Induced pluripotent stem cell-derived vascular networks to screen nano-bio interactions

This jou nal is ©The Royal Socie y o Chemis y 2021 Nanoscale Ho iz. Ci e his: DOI: 10.1039/d0nh00550a Induced plu ipo en s em cell-de i ed ascula ne wo ks o sc een nano–bio in e ac ions† Luı ´s Es onca, a Vi o F ancisco, b Pa ı ´cia Pi ez, b Ine ˆs Hono ´ io, b La a Ca alho, c Helena Vaza ˜o, b Josephine Ble sch, b Akhilesh Rai, a Xa ie Nissan, d Ul ich Simon, e Ma ´ io G a ˜os, b Leono Sau ´de c and Lino Fe ei a * a The ascula bioac i i y/sa e y o nanoma e ials is ypically e alua ed by animal es ing, which is o low h oughpu and does no accoun o biological diffe ences be ween animals and humans such as ageing, me abolism and disease p o iles. The de elopmen o pe so- nalized human in i o pla o ms o e alua e he in e ac ion o nanoma e ials wi h he ascula sys em would be impo an o bo h he apeu ic and egene a i e medicine. A lib a y o 30 nanopa icle (NP) o mula ions, in use in imaging, an imic obial and pha maceu- ical applica ions, was e alua ed in a epo e zeb a ish model o asculogenesis and hen es ed in pe sonalized humanized models composed o human-induced plu ipo en s em cell (hiPSC)-de i ed endo helial cells (ECs) wi h ‘‘young’’ and ‘‘aged’’ pheno ypes in 3 ascula ne wo k o ma s: 2D (in polys y ene dish), 3D (in Ma igel) and in a blood essel on a chip. As a p oo o concep , ascula oxici y was used as he main eadou . The esul s show ha he oxici y p o ile o NPs o hiPSC-ECs was dependen on he ‘‘age’’ o he endo helial cells and ascula ne wo k o ma . hiPSC-ECs we e less suscep ible o he cy o oxici y effec o NPs when cul u ed in low han in s a ic condi ions, he p o ec i e effec being media ed, a leas in pa , by glycocalyx. O e all, he esul s p esen ed he e high- ligh he ele ance o in i o hiPSC-de i ed ascula sys ems o sc een ascula nanoma e ial in e ac ions. 1. In oduc ion Nanoma e ials and nanoma e ial-based echnologies enable he de elopmen o new ma e ials and applica ions ac oss disciplines, including mechanical and elec ical enginee ing, ag icul u e, ene gy gene a ion and medicine. 1–3 The in e - ac ions o hese nanoma e ials wi h he human body a e only pa ially known. 4–6 The de elopmen o new cell-based pla - o ms o he apid p o iling o nanoma e ials in e ms o bioac i i y, oxici y, and biodeg ada ion, among o he aspec s, is in g ea need. 7,8 The majo i y o he nanoma e ials ha en e he human body, independen o he ou e o en y, will ce ainly ci cula e and be anspo ed in he ascula sys em h ough he blood essels and he e o e i is c i ical o s udy hei impac in indi iduals wi h di e ences in hei ascula biology because o hei gene ic backg ound o pa hologies. The e ec s o nanoma e ials in he dis up ion o he in eg i y o endo helial cell–cell communica ion 9 and in he induc ion o endo helial 10 and smoo h muscle 11 cell oxici y ha e been iden i ied. S anda d p o ocols o assessing he effec o nanoma e ials on he ascula sys em in ol e es ing in animals. 5,10 Un o una ely, hese es s a e o low- h oughpu , and expensi e, yield limi ed a Facul y o Medicine, Uni e si y o Coimb a, 3000-548, Coimb a, Po ugal. E-mail: lino@uc-bio ech.p b Cen e o Neu oscience and Cell Biology, Uni e si y o Coimb a, Coimb a, Po ugal c Ins i u o de Medicina Molecula e Ins i u o de His ologia e Biologia do Desen ol imen o, Faculdade de Medicina da Uni e sidade de Lisboa, 1649-028, Lisboa, Po ugal d CECS, I-STEM, AFM, Ins i u e o S em Cell The apy and Explo a ion o Monogenic diseases, E y cedex, F ance e Ins i u e o Ino ganic Chemis y, RWTH Aachen Uni e si y, Ge many †Elec onic supplemen a y in o ma ion (ESI) a ailable. See DOI: 10.1039/d0nh00550a Recei ed 17 h Sep embe 2020, Accep ed 21s Decembe 2020 DOI: 10.1039/d0nh00550a sc.li/nanoscale-ho izons New concep s The exposu e o humans o en i onmen s wi h inc eased le els o nano- pa icles in he ai as well as o pha maceu ical nano o mula ions o egene a i e and he apeu ic medicine equi es a be e knowledge o hei bioac i i y/sa e y. In he pas , hese es s we e pe o med in low h oughpu in i o es s (e.g. mice) ha did no accoun o diffe ences wi h he human sys em and in non-pe sonalized human cells, i.e.,incells ha didno ha e pa ien -speci ic in o ma ion and ageing signa u es. In his s udy, we ha e de eloped pe sonalized ascula ne wo ks wi h a iable complexi y (2D, 3D, and a blood essel on a chip) o med by human induced plu ipo en s em cell-de i ed endo helial cells wi h a ‘‘young’’ o ‘‘aged’’ pheno ype, o sc een ascula –nanoma e ial in e ac ions. We ha e used a lib a y o 30 nanoma- e ials wi h di e en physico-chemical p ope ies and ele ance o clinical molecula imaging, p o ec i e o mula ions, an imic obial coa ings, ca alysis and pha maceu ical applica ions. The complexi y o he ascula ne wo k, and pa icula ly i s abili y o exp ess glycocalyx, as well as he age o he cells in luenced la gely hei sensi i i y o he nanoma e ials. The pla o m p esen ed in his s udy is e y p omising o high- h oughpu sc eening o nano–bio in e ac ions and o he iden i ica ion o nanoma e - ials able o elimina e mo e speci ically aged ascula cells. Nanoscale Ho izons COMMUNICATION Open Access A icle. Published on 12 Feb ua y 2021. Downloaded on 2/15/2021 11:13:43 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online View Jou nal Nanoscale Ho iz. This jou nal is ©The Royal Socie y o Chemis y 2021 mechanis ic in o ma ion, do no add ess he cu en policies o egula o y agencies o use al e na i es o animal es ing, and do no accoun o diffe ences be ween species. Mo eo e , wi h he ad en o pe sonalized medicine, new cell echnologies a e equi ed o p o ide pa ien -speci ic in o ma ion abou nanoma e ial bioac i i y/sa e y. In his espec , human-induced plu ipo en s em cells (hiPSCs) ep esen a po en ial sou ce o endo helial cells (ECs) and smoo h muscle cells (SMCs). 