Delaware corporation having a principal place of business at 399 Binney Street Cambridge, MA 02139 United States. E.B. is co-founder of Venthera; PI of the clinical trial NCT04589650 (Novartis) and Advisor for Novartis. B.V. is a consultant for Pharming (Leiden, The Netherlands) and iOnctura (Geneva, Switzerland) and a shareholder of Open Orphan (Dublin, Ireland). Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work, the author(s) used Grammarly to proofread English grammar and ChatGPT to address doubts regarding English usage. After using this tool, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication. Material and methods Mice The in vivo experiments were performed in agreement with the guidelines and legislations of the Catalan Ministry of Agriculture, Livestock, Fisheries and Food (Catalonia, Spain), following protocols approved by the local Ethics Committees of the IGTP CEEAs. Mice were kept in individually ventilated cages under specific pathogen-free conditions. All mice were crossed onto the C57BL/6J genetic background. The following mouse strains were used in this study: Pdgfb-CreER T2 -EGFP (pan-endothelial Cre line) ( Claxton et al. 2008 ); Cdh5(PAC)CreER T2 (pan-endothelial Cre line) ( Sörensen, Adams, and Gossler 2009 ), Bmx(PAC)CreER T2 (arterial-endothelial Cre line) ( Ehling et al. 2013 ) and Esm1(BAC)CreER T2 (tip-endothelial Cre line) ( Rocha et al. 2014 ), all obtained from Ralf Adams; R26-mTmG ( Muzumdar et al. 2007 ) that express membrane Tomato in the absence of Cre activity and switch to membrane GFP expression upon Cre activation, allowing for the visualization and tracking of specific cell lineages; the endogenous Pik3ca tm1.1Waph/+ mice, carry a germline Cre-inducible point mutation (H1047R) in one of its endogenous Pik3ca alleles ( Kinross et al. 2012 ). LoxP sites flank exon 20 of the Pik3ca gene, and upon Cre recombination, this exon is replaced with a downstream version of exon 20 that contains a CAT to AGG alteration at codon 1047; the newly developed Rosa26 CAG-LSL-H2B-Cherry/H2B-GFP-2A-HAHuPIK3CAH1047R mice (further referred to as iFluMosaic-PIK3CA H1047R ), which overexpressed the human PIK3CA H1047R cDNA together with the expression of a H2B-EGFP-V5 (details in next section and Figure S4); and the Nr2f2 flox/flox mouse line ( Takamoto et al. 2005 ) kindly provided by Sophia Y. Tsai, that was used to delete Nr2f2 . To induce CreER T2 activity in pups, 4-OHT (Sigma, H7905) was dissolved in ethanol to prepare a stock solution at a concentration of 10 mg/mL. The solution was aliquoted and stored at-20°C. Pups received intragastric injections of the corresponding volume, adjusted to deliver the doses specified in the figures. To induce mosaic Pik3ca H1047R expression in the postnatal endothelium, Pik3ca H1047R /+ mice were mated with Pdgfb-CreERT2 T/+ R26-mTmG T/T mice. Pups were injected once at P1 with 0.5 mg/kg or 0.05 mg/kg 4-OHT and analysed at P4 or P6 respectively. To specifically target mature arteries, mice carrying the alleles Pik3ca H1047R /+ , Bmx-CreERT2 T /+ and R26-mTmG T/T were mated. The offspring were injected with 10 mg/kg of 4-OHT once at P1, P4, or P7, and subsequently analyzed at P4, P9, P12, or P21, depending on the experimental setup, as detailed in the text and figure legends. Additionally, mice carrying the iFluMosaic-PIK3CA T/T and Bmx-CreERT2 T/+ alleles were interbred, and the pups were injected daily from P7 to P9 with a higher dose of 4-OHT (50 mg/kg) due to the reduced recombination efficiency of this allele. To achieve tip EC-specific Pik3ca overactivation, mice carrying the alleles Pik3ca H1047R /+ , Esm1-CreERT2 T/+ , and R26-mTmG T/T were interbred. Pups were injected with 10 mg/kg of 4-OHT at P1, and retinal analysis was performed at P2, P6, P8, or P21. Mice carrying the R26-iFluMosaic-PIK3CA T/T and Esm1-CreERT2 T/+ alleles were mated, and the offspring were injected with a higher dose of 4-OHT (50 mg/kg) at P1 and P2 due to the reduced recombination efficiency of this allele. For Nr2f2 deletion specifically in tip ECs to perform rescue experiments, Nr2f2 flox/flox mice were interbred with Pik3ca H1047R /+ , Esm1-CreERT2 T/+ and R26 - mTmG T/T . Pups were injected with 25 mg/kg of 4-OHT at P1 and P2, followed by analysis of the vasculature at P6. Also, mice carrying Nr2f2 flox/flox allele were mated with iFluMosaic-PIK3CA T/T and Esm1-CreERT2 T/+ . In this case, mutant offspring were compared with the one resulted from the breeding of iFluMosaic-PIK3CA T/T and Esm1-CreERT2 T/+ mice alone. In both cases, progeny was injected with 50 mg/kg of 4-OHT at P1 and retina analysis was done at P8. To evaluate COUP-TFII protein expression in the mouse retina, offspring resulting from the cross between Nr2f2 flox/flox , Pik3ca H1047R /+ and Esm1-CreERT2 T/+ mice were injected with 25 mg/kg of 4-OHT at P1 and P2, and retinal analysis was performed at P6. Cre-negative mice served as the control group. COUP-TFII staining worked optimally with a secondary antibody conjugated to a fluorophore excited in the 488 nm range, so the R26-mTmG alelle was excluded from this cross due to fluorescence compatibility issues. All mouse lines and primer sequences required for genotyping are provided in Table S1 . Generation of the new iFluMosaic-PIK3CA H1047R mouse line To generate the iFluMosaic-PIK3CA H1047R (Rosa26 CAG-LSL-H2B-Cherry/H2B-GFP-2A- HAHuPIK3CAH1047R ) mouse line, we introduced the DNA insert containing the H2B-EGFP-V5-2A-HA-humanPIK3CA H1047R gene, by CRISPR/Cas9-mediated homologous-dependent repair (HDR), in the previously targeted iChr2-Mosaic mouse embryonic stem (ES) cell line ( Pontes-Quero et al. 2017 ). To generate the donor construct of interest, we first obtained the HA-huPIK3CA H1047R sequence from the plasmid pBabe-puro-HA-huPIK3CA H1047R (Addgene #12524) and cloned it into plasmid SO107 (Addgene #99617)( Pontes-Quero et al. 2017 ), which contained the LoxP2-H2B-EGFP-V5-2A cassette. We then inserted the H2B-EGFP-V5-2A-HA-huPIK3CA H1047R cassette into the donor vector JP6 (Addgene #99627), which contains the hygromycin resistance gene. This resulted in the generation of the donor vector NA453, containing the complete donor construct of interest ( Extended Data Fig. 4 ). The ES cells have the G4 background ( George et al. 2007 ) and were cultured in standard ES cell media consisting of DMEM supplemented with Glutamax (31966-047, Gibco), 15% fetal bovine serum (FBS, tested for germline transmission), 1x non-essential amino acids (NEAA, Hyclone, SH3023801), 0.1% ß-mercaptoethanol (Sigma, M7522), 1x penicillin-streptomycin (Pen/Strep, Lonza, DE17-602E), and leukemia inhibitory factor (LIF). The ES cells were maintained in dishes coated with a feeder layer of mouse embryonic fibroblasts (MEFs). For each nucleofection we resuspended 2.5 million ES cells in 100 µl volume containing 1.5 µg of circular px330 plasmid (NA505) including the guide RNA sequence (Guide3, GTTGCCTATGAGAGGCTAGAC) and 3.5 µg of the circular donor plasmid (NA453) containing the H2B-EGFP-V5-2A-HA-huPIK3CA H1047R sequence and the PGK-Hygromycin resistance cassette ( Extended Data Fig. 4 ). After nucleofection we plated 5 µl or 30 µl of the mix in two different wells on a 6-well plate with MEFs. 7 days after hygromycin selection, 24 isolated ES cell colonies were picked for storage and further screening. PCR with the flanking primers allowed us to identify ES cell clones with precise homologous recombination and insertion. After the identification of clones with precise gene targeting, we further validated the functionality by transfecting the ES cells with a Cre-expressing plasmid. Positive ES cell clones were expanded and subsequently used for microinjection into host blastocysts from the C57Bl/6J strain. Chimeric mice displaying a high percentage of agouti coat color were then bred to achieve germline transmission of the targeted insertion. Mouse retina isolation and whole-mount immunostaining Mice were sacrificed by decapitation, and their eyes were carefully isolated, and incubated on ice in 4% paraformaldehyde (PFA; Sigma, 158127) in PBS for 1 h. Retinas were then dissected and fixed in 4% PFA for an additional hour