cance s
A icle
Immunopheno ype o Gas ic Tumo s Un eils a Pleio opic
Role o Regula o y T Cells in Tumo De elopmen
Sa a Rocha 1,2,3 , A onso P Bas o 4,5, Ma ieke E Ijssels eijn 6, Sa a P Teles 1,2 , Ma ia M Aze edo 1,
Gilza Gonçal es 7, I ene Gullo 1,2,7,8 , Gab iela M Almeida 1,2,7 , Joaquín J Maqueda 1,2 , Ma a I Oli ei a 9,
Fá ima Ca nei o 1,2,7,8 , João T Ba a a 4, Luís G aça 4,5, Noel F C C de Mi anda 6, Joana Ca alho 1,2
and Ca la Oli ei a 1,2,7,*


Ci a ion: Rocha, S.; Bas o, A.P.;
Ijssels eijn, M.E.; Teles, S.P.; Aze edo,
M.M.; Gonçal es, G.; Gullo, I.;
Almeida, G.M.; Maqueda, J.J.;
Oli ei a, M.I.; e al.
Immunopheno ype o Gas ic Tumo s
Un eils a Pleio opic Role o
Regula o y T Cells in Tumo
De elopmen . Cance s 2021,13, 421.
h ps://doi.o g/10.3390/cance s
13030421
Academic Edi o :
F ancois Ghi inghelli
Recei ed: 10 Decembe 2020
Accep ed: 20 Janua y 2021
Published: 23 Janua y 2021
Publishe ’s No e: MDPI s ays neu al
wi h ega d o ju isdic ional claims in
published maps and ins i u ional a il-
ia ions.
Copy igh : © 2021 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
1i3S—Ins i u o de In es igação e Ino ação em Saúde, Uni e sidade do Po o, 4200-135 Po o, Po ugal;
[email p o ec ed] (S.R.); sa a. eles@c uk.cam.ac.uk (S.P.T.); [email p o ec ed] (M.M.A.);
[email p o ec ed] (I.G.); [email p o ec ed] (G.M.A.); [email p o ec ed] (J.J.M.);
[email p o ec ed] (F.C.); [email p o ec ed] (J.C.)
2
Ipa imup—Ins i u e o Molecula Pa hology and Immunology o Uni e si y o Po o, 4200-135 Po o, Po ugal
3
Doc o al P og am on Cellula and Molecula Bio echnology Applied o Heal h Sciences, ICBAS—Ins i u o de
Ciências Biomédicas Abel Salaza , Uni e sidade do Po o, 4050-313 Po o, Po ugal
4iMM—Ins i u o de Medicina Molecula João Lobo An unes, Faculdade de Medicina da Uni e sidade de
Lisboa, 1649-028 Lisbon, Po ugal; abas o@ m .ulisboa.p (A.P.B.); [email p o ec ed] (J.T.B.);
[email p o ec ed] (L.G.)
5Ins i u o Gulbenkian de Ciência, 2780-156 Oei as, Po ugal
6Depa men o Pa hology, Leiden Uni e si y Medical Cen e , 2333 ZA Leiden, The Ne he lands;
[email p o ec ed] (M.E.I.); [email p o ec ed] (N.F.C.C.d.M.)
7Depa men o Pa hology, Facul y o Medicine o he Uni e si y o Po o (FMUP), 4200-319 Po o, Po ugal;
[email p o ec ed]
8Depa men o Pa hology, Cen o Hospi ala Uni e si á io de São João, 4200-319 Po o, Po ugal
9In e na ional Ibe ian Nano echnology Labo a o y, 4715-330 B aga, Po ugal; [email p o ec ed]
*Co espondence: [email p o ec ed]; Tel.: +351-225-570-785
Simple Summa y:
The ole o egula o y T cells (T egs) in gas ic cance (GC) is s ill con o e sial
and poo ly unde s ood. GC pa ien s ha e inc eased numbe s o T egs in pe iphe al blood and among
umo in il a ing lymphocy es; howe e , hei p ognos ic alue depends on speci ic umo ea u es
(e.g., umo loca ion and/o mic osa elli e ins abili y s a us). We ound ha T egs migh induce
memb ane exp ession o IL2R
α
in in es inal- ype GC cells, which associa es wi h MAPK signaling
pa hway ac i a ion and sphe oid g ow h. Mo eo e , T egs accumula e a ea ly s eps o in es inal- ype
GCs p og ession, when umo s a e s a ing o g ow h ough he s omach wall, and do no p esen
ascula and pe ineu al in asion. Ou indings sugges a no el non-immunosupp essi e ole o T eg
cells in in es inal- ype GC, which may unlock no el he apeu ic immuno-oncology s a egies o
in es inal- ype GC o o he umo s wi h simila immune con ex .