12–14 These humanized sys ems a e, in some aspec s, supe io o animal models because hey ecapi ula e human ageing, me abolism and disease p o iles. Un o una ely, hiPSC-de i ed ECs and SMCs ha e no been used o s udy he bioac i i y/sa e y o nanoma e ials among di e en indi iduals. These cells may be cul u ed unde low shea condi- ions in mic o luidic sys ems o be e mimic he in i o condi ions. In his wo k, he impac o nanoma e ials in iPSC-de i ed ascula ne wo ks was e alua ed in 2D, 3D and blood essel on a chip in i o models. While he i s wo models un in s a ic condi ions, he blood essel on a chip model uns in low condi ions, simila o in i o condi ions, which allows he o ma ion o a unc ional glycocalyx laye on op o endo helial cells. Thi y nanoma e ials ha a e no mally used o diffe en applica ions such as clinical molecula imaging, p o ec i e o mula ions, an imic obial coa ings, pha maceu ical o mula ions and ca alys s ha e been selec ed (Fig. 1). The nanoma e ials wi h a me allic o polyme ic composi ion and sizes anging be ween 1.4 and 400 nm we e ini ially es ed in he zeb a ish emb yo asculogenesis/angiogenesis o in es iga e hei in i o effec . Subsequen ly, he nanoma e ials we e es ed in he in i o human iPSC-de i ed ascula ne wo ks om wo indi iduals in he same ime ame as he zeb a ish es s (i.e. 24 h) o e alua e hei acu e oxici y. One o he iPSC lines was de i ed om e al cells (co d blood o a heal hy newbo n), while he o he was gene a ed om ib oblas s o a 14 yea -old emale pa ien wi h Hu chinson–Gil o d p oge ia synd ome (HGPS). 15 HGPS is a a e and a al disease caused by a single poin mu a ion o he LMNA gene leading o he accumula ion o an abno mally unca ed lamin A p o ein called p oge in, which in u n leads o accele a ed ageing. 16 The accumula ion o p oge in is also obse ed du ing physiological ageing, al hough a much lowe le els. 17 The mo i a ion he e was o in es iga e di e ences in he in e ac ion o nanoma e ials wi h ascula cellswi ha‘‘young’’and‘‘aged’’pheno ype.Asap oo o concep , we ha e selec ed ascula oxici y as he main eadou because i is ela i ely easy o quan i y; howe e , cellula NP in e naliza ion was also quan i ied in some expe imen s. We ha e used monocul u e o bo h iPSC-de i ed ECs ei he in polys y ene cul u e dishes (2D) o in Ma igel (3D) o sc een, in s a ic condi ions. Some o he oxic o mula ions we e u he es ed in a blood essel on a chip o med by a co-cul u e o iPSC-ECs and iPSC-SMCs. 2. Resul s Cha ac e iza ion o nanoma e ials We ha e selec ed 30 nanoma e ials, bo h o ganic and ino ganic, wi h diffe en sizes and ze a po en ials o e alua e in ou ascula sc eening pla o m (Table 1). We ocused on nano- ma e ials used o : (i) molecula imaging (NP4), (ii) sunsc een (NP15 and NP16), (iii) an imic obial applica ions (NP29, NP9 18 o NP12 19 ), (i ) pha maceu ical applica ions (NP20 and NP21, 20 NP24, 21 and ligh - igge able o mula ions ecen ly syn hesized by us such as NP30, NP23, NP26, NP27, NP18 (all desc ibed in e e ence e . 22), NP22, 23 NP14, 23 and NP25 24 ) and ( ) eac ion ca alysis o biomedical applica ions (NP1, NP2, NP3, NP5, NP6, NP7, NP8, o silica-based NPs such as NP13 and NP28). Many o hese NPs may ge inside o he human body and in e ac wi h blood essels in he ci cula ion. Dynamic ligh sca e ing (DLS) analyses showed ha he a e age sizes o he ino ganic nano- ma e ials suspended in PBS a ied be ween 1 (NP1) and 360 nm (NP29), while hose o o ganic nanoma e ials a ied be ween 32 (NP10) and 427 nm (NP30) (Fig. 2A.1 and A.2). Mos o he nanoma e ials main ained hei size a e 24 h in suspension wi h a ew excep ions (NP8 and NP16 sligh ly inc eased hei diame e s while NP28 dec eased signi ican ly hei diame e s, likely due o a poo ini ial dispe sion o he NPs as shown by Fig. 1 NP lib a y and in i o/in i o models o sc een hei oxici y. (A) P ope ies o he NPs. The NPs can be ca ego ized based on hei size as: ul asmall (3 NPs) o small (27 NPs); based on hei composi ion as: ino ganic (15 NPs) o o ganic (15 NPs); and by he ype o he coa ing: pep ides (3), polyme s (2), unc ional g oups (4) o an i-oxidan s (2). (B) In i o and in i o models o sc een NP oxici y. In i o model: 30 NPs we e es ed a 1 mgmL 1 in 4 hp zeb a ish emb yos (manually decho iona ed) o 24 h. The eadou was asculogenesis. 2D model (in cul u e pla es, s a ic condi ions): 30 NPs we e incuba ed a 4 diffe en concen a ions (6, 12.5, 25 and 50 mgmL 1 ) o 24 h wi h N-iPSC ECs o HGPS-iPSC ECs. The eadou s we e: (i) cell su i al and (ii) nanoma e ial in e naliza ion. 3D model (in Ma igel, s a ic condi ions): 30 NPs we e incuba ed a 2 concen a ions (12.5 and 25 mgmL 1 ) oge he wi h iPSC ECs in cul u e pla es wi h Ma igel o 24 h. The eadou s we e: (i) leng h o mic o essels o med, (ii) cell su i al and (iii) gene exp ession. Blood essel on a chip (mic o luidic sys em; low condi ions): 10 NPs we e es ed a 50 mgmL 1 in a e ial low condi ions in a co-cul u e o N-iPSC SMCs and N-iPSC ECs o 24 h. The eadou was cell su i al. Communica ion Nanoscale Ho izons Open Access A icle. Published on 12 Feb ua y 2021. Downloaded on 2/15/2021 11:13:43 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online This jou nal is ©The Royal Socie y o Chemis y 2021 Nanoscale Ho iz. TEM analyses, see below). The hyd odynamic diame e o he nanoma e ials anged be ween 1.1 and 427 nm. T ans- mission elec on mic oscopy (TEM) analyses we e pe o med o con i m he sizes and shapes o some o he nanoma e ials (Fig. S1, ESI†). To s udy he effec o sal s and p o eins on NP size and s abili y, he hyd odynamic diame e s o NPs (suspended a a concen a ion o 50 mgmL 1 ) was e alua ed by DLS in a cell cul u e medium wi h se um (2% e al bo ine se um (FBS); concen a ion o se um used o cul u e ECs) o 24 h (du a ion o mos in i o expe imen s) (Fig. 2A.3 and A.4). In he cell cul u e medium, wi h some excep ions (NP12, NP14, NP18, NP29 and NP30 inc eased hei diame e s, while NP16 and NP17 sligh ly dec eased hei diame e s), he NPs main ained hei diame e s o a leas 24 h. The hyd odynamic diame e s o he NPs anged be ween 1.1 and 357 nm. All he NPs showed a nega i e ze a po en ial in cell cul u e medium, which indica es ha p o eins and sal s ha e adso bed on hei su aces (Fig. 2B). The mos nega i ely cha ged NP was NP29. O e all, we ha e selec ed a lib a y o NPs ha is ele an o biomedical and en i onmen al applica ions. These NPs main ained a easonable le el o s abili y in he cell cul u e medium o a leas 24 h. Nanoma e ial sc eening in zeb a ish We e alua ed he effec o he nanoma e ials in asculogenesis and angiogenesis using he zeb a ish ansgenic Tg( li1a: EGFP) y1 line. 26 In his ansgenic line, he p omo e o he endo