1
Scien i ic RepoR s | 5:14684 | DOi: 10.1038/s ep14684
www.na u e.com/scien i ic epo s
Compa a i e p o eomics e eals
human plu ipo en s em cell-
de i ed limbal epi helial s em cells
a e simila o na i e ocula su ace
epi helial cells
Alexand a Mikhailo a1, An i Jylhä2, Jochen Rieck1, Janika Nä inen2, Tanja Ilma inen1,
Zol án Ve éb3, Ulla Aapola2, Roge Beue man2,4, Go an Pe o ski3, Hannu Uusi alo2,5 &
Heli Sko man1
Limbal epi helial s em cells (LESCs) a e issue-speci ic s em cells esponsible o enewing he co neal
epi helium. Acu e auma o ch onic disease a ec ing LESCs may dis up co neal epi helial enewal,
causing ision h ea ening and pain ul ocula su ace diso de s, collec i ely e e ed o as LESC
de iciency (LESCD). These diso de s canno be ea ed wi h adi ional co neal ansplan a ion and
he e o e al e na i e cell sou ces o success ul cell-based he apy a e needed. LESCs de i ed om
human plu ipo en s em cells (hPSCs) a e a p ospec i e sou ce o ocula su ace econs uc ion,
ye c i ical e alua ion o hese cells is c ucial be o e conside ing clinical applica ions. In o de o
quan i a i ely e alua e hPSC-de i ed LESCs, we compa ed p o ein exp ession in na i e human
co neal cells o ha in hPSC-de i ed LESCs using isoba ic ag o ela i e and absolu e quan i a ion
(iTRAQ) echnology. We iden i ied 860 unique p o eins p esen in all samples, including p o eins
in ol ed in cell cycling, p oli e a ion, di e en ia ion and apop osis, a ious LESC niche componen s,
and limbal and co neal epi helial ma ke s. P o ein exp ession p o iles we e nea ly iden ical in LESCs
de i ed om wo di e en hPSC lines, indica ing ha he di e en ia ion p o ocol is ep oducible,
yielding homogeneous cell popula ions. Thei p o ein exp ession p o ile sugges s ha hPSC-de i ed
LESCs a e simila o he human ocula su ace epi helial cells, and possess LESC-like cha ac e is ics.
Co neal epi helium, he ou e mos laye o he anspa en and a ascula co nea, is a apidly- egene a ing
s a i ied squamous epi helium. I s in eg i y and main enance a e essen ial o co neal anspa ency and
no mal ision. Limbal epi helial s em cells (LESCs) a e a ype o issue-speci ic s em cells loca ed a
he co neoscle al junc ion wi hin niche egions o he palisades o Vog 1,2. These s em cells a e c ucial
o main aining he ocula su ace in wo ways: i s , hey cons an ly enew he co neal epi helium, as
he opmos laye s a e shed o in o he ea ilm; and second, hey se e as a physical ba ie be ween
he co neal and conjunc i al epi helia3,4. Like o he issue-speci ic s em cells, LESCs a e hough o be
slow cycling, ye wi h a po en ial o sel - enewal, apid p oli e a ion, and di e en ia ion in esponse o
app op ia e s imuli5,6. LESCs gi e ise o ansien ampli ying cells (TACs) ha ha e a highe capaci y o
p oli e a ion and di e en ia ion. TACs help p ese e he no mal homeos asis o he co neal epi helium
1BioMediTech, Uni e si y o Tampe e, Finland. 2Depa men o Oph halmology, School o Medicine, Uni e si y o
Tampe e, Finland. 3S em Cells and Eye Resea ch Labo a o y, Depa men o Oph halmology, Facul y o Medicine,
Uni e si y o Szeged, Hunga y. 4Singapo e Eye Resea ch Ins i u e and School o Medicine, Singapo e. 5Tampe e
Uni e si y Hospi al Eye Cen e , Uni e si y o Tampe e, Finland. Co espondence and eques s o ma e ials should
be add essed o A.M. (email: [email p o ec ed])
Recei ed: 29 June 2015
accep ed: 08 Sep embe 2015
Published: 01 Oc obe 2015
OPEN
www.na u e.com/scien i ic epo s/
2
Scien i ic RepoR s | 5:14684 | DOi: 10.1038/s ep14684
by mig a ing apically owa ds he cen e o he co nea and eplacing he los co neal epi helial cells
(CECs). Acu e auma o ch onic disease a ec ing LESCs may cause a dis up ion in his homeos asis
and allow he neighbo ing conjunc i al epi helial cells, along wi h blood essels, o mig a e o e he
ocula su ace7. Such ocula su ace diso de s a e collec i ely e e ed o as LESC de iciency (LESCD),
and hey a e di icul o ea wi h con en ional co neal ansplan a ion, as co neal g a s do no eplace
he damaged limbus8.