Abs ac :
Gas ic cance (GC) pa ien s display inc eased egula o y T cell (T egs) numbe s in pe-
iphe al blood and among umo -in il a ing lymphocy es. Ne e heless, he ole o T egs in GC
p og ession emains con o e sial. He e, we sough o explo e he impac o T egs in GCs wi h
dis inc his ology, and whe he T egs can di ec ly in luence umo cell beha io and GC de elop-
men . We pe o med a comp ehensi e immunopheno yping o 82 human GC cases, h ough an
in eg a ed analysis o mul ispec al immuno luo escence de ec ion o T cells ma ke s and pa ien
clinicopa hological da a. Mo eo e , we de eloped 3D
in i o
co-cul u es wi h T egs and umo cells
ha we e ollowed by high- h oughpu and ligh -shee imaging, and hei biological ea u es s udied
wi h con en ional/imaging low cy ome y and Wes e n blo ing. We showed ha T egs loca ed a
he umo nes we e equen in in es inal- ype GCs bu did no associa e wi h inc eased le els o
e ec o T cells. Ou
in i o
esul s sugges ed ha T egs p e e en ially in il a ed in es inal- ype GC
sphe oids, induced he exp ession o IL2R
α
and ac i a ion o MAPK signaling pa hway in umo
cells, and p omo ed sphe oid g ow h. Accumula ion o T egs in in es inal- ype GCs was inc eased a
ea ly s ages o he s omach wall in asion and in he absence o ascula and pe ineu al in asion. In
Cance s 2021,13, 421. h ps://doi.o g/10.3390/cance s13030421 h ps://www.mdpi.com/jou nal/cance s
Cance s 2021,13, 421 2 o 20
his s udy, we p oposed a non-immunosupp essi e mechanism h ough which T egs migh di ec ly
modula e GC cells and he eby p omo e umo g ow h. Ou indings hold insigh ul implica ions o
he apeu ic s a egies a ge ing in es inal- ype GCs and o he umo s wi h simila immune con ex .
Keywo ds: umo -in il a ing CD4 T cells; egula o y T cells; molecula egula ion; gas ic cance
1. In oduc ion
Recen ad ances on he molecula aspec s o gas ic cance (GC) ha e p o ided in alu-
able knowledge ha led o he iden i ica ion o new ac ionable a ge s and he apies [
1
].
Ta ge ed ea men s and immuno he apies a e mainly o e ed o ad anced cance pa ien s
and ha e, so a , minimally imp o ed GC p ognosis, jus i ying why GC anks as he hi d
leading cause o cance - ela ed dea hs wo ldwide [
2
]. Gas ic umo s wi h high mu a ion
load, mic osa elli e ins abili y (MSI) o posi i e o Eps ein–Ba i al in ec ion a e po en-
ially immunogenic, and hus amenable o immuno he apy based on checkpoin inhibi o s
(e.g., an i- PD-1/PD-L1) [
3
,
4
]. Finding addi ional con ex s o immuno- ela ed he apies is
an oppo uni y ha is wo h in es iga ing u he .