helial ma ke li1 d i es he exp ession o EGFP in blood essels, hus allowing he in i o analysis o he ascula u e. In zeb a ish emb yos, he ci cula ion begins a e 24 h-pos - e iliza ion (hp ) in a single ci cula o y loop. 27 Vasculogenesis o he do sal ao a, ca dinal ein and a p imi i e c anial ascula u e occu s a his s age. 28 A e he o ma ion o his p imi i e ascula u e, he o ma ion o new blood essels occu s ia angiogenesis, he in e segmen al essels (ISVs) o he unk being among he i s angiogenic essels o be o med. 29 The ISVs sp ou om he do sal ao a along he somi e bounda ies, which ul ima ely in e connec be ween each o he o o m he do sal longi udinal anas omo ic essel (DLAV) (Fig. S2A, ESI†). Ha ing his in mind, we e alua ed he effec o nanoma e ials in he ascula de elopmen o ISVs du ing he i s 24–28 hp . Fe ilized emb yos we e selec ed and manually decho iona ed o a oid any in e e ence o he cho ion in he in e naliza ion o he NPs. Indeed, p e ious s udies ha e demons a ed ha he p esence o he cho ion may a ec he oxici y o some NPs. 30 Fo example, Ag NPs caused highe mo ali y and mal o ma- ions in emb yos wi hou cho ion as compa ed wi h emb yos wi h cho ion a he same concen a ions. 30 In ou s udy, he decho iona ion caused highe sensi i i y o he emb yos o hei Table 1 Summa y o he NP sou ces and p ope ies NP numbe Co e Ligand Size (nm) Ze a (mV) Sou ce/ e . NP numbe Co e Ligand Size (nm) Ze a (mV) Sou ce/ e . NP1 Au MS 1.4 — 25 NP16 TiO 2 — 10–30 — ssnano.com NP2 Au GSH 1.5 — 25 NP17 PS COOH 100 — mic omod.de NP3 Au HA 5 1 a 25 2 a N.A. NP18 C11 — 65 514122 NP4 SPION DEX 40 31 biopal.com NP19 PS NH 2 100 — mic omod.de NP5 Au MS 13.9 — NP20 PLGA — 170 79320 NP6 Au GSH 12 — NP21 PLGA–PS — 218 97220 NP7 Au MS 15 1.5 — 25 NP22 P1C5 — 350 35 21 223 NP8 Au HA 20 3 a 26 3 a N.A. NP23 A9 — 265 23 21 222 NP9 Au LL37 21 8152 18 NP24 PEI:DS — 91 615321 NP10 PM COOH 25 — mic omod.de NP25 PEI–DMNC:DS — 108 10 27 224 NP11 PM NH2 25 — mic omod.de NP26 E1 — 330 24 7 122 NP12 Au CM 14 1282 19 NP27 E2 — 260 18 2222 NP13 SiO 2 — 7–14 — plasmachem.com NP28 SiO 2 — 10–20 — ssnano.com NP14 P1C7 — 74 10 19 1 23 NP29 Ag — o6000 — sciessen .com NP15 ZnO — 10–30 — ssnano.com NP30 A1 — 420 14 1122 a De e mined in wa e by DLS (n= 3). Ligands: MS – sodium 3-(diphenylphosphino)benzene sul ona e; GSH – glu a hione; HA – hyalu onic acid; DEX – dex an; LL37 – an imic obial pep ide; CM – cec opin meli in. Fig. 2 Physicochemical cha ac e iza ion o he NP lib a y. (A) NP sizes and coun s measu ed by dynamic ligh sca e ing a 0 and 24 h in PBS (A.1 and A.2) and EGM-2 medium (A.3 and A.4). (B) Ze a po en ials o NP lib a y a 0 h (in molecula biology g ade wa e ) and 24 h (in EGM-2 medium). A e 24 h in EGM-2, he NPs we e cen i uged and esuspended in H 2 O o ze a po en ial measu emen s. In A and B, esul s a e exp essed as mean SEM (n=3). Nanoscale Ho izons Communica ion Open Access A icle. Published on 12 Feb ua y 2021. Downloaded on 2/15/2021 11:13:43 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online Nanoscale Ho iz. This jou nal is ©The Royal Socie y o Chemis y 2021 en i onmen , so we used NPs a a concen a ion o 1 mgmL 1 . The numbe o ISVs, he a e age leng h o he i s 10 ISVs and he o al leng hs o he ISVs we e e alua ed (Fig. 3 and Fig. S2B, ESI†). Ou esul s showed ha some NPs had a nega i e impac on angiogenesis by dec easing he leng h o ISVs (NP1, NP3, NP4, NP7, NP8, NP14, NP17, NP18, NP21, NP24, NP25 and NP26) whe eas one nanoma e ial (NP9) had a posi i e impac on angiogenesis by inc easing he leng h and numbe o ISVs (al hough no s a is ically signi ican ela i e o con ol). The p o-angiogenic ac i i y o LL37–Au NPs is aligned wi h ou p e ious esul s showing ha hese NPs p omo ed neo ascula - iza ion in a wound healing animal model. 18 De elopmen o a ascula sc eening pla o m based on iPSC-de i ed ascula cells: de i a ion o endo helial cells om hiPSCs iPSC lines we e diffe en ia ed in o ECs acco ding o a p o ocol p e iously epo ed by us in e . 13 (Fig. 4A). Undiffe en ia ed cells we e cul u ed in chemically de ined medium supplemen ed wi h BMP4 and FGF-basic o 5 days o gene a e mesode m p ogeni o cells ollowed by hei diffe en ia ion in o endo helial p ogeni o cells (PECAM1 + cells) in a medium con aining VEGF and Tb4 o an addi ional 5 days. PECAM1 + cells we e hen selec ed using magne ism-ac i a ed cell so ing and cul u ed o up o 7 passages in an endo helial medium supplemen ed wi h VEGF 165 and TGF-binhibi o (SB431542). Gene exp ession analysis in cells diffe en ia ed o 7 passages (be ween 24 and 28 days a e cell seeding) indica ed ha bo h N-iPSC ECs ( om non-disease co d blood cells) and HGPS-iPSC ECs exp essed EC ma ke s such as PECAM1,CDH5, KDR and WF as human umbilical a e y ECs (HUAECs), al hough wi h some luc ua ions in he exp ession o hese ma ke s (Fig. 4B). In gene al, he mRNA le els o he EC ma ke s we e highe in HGPS-iPSC ECs han in N-iPSC ECs. Fig. 3 Impac o nanoma e ials on he emb yonic ascula de elopmen o zeb a ish emb yos. (A) Rep esen a i e luo escence mic oscopy images (all emb yos and magni ica ion o he i s 10 ISVs) o zeb a ish emb yos wi h B28 hp in con ol condi ions ( ehicle) and a e 24 h ea men wi h NP9 and NP14. Scale ba is 200 mm. (B.1) Impac o 24 h incuba ion o nanoma e ials on he a e age leng h o he i s 10 ISVs o zeb a ish emb yos wi h B28 hp . (B.2) Impac o 24 h incuba ion o nanoma e ials on he o al ISV leng h o zeb a ish emb yos wi h B28 hp . (B.3). Impac o 24 h incuba ion o nanoma e ials on he o al numbe o ISVs o zeb a ish emb yos wi h B28 hp . In B.1, B.2 and B.3, esul s a e mean SEM (n= 6–24) and s a is ical analyses we e pe o med by an unpai ed - es . *, **, and **** deno e s a is ical signi icance (Po0.05; Po0.01; Po0.0001) ela i e o con ol. Fig. 4 Cha ac e iza ion o iPSC-ECs. (A) P o ocol o he diffe en ia ion o N-iPSCs and HGPS-iPSCs in o EC cells. Undiffe en ia ed iPSCs we e ans e ed o Pe i dishes coa ed wi h ib onec in and we e cul u ed in a diffe en ia ion medium o a pe iod o 10 days (see Me hods sec ion o mo e de ails), a e which he PECAM1 + cells we e selec ed by magne ism- ac i a ed cell so ing and cul u ed wi h ull cul u e medium supplemen ed wi h VEGF 165 and SB431542 o 7 passages (B25 days) wi h he medium changed e e