Va ious s a egies ha e been in es iga ed o acili a e he econs uc ion o damaged ocula su ace,
such as ansplan a ion o au ologous o allogeneic limbal issue, o mo e ecen ly cul i a ed limbal epi-
helial ansplan a ion (CLET), whe e a small amoun o au ologous o allogeneic LESCs is expanded
in i o be o e being ansplan ed o he ocula su ace4,5,9,10. In ac , Holocla , he i s ad anced he -
apy medicinal p oduc con aining au ologous LESCs was ecen ly g an ed condi ional app o al by he
Eu opean Medicines Agency. Despi e he gene ally p omising esul s o CLET, i is limi ed by a ia ion
in long- e m success a es, use o xenogeneic and unde ined cul u e componen s, and sca ci y o dono
issue11,12. This is especially impo an in bila e al LESCD cases, whe e au ologous issue is unsui able
o CLET, and al e na i e solu ions a e needed. Human plu ipo en s em cells (hPSCs), namely human
emb yonic s em cells (hESCs) and human induced plu ipo en s em cells (hiPSCs), a e eadily a ail-
able in limi less supply, and ha e a as di e en ia ion po en ial. They p o ide new oppo uni ies o
cell-based issue enginee ing and d ug disco e y, and o e no el ways o s udy human de elopmen .
Success ul di e en ia ion o co neal epi helial lineages has been epo ed using bo h hESCs and hiP-
SCs13–17. We ha e p e iously desc ibed an e icien di e en ia ion me hod om hPSCs owa ds LESC-like
co neal epi helial p ogeni o cells in eede - ee and se um- ee condi ions18.
Tho ough cha ac e iza ion o di e en ia ed cells is an essen ial s ep owa ds clinical applica ions, as
i is impo an o e i y he au hen ici y o cell popula ions p io ansplan a ion o he ocula su ace.
In his s udy, we compa ed p o ein exp ession in human CECs and limbal epi helial cells (LECs) o
ha in hESC-de i ed LESCs (hESC-LESCs) and hiPSC-de i ed LESCs (hiPSC-LESCs) using isoba ic
ag o ela i e and absolu e quan i a ion (iTRAQ) echnology. Addi ionally, p o ein exp ession o se e al
pu a i e LESC ma ke s was e i ied using low cy ome y and immuno luo escence. Recen ad ances in
mass spec ome y (MS) echniques ha e p o en ha MS-based app oaches wi h quan i a i e analyses
can con ibu e o iden i ica ion o p o eins in ol ed in s em cell p oli e a ion and di e en ia ion19–21. The
main ad an age o using iTRAQ p o eomics is ha i allows mul iplexing wi h ou o eigh di e en
isoba ic ags. Bo h 4-plex and 8-plex iTRAQ me hods ha e been u ilized o analyze and s udy p o ein
exp ession du ing di e en ia ion o mouse o human PSCs22–24. We used 4-plex iTRAQ o enable ou
samples o be compa ed di ec ly wi hin he same measu emen . To ou knowledge, his is he i s s udy
compa ing hPSC-LESCs wi h hei na i e coun e pa s using MS-based p o eomics.
Resul s
In his s udy, na i e human CECs and LECs we e compa ed o hESC-LESCs and hiPSC-LESCs using
iTRAQ p o eomics (Fig.1). Cen al co neal (CEC) and limbal (LEC) epi helial samples we e collec ed
by posi ional sc aping o cells di ec ly om he su ace o cada e eyes ( h ee sepa a e dono s), wi hou
addi ional pu i ica ion. Di e en ia ion o hESCs and hiPSCs owa ds LESC-like cells was ca ied ou wi h
h ee biological eplica es.
Dis ibu ion o iden i ied p o eins. A o al o 860 unique p o eins exp essed in all ou samples
we e iden i ied (Fig.2). The comple e lis o iden i ied p o eins and hei espec i e exp ession le els is
p o ided in Supplemen a y Table S1. Fo a mo e de ailed analysis, p o eins ha we e de ec ed in only one
o he biological eplica es, as well as obsole e p o ein IDs we e disca ded. A e his il e ing, 497 p o-
eins we e le when CECs and LECs we e compa ed o hESC-LESCs, and 485 p o eins when compa ed
o hiPSC-LESCs (Fig.2). Di e ences in p o ein exp ession g ea e han 2- old we e conside ed as o e
o unde -exp ession. A la ge po ion o he iden i ied p o eins we e simila ly exp essed (be ween -2 and
2- old) in all samples (Fig.3a). Wi h he help o PANTHER classi ica ion, he il e ed se o p o eins was
ca ego ized by hei molecula unc ion (Fig.3b). The la ges p o ein class in he da a se was enzymes,
con aining he ollowing sub-g oups: hyd olases, isome ases, kinases, ligases, lyases, oxido educ ases,
phospha ases, p o eases and ans e ases. Some o he mos in e es ing p o ein classes in he case o
co neal and limbal epi helium a e adhesion and junc ion p o eins, s uc u al and cy oskele al p o eins
and signaling molecules. These p o ein classes a e commonly in ol ed in main aining s em cell o TAC
beha io and he niche mic oen i onmen . The e o e, hese g oups o p o eins and he di e ences in
hei exp ession we e examined in mo e de ail.
P o eins ela ed o s em cell o TAC beha io . All s em cells, including LESCs, a e gene ally quies-
cen and slow cycling, while TACs unde go ac i e cell cycling, p oli e a ion, di e en ia ion and apop osis.