on ice. After three washes with PBS, retinas were incubated overnight at 4°C in blocking buffer (1% BSA, 0.3% Triton X-100 in PBS). For COUP-TFII staining, retinas were boiled for 15 minutes in 0.01 M citrate buffer (pH 6, Sigma-Aldrich, W305600) before permeabilization. Retinas were then incubated overnight at 4°C with specific primary antibodies diluted in blocking buffer. Primary antibodies used included: anti-GFP (Abcam, ab13970; 1:100), anti-GFP (Origene, R1091P; 1:500), anti-GFP AlexaFluor 488 (ThermoFisher Scientific, A-21311; 1:200), anti-RFP CF594 (BIOTIUM, 20422; 1:200), anti-p-S6 Ser235/236 (Cell Signaling, 4857; 1:100), anti-Ki67 (Invitrogen, 14-5698-82; 1:200), anti-αSMA Cy3 (Sigma-Aldrich, C6198; 1:200), anti-Erg (Abcam, AB92513; 1:100), and anti-COUP-TFII (R&D Systems, PP-H7147-00; 1:100). After washing 3 times with 0.1% Tween 20 in PBS (PBST), retinas were incubated for 30 minutes at room temperature in PBLEC buffer [1% Triton X-100, 1 mM CaCl2, 1 mM MgCl2, and 1 mM MnCl2 in PBS (pH 6.8)]. The following secondary AlexaFluor-conjugated antibodies (1:300, ThermoFisher Scientific: A11001, A11011, A11039, A21094, A21039, A31573, S32351) were added to the retinas in PBLEC for 2 h at room temperature. Blood vessels were visualized using AlexaFluor-conjugated Isolectin GS-B4 (IB4, ThermoFisher Scientific, I21412, I32450) incubated together with the secondary antibodies. After three additional washes with PBST, retinas were flat-mounted on microscope slides with Mowiol (Calbiochem, 475904), combined with anti-fade reagent DABCO (1.4-Diazabicyclo-(2.2.2)octane; Sigma, D27802), or Fluoromount-G (SouthernBiotech, 0100-01). Isolation and culture of mouse lung endothelial cells Mouse endothelial cells were isolated from the lungs of adult mice (both females and males) aged 3 to 6 weeks. Under sterile conditions in a cell culture hood, lungs were minced with a scalpel and digested in 4 U/ml dispase II (Roche, 04942078001) in Hanks’ Balanced Salt Solution (HBSS; HBSS GIBCO14170-112) for 1 h with constant agitation at 37°C. The digested tissue was dissociated by pipetting to create a single-cell suspension, and the enzyme was inactivated with Dulbecco’s Modified Eagle’s Medium (DMEM; Corning, 10-013-CV) supplemented with 10% fetal bovine serum (FBS; South America, S1810-500) and 1% penicillin-streptomycin (GIBCO, 15140-122). Cells were resuspended in PBS and incubated with rat anti-CD144/Cdh5 (BD Biosciences, 55289) antibody-coated magnetic beads for 30 minutes at room temperature. CD144-positive cells were washed with PBS containing 0.5% BSA and sorted using a magnetic separator. The sorted cells were resuspended and cultured in 12-well plates coated with 0.5% gelatin (Sigma, G1890). Cells were maintained in F12/DMEM medium (PromoCell, #C30140) supplemented with 20% fetal bovine serum (FBS), 1% penicillin-streptomycin, and 4 mL endothelial cell growth supplement/heparin (ECGF/H; Promocell C-30120) until they reached 80–90% confluency. A second selection with CD144 antibody-coated magnetic beads was performed for 1 h at room temperature. Cells were trypsinized, magnet-sorted, resuspended in F12 complete medium, and cultured further in a 6-well format. Cells were maintained at 37°C in a 5% CO2 atmosphere and used for experiments up to passage 5. Primary cells were not tested for mycoplasma contamination. To induce CreER T2 -mediated recombination in vitro for Western blot or qPCR studies, ECs were treated with 2 μM 4-OHT for 6 h, while ethanol-treated cells served as controls. Recombined cells were re-seeded for experiments and cultured at 37°C in a 5% CO2 atmosphere. Immunofluorescence of cultured endothelial cells Mouse lung ECs isolated from Cdh5(PAC)CreER T2 ; iFluMosaic-PIK3CA H1047R mice were expanded until passage 3. ECs were treated with 5 μM of 4-OHT or ethanol (control) for 6 h, followed by a 72 h incubation. A day prior to the assay, cells were split and seeded at low confluency onto gelatin-coated coverslips in 12-well plates. Four hours before collecting samples at each time point, the media was switched to starvation media. Then, cells were washed with warm 1% PBS containing Mg²⁺ and Ca²⁺ (GIBCO, 14080-055), fixed for 15 minutes in 4% PFA, and washed again with warm PBS. For staining, cells were permeabilized for 5 minutes with 0.4% Triton X-100 in PBS, followed by a 1 h incubation in blocking buffer (2% BSA in PBS). After washing with PBS, cells were incubated for 1 h at room temperature with primary antibodies diluted in blocking buffer. The primary antibodies used were anti-Ki67 (Invitrogen, 14-5698-82; 1:100), anti-p-S6 Ser235/236 (Cell Signaling, 4857; 1:100), anti-GFP (Abcam, ab13970; 1:100), and anti-RFP CF594 (BIOTIUM, 20422; 1:200). Subsequently, secondary AlexaFluor-conjugated antibodies were diluted 1:300 in blocking buffer and added to the cells for a 1 h incubation at room temperature (ThermoFisher Scientific: A11039, A21094, A21039). After additional washes with PBS, cells were incubated with DAPI (Invitrogen, D1305; 1:500) for 1 minute and then mounted on microscope slides with Fluoromount-G (Southern Biotech, 0100-01). Isolation, culture and sequencing of ECs from patient-derived vascular malformations Human ECs were isolated from patient biopsies of vascular malformations. These biopsies were obtained during therapeutic surgical resection, with informed consent and approval from the Biomedical Committees at Hospital Sant Joan de Deu, Hospital Santa Creu i Sant Pau, and Hospital Universitari de Bellvitge (codes PR264/16 and PIC-96-16). Experiments adhered to the WMA Declaration of Helsinki and the Belmont Report. Data were stored in a secure database maintained by Hospital Sant Joan de Deu. Biopsies were homogenized and digested in dispase II (Roche, 04942078001) and collagenase A (Roche Diagnostics, 10103586001) for up to 1.5 h at 37°C, with continuous movement. The tissue was dissociated by pipetting into a single-cell solution, followed by enzyme inactivation with DMEM (10% FBS, 1% penicillin-streptomycin). Cells were resuspended in 0.5% BSA in PBS and incubated with CD31 antibody (Agilent Dako, M0823, clone JC70A)-coated magnetic beads (ThermoFisher Scientific, 11041) for 1 h at room temperature. The CD31+ fraction was magnetically sorted, resuspended, and cultured in 0.5% gelatin-coated wells in EGM2 medium (PromoCell, C30140) supplemented with 10% FBS, 1% penicillin-streptomycin (EGM2 complete) at 37°C and 5% CO₂ until confluency. Cells were then subjected to a second selection. Mycoplasma testing was not performed on primary cells. To validate the presence of the mutation in the isolated ECs, genomic DNA was isolated according to the manufacturer’s protocol (ThermoFisher Scientific, K182001) and detected by NGS. For the vascular malformation (VM) #64 we obtained from 1572 reads, c.3140 A>G (H1047R) with a 45.86% variant allelic frequency (VAF) and for the VM90 from 461 reads, c.2740 A>G (E545G) with 48.37% VAF. Protein extraction, immunoprecipitation and immunoblotting Cells were lysed in ice-cold lysis buffer containing 50 mM Tris-HCl (pH 7.4), 5 mM EDTA, 150 mM NaCl, and 1% Triton X-100, supplemented with protease (Roche, 11836153001) and phosphatase (Sigma-Aldrich, 4906837001) inhibitors. The protein concentration of the supernatants was measured using the Pierce BCA Protein Assay Kit (ThermoFisher Scientific, 23225) following the manufacturer’s instructions. Total cell lysates were resolved on 10% SDS-polyacrylamide gels and then transferred onto nitrocellulose membranes. The membranes were blocked with 5% milk in TBST (TBS buffer with 0.1% Tween 20) and incubated with appropriate primary antibodies diluted in 2% BSA in TBST. The following primary antibodies were used: anti-p-AKT Ser473 (Cell Signaling Technology, 4060; 1:1,000), anti-AKT (Cell Signaling Technology, 9272; 1:2,000), anti-COUP-TFII (R&D Systems, PP-H7147-00; 1:500), anti-VE-cadherin (Santa Cruz Biotechnology, sc-6458; 1:500), and anti-Vinculin (Sigma-Aldrich, 9131; 1:10,000). Secondary antibodies from DAKO were diluted in 5% milk in TBST (all at 1:5,000): swine anti-rabbit (P0399), rabbit anti-mouse (P0260), rabbit anti-goat (P0449), and rabbit anti-rat (P0450). cDNA synthesis and quantitative polymerase chain reaction RNA was extracted using the Maxwell RSC SimplyRNA Cells Kit (Promega, AS1390) following the manufacturer’s instructions. The quality and quantity of the extracted RNA were assessed using a NanoDrop spectrometer. Reverse transcription was performed using 500 ng of RNA from cell lysate samples with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814). For quantitative polymerase chain reaction (qPCR), the QuantStudio 7 Flex system was utilized with SYBR Green real-time PCR master mix. The following primers were used for amplification: for Nr2f2 in mouse, the forward primer was 5′-GCAAGTGGAGAAGCTCAAGG-3’ and the reverse primer was 5′-TTCCAAAGCACACTGGGACT-3’. For NR2F2 in human samples, the forward primer was 5′-TAGTCCTGTTCACCTCAGATGCC-3’ and the reverse primer was 5′-CAGTTAAAACTGCTGCCGGAC-3’. To normalize gene expression, the following genes were used in mouse and human samples, respectively. The ml32 gene was amplified using the forward primer 5′-ACCCCAGAGGCATTGACAAC-3’ and the reverse primer 5′-ATTGTGGACCAGGAACTTGC-3’, while RPLP0 was amplified with the forward primer 5′-AGCCCAGAACACTGGTCTC-3’ and the reverse primer 5′-ACTCAGGATTTCAATGGTGCC-3’. RNA sequencing and analysis The RNA sequencing data from Kobialka et al. ( Kobialka et al. 2022 ) with public accession number PRJNA780473 (BioProject, NCBI) was reanalyzed. The counts matrix was retrieved and normalization performed using DEseq2 R package (version 1.40.2, ( Love, Huber, and Anders 2014 )). One biological replicate (ethanol and 4-OHT treated samples) was removed for downstream analysis to decrease variability and noise. Differential gene expression was also analyzed using DESeq2, with 0.05 as the significance cut-off, and the obtained P-values were corrected for multiple testing using the Benjamini and Hochberg method. The heatmap of selected genes was generated using ComplexHeatmaps (version 2.16.0 ( Gu, Eils, and Schlesner 2016 ; Gu 2022 )). Imaging analysis and quantification Imaging was performed with Leica Stellaris 8, Leica TCS SP5, and Zeiss LSM900 confocal microscopes. Adobe Photoshop, Adobe Illustrator, ImageJ, CellProfiler 4.2.6 and Imaris 10.1 softwares were used for image editing and quantification. All images shown in the figures are maximum intensity projections. Images were taken from at least three retinal areas in each retina and are representative of at least four retinas analyzed for each genotype unless otherwise stated. The identification of the retina vascular lesions and their location was carried out based on the observation of IB4 in 10X images. Veins, arteries, and capillaries were identified by morphology as previously described in Pitulescu et al ( Pitulescu et al. 2010 ). Vascular malformation incidence was quantified as the percentage of capillary beds, arterial and veins vessels exhibiting overgrowth and displayed a pathological phenotype. Retina vascularity was measured using the IB4 channel by adjusting the threshold to select the IB4+, followed by quantification of the percentage of IB4+ area in the total retinal surface. The width of arterial vessels was determined by measuring the average diameter at three different points along each artery. GFP+ cell expansion was quantified by calculating the percentage of GFP+ area within the IB4+ area. For capillary analysis, images containing only capillary beds were used. For the quantification of GFP+ cells in veins and arteries, an IB4+ area threshold was applied to identify and manually select the vessel-specific regions of interest (ROI). For the quantification of aEC-GFP+ ECs, the analyzed area was defined by a fixed ROI (200 x 630 μm at P9 and 500 x 1200 μm at P12 retinas) to maintain consistent arterial length across replicates and avoid the inclusion of other vascular types. In P21 retinas, aEC-GFP+ cells were spread across various vessel types, so the ROI for quantification included the entire vascular plexus. To quantify vascular p-S6 intensity a manual threshold was set to obtain the IB4+ area and define the ROI. Then, the integrated density of p-S6 was measured within IB4+ area. In the case of the iFluMosaic-PIK3CA H1047R , p-S6 intensity within wild-type or PIK3CA H1047R populated areas was measured by manually determining IB4+ areas containing each recombinant cell type (wild-type or PIK3CA H1047R cell clusters) and measuring the integrated density of p-S6 within ROI. In both cases, the background measurements (mean grey values) were taken from areas near the vasculature but negative for IB4. The corrected total fluorescence (CTF) was calculated using the following equation: CTF = integrated density – (vascular area x mean grey background value). EC proliferation was quantified by calculating the percentage of EC nuclei co-immunostained for both Ki67 and ERG, relative to the total number of ERG+ nuclei. COUP-TFII nuclear mean intensity was quantified through co-immunostaining for COUP-TFII and ERG. The ERG+ area was designated as the ROI, where the mean intensity of COUP-TFII was specifically measured. This mean intensity was then normalized to ERG intensity. Quantification of Cherry-Wild-type and EGFP- PIK3CA H1047R ECs in iFluMosaic-PIK3CA H1047R retinas was performed manually, differentiating their presence in arteries, veins, and capillaries. Lesions were classified based on size as small (35 PIK3CA H1047R cells), through the identification and manual counting of mutant cells. The incidence of arteries populated with PIK3CA H1047R cells was determined as a percentage in relation to the total number of arteries containing any recombinant cell. Quantification of EC proliferation within Cherry-Wild-type and EGFP- PIK3CA H1047R cells was achieved by assessing the rate of EC nuclei immunostained for both Ki67 and either Cherry or EGFP, relative to the total number of Cherry or EGFP, respectively. Arterial vessel width in iFluMosaic-PIK3CA H1047R retinas was assessed by comparing the diameter measurement of a central area containing recombinant cells with the diameter measurement within the same vessel in a surrounding region without any recombined cell. In cultured endothelial cells, maximum intensity projections of confocal images were acquired. Image analysis was performed with CellProfiler 4.2.6. DAPI, EGFP, Cherry and Ki67 signals were used by the software as ‘primary objects’. A ‘secondary object’ dependent on DAPI was created for the quantification of pS6 (nuclei identification). CellProfiler associated pS6 signal to each nucleus and classified each cell as positive or negative for pS6. The number of wild-type and PIK3CA H1047R ECs were quantified by cross-comparison of Cherry-DAPI+ and EGFP-DAPI+ cells. Single-cell RNA sequencing and data processing Mouse lung ECs isolated from Pdgfb-CreER T2 - Pik3ca H1047R /+ mice were treated in vitro overnight with 2 μM 4-OHT to induce recombination and expression of Pik3ca H1047R while ethanol-treated ones were used as a control (wild-type cells for Pik3ca ). The medium was then changed to F12 complete, and cells were collected 48 h post-induction. Cells were detached with trypsin, centrifuged, the pellet was subjected to a debris removal treatment (Miltenyi Biotec, 130-109-398), and finally cells were recollected in 0,04% BSA in PBS. For the single-cell RNA sequencing, three independent biological replicates were pooled for each treatment condition. For scRNA-seq experiments, the single cells were encapsulated in emulsion droplets using the Chromium Controller instrument using the kit 3’ Next GEM (10X Genomics). scRNA-seq libraries were prepared according to the manufacturer’s instructions. The aimed target cell recovery for each port was 10,000 cells. The generated libraries were sequenced in a NovaSeq6000 S4 (Illumina) using 2 × 150 bp paired-end read. The sequencing results for Pik3ca H1047R (4-OHT-treated) and wild-type (ethanol-treated) samples were aligned and quantified against the mouse reference genome mm10 using the Cell Ranger software (version 7.0.1) with default parameters. The output from Cell Ranger and the count matrices were read using the Read10X function from the Seurat library (version 4.9.9). Next, the established Seurat pipeline ( Hao et al. 2021 ) was followed to analyze the samples, which were initially processed independently and later combined. As part of our data analysis, we removed genes with no reads and low-quality cells. Specifically, we excluded cells with: (i) fewer than 250 total genes, (ii) mitochondrial transcript fraction greater than 20%, and (iii) a complexity index greater than 0.8. Our quality control criteria were based on protocols from https://hbctraining.github.io/scRNA-seq/lessons/04_SC_quality_control.html and the Tonsil Atlas ( Massoni-Badosa et al. 2024 ). To estimate doublets, which are pairs of cells sequenced under the same cellular barcode typically captured in the same droplet, we used the DoubletFinder (version 2.0.3) ( McGinnis, Murrow, and Gartner 2019 ) package in R. This tool simulates artificial doublets using existing data and compares them with actual cells to identify potential doublets. Moreover, the DoubletFinder pipeline is integrated into the Seurat pipeline and calculates doublets per sample. After filtering, a total of 7514 cells were retained from the Pik3ca H1047R sample (4OHT-pEC- Pik3ca H1047R ) and 6443 cells from the wild-type sample (ethanol-pEC- Pik3ca WT ). For normalization, we applied the NormalizeData function with the LogNormalize method and a scale factor of 10,000. This process involves dividing the raw gene counts of each cell by the total counts for that cell, multiplying by the scale factor, and then performing log-normalization as log(1+x). To select the number of highly variable genes (HVG) for further analysis, we first calculated the distribution of non-zero counts across all genes using the RNA assay data. This was achieved by applying the colSums function from the Matrix package (v 1.6.0) ( Bates, Maechler, and Jagan 2000 ). We then identified the number of HVG by adding 100 to the third quartile of this distribution, establishing a threshold for variability selection. Following normalization, we performed a z-score transformation on the normalized values using Seurat’s ScaleData with default parameters, followed by principal component analysis using RunPCA. Finally, clustering analysis was performed based on the edge weights between any two cells, utilizing a shared nearest-neighbor graph produced by the Louvain algorithm, implemented in Seurat’s FindNeighbors and FindClusters functions. Louvain clustering was performed at resolution 0.3 and clusters were annotated based on known tip, venous, arterial, and proliferative markers as described in the main text. Taking advantage of the PCA dimensionality reduction performed as part of the pre-processing pipeline, cell types were represented in PCA dimensions 1 and 2 to assess the degree of similarity between different clusters. Using this representation, we conducted a differential gene expression analysis employing the Wilcoxon test to compare Pik3ca H1047R (4OHT-pEC- Pik3ca H1047R ) with the wild-type sample (ethanol-pEC- Pik3ca WT ) within the Tip/Arterial-Venous cluster (C1/C2). This analysis identified the genes differentially expressed in the Pik3ca H1047R ECs. Statistically significant genes (Bonferroni-adjusted p-value < 0.05) with positive or negative log2FC values were selected as the set of significantly regulated genes. Violin plots of the selected genes were generated using the VlnPlot function from the Seurat package, with the split.by argument applied to divide the average expression of the genes. Heatmaps of gene expression were created using the ComplexHeatmap library in R ( Gu, Eils, and Schlesner 2016 ). The heatmap values represented the scaled and normalized expression levels of the genes within the Tip/Arterial-Venous cell cluster (C1/C2). Code availability No new algorithms were developed for this article. All original code has been deposited at GitHub ( https://github.com/anemartinezlarrinaga2898/Sabata_Rocat_et_al.git ). Data availability The scRNA-seq expression data supporting is uploaded to the Gene Expression Omnibus (GEO) repository under the accession ID GSE287780. RNA-seq data was obtained from the BioProject repository (NCBI) under the accession ID PRJNA780473 ( https://www.ncbi.nlm.nih.gov/bioproject/PRJNA780473 ) ( Kobialka et al. 2022 ). Statistics Statistical analysis was performed using Prism 10 (GraphPad Software Inc.). All data are displayed with individual data points that indicate biological replicates and presented as mean ± s.d. or mean ± s.e.m. (as indicated in the figure legends). At least four biological replicates were used. Statistical significance between two groups was assessed using either an unpaired two-tailed Student’s t-test or the nonparametric Mann-Whitney test. For comparisons involving more than two groups, one-way and two-way ANOVA followed by Tukey’s post-hoc test for multiple comparisons were used. No statistical method was employed to predetermine sample sizes; sample sizes were based on prior experiments. Reproducibility was confirmed through multiple independent experiments. The experiments were not randomized, and the investigators were not blinded during the experiments or data analysis. Antibodies anti-GFP - Abcam, ab13970; 1:100 (IF) anti-GFP - Origene, R1091P; 1:500 (IF) anti-GFP AlexaFluor 488 - ThermoFisher Scientific A-21311; 1:200 (IF) anti-RFP CF594 - BIOTIUM, 20422; 1:200 (IF) anti-p-S6 Ser 235/236 - Cell Signalling, 4857; 1:100 (IF) anti-Ki67 - Invitrogen, 14-5698-82; 1:200 (IF) anti-αSMA Cy3 - Sigma-Aldrich, C6198; 1:200 (IF) anti-Erg - Abcam, AB92513; 1:100 (IF) anti-COUP-TFII - R&D Systems, PP-H7147-00; 1:100 (IF) 1:500 (WB) anti-p-AKT Ser 473 - Cell Signaling Technology, 4060; 1:1,000 (WB) anti-AKT - Cell Signaling Technology, 9272; 1:2,000 (WB) anti-VE-cadherin - Santa Cruz Biotechnology, sc-6458; 1:500 (WB) anti-Vinculin - Sigma-Aldrich, 9131, 1:10,000 (WB) Extended Data Figures Download figure Open in new tab Extended Data Fig. 1. Phenotype overview and assessment of EC proliferation in Pik3ca H1047R retinas related to Figure 1 (a) Schematic illustration of the transgenic mouse strategy combining the pan-endothelial (pEC) tamoxifen-inducible Cre ( Pdgfb-CreER T2 ) allele with Pik3ca H1047R allele ( pEC-GFP-Pik3ca H1047R ). (b) Representative confocal images of control (Cre-) and pEC-Pik3ca H1047R P6 retinas. IB4 staining shows that only veins (orange arrowhead) and capillaries, but not arteries (red asterisks) develop vascular abnormalities upon Pik3ca H1047R expression. (c) High-magnification confocal images of arteries (red asterisks), vein (orange arrowhead), and capillary plexus (right panel) from P6 pEC-Pik3ca H1047R retinas stained with IB4. (d) Representative confocal images of pEC-GFP and pEC-GFP-Pik3ca H1047R P4 retinas stained with IB4 (red, blood vessels) and GFP (cyan, tracing recombined ECs). Retinas were isolated at P4, 3 days after 0.5 mg/kg 4-OHT injection at P1. (e) High-magnification confocal images of P4 retinas stained for IB4 (red, blood vessels), GFP (cyan, recombined cells), ERG (blue, EC nuclei), and Ki67 (yellow, proliferative cells). At this stage, pEC-GFP-Pik3ca H1047R retinas show initial vascular lesions. (f) Quantification of EC proliferation within the GFP+ area, represented as the percentage of ERG + Ki67+ nuclei per total number of ERG+ nuclei (n ≥ 8 retinas per genotype). Data are presented as mean ± s.d. Statistical analysis was performed using nonparametric Mann-Whitney test. Scale bars: 150 µm (b, d), 30 µm (c) and 20 µm (e). Download figure Open in new tab Extended Data Fig. 2. Tracing Bmx + ECs at different time points and impact of Pik3ca H1047R expression in αSMA staining related to Figure 2 (a) Schematic illustration of the transgenic mouse strategy combining the arterial EC-specific ( aEC ) tamoxifen-inducible Cre ( Bmx(PAC) -CreER T2 ) allele with the R26 - mTmG reporter allele ( aEC-GFP ). (b) Representative confocal images of aEC-GFP retinas after 10 mg/kg of 4-OHT treatment at P1, P4 or P7, followed by retina isolation and analysis at P6, P9 or P12, respectively. IB4 (red, blood vessels) and GFP (cyan, recombined cells) immunostaining show no GFP+ ECs when induced at P1, indicating