y 2–3 days. (B) qRT-PCR analyses o he exp ession o endo helial and p oge ia genes. Exp ession o EC (PECAM1,CDH5,KDR and WF) and p oge ia (p oge in, which is encoded by LMNA G608G gene) ma ke s in HGPS-iPSC and N-iPSCs a passage 7. Human umbilical ao ic endo helial cells (HUAECs) we e used as con ols. Resul s a e mean SEM (n= 3). **, and **** deno e s a is ical signi icance (Po0.01, and Po0.0001) no malized by he housekeeping gene (GAPDH). S a is ical analyses we e pe o med by an unpai ed - es . (C) Immuno luo escence analysis pe o med on HGPS-iPSC ECs and N-iPSC ECs a passage 7 o ECs (PECAM1, VE-cadhe in, and ZO-1), p oge ia (p oge in) and lamin A/C (inse ) ma ke s. Cell nuclei we e labelled wi h DAPI (blue). Scale ba is 50 mm, excep in he inse whe e scale ba is 10 mm. Communica ion Nanoscale Ho izons Open Access A icle. Published on 12 Feb ua y 2021. Downloaded on 2/15/2021 11:13:43 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online This jou nal is ©The Royal Socie y o Chemis y 2021 Nanoscale Ho iz. N-iPSC-de i ed o HGPS-iPSC-de i ed PECAM1 + cells cul u ed o 7 passages exp essed high le els o EC ma ke s such as PECAM1 [by immunocy ochemis y (Fig. 4C) and low cy o- me y (Fig. S3A, ESI†)], VE-cadhe in and ZO-1 a cell–cell con ac s (Fig. 4C) as well as endoglin (CD105). In con as o N-iPSC ECs, HGPS-iPSC ECs exp essed p oge in, he unca ed o m o lamin A. Indeed, 61 18% (n= 3) o HGPS-iPSC ECs had accumula ion o p oge in, as quan i ied by immuno luo escence. The p esence o nuclea blebs (cha ac e is ics o p oge ia cells) was also obse ed by immunos aining o lamin A/C (inse in Fig. 4C). A he unc ional le el, bo h N-iPSC ECs 13 and HGPS- iPSC ECs esponded o p o-in lamma o y s imuli such as TNF-a o 24 h leading o an inc ease in he exp ession o adhesion molecules, which hen media ed he adhesion o monocy es (Fig. S3C, ESI†), and hiPSC lines. The de i ed cells exp essed EC ma ke s a he gene and p o ein le els and we e unc ional. Vascula sc eening o NPs in a 2D model The impac o he nanoma e ial lib a y was e alua ed in ECs de i ed om N-iPSCs o HGPS-iPSCs cul u ed in s a ic condi ions. Cells we e cul u ed o 24 h, washed and hen exposed o 24 h o diffe en concen a ions o he nanoma e ials (6, 12.5, 25 and 50 mgmL 1 )(Fig.5A).Theeffec o heNPswasquan i iedby wo es s: (i) cell me abolism h ough he measu emen o mi o- chond ial ac i i y by a P es oBlue assay and (ii) cell nec osis by coun ing cell nuclei using a Hoechs 33342/p opidium iodide (PI) s aining (Hoechs o quan i y he o al numbe o cells; PI o quan i y he dead cells). Because some NPs may in e e e wi h he P es oBlue assay, 31 cells a e being incuba ed wi h NPs o 24 h we e washed 3 imes wi h PBS be o e incuba ing wi h P es oBlue. These washing s eps (which we e no necessa y and hus no pe o med in he Hoechs 33342/PI assay) likely induced highe cell de achmen /loss, especially when NPs cause some oxici y, which may lead o an appa en low mi ochond ial ac i i y. This migh be he eason why he P es oBlue esul s appea ed o show mo e oxic effec o some NPs han he co esponding Hoechs 33342/PI assay. The e o e, p ima ily, we ha e used Hoechs /PI s aining o compa e all he expe imen al g oups. F om he hi y nanoma e ials es ed, 7 induced signi ican oxici y (460% a 50 mgmL 1 ; based on PI s aining) owa ds bo h hPSC-de i ed ECs es ed (i.e., NP1, NP3, NP14, NP15, NP18, NP22 and NP30) whe eas 2 nanoma e ials (NP25 andNP28)inducedsigni ican oxici y only owa ds N-iPS ECs (Fig. 5B–E). The high oxici y obse ed o NP1 and NP3 NPs (i.e.,ul asmallAuNPs)isin ag eemen wi h p e ious esul s ob ained in HeLa cells, wi h a epo ed IC50 o abou 50 mM. 25,32 The oxici y o NP15 (ZnO) is in line wi h he oxic p o ile obse ed p e iously in human ao ic ECs (450% o 4 o24hincuba iona 50mgmL 1 ), 33,34 human ca diac mic o ascula ECs (460% o 12 o24hincuba iona 50 mgmL 1 ) 35 and in human umbilical ein ECs (HUVECs) (475% o 24 h incuba ion a 32 mgmL 1 ). 36,37 Rega ding NP28, he esul s a e also in acco dance wi h hose epo ed in he li e a u e o HUVECs exposed o 24 h, wi h oxici ies anging om 20 o 60% o 50 and 75 mgmL 1 SiO 2 concen a ions, espec i ely. 38 NPs showed a diffe en ial oxic effec agains bo h ypes o ECs. HGPS-iPSC ECs we e mo e sensi i e o NP13, NP18, NP22, NP26 and NP29 NPs as compa ed o N-iPSC ECs (Fig. 5B–E). Howe e , NP3, NP7, NP24, NP25, and NP28 we e mo e oxic o N-iPSC-de i ed ECs han o HGPS-iPSC ECs. To in es iga e whe he he nanoma e ial oxici y co ela es wi h i s capaci y o be in e nalized by ECs, we pe o med induc i ely coupled plasma mass spec ome y measu emen s (ICP-MS) on N-iPSC ECs and HGPS-iPSC ECs exposed o 11 diffe en nanoma e ials (25 mgmL 1 ) o 24 h (Fig. 5F). The concen a ion o NPs was selec ed based on a comp omise be ween cell dea h and NP quan i ica ion be o e cell dea h. Fo one o he nanoma e ials es ed (NP1), he accumula ion was e alua ed a e 4 h o incuba ion wi h cells, because o he high oxici y obse ed a e 24 h incuba ion. O e all, ou esul s Fig. 5 Impac o nanoma e ials in iPSC-ECs cul u ed in 2D. (A) Schema ic ep esen a ion o he p o ocol used. Cells we e seeded a densi y o 1  10 5 cells pe cm 2 and we e allowed o es o 24 h, a e which NPs we e added in esh medium a concen a ions be ween 6 and 50 mgmL 1 o a 24 h pe iod. Cell iabili y was e alua ed by Hoechs 33342/PI s aining, while cell me abolism was quan i ied by a P es oBlue assay. (B) Rep e- sen a i e images o Hoechs 33342/PI s aining o N-iPSC ECs wi h (NP1 o NP15, bo h a 50 mgmL 1 ) o wi hou exposu e o NPs o 24 h. Scale ba is 50 mm. (C and D) Hea maps o NP oxici y based on Hoechs 33342/PI s aining (C) and P es oBlue assay (D; no malized by con ol cells, wi hou NPs). (E) Toxici y obse ed by PI s aining in N-iPSC ECs and HGPS-iPSC ECs exposed o NPs a 25 mgmL 1 (E.1) and 50 mgmL 1 (E.2). Resul s a e mean SEM (n=4). *, **, and **** deno e s a is ical signi icance (Po0.05, Po0.01, and Po0.0001). (F) ICP-MS analyses in N-iPSC ECs and HGPS- iPSC ECs a e being exposed o NPs o 24 h (excep o NP1, which was 4 h incuba ion). Resul s a e mean SEM (n=4). **, ***, and **** deno e s a is ical signi icance (Po0.01, Po0.001, and Po0.0001). In E and F, he s a is ical analyses we e pe o med by an unpai ed - es . Nanoscale Ho izons Communica ion