In his s udy, 31 p o eins in ol ed in egula ing cell p oli e a ion we e iden i ied in all samples: 14 posi-
i e egula o s, 13 nega i e egula o s, and 4 p o eins in ol ed in bo h p ocesses (Table1). Fu he mo e,
59 p o eins in ol ed in cell di e en ia ion, 65 p o eins in ol ed in apop osis, and 28 cell cycling p o eins
we e iden i ied (Fig.4). Mos o hese p o eins we e simila ly exp essed in all samples. Se e al p o eins in
hese g oups, o ins ance Annexin 1 (ANXA1), galec ins 1, 3 and 7 (LGALS1, -3 and -7), and s a i in
(SFN), a e especially ele an o he main enance and enewal o he ocula su ace.
www.na u e.com/scien i ic epo s/
3
Scien i ic RepoR s | 5:14684 | DOi: 10.1038/s ep14684
S em cell niche componen s o he ocula su ace. Niche mic oen i onmen o s em o p o-
geni o cells in ol es a ious componen s such as cell adhesion and junc ion p o eins, calcium-binding
S100A p o eins, as well as p o eins wi h an ioxidan ac i i y o in ol emen in angiogenesis and immune
esponse, including a ious g ow h ac o s and cy okines sec e ed by ocula su ace epi helial cells.
Al oge he 14 p o eins in ol ed in cell adhesion, junc ion o ma ion o ex acellula ma ix (ECM) dep-
osi ion we e iden i ied in his s udy (Fig.5a). This g oup includes se e al cadhe ins and in eg ins, and
mos o he p o eins we e simila ly exp essed in all samples, o mildly o e exp essed in CECs and LECs.
One clea excep ion was ca cinoemb yonic an igen- ela ed cell adhesion molecule 7 (CEACAM7): i s
exp ession was highe in LECs and sligh ly lowe in CECs, compa ed o hPSC-LESCs. Addi ionally,
cadhe in 13 (CDH13) was o e exp essed in CECs and LECs compa ed o hiPSC-LESCs. All ou sam-
ples we e ound o exp ess 17 p o eins in ol ed in immune esponse (Fig.5b). Mos o hese p o eins
we e simila ly exp essed in all samples, wi h he ollowing excep ions: immunoglobulin hea y cons an
alpha 1 (IGHA1) was o e exp essed in LECs, CD55, clus e in (CLU) and immunoglobulin lambda con-
s an 2 (IGLC2) we e o e exp essed in CECs and LECs, while mac ophage mig a ion inhibi o y ac o
(MIF) was only o e exp essed compa ed o hiPSC-LESCs. S100A p o eins o m he la ges amily o
calcium-binding p o eins, and a e known o be exp essed a he limbus. Eigh membe s o his p o-
ein amily we e iden i ied in he analyzed da a se (Fig.5c). When compa ed o hPSC-LESCs, p o eins
S100A8 and -A9 we e ound o be exp essed a lowe le els in CECs, and a highe le els in LECs. The
emaining p o eins we e ei he mildly o e exp essed in CECs and LECs, o simila ly exp essed in all
samples. Mo eo e , 9 p o eins wi h an ioxidan p ope ies and 10 p o eins wi h a ole in angiogenesis
we e iden i ied, mos o which we e simila ly exp essed in all samples (Fig.5d,e).
Co neal and limbal ma ke s. A o al o 21 p o eins impo an o he s uc u e and unc ion o co -
neal and limbal epi helial cells we e iden i ied using iTRAQ p o eomics (Fig.6a). Six o hese p o eins
a e known o be p e alen in he cen al co neal epi helium: aldehyde dehyd ogenase 3, amily membe
1 (ALDH3A1), deco in (DCN), cy oke ains 3 and 12 (KRT3 and KRT12), mucin 16 (MUC16), and
ans o ming g ow h ac o -be a-induced p o ein (TGFBI). These p o eins we e ound o be exp essed
a a highe le el in CECs han in LECs and hPSC-LESCs. Mo eo e , KRT3 was exp essed a a lowe
le el in LECs han in hPSC-LESCs, and KRT12 was simila ly exp essed in LECs and hPSC-LESCs.
Figu e 1. Wo k low o he s udy. Human CECs we e collec ed om he su ace o he cen al co nea,
while LECs we e collec ed om he su ace o he limbus (1–2 mm wide zone be ween he co nea and
he scle a) o h ee cada e ic dono s. Th ee biological eplica es o hESC-LESCs and hiPSC-LESCs we e
ob ained using a di ec ed di e en ia ion me hod. To al p o ein was ex ac ed om all samples and diges ed
wi h ypsin. Pep ides we e labeled wi h iTRAQ 4-plex eagen s and analyzed using Nano-RPLC-T iple TOF
ins umen a ion o ob ain ou sepa a e compa isons om each se o biological eplica es. Compa a i e
p o eomics using he iTRAQ echnology is only capable o iden i ying p o eins ha a e exp essed in all
analyzed samples.