Regula o y T cells (T egs) a e CD4
+
T cells cha ac e ized by he su ace exp ession o
IL2R
α
, and nuclea exp ession o he ansc ip ion ac o FoxP3. T egs a e endowed wi h
immunosupp essi e ac i i y ha en o ces pe iphe al ole ance and main ains immunologi-
cal homeos asis [
5
]. Howe e , in cance , he immunosupp essi e en i onmen p omo ed by
T egs wi hholds he an i umo immune esponse, hence p omo ing umo p og ession and
dissemina ion [
6
,
7
]. In GC pa ien s, pe iphe al blood and umo -in il a ing lymphocy es
a e en iched in T egs [
8
–
12
], which associa es wi h inc eased umo s age, poo p ognosis
and educed pa ien su i al [
13
–
17
]. Ne e heless, o he s udies ha e shown ha umo
in il a ing T egs may also be associa ed wi h a o able p ognosis, speci ically in pa ien s
ca ying umo s om he ca dia o wi h MSI [
18
–
20
]. Hence, he ole o T egs in GC p o-
g ession emains poo ly unde s ood and highly con o e sial. Fu he , i is also unexplo ed
in GC whe he T egs may p omo e umo p og ession ia non-immunological mechanisms,
as desc ibed in o he cance models [
21
,
22
]. Thus, we sough o explo e he con ibu ion o
T egs in GCs wi h dis inc his ology, and whe he T egs can di ec ly impac umo cells o
p omo e GC p og ession.
To add ess his, we in eg a ed he esul s o T cell immunopheno yping analysis wi h
clinicopa hological ea u es o 82 GC pa ien s. This analysis un eiled an en ichmen o
T egs speci ically in in es inal- and inde e mina e- ype GC, as compa ed o di use- ype GC.
Fu he mo e, ou da a sugges s ha a popula ion o T egs is p esen a he umo nes o
in es inal- ype GC independen ly o he p e alence o e ec o T cells. Gi en hese esul s,
and o unde s and whe he T egs may ac i ely modula e he pheno ype o umo cells, we
es ablished and explo ed 3D co-cul u es o T egs wi h in es inal- o di use- ype GC cell
lines. We ound ha T egs ac i ely in il a e in es inal- ype GC sphe oids. Upon co-cul u e
wi h T egs, in es inal- ype GC cells acqui e exp ession o IL2R
α
a he cell memb ane,
ha e inc eased ac i a ion o MAPK signaling pa hway and sphe oid g ow h. Fu he mo e,
we ound an en ichmen o T egs in ea ly-s age in es inal- ype GC, and in he absence o
ascula and pe ineu al in asion.
Al oge he , ou da a sugges s a di ec e ec o T egs on umo cells ha may be
pa icula ly impo an in ea ly s ages o in es inal- ype GC p og ession.
2. Resul s
This sec ion may be di ided by subheadings. I should p o ide a concise and p ecise
desc ip ion o he expe imen al esul s, hei in e p e a ion as well as he expe imen al
conclusions ha can be d awn.
Cance s 2021,13, 421 3 o 20
2.1. In es inal-Type GC Main ain a Popula ion o T egs a he Tumo Nes Independen ly on he
P e alence o E ec o T Cells
To in es iga e he immune T cell landscape wi hin GC, we pe o med immunophe-
no yping o 82 pa ien samples. Each umo sec ion was s ained simul aneously o a
panel o se en ma ke s, including CD3, CD8, and FoxP3 T cell-associa ed ma ke s, cy-
oke a in o label epi helial umo cells and DAPI o nuclei de ec ion (Figu e 1a,b). We
ocused he analysis on h ee majo pheno ypes, de ined as ollows: (1) Helpe T cells,
CD3
+
CD8
-
FoxP3
-
cells (Figu e 1c); (2) Cy o oxic T cells, CD3
+
CD8
+
cells (Figu e 1d); and
(3) T egs, CD3
+
CD8
-
FoxP3
+
cells (Figu e 1e). Cells exp essing cy oke a in we e excluded
when building he T cell p o iles. Poo pe o mance o an i-IL2R
α
an ibodies in he im-
muno luo escence s aining hampe ed he analysis o IL2Rαexp ession in issue samples.