Open Access A icle. Published on 12 Feb ua y 2021. Downloaded on 2/15/2021 11:13:43 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online Nanoscale Ho iz. This jou nal is ©The Royal Socie y o Chemis y 2021 indica e ha he oxici y o he NPs was no dependen on he concen a ion in e nalized by he cells. Fo example, NP24, NP25 and NP30 had simila anges o cy o oxici y (Fig. 5E.1) agains N-iPSC ECs bu signi ican diffe ences in NP up ake (Fig. 5F). On he o he hand, he cellula up ake o NP1 and NP7 was e y low compa ed o ha o o he NP o mula ions; howe e , hey exe ed a highe cy o oxic effec agains ECs. In addi ion, one (NP12) showed high cellula in e naliza ion (Fig. 5F), mos ly by N-iPSC ECs, wi hou showing measu able oxici y (Fig. 5C). In e es ingly, some nanoma e ials we e mo e in e nalized by HGPS-iPSC ECs han by N-iPSC ECs and his is su p ising aking in o accoun ha HGPS-iPSC ECs showed lowe cell doublings han N-iPSC ECs (Fig. S3B, ESI†). O e all, ou esul s indica e ha : (i) NPs showed a diffe en ial oxic effec agains bo h ypes o ECs and (ii) he e is no a di ec co ela ion be ween he concen a ion o NPs in e nalized by cells and he obse ed oxici y in i o unde s a ic condi ions. Rega ding his las poin , some NPs showed high in e naliza ion efficiency wi hou causing any oxic effec on cells, whe eas o he s caused high oxici y wi h e y low in e naliza ion efficiency. Vascula sc eening o NPs in a 3D model The ascula impac o he NPs was hen sc eened by a Ma igel assay 39 comp ising he o ma ion o mic o essels wi h a pa en lumen, as p e iously demons a ed by us in e . 12. The op imiza ion o he assay (e.g. cell exposu e ime o nanoma e ials, concen a ion o nanoma e ials, and ype o eadou s) was ca ied ou wi h soma ic ECs (human umbilical a e ial ECs) (Fig. S4, ESI†). Cells we e cul u ed on op o Ma igel in he p esence o NPs o 24 o 48 h a e which EC iabili y and ube leng h was measu ed. The op ion o mixing he NPs du ing he o ma ion o mic o essels, and no a e , was inspi ed by he in i o s a egy desc ibed below in zeb a ish emb yos in which he effec o he NPs was e alua ed du ing, and no a e , emb yonic ascula u e o ma ion. Ou esul s indica e ha cells exposed o NP1 a 1 mgmL 1 o 24 h had no signi ican effec , while a 12.5 mgmL 1 showed app oxima ely 12% and 14% o apop osis and nec osis, espec i ely, and we e unable o o m ascula mic o essels in Ma igel (cells showed clea al e ed mo phology) (Fig. S4, ESI†). The e o e, o he sc eening assay in Ma igel we ha e used N-iPSC-ECs, a a concen a ion o nanoma e ials o 12.5 mgmL 1 and we e alua ed he leng h o he mic o essels a e 24 h (Fig. 6A). In e es ingly, only h ee o mula ions could nega i ely affec mic o essel o ma ion, speci ically NP1, NP3 and NP29 ha educed o e en abolished he o ma ion o mic o essels (Fig. 6B.1). In e es ingly, NP9 enhanced angiogenesis, as LL37 pep ide is a ecognized p o-angiogenic pep ide. 18 Nex , o compa e he ascula esponse in 3D e sus 2D (see sec ion abo e), we ha e mixed NP o mula ions (a a concen a ion o 25 mgmL 1 ) wi h N-iPSC ECs o HGPS-iPSC ECs, cul u ed he cells in Ma igel o 24 h and measu ed cell oxici y by Hoechs /PI s aining. We ha e selec ed 6 NP o mula ions o hese analyses including NP1, NP3, NP4, NP14, NP15 and NP29 since hey showed cy o oxici y in a 2D model (Fig. 5C). The oxici y le els (based on PI s aining) o NP1 and NP3 we e lowe han he ones obse ed in a 2D model (NP1: 60 s. 90%; NP3: 40 s. 90%, all based in N-iPSC ECs); howe e , i was highe o NP29 (N-iPSC ECs: 20 s. 10%; HGPS-iPSC ECs: 40 s. 20%). Ou esul s u he showed diffe ences in NP oxici y agains N-iPSC ECs and HGPS- iPSC ECs (Fig. 6C). Fo example, NP4 and NP14 induced highe oxici y owa ds N-iPSC ECs han HGPS-iPSC ECs. In con as , NP29 induced highe oxici y owa ds HGPS-iPSC ECs han N-iPSC ECs, simila o he esul s ob ained wi h he 2D model (Fig. 5). P e ious s udies ha e shown ha he accumula ion o p oge in in ECs induced a p o-in lamma o y p og am o EC ac i a ion cha ac e ized by an o e exp ession o leukocy e adhesion mole- cules (VCAM1, E-selec in) 40 and p o-in lamma o y cy okines (IL8) 40 and a down egula ion o he ansc ip ion ac o K uppel-like ac o (KLF2), 40 which egula es EC p o-in lamma o y ac i a ion, bu also a dec ease in he an i-oxida i e s ess esponse p og am Fig. 6 Impac o nanoma e ials on iPSC-ECs cul u ed in 3D. (A) Schema ic ep esen a ion o he p o ocol o e alua e he effec o NPs in he o ma ion o capilla y-like ne wo ks by iPSC-ECs on Ma igel. Cells we e seeded on op o Ma igel a a densi y o 2 10 3 cells pe well oge he wi h NPs (12.5 mgmL 1 ) o 24 h, a e which cell iabili y and ube o ma ion was analyzed. (B.1) Impac o NP lib a y on he leng h o mic o essels o med by N-iPSC ECs. Resul s a e a e age SEM (n= 4). **, and **** deno e s a is ical signi icance (Po0.01, and Po0.0001, espec i ely) ela i e o con ol. (B.2) Rep esen a i e images o capilla y-like ne wo ks o med in Ma igel by N-iPSC ECs ea ed o 24 h wi h 12.5 mgmL 1 o NP1, NP9 o wi hou NPs (con ol). Scale ba is 200 mm. (C) Toxici y was measu ed by Hoechs 33342/PI s aining a 24 h, in HGPS-iPSC ECs o N-iPSC ECs cul u ed in he p esence o NP1, NP3, NP4, NP14, NP15 and NP29 in Ma igel a 25 mgmL 1 . *, ***, and **** deno e s a is ical signi icance (Po 0.05, Po0.001, and Po0.0001). (D) Rela i e gene exp ession o genes in ol ed in oxida i e s ess (NRF2), in lamma ion (ICAM1,SELE,VCAM1 and IL8) and endo helial homeos asis (KLF2,andVEGF) in N-iPSC ECs and HGPS-iPSC ECs be o e (D.1) and a e exposu e o 25 mgmL 1 o NP29 (D.2) o NP14 (D.3). *, **, and *** deno e s a is ical signi icance (Po0.05, Po0.01, and Po0.001). In B.1, C and D, he s a is ical analyses we e pe o med by an unpai ed - es . Communica ion Nanoscale Ho izons Open Access A icle. Published on 12 Feb ua y 2021. Downloaded on 2/15/2021 11:13:43 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online This jou nal is ©The Royal Socie y o Chemis y 2021 Nanoscale Ho iz. cha ac e ized by an impai ed ansc ip ional ac i i y o he ansc ip ion ac o NRF2. 