www.na u e.com/scien i ic epo s/
4
Scien i ic RepoR s | 5:14684 | DOi: 10.1038/s ep14684
The emaining 15 p o eins end o be mo e p e alen in na i e limbal epi helium: cadhe in 1 (CDH1),
α -enolase (ENO1), hea shock 70 kDA p o ein 1A/1B (HSP70), in eg ins α 6, β 1 and β 4 (ITGA6, ITGB1,
ITGB4), supe oxide dismu ase 1 and 2 (SOD1 and SOD2), cy oke a ins 5, 7, 8, 14, and 19, SERPINA3,
and imen in (VIM). O hese p o eins, KRT7, KRT8, KRT19, and VIM we e exp essed a lowe le els
in CECs and LECs compa ed o hPSC-LESCs. KRT5 exp ession le els we e simila o ha o KRT3, and
he emaining 11 p o eins we e ei he simila ly exp essed in all samples, o mildly o e exp essed in CECs
and LECs. Se e al key LESC ma ke s could no be iden i ied using compa a i e iTRAQ p o eomics, hus
hei exp ession in hPSC-LESCs was e i ied using low cy ome y o immuno luo escence. BMI-1 was
exp essed in 80% o hESC-LESCs and 84% o hiPSC-LESCs, p63 and TCF4 we e co-exp essed in mos
cells, and ABCG2 was localized o he cell memb anes (Fig.6b–d).
Figu e 2. O e iew o p o ein exp ession p o iles. Only he p o eins iden i ied in a leas wo o he h ee
biological eplica es we e conside ed eliable, and we e kep o u he analyses. (a,c,e,g) un il e ed esul s;
(b,d, ,h) il e ed esul s. Each do ep esen s a single iden i ied p o ein, p esen ed as mean old changes
on a loga i hmic scale, whe e y = 1 signi ies equal p o ein exp ession. CECs (black do s) and LECs
(g ay do s) compa ed o (a,b) hESC-LESCs and (c,d) hiPSC-LESCs ha e e y simila exp ession p o iles.
(e, ) Compa ison o hiPSC-LESCs and hESC-LESCs shows ha hey ha e consis en p o ein exp ession
p o iles. (g,h) Compa ison o CECs and LECs e eals a high le el o simila i y in hei exp ession p o iles.
Comple e lis o iden i ied p o eins is p o ided in Supplemen a y Table S1.
www.na u e.com/scien i ic epo s/
5
Scien i ic RepoR s | 5:14684 | DOi: 10.1038/s ep14684
Discussion
Di e en ia ion o hPSCs owa ds LESC-like cells o e s a no el and unlimi ed cell sou ce o he ea -
men o se e e ocula su ace diso de s. Howe e , be o e p oceeding u he owa ds clinical applica ions,
i is impo an o ho oughly cha ac e ize he di e en ia ed cell popula ions. We ha e p e iously shown
ha hPSC-LESCs possess he app op ia e cell mo phology, exp ess se e al key LESC ma ke s, and mos
impo an ly lack plu ipo ency ma ke s18. Ne e heless, a high- h oughpu cha ac e iza ion me hod would
be mo e in o ma i e when dealing wi h hPSC-de i ed cells. In his s udy, iTRAQ p o eomics we e used
o compa e na i e co neal and limbal epi helia ob ained om cada e ic human dono s wi h hPSC-LESCs
di e en ia ed in he absence o se um and eede cells. A o al o 860 unique p o eins exp essed in all
samples we e iden i ied, and abou 57% o hese p o eins we e p esen in a leas wo biological eplica es.
In gene al, p o ein exp ession le els we e e y simila in hESC-LESCs and hiPSC-LESCs, meaning ha
he di e en ia ion p o ocol is highly ep oducible, yielding homogeneous cell popula ions. Meanwhile,
na i e human CECs and LECs collec ed om he ocula su ace also showed simila p o ein exp ession
p o iles, sugges ing hey we e likely mixed popula ions o co neal epi helial cells a a ious s ages o
ma u i y, anging om LESCs o e minally-di e en ia ed cells. The aim o his s udy was o ocus mo e
on o e all exp ession p o iles, a he han di e ences in exp ession o speci ic p o eins. He e we high-
ligh he mos in e es ing indings ega ding s em cell and TAC beha io , limbal niche componen s and
p o eins speci ic o co neal and limbal epi helia.
LESCs, like o he issue-speci ic s em cells, end o be me abolically do man and slow cycling, g ad-
ually changing as hey ma u e owa ds TACs and u he in o e minally-di e en ia ed CECs3. In his
s udy, a ious p o eins in ol ed in cell cycling, di e en ia ion, p oli e a ion and apop osis we e iden i-
ied. Mos p o eins we e simila ly exp essed in all samples, and some p o eins a e known o be in ol ed
in mo e han one p ocess. Fo example, ANXA1, a calcium- and phospholipid- binding p o ein, has
an i-in lamma o y and an i-mig a o y p ope ies, while being in ol ed in cell di e en ia ion, cell cycling
and nega i e egula ion o apop osis25. ANXA1 was p e iously shown o be ou imes mo e p e alen
in limbal epi helium compa ed o co neal epi helium26. He e, ANXA1 was indeed exp essed a a highe
le el in LECs han CECs. Fu he mo e, ANXA1 was exp essed a simila le els in LECs and hPSC-LESCs,
sugges ing ha hPSC-LESCs a e less ma u e han CECs. Galec ins a e a amily o β -galac oside-binding
p o eins implica ed in s em cell and TAC beha io by modula ing cell-cell and cell-ma ix in e ac ions.