a lack of Bmx expression at this stage. In contrast, GFP+ cells are observed after induction at P4 and P7. (c) High-magnification images of arteries from aEC-GFP and aEC-GFP - Pik3ca H1047R P9 (left) and P12 (right) retinas stained with IB4 (red, blood vessels), anti-GFP (cyan, recombined cells) and anti-αSMA (white, marker of smooth muscle cells). αSMA staining is used to identify arteries and assess their structural integrity. Scale bars: 150 µm (b) and 30 µm (c). Download figure Open in new tab Extended Data Fig. 3. Assessment of EC proliferation and long-term clonal dynamics of tip ECs related to Figure 3 (a) Representative confocal images of P4 tEC-GFP and tEC-GFP-Pik3ca H1047R retinas treated with 10 mg/kg 4-OHT at P1, stained for IB4 (red, blood vessels) and GFP (cyan, recombined cells). At P4, incipient vascular malformations are detected. (b) High-magnification confocal images of P4 retinas stained for IB4 (red, blood vessels), GFP (cyan, recombined cells), ERG (blue, EC nuclei), and Ki67 (yellow, proliferative cells). (c) Quantification of EC proliferation within the GFP+ area in tEC-GFP and tEC-GFP-Pik3ca H1047R P4 retinas, represented as the percentage of ERG+Ki67+ nuclei per total number of ERG+ nuclei (n ≥ 5 retinas per genotype). (d) Representative confocal images of P21 tEC-GFP and tEC-GFP-Pik3ca H1047R retinas treated with 10 mg/kg 4-OHT at P1, stained for IB4 (red, blood vessels) and GFP (cyan, recombined cells). At P21, the long-term behavior of P1 Esm1 -derived progeny can be assessed. (e) High-magnification confocal images of capillaries, veins and arteries in P21 retinas stained for IB4 (red, blood vessels) and GFP (cyan, recombined cells). (f) Quantification of vessel width (top) and percentage of GFP+ area (bottom) in tEC-GFP and tEC-GFP - Pik3ca H1047R P21 retinas, specifically in capillaries, veins, and arteries (n ≥ 4 retinas per genotype). (g) Quantification of the percentage of GFP+ area in tEC-GFP and tEC-GFP - Pik3ca H1047R P6, P8 and P21 retinas, specifically in capillaries, veins, and arteries. Data previously presented in Fig. 3f and Extended Data Fig. 3f , now represented differently to illustrate temporal changes. Data are presented as mean ± s.d. Statistical analysis was performed using nonparametric Mann-Whitney test (c, f) and 2-way ANOVA with Sidak’s correction for multiple comparisons (g). ns, non-statistical. Scale bars: 150 µm (a, d) and 30 µm (b, e). Download figure Open in new tab Extended Data Fig. 4. Summary of the steps performed to generate the new Rosa26-iFluorescent-Mosaic-PIK3CA H1047R mouse model, related to Figure 4 All details in M&M. PGK-hygro-pA, resistance marker for selection; WPRE-pA, viral sequence to enhance the stability and efficiency of transgene expression by improving the export of mRNA from the nucleus to the cytoplasm; CAG, Strong and ubiquitous promoter; PGK-Neo-pA, resistance marker for ES cell selection; 2A, viral peptide allowing equimolar expression of multiple independent proteins from a single ORF; HA, V5, and His are small epitopes that can be used for specific antibody detection; H2B, histone tag that targets proteins to the chromatin/nucleus; N-PhiM, non-fluorescent protein that is used as a reporter of promoter expression; INS, Insulators, are DNA sequences that regulate gene expression by preventing enhancers from activating nearby genes. Download figure Open in new tab Extended Data Fig. 5. In vitro validation of the iFluMosaic-PIK3CA H1047R model and representative images of iFluMosaic-PIK3CA H1047R mouse retinas stained for different markers, related to Figure 4 (a) Schematic illustration of the transgenic mouse strategy combining the pan EC-specific (pEC) tamoxifen-inducible Cre ( Cdh5(PAC)-CreER T2 ) allele with the iFluMosaic-PIK3CA H1047R allele. (b) Strategy for in vitro EC expansion, 4-OHT administration and analysis. 6 h 4-OHT treatment was followed by 72h cell expansion. Daily (every 24h) collections were preceded by 4h of complete serum starvation. (c, d) Representative confocal images of primary mouse lung ECs isolated from pEC-iFluMosaic-PIK3CA H1047R mice collected at 24, 48, and 72 h after seeding. Cells were stained for DAPI (nuclei, blue), EGFP ( PIK3CA H1047R expressing ECs, yellow), Cherry (wild-type ECs for Pik3ca , cyan), and Ki67 (proliferative cells, magenta) (c) or phospho (p)-S6 (Ser235/236) (grey) (d). (e, f, g) Quantification of Cherry-Wild-type (cyan) and EGFP- PIK3CA H1047R (yellow) recombined ECs at 24, 48, and 72 h post seeding (e). Shown as percentage of PIK3CA H1047R and wild-type cell sum. Within each population, quantification of the percentage of cells positive for Ki67 (f) or p-S6 (g) (n = 6 biological replicates). (h,i) High-magnification images of capillaries and arteries from P8 tEC-iFluMosaic-PIK3CA H1047R retinas showing immunodetection of IB4 (magenta), Cherry (cyan, wild-type ECs for Pik3ca ), EGFP (yellow, PIK3CA H1047R expressing ECs) and Ki67 (red, proliferative cells) (h) or p-S6 (white) (i). Data are presented as mean ± s.d. 2-way ANOVA statistical analysis was performed followed by Tukey test for multiple comparisons (e, f, g). Scale bars: 200 µm (c, d) and 40 µm (h, i). Download figure Open in new tab Extended Data Fig. 6. Extended analysis supporting Figure 5 (a) Heatmap representing Z-normalized gene expression levels of the top 5 most highly expressed markers of each cluster. (b) Feature plots showing the gene expression of known endothelial-specific marker genes and key markers of each cluster. (c) Dot plot showing the average expression of lymphatic markers for each cluster. (d) PCA plot of merged Pdgfb-CreER T2 (pEC)- Pik3ca H1047R mouse lung endothelial cells (ECs) 48 h after treatment with ethanol (vehicle, wild type for Pik3ca ) and 4-OHT (expressing Pik3ca H1047R ) derived from scRNA-seq data. Cells are colored based on clusters identified in the previous UMAP representation in Fig. 5a . C1 (tip-like ECs) and C2 (venous-like ECs) clusters which were found enriched in Pik3ca WT and Pik3ca H1047R ECs respectively, clustered together here (C1/ C2), allowing for the following comparisons between genotypes. (e) Volcano plot of genes analyzed by scRNAseq in C1 and C2 clusters upon Pik3ca H1047R expression. Differentially expressed genes (DEGs) (Padj < 0.05) in Pik3ca H1047R over Pik3ca WT ECs from C1 and C2 (grey, unchanged; blue, downregulated; red, upregulated). (f) Heatmap showing the gene expression of selected DEG per cells from C1/C2 cluster in Pik3ca WT and Pik3ca H1047R ECs. (g) Violoin plots of selected DEG between Pik3ca WT vs Pik3ca H1047R conditions in C1/C2 cluster. (h) Quantification of phopsho(p)-AKT (Ser473) protein levels normalized to total AKT levels (related to Figure 5G ) (n = 6 biological replicates). (i) Quantification of p-AKT (Ser473) protein levels normalized to total AKT levels (related to Figure 5J ) (n = 6 biological replicates). Data are presented as mean ± s.d. Statistical analysis was performed using an unpaired t-test (h) and 2-way ANOVA followed by Tukey test for multiple comparisons (i). Download figure Open in new tab Extended Data Fig. 7. Validation of effective Nr2f2 deletion and its impact on vascular area (a) Representative confocal images of control, tEC-GFP-Pik3ca H1047R and tEC-Pik3ca H1047R - Nr2f2 ΔEC/ΔEC P6 mouse retinas after 25 mg/kg 4-OHT induction at P1 and P2. Immunostaining for IB4 (white), ERG (blue, EC nuclei,), and COUP-TFII (yellow) showing upregulation of COUP-TFII in Pik3ca H1047R vessels. (b) Quantification of COUP-TFII intensity within ERG+ areas (EC nuclei) in capillary vessels from control, tEC-Pik3ca H1047R and tEC -Pik3ca H1047R - Nr2f2 ΔEC/ΔEC P6 retinas (n ≥ 6 retinas per genotype). (c) Representative confocal images of tEC-GFP , tEC-GFP-Nr2f2 ΔEC/ΔEC , tEC-GFP-Pik3ca H1047R , and tEC-GFP-Pik3ca H1047R - Nr2f2 ΔEC/ΔEC P6 mouse retinas, stained for IB4 (red, blood vessels) and GFP (cyan, P1 and P2 Esm1 -derived progeny). (d) Quantification of total vessel area (IB4+ area) comparing tEC-GFP , tEC-GFP-Nr2f2 ΔEC/ΔEC , tEC-GFP-Pik3ca H1047R , and tEC-GFP-Pik3ca H1047R - Nr2f2 ΔEC/ΔEC of P6 retinal vasculature (n ≥ 9 retinas per genotype). (e) Representative