41 Indeed, HGPS-iPSC ECs ha e highe eac i e oxida i e species le els han N-iPSC ECs (Fig. S5, ESI†). To iden i y he mechanism behind he diffe en ial oxici y o NPs agains N-iPSC ECs and HGPS-iPSC ECs, we e alua ed he exp es- sion o genes in ol ed in oxida i e s ess (NRF2), in lamma ion (SELE,ICAM1,VCAM1 and IL-8) and endo helial homeos asis (KLF2,andVEGF) in cells exposed o NP14 and NP29 (bo h a 25 mgmL 1 ) o 24 h (Fig. 6D). In he absence o NPs, HGPS-iPSC ECs showed highe exp ession o in lamma o y and oxida i e s ess bu lowe KLF2 mRNA ansc ip s han N-iPSC ECs. A e exposu e o NPs, he in lamma o y esponse o bo h cells was signi ican ly diffe en : only N-iPSC ECs showed a signi ican up- egula ion o p o-in lamma o y mRNA and he esponse seemed highe o NP14 han o NP29. Mo eo e , N-iPSC ECs, bu no HGPS-iPSC ECs, showed an up egula ion o mRNA o he NRF2 gene ha encodes nuclea ac o e y h oid 2- ela ed ac o in ol ed in he cellula an ioxidan esponse. Al oge he , ou esul s indica e ha : (i) a small numbe o NPs (3 ou o 30) in e e ed wi h mic o essel o ma ion in Ma igel a he con- cen a ion es ed (12.5 mgmL 1 ); (ii) he Ma igel assay showed diffe ences in he ascula oxici y p o ile (e alua ed by PI s aining) o NPs ela i e o he 2D model, bu in bo h cases he impo ance o EC backg ound was c i ical o he biological esponse; (iii) he diffe en ial biological esponse o N-iPSC ECs and HGPS-iPSC ECs o NPs in ol ed diffe en ial egula ion o genes ela ed wi h in lamma ion and oxida i e s ess. Sc eening o NPs in a blood essel on a chip The ascula models desc ibed abo e ha e se e al limi a ions, including he ac ha hey a e no pe used and do no ake in o accoun he he e o ypic in e ac ions be ween ECs and SMCs. Cells exposed o low change hei mo phology and mo e impo an ly hey exp ess a laye o p o eins, glycop o eins, glyco- lipids, p o eoglycans and glycosaminoglycans called glycocalyx. 42 The glycocalyx is in ol ed in se e al unc ions, such as egula ion o ascula pe meabili y, ac ing as a selec i e ba ie o mac o- molecules (based on size and cha ge), as well as a mechanosen- so o luid shea s ess. Thus, we decided o p epa e a blood essel on a chip o sc een nano–bio in e ac ions. In gene al, hese chips ha e only one laye o endo helial cells o e alua e nano– bio in e ac ions. 43 He e, we ha e success ully de eloped a chip wi h wo-cell monolaye s o SMCs and ECs. The essel was pe used wi h medium, a shea a es simila o in i o,and he EC monolaye was able o sec e e a unc ional glycocalyx laye on op. Fo his pu pose, SMCs ( o ini ial s udies we ha e used soma ic cells) we e cul u ed in a mic o luidic chambe o 12 h o o m a monolaye ollowed by he pla ing o ECs on op o SMCs o 4 h and inally by he pe usion o bo h cells o 5 days (Fig. S6, ESI†). Unde hese condi ions, SMCs we e in he ou e pa o he essel, while ECs emained in he luminal side (Fig. S6A and B, ESI†). In e es ingly, ECs co-cul u ed wi h SMCs unde low condi ions showed highe le els o glycocalyx han ECs alone cul u ed unde low condi ions (Fig. S7A, ESI†). To e i y ha he glycocalyx was indeed loca ed in he luminal side o ECs, we ha e used GFP-exp essing ECs, which we e co-cul u ed wi h SMCs o 5 days unde low as p e iously desc ibed (Fig.S7BandC,ESI†). As expec ed, he laye o glycocalyx was loca ed on op o he EC laye in he luminal side, showing he pola iza ion o he glycocalyx. The o ma ion o he glycocalyx was also demons a ed by an EC up ake assay o DiI-LDL (Fig. S7D, ESI†). In his case, cells cul u ed unde low condi ions o 5 days had lowe up ake o Dil-LDL han cells cul u ed o 5 days and exposed o hepa inase III o emo e he hepa in sul a e in glycocalyx. In o de o e i y he effec o lux in he o ma ion o glycocalyx, SMCs and ECs we e co-cul u ed as p e iously desc ibed and he hepa an sul a e in ensi y was quan i ied a diffe en ime poin s (Fig.S8A,ESI†). The amoun o hepa an sul a e inc eased mo e han 2.5- old up o 5 days in low, a 20 dyn cm 2 (Fig. S8B and C, ESI†). The eco e y o hepa an sul a e was also e alua ed. Cells cul u ed unde low o 5 days we e ea ed wi h hepa inase III o deg ade he hepa an sul a e, and i s eco e y was ollowed o e ime(Fig.S8D,ESI†). The kine ics o he eco e y was simila o he one obse ed o he hepa an sul a e o ma ion induced by he shea s ess. The e o e, o subsequen expe imen s wi h iPSC-de i ed cells, hey we e co-cul u ed o 5 days a 20 dyn cm 2 . We pe o med expe imen s wi h N-iPSC SMCs and N-iPSC ECs (Fig. 7A) bu no wi h HGPS-iPSC-de i ed cells because he HGPS-iPSC SMCs we e sensi i e o low and de ached o e ime. 44 N-iPSC SMCs ob ained using p o ocols p e iously published by us in e . 45 we e cul u ed in a mic o- luidic chambe ollowed by he cul u e o N-iPSC ECs on op o he SMCs (Fig. 7B). Cells we e hen cul u ed unde low condi ions (20 dyn cm 2 ) o 5 days. Unde hese condi ions, N-iPSC SMCs we e loca ed in he ou e pa o he blood essel and posi i ely s ained o a-smoo h muscle ac in (a-SMA) whe eas he iPSC-ECs, which we e loca ed in he inne pa o he blood essel, we e posi i ely s ained o VE-cadhe in (Fig. 7B.1). As expec ed, ECs exp essed glycocalyx, he exp ession being highe in low han in s a ic condi ions (Fig. 7B.2) and in co-cul u e wi h SMCs e sus ECs alone (Fig. 7C). Mo eo e , he enzyma ic emo al o hepa an sul a e by hepa inase in ascula cells cul u ed unde low condi ions o 5 days allowed highe accumula ion o Dil LDL (Fig. 7D) han cells wi hou enzyma ic ea men . Al oge he , we ha e de eloped a blood essel on a chip om ECs and SMCs de i ed om iPSCs. ECs co-cul u ed wi h SMCs unde low condi ions exp essed highe le els o hepa an sul a e han ECs cul u ed in s a ic condi ions and hus ep esen a be e ma ch o he in i o blood essel physiology. Nex , we ha e es ed he 10 NPs iden i ied in p e ious ascula models wi h highe cy o oxici y effec s in a blood essel on a chip (Fig. 7E). In his model, we used ECs and SMCs de i ed om N-iPSCs. Ten NPs we e exposed o 24 h unde low condi ions o he blood essel on a chip. Fo wo o he o mula ions (NP15 and NP25) we ha e also quan i ied he cellula in e naliza ion o he NPs by ICP-MS analyses. The esul s show ha NP in e naliza ion was 34.6 pg pe cell and 1.1 pg pe cell (assuming ha all he NPs we e aken up by ECs alone) o NP15 and NP25, espec i ely, which means app oxi- ma ely 3- old lowe han in s a ic condi ions (Fig. S9, ESI†). In e es ingly, all he NPs es ed showed lowe oxici y han in s a ic condi ions agains N-iPSC ECs (Fig. 7F). The dec ease in Nanoscale Ho izons Communica ion Open Access A icle. Published