Th ee membe s o his amily we e iden i ied in ou da a se : LGALS1, -3 and -7. LGALS1 may egula e
apop osis, cell p oli e a ion and cell di e en ia ion, while LGALS3 is a galac ose-speci ic lec in which
binds IgE, and is known o o m a complex wi h ITGA3, ITGB1 and chond oi in sul a e p o eoglycan
Figu e 3. Dis ibu ion o iden i ied p o eins. The il e ed da a se s we e g ouped acco ding o (a) p o ein
exp ession di e ences in each o he ou compa isons, o (b) p o ein class in human CECs and LECs
compa ed o hESC-LESCs o hiPSC-LESCs. Di e ences in p o ein exp ession g ea e han 2- old we e
conside ed as biologically signi ican (highe exp ession: g ea e han 2- old; simila exp ession: be ween
-2 and 2- old; lowe exp ession: less han -2- old). Comple e lis o iden i ied p o eins is p o ided in
Supplemen a y Table S1.

www.na u e.com/scien i ic epo s/
6
Scien i ic RepoR s | 5:14684 | DOi: 10.1038/s ep14684
4 (CSPG4)27. LGALS7 is mainly exp essed in s a i ied squamous epi helia and has been shown o play
a ole in co neal epi helial cell mig a ion and e-epi helializa ion o co neal wounds28. In his s udy,
LGALS1 was ound o be exp essed a lowe le els in CECs and LECs as compa ed o hPSC-LESCs,
LGALS3 was exp essed a simila le els in all samples, and LGALS7 was o e exp essed in na i e CECs
and LECs. The gene al simila i y in exp ession o p o eins in ol ed in cell cycling, di e en ia ion, p oli -
e a ion and apop osis suppo s ou hypo hesis ha hPSC-LESCs a e simila o hei na i e coun e pa s.
LESCs a e known o eside in niche egions o he palisades o Vog a he co neoscle al junc ion.
The e a e a ious niche componen s c ucial o he p ope unc ion o LESCs, including ECM compo-
nen s and a ious cell adhesion p o eins such as cadhe ins and in eg ins29,30. Two cadhe ins (CDH1
and CHD13) and h ee in eg ins (ITGA6, ITGB1 and ITGB4) we e iden i ied in his s udy. CDH13
P o ein CEC:hESC-LESC CEC:hiPSC-LESC LEC:hESC-LESC LEC:hiPSC-LESC
Posi i e egula ion o cell p oli e a ion
CALR − 1.4 (± 0.2) 1.2 (± 1.2) − 1.1 (± 0.3) 1.3 (± 0.9)
CAPN1 1.9 (± 0.6) 1.3 (± 0.3) 1.9 (± 0.3) 1.3 (± 0.3)
CAPNS1 2.5 (± 1.0) 2.5 (± 1.4) 2.4 (± 1.1) 2.4 (± 1.2)
CDH13* 1.5 (± 0.4) 11.3 (± 15.1) 1.9 (± 0.7) 12.7 (± 16.9)
CLU 3.6 (± 1.3) 3.7 (± 0.9) 2.7 (± 1.0) 2.8 (± 0.8)
COL18A1* 5.4 (± 3.8) 2.1 (± 0.3) 7.4 (± 7.2) 2.9 (± 1.5)
EIF5A − 2.8 (± 0.2) − 2.0 (± 0.2) − 3.3 (± 0.1) − 2.5 (± 0.04)
GNAI2 1.1 (± 0.2) 1.1 (± 0.2) − 1.0 (± 0.3) − 1.3 (± 0.3)
ITGB1 1.1 (± 0.4) 1.3 (± 0.7) − 1.1 (± 0.3) 1.1 (± 0.4)
NME2* − 2.2 (± 0.1) − 2.3 (± 0.2) − 2.3 (± 0.1) − 2.7 (± 0.1)
NPM1* − 6.4 (± 0.03) − 4.9 (± 0.1) − 3.1 (± 0.2) − 3.1 (± 0.1)
PRDX3 − 1.5 (± 0.1) 1.2 (± 0.4) − 1.1 (± 0.2) 1.6 (± 0.8)
RPS15A − 3.7 (± 0.04) − 1.2 (± 0.4) − 3.1 (± 0.02) 1.1 (± 0.6)
RPS4X − 3.5 (± 0.1) − 3.3 (± 0.1) − 2.7 (± 0.2) − 2.9 (± 0.1)
RPS9 − 1.7 (± 0.3) − 2.3 (N/A) − 1.4 (± 0.5) − 2.7 (N/A)
S100A6 2.1 (± 0.9) 2.3 (± 1.1) 1.8 (± 0.6) 1.9 (± 0.8)
SSR1 2.8 (± 0.9) 2.9 (± 1.9) 2.6 (± 1.6) 2.2 (± 1.0)
TGM2 − 1.0 (± 0.3) − 1.8 (± 0.1) 1.9 (± 0.6) 1.1 (± 0.2)
Nega i e egula ion o cell p oli e a ion
ASPH 2.3 (± 0.4) 2.5 (± 0.4) 1.8 (± 0.3) 2.0 (± 0.4)
ATP5A1 1.7 (± 0.3) 1.6 (± 0.5) 1.6 (± 0.4) 1.6 (± 0.4)
CAV1 − 1.6 (± 0.2) − 2.8 (N/A) − 1.1 (± 0.1) − 2.7 (N/A)
CDH13* 1.5 (± 0.4) 11.3 (± 15.1) 1.9 (± 0.7) 12.7 (± 16.9)