confocal images of tEC-iFluMosaic-PIK3CA H1047R and tEC-iFluMosaic-PIK3CA H1047R -Nr2f2 ΔEC/ΔEC P8 retinas stained for IB4 (magenta, blood vessels), Cherry (cyan, wild-type ECs for Pik3ca ) and EGFP (yellow, PIK3CA H1047R expressing ECs) at the capillary front. (f) Quantification of the total number of Cherry-Wild-type and EGFP- PIK3CA H1047R ECs in tEC-iFluMosaic-PIK3CA and tEC-iFluMosaic-PIK3CA-Nr2f2 ΔEC/ΔEC P8 retinas (n = 16 retinas per genotype). Data are presented as mean ± s.d. Statistical analysis was performed using one-way ANOVA, followed by Tukey test for multiple comparisons (b, d) and 2-way ANOVA with Sidak’s correction for multiple comparisons (f). Scale bars: 20 µm (a), 500 µm (c), and 50 µm (e). View this table: View inline View popup Download powerpoint Table S1. Mouse lines and genotying primers Acknowledgments We thank members of the Endothelial Pathobiology and Microenvironment Group for helpful discussions. We thank the Microscopy Core Facility and Single Cell Unit at the IJC for their support. We thank CERCA Program/Generalitat de Catalunya and the Josep Carreras Foundation for institutional support. The research leading to these results has received funding by the Spanish Ministry of Science and Innovation MICINN (PID2020-116184RB-I00 /AEI/10.13038/501100011033) and from “la Caixa” Banking Foundation under the project code LCF/PR/HR19/52160023 (also, to E.B.). The M.G. laboratory is also supported by PTEN RESEARCH Foundation (IJC-21-001); la Caixa Banking Foundation (LCF/PR/HR22/52420010 and LCF/PR/HR23/52430009); by la Asociación Española Contra el Cancer (AECC)-Grupos Traslacionales (GCTRA18006CARR); by la Fundación BBVA (Ayuda Fundación BBVA a Equipos de Investigación Científica 2019); World Cancer Research (21-0159). A.A-U. was a recipient of a fellowship from the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement no. 101026227. L.G. was funded by the Swedish Research Council (2018-06591). H.S. is supported by the grant PRE2018-084283-MCIN/AEI and FSE, and J.D. is supported by the grant PRE2021-099260-MCIN/AEI/10.13039/501100011033 and FSE. S.D.C. was funded by la Caixa Banking Foundation Junior Leader project (LCF/BQ/PR20/11770002). R.B. was supported by the European Research Council (ERC) Consolidator Grant AngioUnrestUHD (101001814), the Ministerio de Ciencia e Innovación (PID2020-120252RB-I00), and “la Caixa” Banking Foundation (HR19-00120). M.D-L. was supported by a PhD fellowship from CNIC Severo Ochoa program (CX_E-2015-01). The CNIC is supported by the Instituto de Salud Carlos III (ISCIII), the Ministerio de Ciencia e Innovación (MCIN), and the Pro CNIC Foundation and is a Severo Ochoa Center of Excellence (grant CEX2020-001041-S funded by MICIN/AEI/10.13039/501100011033). M.A. acknowledges funding from the Wellcome Trust and The Royal Society (105942/Z/14/A). E.B. is funded by the Agencia Estatal de Investigación (Proyectos de investigación en salud PI20/00102). The authors thank the Xarxa de Bancs de Tumors de Catalunya (XBTC; sponsored by Pla Director d’Oncologia de Catalunya). We are grateful to the Band of Parents at Hospital Sant Joan de Déu for supporting the overall research activities of the Developmental Tumor laboratory (Pediatric Cancer Center Barcelona). References ↵ Ang , Lay Teng , Alana T. Nguyen , Kevin J. Liu , Angela Chen , Xiaochen Xiong , Matthew Curtis , Renata M. Martin , et al. 2022 . ‘ Generating Human Artery and Vein Cells from Pluripotent Stem Cells Highlights the Arterial Tropism of Nipah and Hendra Viruses ’. Cell 185 ( 14 ): 2523 – 2541 .e30. doi: 10.1016/j.cell.2022.05.024 . OpenUrl CrossRef PubMed ↵ Angulo-Urarte , Ana , and Mariona Graupera . 2022 . ‘ When, Where and Which PIK3CA Mutations Are Pathogenic in Congenital Disorders ’. Nature Cardiovascular Research 2022 1:8 1 ( 8 ): 700–714. doi: 10.1038/S44161-022-00107-8 . OpenUrl CrossRef ↵ Aranguren , Xabier L. , Manu Beerens , Giulia Coppiello , Cornelia Wiese , Ine Vandersmissen , Antonio Lo Nigro , Catherine M. Verfaillie , Manfred Gessler , and Aernout Luttun . 2013 . ‘ COUP-TFII Orchestrates Venous and Lymphatic Endothelial Identity by Homo-or Hetero-Dimerisation with PROX1 ’. Journal of Cell Science 126 ( 5 ): 1164 – 75 . doi: 10.1242/JCS.116293/263400/AM/COUP-TFII-ORCHESTRATES-VENOUS-AND-LYMPHATIC . OpenUrl Abstract / FREE Full Text ↵ Bates , Douglas , Martin Maechler , and Mikael Jagan . 2000 . ‘ Matrix: Sparse and Dense Matrix Classes and Methods ’. CRAN: Contributed Packages . doi: 10.32614/CRAN.package.Matrix . OpenUrl CrossRef ↵ Blanco , Raquel , and Holger Gerhardt . 2013 . ‘ VEGF and Notch in Tip and Stalk Cell Selection ’. Cold Spring Harbor Perspectives in Medicine 3 ( 1 ): 1 – 20 . doi: 10.1101/cshperspect.a006569 . OpenUrl CrossRef ↵ Castel , Pau , Katherine A. Rauen , and Frank McCormick . 2020 . ‘ The Duality of Human Oncoproteins: Drivers of Cancer and Congenital Disorders ’. Nature Reviews Cancer . Nature Publishing Group . doi: 10.1038/s41568-020-0256-z . OpenUrl CrossRef ↵ Chavkin , Nicholas W. , Gael Genet , Mathilde Poulet , Erin D. Jeffery , Corina Marziano , Nafiisha Genet , Hema Vasavada , et al. 2022 . ‘ Endothelial Cell Cycle State Determines Propensity for Arterial-Venous Fate ’. Nature Communications 13 ( 1 ): 5891 . doi: 10.1038/s41467-022-33324-7 . OpenUrl CrossRef PubMed ↵ Chen , Xinpu , Jun Qin , Chiang-Min Cheng , Ming-Jer Tsai , and Sophia Y. Tsai . 2012 . ‘ COUP-TFII Is a Major Regulator of Cell Cycle and Notch Signaling Pathways ’. Molecular Endocrinology 26 ( 8 ): 1268 – 77 . doi: 10.1210/ME.2011-1305 . OpenUrl CrossRef PubMed ↵ Chu , Man , Taotao Li , Bin Shen , Xudong Cao , Haoyu Zhong , Luqing Zhang , Fei Zhou , et al. 2016 . ‘ Angiopoietin Receptor Tie2 Is Required for Vein Specification and Maintenance via Regulating COUP-TFII ’. ELife 5 ( December ). doi: 10.7554/eLife.21032 . OpenUrl CrossRef ↵ Ciwinska , Marta , Hendrik A Messal , Hristina R Hristova , Catrin Lutz , Laura Bornes , Theofilos Chalkiadakis , Rolf Harkes , et al. 2024 . ‘ Mechanisms That Clear Mutations Drive Field Cancerization in Mammary Tissue ’. Nature 633 ( 8028 ): 198 – 206 . doi: 10.1038/s41586-024-07882-3 . OpenUrl CrossRef ↵ Claxton , Suzanne , Vassiliki Kostourou , Shalini Jadeja , Pierre Chambon , Kairbaan Hodivala-Dilke , and Marcus Fruttiger . 2008 . ‘ Efficient, Inducible Cre-Recombinase Activation in Vascular Endothelium ’. Genesis 46 ( 2 ): 74 – 80 . doi: 10.1002/dvg.20367 . OpenUrl CrossRef PubMed Web of Science ↵ Colom , B. , A. Herms , M. W.J. Hall , S. C. Dentro , C. King , R. K. Sood , M. P. Alcolea , et al. 2021 . ‘ Mutant Clones in Normal Epithelium Outcompete and Eliminate Emerging Tumours ’. Nature 2021 598 : 7881 598 (7881): 510–14. doi: 10.1038/s41586-021-03965-7 . OpenUrl CrossRef ↵ Ehling , Manuel , Susanne Adams , Rui Benedito , and Ralf H. Adams . 2013 . ‘ Notch Controls Retinal Blood Vessel Maturation and Quiescence ’. Development (Cambridge ) 140 ( 14 ): 3051 – 61 . doi: 10.1242/dev.093351 . OpenUrl Abstract / FREE Full Text ↵ George , Sophia H. L. , Marina Gertsenstein , Kristina Vintersten , Ella Korets-Smith , John Murphy , Mary E. Stevens , Jody J. Haigh , and Andras Nagy . 2007 . ‘ Developmental and Adult Phenotyping Directly from Mutant Embryonic Stem Cells ’. Proceedings of the National Academy of Sciences 104 ( 11 ): 4455 – 60 . doi: 10.1073/pnas.0609277104 . OpenUrl Abstract / FREE Full Text ↵ Graupera , Mariona , Julie Guillermet-Guibert , Lazaros C. Foukas , Li Kun Phng , Robert J. Cain , Ashreena Salpekar , Wayne Pearce , et al. 2008 . ‘ Angiogenesis Selectively Requires the P110α Isoform of PI3K to Control Endothelial Cell Migration ’. Nature 2008 453 : 7195 453 (7195): 662–66. doi: 10.1038/NATURE06892 . OpenUrl CrossRef ↵ Gu , Zuguang . 2022 . ‘ Complex Heatmap Visualization ’. IMeta 1 ( 3 ): 1 – 15 . doi: 10.1002/imt2.43 . OpenUrl CrossRef ↵ Gu , Zuguang , Roland Eils , and Matthias Schlesner . 