on 12 Feb ua y 2021. Downloaded on 2/15/2021 11:13:43 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online Nanoscale Ho iz. This jou nal is ©The Royal Socie y o Chemis y 2021 oxici y may be explained by he p o ec i e ole o glycocalyx exp essed in he blood essel on a chip. To demons a e his hypo hesis, he oxici ies o NP1 and NP2 we e e alua ed in 3 diffe en models: (i) a co-cul u e o N-iPSC SMCs and N-iPSC ECs (Fig. 7G), (ii) a monocul u e o N-iPSC ECs (Fig. 7H.1) and (iii) a monocul u e o N-iPSC SMCs (Fig. 7H.2). As shown be o e (Fig. 7C and Fig. S7A, ESI†), a co-cul u e o SMCs and ECs led o an inc ease o he p oduc ion o he glycocalyx laye shown by a signi ican inc ease in in ensi y o hepa an sul a e, one impo - an and main componen o he glycocalyx. Ou esul s showed ha when ECs had his p o ec i e glycocalyx laye he oxici y o NP1 was conside ably lowe han in cells wi hou his laye (Fig. 7G). In addi ion, when he glycocalyx laye was impai ed, ei he by enzyma ic deg ada ion (Fig. 7G) o by using he monocul u e cell model o ECs ( ha p oduces less glycocalyx; Fig. 7H.1), he oxici y was highe ela i e o cells wi h a glycocalyx laye . iPS-SMCs showed no oxici y owa ds hese NPs a he concen a ions es ed (Fig. 7H.2). The e o e, he p esence o a glycocalyx laye in luenced he impac o NPs on ECs and migh explain diffe ences in he oxici ies o nanoma e ials in s a ic s. low condi ions. O e all, ou esul s indica e ha mos o he nanoma e ials ha had a nega i e impac on he angiogenesis o blood essels in he ansgenic zeb a ish model showed ascula oxici y in he 2D and 3D models bu wi h diffe ences acco ding o he ype o ascula cells and hei o ganiza ional complexi y. In addi ion, he oxici ies o he es ed NPs we e lowe in he blood essel on a chip han in he 2D model. The esul s u he indica e ha he dec ease in he NP oxici y, a leas in pa , was media ed by he exp ession o a p o ec i e glycocalyx laye in ECs cul u ed in he blood essel on a chip. 3. Discussion The cu en wo k desc ibes a pla o m o s udy he bioac i i y/ oxici y o nanoma e ials based on ascula cells de i ed om hiPSCs, which we e cul u ed in 3 diffe en pla o ms wi h inc easing le el o complexi y: (i) monocul u e in poly(s y ene) (named ‘‘2D’’); (ii) monocul u e in Ma igel (named ‘‘3D’’) and (iii) co-cul u e (bo h ECs and SMCs) unde low condi ions (named ‘‘blood essel on a chip’’). These 3 pla o ms cap u ed diffe en aspec s o he impac o he nanoma e ials and he esul s we e alida ed in zeb a ish emb yos, mo e speci ically, in he capaci y o he NPs o in e e e wi h emb yonic asculogenesis. The2Dmodelis hemos sui ablemodel o high- h oughpu sc eening; howe e , i does no ecapi ula e he h ee-dimensional o ganiza ion o he blood essel as well as he dynamic en i on- men o he bloods eam, which may induce pheno ype al e a ions in ECs (e.g. absence o glycocalyx exp ession) and NPs may deposi a highe le els on o he cells, which inc eases NP up ake. The 3D model is sui able o ep oduce he geome y o he blood essel in i o since ECs associa e in o 3D ubes wi h lumen o ma ion; 12,39 howe e , as a 2D model, i does no ake in o accoun he complexi y o he blood essel o he effec o low. The blood essel on a chip is sui able o ep oduce he Fig. 7 Impac o nanoma e ials in a blood essel on a chip composed by iPSC-ECs and iPSC-SMCs. (A) P o ocol o gene a e a blood essel on a chip. A suspension o SMCs (4.1 10 4 cells pe cm 2 ) was applied in each channel o an Ibidi m-Slide VI 0,4 pla e. SMCs we e main ained in s a ic condi ions o 12 h, a e which a suspension o ECs (12.5 10 4 cells pe cm 2 )wasappliedon opo SMCs and he co-cul u e was main ained in s a ic condi ions o 4 h, a e which cells we e pe used wi h EGM-2 medium a 20 dyn cm 2 o 5 days. (B) Immuno luo escence analyses pe o med on a co-cul u e o N-iPSC SMCs and N-iPSC ECs o : SMC (a-SMA), EC (VE-cadhe in) (B1) and glycocalyx (hepa an sul a e) (B2) ma ke s. Scale ba is 50 mm. (C) Hepa an sul a e in ensi y in a co-cul u e o N-iPSC SMC and N-iPSC EC cells s. monocul u e o N-iPSC ECs a e 5 days unde a e ial low condi ions (20 dyn cm 2 ). **** deno es s a is ical signi icance (Po0.0001). (D) Effec o glycocalyx impai men (by hepa inase III ea men ) on he up ake o DiI-LDL by N-iPSC ECs a e 5 days in co-cul u e wi h N-iPSC SMCs unde a e ial low condi ions (20 dyn cm 2 ). **** deno es s a is ical signi icance (Po0.0001). In C and D, he s a is ical analyses we e pe o med by an unpai ed - es . (E) Schema ic ep esen a ion o he p o ocol. Cells (co-cul u e o N-iPSC SMCs and N-iPSC ECs) we e cul u ed o 5 days in a e ial low condi ions (20 dyn cm 2 ), a e which NPs we e incuba ed o 24handcha ac e ized o cellme abolismanddea h.(F)Cell iabili y a e 24 h incuba ion in low condi ions measu ed by Hoechs 33342/PI s aining. A ows show he diffe ence in oxici y be ween s a ic (monocul u e o N-iPSC ECs) and low condi ions (co-cul u e N-iPSC SMCs and N-iPSC ECs) o he same NPs. Resul s a e mean SEM (n= 4–6). *, and **** deno e s a is ical signi icance (Po0.05, and Po0.0001) agains con ol (cells wi hou NPs). (G) E ec o he emo al o endo helial glycocalyx, by hepa inase III ea men , on he me abolism o N-iPSC ECs and N-iPSC SMCs cul u ed in he mic o luidic chip o 24 h wi h NP1 o NP2. ** deno es s a is ical signi icance (Po0.01) agains con ol (cells wi hou NPs). (H.1) Cell me abolism o N-iPSC ECs a e 24 h incuba ion wi h NP1 o NP2 unde a e ial low condi ions. **** deno es s a is ical signi icance (Po0.0001) agains con ol (cells wi hou NPs). (H.2) Cell me abolism o N-iPSC SMCs a e 24 h incuba ion wi h NP1 o NP2 unde a e ial low condi ions. In F, G and H, esul s a e mean SEM (n= 3) and s a is ical analyses we e pe o med by an unpai ed - es . Communica ion Nanoscale Ho izons Open Access A icle. Published on 12 Feb ua y 2021. Downloaded on 2/15/2021 11:13:43 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online This jou nal is ©The Royal Socie y o Chemis y 2021 Nanoscale Ho iz. complexi y (as i includes ECs and SMCs) and low cul u e condi ions as obse ed in he in i o blood essels; howe e , i is complex o use in high- h oughpu sc eening. Ou esul s indica e ha he h ee pla o ms we e able o iden i y, o diffe en ex en s, he NPs ha showed oxici y agains zeb a ish emb yos. F om he 18 NP o