COL18A1* 5.4 (± 3.8) 2.1 (± 0.3) 7.4 (± 7.2) 2.9 (± 1.5)
COMT 1.3 (± 0.7) 2.2 (± 1.4) 1.6 (± 0.8) 2.6 (± 1.5)
FTH1 1.5 (± 0.5) 1.7 (± 0.5) 1.0 (± 0.4) 1.3 (± 0.9)
GPNMB 1.1 (± 0.5) 1.1 (± 0.7) 2.7 (± 0.8) 3.2 (± 2.9)
KRT4 1.9 (± 1.5) 1.6 (± 0.4) 1.5 (± 0.9) 1.3 (± 0.1)
KRT5 2.5 (± 2.4) 2.1 (± 1.9) − 1.9 (± 0.3) − 2.4 (± 0.3)
NME2* − 2.2 (± 0.1) − 2.3 (± 0.2) − 2.3 (± 0.1) − 2.7 (± 0.1)
NPM1* − 6.4 (± 0.03) − 4.9 (± 0.1) − 3.1 (± 0.2) − 3.1 (± 0.1)
PHB − 1.4 (± 0.2) − 1.3 (± 0.2) − 1.1 (± 0.2) − 1.0 (± 0.3)
PTGES 4.3 (± 1.5) 1.5 (± 0.8) 3.3 (± 0.8) 1.1 (± 0.5)
S100A11 1.2 (± 0.5) 1.3 (± 1.6) 1.6 (± 0.7) 1.6 (± 1.4)
SFN − 1.1 (± 0.2) − 1.2 (± 0.2) − 1.0 (± 0.4) − 1.1 (± 0.2)
SOD2 − 1.2 (± 0.2) 1.1 (± 0.4) 2.0 (± 1.3) 2.3 (± 0.9)
Table 1. Cell p oli e a ion p o eins. Exp ession di e ences in human na i e CECs and LECs compa ed o
hESC-LESCs and hiPSC-LESCs, p esen ed as mean old changes ± s anda d de ia ion. As e isks (*) deno e
p o eins ha pa icipa e bo h in posi i e and nega i e egula ion o cell p oli e a ion.
www.na u e.com/scien i ic epo s/
7
Scien i ic RepoR s | 5:14684 | DOi: 10.1038/s ep14684
was o e exp essed in CECs and LECs compa ed o hiPSC-LESCs, while he emaining p o eins we e
exp essed a simila le els in all samples. CDH1 is impo an in desmosomal junc ion o ma ion and
s a i ied epi helium ans o ma ion, and was p e iously ound o be sligh ly up egula ed in LESCs com-
pa ed o CECs a he gene le el31. In eg in α 6/β 4 is a ecep o o laminin in epi helial cells and i plays
a c i ical s uc u al ole in he hemidesmosome32. Taken oge he , he wo cadhe ins and h ee in eg ins
de ec ed in all he analyzed samples e i y he epi helial na u e o hPSC-LESCs. TGFBI p o ein binds
o ype I, II, and IV collagens and may play an impo an ole in cell-collagen in e ac ions33. I is highly
exp essed in he co neal epi helium and mu a ions in he TGFBI gene a e associa ed wi h mul iple ypes
o co neal dys ophy34. This p o ein was ound o be o e exp essed in CECs, and simila ly exp essed in
LECs as compa ed o hPSC-LESCs in ou da a se . In e es ingly, despi e he lack o in e ac ion wi h o he
cell ypes and simpli ied cul u e en i onmen in which hPSC-LESCs a e main ained, he o e all le els o
cell adhesion p o ein exp ession we e simila o ha in na i e CECs and LECs.
Ano he LESC niche componen a e he S100A p o eins, known o be in ol ed in he egula ion o
many cellula p ocesses such as calcium homeos asis, cy oskele on o ganiza ion, s ess esponse, cell
mo ili y, p oli e a ion and di e en ia ion35. Se e al S100A p o eins, including S100A6, S100A10 and
S100A11 bind annexins and a e in ol ed in cell memb ane o ganiza ion, ion channel modula ion and
ke a inocy e di e en ia ion35. Common ocula su ace diseases such as d y eye, p e ygium and co -
neal angiogenesis in ol e S100 amily o p o eins, pa icula ly S100A8 and S100A936. Mos impo an ly,
S100A4 and A9 p o eins ha e been ound o be po en ma ke s o limbal epi helial c yp cells, wi h a
p obable in ol emen in cell di e en ia ion, egula ion o g ow h and cellula s uc u e37. S100A8 has
also been iden i ied as a pu a i e LESC ma ke 26. He e in ou s udy, S100A8 and A9 we e exp essed a
highe le els in LECs, and a lowe le els in CECs, as compa ed o hPSC-LESCs. These esul s sugges
ha hESC-LESCs and hiPSC-LESCs a e mo e ma u e han LECs, bu less ma u e han CECs.