2016 . ‘ Complex Heatmaps Reveal Patterns and Correlations in Multidimensional Genomic Data ’. Bioinformatics 32 ( 18 ): 2847 – 49 . doi: 10.1093/bioinformatics/btw313 . OpenUrl CrossRef PubMed ↵ Hao , Yuhan , Stephanie Hao , Erica Andersen-Nissen , William M. Mauck , Shiwei Zheng , Andrew Butler , Maddie J. Lee , et al. 2021 . ‘ Integrated Analysis of Multimodal Single-Cell Data ’. Cell 184 ( 13 ): 3573 – 3587 .e29. doi: 10.1016/j.cell.2021.04.048 . OpenUrl CrossRef PubMed ↵ Hassanein , Aladdin H. , John B. Mulliken , Steven J. Fishman , Ahmad I. Alomari , David Zurakowski , and Arin K. Greene . 2012 . ‘ Venous Malformation: Risk of Progression during Childhood and Adolescence ’. Annals of Plastic Surgery 68 ( 2 ): 198 – 201 . doi: 10.1097/SAP.0b013e31821453c8 . OpenUrl CrossRef PubMed ↵ Herms , Albert , Bartomeu Colom , Gabriel Piedrafita , Argyro Kalogeropoulou , Ujjwal Banerjee , Charlotte King , Emilie Abby , et al. 2024 . ‘ Organismal Metabolism Regulates the Expansion of Oncogenic PIK3CA Mutant Clones in Normal Esophagus ’. Nature Genetics 2024 , August, 1–14. doi: 10.1038/s41588-024-01891-8 . OpenUrl CrossRef ↵ Herms , Albert , and Philip H Jones . 2023 . ‘ Somatic Mutations in Normal Tissues: New Perspectives on Early Carcinogenesis ’. Annual Review of Cancer Biology . doi: 10.1146/annurev-cancerbio-061421-012447 . OpenUrl CrossRef ↵ Hoxhaj , Gerta , and Brendan D. Manning . 2020 . ‘ The PI3K–AKT Network at the Interface of Oncogenic Signalling and Cancer Metabolism ’. Nature Reviews Cancer 20 ( 2 ): 74 – 88 . doi: 10.1038/s41568-019-0216-7 . OpenUrl CrossRef PubMed ↵ International Society for the Study of Vascular Anomalies . 2018 . ‘ ISSVA Classification for Vascular Anomalies - 2018 ’. 20th ISSVA Workshop . http://www.issva.org/classification . ↵ Keymeulen , Alexandra Van , May Yin Lee , Marielle Ousset , Sylvain Brohée , Sandrine Rorive , Rajshekhar R. Giraddi , Aline Wuidart , et al. 2015 . ‘ Reactivation of Multipotency by Oncogenic PIK3CA Induces Breast Tumour Heterogeneity ’. Nature 525 ( 7567 ): 119 – 23 . doi: 10.1038/nature14665 . OpenUrl CrossRef PubMed ↵ Kinross , Kathryn M. , Karen G. Montgomery , Margarete Kleinschmidt , Paul Waring , Ivan Ivetac , Anjali Tikoo , Mirette Saad , et al. 2012 . ‘ An Activating Pik3ca Mutation Coupled with Pten Loss Is Sufficient to Initiate Ovarian Tumorigenesis in Mice ’. The Journal of Clinical Investigation 122 ( 2 ): 553 – 57 . doi: 10.1172/JCI59309 . OpenUrl CrossRef PubMed Web of Science ↵ Kobialka , Piotr , Helena Sabata , Odena Vilalta , Leonor Gouveia , Ana Angulo-Urarte , Laia Muixí , Jasmina Zanoncello , et al. 2022 . ‘ The Onset of PI3K-related Vascular Malformations Occurs during Angiogenesis and Is Prevented by the AKT Inhibitor Miransertib ’. EMBO Molecular Medicine 14 ( 7 ): e15619 . doi: 10.15252/emmm.202115619 . OpenUrl CrossRef ↵ Koren , Shany , Linsey Reavie , Joana Pinto Couto , Duvini De Silva , Michael B. Stadler , Tim Roloff , Adrian Britschgi , et al. 2015 . ‘ PIK3CAH1047R Induces Multipotency and Multi-Lineage Mammary Tumours ’. Nature 525 (7567): 114 – 18 . doi: 10.1038/nature14669 . OpenUrl CrossRef PubMed ↵ Love , Michael I. , Wolfgang Huber , and Simon Anders . 2014 . ‘ Moderated Estimation of Fold Change and Dispersion for RNA-Seq Data with DESeq2 ’. Genome Biology 15 ( 12 ): 1 – 21 . doi: 10.1186/s13059-014-0550-8 . OpenUrl CrossRef PubMed ↵ Luo , Wen , Irene Garcia-Gonzalez , Macarena Fernández-Chacón , Verónica Casquero- Garcia , Maria S. Sanchez-Muñoz , Severin Mühleder , Lourdes Garcia-Ortega , Jorge Andrade , Michael Potente , and Rui Benedito . 2020 . ‘ Arterialization Requires the Timely Suppression of Cell Growth ’. Nature 2020 589 : 7842 589 (7842): 437–41. doi: 10.1038/S41586-020-3018-X . OpenUrl CrossRef ↵ Madsen , Ralitsa R. , Rachel G. Knox , Wayne Pearce , Saioa Lopez , Betania Mahler- Araujo , Nicholas McGranahan , Bart Vanhaesebroeck , and Robert K. Semple . 2019 . ‘ Oncogenic PIK3CA Promotes Cellular Stemness in an Allele Dose-Dependent Manner ’. Proceedings of the National Academy of Sciences of the United States of America 116 ( 17 ): 8380 – 89 . doi: 10.1073/PNAS.1821093116 . OpenUrl Abstract / FREE Full Text ↵ Madsen , Ralitsa R. , Bart Vanhaesebroeck , and Robert K. Semple . 2018 . ‘ Cancer-Associated PIK3CA Mutations in Overgrowth Disorders ’. Trends in Molecular Medicine. Elsevier Current Trends . doi: 10.1016/j.molmed.2018.08.003 . OpenUrl CrossRef PubMed ↵ Manning , Brendan D. , and Alex Toker . 2017 . ‘ AKT/PKB Signaling: Navigating the Network ’. Cell 169 ( 3 ): 381 – 405 . doi: 10.1016/J.CELL.2017.04.001 . OpenUrl CrossRef PubMed ↵ Martincorena , Iñigo , Joanna C. Fowler , Agnieszka Wabik , Andrew R.J. Lawson , Federico Abascal , Michael W.J. Hall , Alex Cagan , et al. 2018 . ‘ Somatic Mutant Clones Colonize the Human Esophagus with Age ’. Science 362 ( 6417 ): 911 – 17 . doi: 10.1126/science.aau3879 . OpenUrl Abstract / FREE Full Text ↵ Martinez-Corral , Ines , Yan Zhang , Milena Petkova , Henrik Ortsäter , Sofie Sjöberg , Sandra D. Castillo , Pascal Brouillard , et al. 2020 . ‘ Blockade of VEGF-C Signaling Inhibits Lymphatic Malformations Driven by Oncogenic PIK3CA Mutation ’. Nature Communications 11 ( 1 ). doi: 10.1038/s41467-020-16496-y . OpenUrl CrossRef PubMed ↵ Massoni-Badosa , Ramon , Sergio Aguilar-Fernández , Juan C. Nieto , Paula Soler-Vila , Marc Elosua-Bayes , Domenica Marchese , Marta Kulis , et al. 2024 . ‘ An Atlas of Cells in the Human Tonsil ’. Immunity 57 ( 2 ): 379 – 399 .e18. doi: 10.1016/j.immuni.2024.01.006 . OpenUrl CrossRef PubMed ↵ Mauri , Federico , Corentin Schepkens , Gaëlle Lapouge , Benjamin Drogat , Yura Song , Ievgenia Pastushenko , Sandrine Rorive , et al. 2021 . ‘ NR2F2 Controls Malignant Squamous Cell Carcinoma State by Promoting Stemness and Invasion and Repressing Differentiation ’. Nature Cancer 2 ( 11 ): 1152 – 69 . doi: 10.1038/s43018-021-00287-5 . OpenUrl CrossRef PubMed ↵ McCracken , Ian R. , Andrew H. Baker , Nicola Smart , and Sarah De Val . 2023 . ‘ Transcriptional Regulators of Arterial and Venous Identity in the Developing Mammalian Embryo ’. Current Opinion in Physiology 35 : 100691 . doi: 10.1016/j.cophys.2023.100691 . OpenUrl CrossRef ↵ McGinnis , Christopher S. , Lyndsay M. Murrow , and Zev J. Gartner . 2019 . ‘ DoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors ’. Cell Systems 8 ( 4 ): 329 – 337 .e4. doi: 10.1016/j.cels.2019.03.003 . OpenUrl CrossRef PubMed ↵ Muzumdar , Mandar Deepak , Bosiljka Tasic , Kazunari Miyamichi , Ng Li , and Liqun Luo . 2007 . ‘ A Global Double-Fluorescent Cre Reporter Mouse ’. Genesis (United States ) 45 ( 9 ): 593 – 605 . doi: 10.1002/dvg.20335 . OpenUrl CrossRef PubMed Web of Science ↵ Neal , Alice , Svanhild Nornes , Sophie Payne , Marsha D. Wallace , Martin Fritzsche , Pakavarin Louphrasitthiphol , Robert N. Wilkinson , et al. 2019 . ‘ Venous Identity Requires BMP Signalling through ALK3 ’. Nature Communications 10 ( 1 ): 1 – 18 . doi: 10.1038/s41467-019-08315-w . OpenUrl CrossRef PubMed ↵ Nussinov , Ruth , Chung Jung Tsai , and Hyunbum Jang . 2022 . ‘How Can Same-Gene Mutations Promote Both Cancer and Developmental Disorders ?’ Science Advances 8 ( 2 ): 1–11. doi: 10.1126/sciadv.abm2059 . OpenUrl CrossRef PubMed ↵ Orsenigo , Fabrizio , Lei Liu Conze , Suvi Jauhiainen , Monica Corada , Francesca Lazzaroni , Matteo Malinverno , Veronica Sundell , et al. 2020 . ‘ Mapping Endothelial-Cell Diversity in Cerebral Cavernous Malformations at Single-Cell Resolution ’. ELife 9 ( October ): 1 – 34 . doi: 10.7554/ELIFE.61413 . OpenUrl CrossRef PubMed ↵ Park , Hyojin , Jessica Furtado , Mathilde Poulet , Minhwan Chung , Sanguk Yun , Sungwoon Lee , William C. Sessa , Claudio A. Franco , Martin A. Schwartz , and Anne Eichmann . 2021 . ‘ Defective Flow-Migration Coupling Causes Arteriovenous Malformations in Hereditary Hemorrhagic Telangiectasia ’. Circulation 144 ( 10 ): 805 – 22 . doi: 10.1161/CIRCULATIONAHA.120.053047 . OpenUrl CrossRef PubMed ↵ Payne , Sophie , Alice Neal , and Sarah De Val . 2024 . ‘ Transcription Factors Regulating Vasculogenesis and Angiogenesis ’. Developmental Dynamics 253 ( 1 ): 28 – 58 . doi: 10.1002/dvdy.575 . OpenUrl CrossRef ↵ Petkova , Milena , Marle Kraft , Simon Stritt , Ines Martinez-Corral , Henrik Ortsäter , Michael Vanlandewijck , Bojana Jakic , et al. 2023 . ‘ Immune-Interacting Lymphatic Endothelial Subtype at Capillary Terminals Drives Lymphatic Malformation ’. Journal of Experimental Medicine 220 ( 4 ). doi: 10.1084/JEM.20220741 . OpenUrl CrossRef ↵ Pitulescu , Mara E. , Inga Schmidt , Rui Benedito , and Ralf H. Adams . 