mula ions ha in e e ed wi h he asculogenesis p ocess in zeb a ish emb yos (Fig. 3), 15 NP o mula ions ha e been iden i ied by he in i o models (2D and 3D models; he numbe o PI + cells was abo e 20%), hei oxici y le el being dependen on he hiPSC-EC ype. Some o he o mula ions we e also es ed in he blood essel on a chip and he esul s u he con i med hei oxici y; howe e , in gene al, a lowe in ensi y. Table S1 (ESI†) summa izes he main indings ob ained among he ou sys ems as well as oden da a ound in he li e a u e. P e ious s udies ha e used 2D models based on soma ic ECs, bu no on ECs de i ed om hiPSCs gene a ed om diffe en indi iduals, o sc een mul iple NP o mula ions. 46 In addi ion, hiPSC-ECs ha e been used o sc een small molecules 12 bu no nanoma e ials. The use o hiPSCs enables he in i o modelling o diffe en indi iduals wi h o wi hou diseases. The de elopmen o iPSC echnology allows he access o a i ually unlimi ed numbe o cells om any indi idual, which can hen be diffe en ia ed in o any kind o cells and be used o high- h oughpu sc eening s udies. This echnology has allowed, o ins ance, he sc eening o la ge lib a ies o small compounds o ind hi s ha escue gene ic diseases 47,48 o in e e e wi h human de elopmen . 12 We ha e used hiPSCs as a sou ce o ECs o s udy nanoma e ial in e ac ions, which ha e no ye been in es iga ed. We ha e s udied he impac o nanoma e ials on hiPSC-ECs ha ing emb yonic (N-iPSC ECs) 12 and aged pheno- ypes (HGPS-iPSC ECs), as con i med by he accumula ion o p oge in and he exis ence o dysmo phic nuclei in he la e . The selec ion o HGPS-iPSC ECs in he con ex o his wo k was o in es iga e he biological esponse o aged cells o NPs, which is a opic ha emains la gely unexplo ed. Ou esul s indica e ha HGPS-iPSC ECs and N-iPSC ECs had diffe en ial oxici y p o iles agains NPs, bo h in 2D and 3D models. The ul asmall NP1 and NP3 we e shown o be he mos oxic o all NPs o he lib a y o bo h ypes o ECs, e en o concen a ions a 6 mgmL 1 (2D model); i e NPs (NP14, NP15, NP18, NP22 and NP30) showed oxici y le els highe han 60% (measu ed by a PI assay) o bo h EC-de i ed cells; h ee NPs (2D sys em: NP25 and NP28; 3D sys em: NP14) showed highe oxici ies agains N-iPSC ECs han HGPS-iPSC ECs; one NP o mula ion (NP29) showed highe oxici y (3- old) o HGPS-iPSC ECs han o N-iPSC ECs. The ascula oxici ies obse ed o some NPs in he 2D model a e in good acco dance wi h hose epo ed in he li e a u e using soma ic cells. Fo example, NP1 and NP3 showed high oxici ies agains HeLa cells. 32 Zinc oxide NPs we e obse ed o be oxic (450%) a 50 mgmL 1 in human ao ic ECs, 34 human ca diac mic o ascula ECs 35 and in HUVECs; 36,37 all hese esul s a e in good acco - dance wi h he 60% oxici y obse ed in ou s udy o ECs de i ed om iPSCs. SiO 2 NPs ha e also been epo ed o induce oxici ies om 20 o 60% in HUVECs o NP concen a ions be ween 50 and 75 mgmL 1 , 38 while Ag NPs ha e been epo ed o induce oxici y in a b ain mic o essel ECs 49 and o demon- s a e an i-angiogenic p ope ies. 50 I is possible ha he highe oxici y o Ag NPs agains HGPS-iPSC ECs, as compa ed o ha o N-iPSC ECs, is ela ed o a dec ease in he an ioxida i e s ess esponse p og am in HGPS-iPSC ECs. I is known ha Ag NPs induce cy o oxici y h ough oxida i e s ess, leading o he p oduc ion o eac i e oxygen species (ROS) in hei su ace o hough he di ec in e ac ion wi h cell mi ochond ia. 51 On he o he hand, he nuclea accumula ion o p oge in impai s he ansc ip ional ac i a ion o NRF2, due o he physical in e ac ion o bo h p o eins, which educes he an i-oxida i e s ess p og am o HGPS cells. 41 The e o e, i is expec ed ha HGPS-iPSC ECs a e mo e sensi i e o NPs ha induce oxici y by oxida i e s ess such as Ag NPs. In con as , he highe oxici y o NP14 agains N-iPSC ECs as compa ed o HGPS-iPSC ECs may be ela ed o he diffe en ial in lamma o y p og am be ween cells. Ou esul s show ha HGPS-iPSCs ha e a p o- in lamma o y p og am be o e NP exposu e, cha ac e ized by highe le els o VCAM1,SELE,IL8 and ICAM1 mRNA ansc ip s as compa ed o N-iPSC ECs. I is possible ha he p o-in lamma o y s a us o HGPS-iPSC ECs makes hese cells less sensi i e o in lamma o y cy okines han N-iPSC ECs. Indeed, he p o-in lamma o y p og am o HGPS-iPSC ECs is no signi ican ly affec ed a e exposu e o NPs, while he opposi e was obse ed in N-iPSC ECs. The EC esponse o NPs is la gely in luenced by he low shea s ess and he in e ac ion wi h SMCs. EC geno ype/ pheno ype as well as endocy ic capaci y a e in luenced by low shea s ess. Fo example, ECs cul u ed unde inc eased shea s ess ha e dec eased oxida i e s ess and in lamma ion 52 and showed diffe en ial NP up ake as compa ed o ECs cul u ed unde s a ic condi ions. 43,53 Impo an ly, EC esponse o low shea s ess is in luenced by he neighbou ing SMCs. 54,55 Al hough in i o co-cul u e sys ems o ECs and SMCs unde low condi ions ha e been es ablished o soma ic ECs and SMCs, 54,55 he gene a ion o co-cul u e sys ems wi h hiPSC-de i ed ascula cells emains unexplo ed. The model de eloped in he cu en wo k allows he pe o mance o assays a a e ial condi ions by submi ing cells o shea s esses o 20 dyn cm 2 . In his model, cells aligned pa allel o he low di ec ion, and ECs p oduced a glycocalyx (a glycop o ein–polysaccha ide meshwo k) laye in he apical egion as i occu s in i o. P e ious s udies ha e shown ha glycocalyx in luences NP up ake by soma ic ECs. 56,57 Ou esul s showed ha NP oxici y agains ECs is signi ican ly lowe in cells cul u ed unde low condi ions and in co-cul u e wi h SMCs. Ou esul s u he showed ha glycocalyx media ed, a leas in pa , he educ ion in NP in e naliza ion, since he deg ada ion o glycocalyx by hepa inase signi ican ly educed cell me abolism/ iabili y. In many cases, he impac o nanoma e ials is sc eened in oden s; 4–6 howe e , hese expe imen s a e expensi e, ime consuming and low h oughpu . Zeb a ish emb yos can be used o in es iga e ascula oxici y o NPs because hey a e ela i ely cheap, he de elopmen al p ocesses a e well s udied and he oxici y o he NPs can be moni o ed in eal ime using Nanoscale Ho izons Communica ion Open Access A icle. Published on 12 Feb ua y 2021. Downloaded on 2/15/2021 11:13:43 AM. This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence. View A icle Online