The co nea is in cons an con ac wi h ai and ex e nal en i onmen , and he e o e highly exposed
o UV- adia ion and a ious pa hogens. Consequen ly, he ocula su ace equi es p o ec ion agains he
Figu e 4. P o eins in ol ed in cell cycling, di e en ia ion and apop osis. Exp ession di e ences o (a) cell
cycle p o eins, (b) p o eins in ol ed in cell di e en ia ion, o (c) apop osis, in na i e CECs (black ma ke s)
and LECs (g ay ma ke s) as compa ed o hESC-LESCs and hiPSC-LESCs. Each ma ke ep esen s a single
unique p o ein, p esen ed as mean old changes on a loga i hmic scale, whe e y = 1 signi ies equal p o ein
exp ession. Lis s o p o eins and hei exp ession in nume ic alues a e p o ided in Supplemen a y Table S2.
www.na u e.com/scien i ic epo s/
8
Scien i ic RepoR s | 5:14684 | DOi: 10.1038/s ep14684
Figu e 5. Niche componen s o he ocula su ace. Exp ession di e ences o (a) cell adhesion p o eins,
(b) p o eins in ol ed in immune esponse, (c) S100A p o eins, (d) an ioxidan s, and (e) angiogenic p o eins,
in na i e CECs (black ma ke s) and LECs (g ay ma ke s) as compa ed o hESC-LESCs and hiPSC-LESCs.
Each ma ke ep esen s a single unique p o ein, p esen ed as mean old changes on a loga i hmic scale,
whe e y = 1 signi ies equal p o ein exp ession. Lis s o p o eins and hei exp ession in nume ic alues a e
p o ided in Supplemen a y Table S3.
Figu e 6. Co neal and limbal ma ke s. (a) Exp ession di e ences in na i e CECs (black ma ke s)
and LECs (g ay ma ke s) as compa ed o hESC-LESCs and hiPSC-LESCs. Each ma ke ep esen s a
single unique p o ein, p esen ed as mean old changes on a loga i hmic scale, whe e y = 1 signi ies equal
p o ein exp ession. Lis o p o eins and hei exp ession in nume ic alues a e p o ided in Supplemen al
Table S4. (b) P o ein exp ession o BMI-1 in hPSC-LESCs quan i ied by low cy ome y. Whi e his og ams
ep esen iso ype con ols, and black his og ams ep esen BMI-1 s aining. (c) Rep esen a i e images o
TCF4 and p63 p o ein exp ession in hiPSC-LESCs, isualized using immuno luo escence.
(d) Rep esen a i e image o ABCG2 p o ein exp ession in hiPSC-LESCs. Scale ba s 50 μ m.
www.na u e.com/scien i ic epo s/
9
Scien i ic RepoR s | 5:14684 | DOi: 10.1038/s ep14684
o ma ion o eac i e oxygen species (ROS) in he o m o p o eins wi h an ioxidan p ope ies, as well
as p o eins in ol ed in immune esponse. Ou da a se con ains 9 an ioxidan s, which include pe oxi e-
doxins and supe oxide dismu ases, and 19 immune esponse p o eins, including in e leukin enhance
binding ac o 2 (ILF2), IGHA1, IGLC2 and MIF. The ocula su ace is known o be a e sa ile ba -
ie o inna e immuni y, in ol ing CECs, ib oblas s and Lange hans cells38. In his s udy, p o ein was
ex ac ed om cell pelle s, and he e o e cy okines and g ow h ac o s known o be sec e ed by he co -
neal and limbal epi helia, including a ious in e leukins and chemokines such as chemokine (C-C mo i )
ligand 5 (CCL5), chemokine (C-X-C mo i ) ligand 2 (CXCL2), o glial cell-de i ed neu o ophic ac o
(GDNF)31,38,39, we e no de ec ed. In addi ion, he e is e idence sugges ing ha hPSC-de i ed cells pos-
sess a ce ain deg ee o immune p i ilege40,41, which would also explain he sca ci y o iden i ied immune
esponse p o eins. O e all, conside ing ha hPSC-LESCs cul u ed in i o a e unlikely o su e om
exposu e o pa hogens, as well as UV- adia ion and he esul ing ROS, i is no su p ising ha ela i ely
ew p o ec i e p o eins we e de ec ed. Ano he impo an componen o he limbal niche is he blood
essel ne wo k o he limbal s oma, which con ibu es o angiogenesis du ing wound healing42. LESCs
ha e been shown o exp ess bo h p o- and an i-angiogenic genes, while limbal s omal niche cells possess
he abili y o di e en ia e owa ds angiogenesis p ogeni o s and p e en co neal epi helial di e en ia-
ion31,42. Only 10 p o eins ha may be in ol ed in angiogenesis we e iden i ied in ou da a se , which is
no su p ising, as he in i o cell cul u e has no su ounding issues and blood essels o in e ac wi h,
making angiogenesis unnecessa y and hence down- egula ing he angiogenic ma ke s.