2010 . ‘ Inducible Gene Targeting in the Neonatal Vasculature and Analysis of Retinal Angiogenesis in Mice ’. Nature Protocols 5 ( 9 ): 1518 – 34 . doi: 10.1038/nprot.2010.113 . OpenUrl CrossRef PubMed ↵ Pitulescu , Mara E. , Inga Schmidt , Benedetto Daniele Giaimo , Tobiah Antoine , Frank Berkenfeld , Francesca Ferrante , Hongryeol Park , et al. 2017 . ‘ Dll4 and Notch Signalling Couples Sprouting Angiogenesis and Artery Formation ’. Nature Cell Biology 19 ( 8 ): 915 – 27 . doi: 10.1038/ncb3555 . OpenUrl CrossRef PubMed ↵ Pontes-Quero , Samuel , Luis Heredia , Verónica Casquero-García , Macarena Fernández-Chacón , Wen Luo , Ana Hermoso , Mayank Bansal , et al. 2017 . ‘ Dual IfgMosaic: A Versatile Method for Multispectral and Combinatorial Mosaic Gene-Function Analysis ’. Cell 170 ( 4 ): 800 – 814 .e18. doi: 10.1016/j.cell.2017.07.031 . OpenUrl CrossRef PubMed ↵ Potente , Michael , Holger Gerhardt , and Peter Carmeliet . 2011 . ‘ Basic and Therapeutic Aspects of Angiogenesis ’. Cell 146 ( 6 ): 873 – 87 . doi: 10.1016/j.cell.2011.08.039 . OpenUrl CrossRef PubMed Web of Science ↵ Qin , Jun , San Pin Wu , Chad J. Creighton , Fangyan Dai , Xin Xie , Chiang Min Cheng , Anna Frolov , et al. 2013 . ‘ COUP-TFII Inhibits TGF-β-Induced Growth Barrier to Promote Prostate Tumorigenesis ’. Nature 493 ( 7431 ): 236 – 40 . doi: 10.1038/nature11674 . OpenUrl CrossRef PubMed Web of Science ↵ Rocha , Susana F. , Maria Schiller , Ding Jing , Hang Li , Stefan Butz , Dietmar Vestweber , Daniel Biljes , et al. 2014 . ‘ Esm1 Modulates Endothelial Tip Cell Behavior and Vascular Permeability by Enhancing VEGF Bioavailability ’. Circulation Research 115 ( 6 ): 581 – 90 . doi: 10.1161/CIRCRESAHA.115.304718 . OpenUrl Abstract / FREE Full Text ↵ Samuels , Yardena , Zhenghe Wang , Alberto Bardelli , Natalie Silliman , Janine Ptak , Steve Szabo , Hai Yan , et al. 2004 . ‘ High Frequency of Mutations of the PIK3CA Gene in Human Cancers ’. Science 304 ( 5670 ): 554 . doi: 10.1126/SCIENCE.1096502/SUPPL_FILE/SAMUELS.SOM.PDF . OpenUrl FREE Full Text ↵ Selvam , Senthil , Tejas Kumar , and Marcus Fruttiger . 2018 . ‘ Retinal Vasculature Development in Health and Disease ’. Progress in Retinal and Eye Research 63 (November 2017): 1–19. doi: 10.1016/j.preteyeres.2017.11.001 . OpenUrl CrossRef PubMed ↵ Shah , Aarti V. , Graeme M. Birdsey , and Anna M. Randi . 2016 . ‘ Regulation of Endothelial Homeostasis, Vascular Development and Angiogenesis by the Transcription Factor ERG ’. Vascular Pharmacology 86 : 3 – 13 . doi: 10.1016/j.vph.2016.05.003 . OpenUrl CrossRef PubMed ↵ Sörensen , Inga , Ralf H. Adams , and Achim Gossler . 2009 . ‘ DLL1-Mediated Notch Activation Regulates Endothelial Identity in Mouse Fetal Arteries ’. Blood 113 ( 22 ): 5680–88. doi: 10.1182/blood-2008-08-174508 . OpenUrl Abstract / FREE Full Text ↵ Stewen , Jonas , Kai Kruse , Anca T. Godoi-Filip , Zenia , Hyun-Woo Jeong , Susanne Adams , Frank Berkenfeld , et al. 2024 . ‘ Eph-Ephrin Signaling Couples Endothelial Cell Sorting and Arterial Specification ’. Nature Communications 15 ( 1 ): 2539 . doi: 10.1038/s41467-024-46300-0 . OpenUrl CrossRef PubMed ↵ Strittmatter , Brady G. , Travis J. Jerde , and Peter C. Hollenhorst . 2021 . ‘ Ras/ERK and PI3K/AKT Signaling Differentially Regulate Oncogenic ERG Mediated Transcription in Prostate Cells ’. PLoS Genetics 17 ( 7 ): 1 – 20 . doi: 10.1371/journal.pgen.1009708 . OpenUrl CrossRef ↵ Su , Tianying , Geoff Stanley , Rahul Sinha , Gaetano D’Amato , Soumya Das , Siyeon Rhee , Andrew H. Chang , et al. 2018 . ‘ Single-Cell Analysis of Early Progenitor Cells That Build Coronary Arteries ’. Nature 559 ( 7714 ): 356 – 62 . doi: 10.1038/s41586-018-0288-7 . OpenUrl CrossRef PubMed ↵ Swift , Matthew R. , Van N. Pham , Daniel Castranova , Kameha Bell , Richard J. Poole , and Brant M. Weinstein . 2014 . ‘ SoxF Factors and Notch Regulate Nr2f2 Gene Expression during Venous Differentiation in Zebrafish ’. Developmental Biology 390 ( 2 ): 116 – 25 . doi: 10.1016/J.YDBIO.2014.03.018 . OpenUrl CrossRef PubMed ↵ Takamoto , Norio , Li Ru You , Kelvin Moses , Chin Chiang , Warren E. Zimmer , Robert J. Schwartz , Francesco J. DeMayo , Ming Jer Tsai , and Sophia Y. Tsai . 2005 . ‘ COUP-TFII Is Essential for Radial and Anteroposterior Patterning of the Stomach ’. Development 132 ( 9 ): 2179 – 89 . doi: 10.1242/dev.01808 . OpenUrl Abstract / FREE Full Text ↵ Vanhaesebroeck , Bart , Julie Guillermet-Guibert , Mariona Graupera , and Benoit Bilanges . 2010 . ‘ The Emerging Mechanisms of Isoform-Specific PI3K Signalling ’. Nature Reviews Molecular Cell Biology 2010 11:5 11 ( 5 ): 329 – 41 . doi: 10.1038/nrm2882 . OpenUrl CrossRef PubMed Web of Science ↵ Wu , San Pin , Cheng Tai Yu , Sophia Y. Tsai , and Ming Jer Tsai . 2016 . ‘ Choose Your Destiny: Make a Cell Fate Decision with COUP-TFII ’. The Journal of Steroid Biochemistry and Molecular Biology 157 ( March ): 7 – 12 . doi: 10.1016/J.JSBMB.2015.11.011 . OpenUrl CrossRef PubMed ↵ Xu , Cong , Sana S. Hasan , Inga Schmidt , Susana F. Rocha , Mara E. Pitulescu , Jeroen Bussmann , Dana Meyen , Erez Raz , Ralf H. Adams , and Arndt F. Siekmann . 2014 . ‘ Arteries Are Formed by Vein-Derived Endothelial Tip Cells ’. Nature Communications 2014 5:1 5 ( 1 ): 1–11. doi: 10.1038/ncomms6758 . OpenUrl CrossRef PubMed ↵ You , Li Ru , Fu Jung Lin , Christopher T. Lee , Francesco J. DeMayo , Ming Jer Tsai , and Sophia Y. Tsai . 2005 . ‘ Suppression of Notch Signalling by the COUP-TFII Transcription Factor Regulates Vein Identity ’. Nature 435 ( 7038 ): 98 – 104 . doi: 10.1038/nature03511 . OpenUrl CrossRef PubMed Web of Science ↵ Yum , Min Kyu , Seungmin Han , Juergen Fink , Szu Hsien Sam Wu , Catherine Dabrowska , Teodora Trendafilova , Roxana Mustata , et al. 2021 . ‘ Tracing Oncogene-Driven Remodelling of the Intestinal Stem Cell Niche ’. Nature 2021 594 :7863 594 ( 7863 ): 442–47. doi: 10.1038/s41586-021-03605-0 . OpenUrl CrossRef PubMed View the discussion thread. Back to top Previous Next Posted February 25, 2025. Download PDF Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Context-dependent response of endothelial cells to PIK3CA mutation Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Context-dependent response of endothelial cells to PIK3CA mutation Helena Sabata , Ariadna Roca-Coll , Jose A. Dengra , Leonor Gouveia , Ane Martinez-Larrinaga , Alberto Collado-Remacha , Sandra D. Castillo , Elena Castillo , Nathalie Tisch , Macarena De Andrés-Laguillo , Svanhild Nornes , Judith Llena , Pilar Villacampa , Bart Vanhaesebroeck , María P. Alcolea , Sarah De Val , Susana Lopez-Fernandez , Katrien De Bock , Eulalia Baselga , Rui Benedito , Ana Angulo-Urarte , Mariona Graupera bioRxiv 2025.02.25.640041; doi: https://doi.org/10.1101/2025.02.25.640041 Share This Article: Copy Citation Tools Context-dependent response of endothelial cells to PIK3CA mutation Helena Sabata , Ariadna Roca-Coll , Jose A. Dengra , Leonor Gouveia , Ane Martinez-Larrinaga , Alberto Collado-Remacha , Sandra D. Castillo , Elena Castillo , Nathalie Tisch , Macarena De Andrés-Laguillo , Svanhild Nornes , Judith Llena , Pilar Villacampa , Bart Vanhaesebroeck , María P. Alcolea , Sarah De Val , Susana Lopez-Fernandez , Katrien De Bock , Eulalia Baselga , Rui Benedito , Ana Angulo-Urarte , Mariona Graupera bioRxiv 2025.02.25.640041; doi: https://doi.org/10.1101/2025.02.25.640041 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Developmental Biology Subject Areas All Articles Animal Behavior and Cognition (7622) Biochemistry (17648) Bioengineering (13870) Bioinformatics (41880) Biophysics (21423) Cancer Biology (18553) Cell Biology (25458) Clinical Trials (138) Developmental Biology (13364) Ecology (19866) Epidemiology (2067) Evolutionary Biology (24290) Genetics (15589) Genomics (22475) Immunology (17711) Microbiology (40327) Molecular Biology (17145) Neuroscience (88472) Paleontology (666) Pathology (2826) Pharmacology and Toxicology (4815) Physiology (7635) Plant Biology (15114) Scientific Communication and Education (2044) Synthetic Biology (4286) Systems Biology (9815) Zoology (2268)