Dy lund e al. ha e ca ied ou an ex ensi e p o eomic analysis o all h ee laye s in he human co -
nea: hey iden i ied 2737 p o eins in he co neal epi helium, and quan i ied ei he 110 o 663 p o eins,
depending on he quan i ica ion me hod43. Ou o he 55 mos abundan p o eins iden i ied in hei
s udy o he co neal epi helium, we de ec ed 44 (Supplemen a y Table S5). Cy oke a ins 3 and 12, as
well as MUC16, ALDH3A1, DCN and TGFBI ha e been p e iously desc ibed o be mo e p e alen
in ma u e CECs a he han LESCs31,44,45. The esul s o ou s udy a e in line wi h his: hese p o eins
we e de ec ed a highe le els in CECs han in LECs. Mo eo e , exp ession p o iles o KRT3 and KRT12
sugges ha hPSC-LESCs a e mo e ma u e han LECs, bu less ma u e han CECs. Cy oke a ins 5, 7, 8,
14, and 19, HSP70, SERPINA3, and VIM ha e been de ec ed in p ima y LESCs in se e al s udies45–47.
KRT8 is known o be co-localized wi h KRT15 and VIM, and may be exp essed in ac i a ed basal cells
ha a e eady o di ide and di e en ia e48. HSP70 has been epo ed o be in ol ed in LEC p oli e a ion,
di e en ia ion and mig a ion, co-localizing wi h p6347. In ou s udy, KRT7, KRT8, KRT19 and VIM we e
exp essed a lowe le els in CECs and LECs compa ed o hPSC-LESCs, sugges ing ha hPSC-LESCs may
be mo e homogeneous cell popula ions han he na i e CECs and LECs.
Rela i e p o eomics using he iTRAQ echnology is only capable o iden i ying p o eins ha a e
exp essed in all analyzed samples. The e o e, iTRAQ me hods a e known o wo k bes when he di e -
ences be ween samples a e minimal, o ins ance in ime-poin s udies. In his s udy we compa ed na i e
issue om h ee di e en human dono s wi h LESC-like cells di e en ia ed om hESCs and hiPSCs.
In his case, he di e ences be ween he samples a e almos ce ainly high, esul ing in a ai ly low
amoun o p o eins iden i ied in all samples, hus making he compa ison mo e di icul . Ne e heless,
we can clea ly see ha all samples ha e simila p o ein exp ession p o iles, indica ing good eliabili y o
he analysis me hod. Se e al pu a i e LESC ma ke s, mos impo an ly p63 and ABCG2 could no be
iden i ied by means o iTRAQ, meaning ha hey we e absen om a leas one o he samples. Mos
likely, hese ma ke s we e no p esen in CECs in de ec able amoun s, as he cen al co neal epi helium
mainly con ains ma u e cells. The e o e, we e i ied p o ein exp ession o se e al pu a i e LESC ma ke s
ha we e no iden i ied in he p o eomics analysis, namely p63, TCF4, ABCG2 and BMI-1, using low
cy ome y and immuno luo escence. Bo h hESC-LESCs and hiPSC-LESCs we e shown o exp ess hese
p o eins, con i ming ha hey do possess LESC-like cha ac e is ics.
O e all, i is di icul o d aw conclusions ega ding he di e en ia ion s a e o hPSC-LESCs. We
pos ula e ha he na i e human samples collec ed om he ocula su ace we e ai ly he e ogeneous,
con aining sub-popula ions o cells in a a ie y o di e en ia ion s a es. Fu he mo e, a ia ion be ween
dono s likely a ec s he ou comes o he s udy. Signi ican in e -dono a ia ion was also obse ed in a
mic oa ay s udy using p ima y cells collec ed om he ocula su ace e en a e unde going expansion
in i o31. In con as , se e al high- h oughpu mic oa ay s udies ha e shown subs an ial di e ences in
gene exp ession be ween pu i ied popula ions o CECs and LESCs6,31,49,50. Using human cells ha ha e
no been expanded in i o has i s ad an ages and disad an ages. On he one hand, hey ep esen he in
i o s a e be e , wi hou he isk o p o eome al e a ion caused by p olonged cul i a ion. On he o he
hand, i is di icul o ob ain a pu e popula ion o he desi ed cell ype, especially knowing ha only a
small pe cen age o cells in he limbal egion a e ac ual s em cells–abou 5% o cells in he human limbus
a e conside ed LESCs, while he es a e likely TACs a a ious s ages o di e en ia ion51. I emains o
be seen whe he pu i ica ion o ansplan able cell popula ions ia cell so ing becomes a quali y equi e-
men o clinical use. In ac , i may be bene icial o include LESCs, TACs and CECs, o be e mimic he
ange o cell ypes a he ocula su ace.
In conclusion, he esul s o his s udy show ha hPSC-LESCs exp ess a leas 485 p o eins ound
in na i e ocula su ace epi helia. Mos impo an ly, p o eins in ol ed in cell cycling, p oli e a ion, di -
e en ia ion and apop osis, a ious LESC niche componen s, and 21 co neal and limbal ma ke s we e
iden i ied. The o e all p o ein exp ession p o iles o hESC-LESCs and hiPSC-LESCs sugges ha he