Cance s 2021, 13, x FOR PEER REVIEW 4 o 23
Figu e 1. Cha ac e iza ion o he immune T cell landscape in gas ic cance (GC) issue sec ions. (a,b) Rep esen a i e com-
posi e and single-s aining immuno luo escence images o he wo main his ological GC ypes, in es inal- ype (a) and di -
use- ype (b), ob ained by mul ispec al imaging. Scale ba : 100 μm. (c–e) Main T cell popula ions iden i ied in he GC
mic oen i onmen based on he posi i i y o CD3, CD8, and FoxP3 cell ma ke s. ( ) Rela i e numbe s o helpe , cy o oxic
T cells and T egs, acco ding o he his ologic p ope ies o he umo : in es inal- ype, n = 41; di use- ype, n = 15; mixed-
ype, n = 11; inde e mina e- ype, n = 15 pa ien s. Box and whiske s ep esen median ± 10 o 90 pe cen ile. * p < 0.05, ** p <
0.01; K uskal–Wallis es wi h Dunn’s mul iple compa ison es .
We u he assessed whe he hese di e ences could be explained h ough a di e -
en ial p e alence o MSI cases wi hin each his o ype. Whils 37% and 40% o in es inal-
and inde e mina e- ype GC cases we e MSI, only 13% and 18% o di use- and mixed- ype
GC had his pheno ype (Figu e 2a). Gi en hese di e ences, we e-assessed he dis ibu-
ion o T cell popula ions conside ing bo h GC his o ype and MSI s a us (Figu e 2b– ). We
ound ha inde e mina e- ype MSS cases ha e mo e helpe , cy o oxic cells and T egs as
compa ed o in es inal- ype MSS cases (p = 0.0261, p = 0.0349 and p = 0.0026, espec i ely);
display inc eased numbe s o helpe T cells and T egs in compa ison o di use- ype MSS
(p = 0.0297 and p = 0.0002, espec i ely); and inc eased numbe o T egs in compa ison o
mixed- ype MSS (p = 0.0065; Figu e 2b–d). Al hough, inde e mina e- ype MSS cases had
inc eased numbe o T egs as compa ed o MSI cases (p = 0.0444; Figu e 2d), he T eg/cy-
o oxic T cell and T eg/helpe T cell a ios we e no signi ican ly changed (Figu e 2e, ).
O e all, hese esul s sugges ed ha inde e mina e- ype MSS cases elici s onge immune
esponses, as compa ed o he o he his o ypes, and ha he inc eased accumula ion o
Figu e 1.
Cha ac e iza ion o he immune T cell landscape in gas ic cance (GC) issue sec ions. (
a
,
b
) Rep esen a i e
composi e and single-s aining immuno luo escence images o he wo main his ological GC ypes, in es inal- ype (
a
) and
di use- ype (
b
), ob ained by mul ispec al imaging. Scale ba : 100
µ
m. (
c
–
e
) Main T cell popula ions iden i ied in he GC
mic oen i onmen based on he posi i i y o CD3, CD8, and FoxP3 cell ma ke s. (
) Rela i e numbe s o helpe , cy o oxic T
cells and T egs, acco ding o he his ologic p ope ies o he umo : in es inal- ype, n= 41; di use- ype, n= 15; mixed- ype,
n= 11; inde e mina e- ype, n= 15 pa ien s. Box and whiske s ep esen median
±
10 o 90 pe cen ile. * p< 0.05, ** p< 0.01;
K uskal–Wallis es wi h Dunn’s mul iple compa ison es .
Cance s 2021,13, 421 4 o 20
To disclose a po en ial associa ion be ween T cell pheno ypes and GC his ology, we
e alua ed he densi y o each cell popula ion acco ding o GC his ological ypes (Figu e 1 ;
Supplemen a y File S1). We obse ed ha whe eas he p esence o helpe and cy o oxic T
cells only sligh ly changed ac oss GC his ological ypes, T egs we e signi ican ly en iched
in in es inal- and inde e mina e- ype GC, as compa ed o di use- ype GC (p= 0.0175 and
p= 0.0028, espec i ely; Figu e 1g).
We u he assessed whe he hese di e ences could be explained h ough a di e en-
ial p e alence o MSI cases wi hin each his o ype. Whils 37% and 40% o in es inal- and
inde e mina e- ype GC cases we e MSI, only 13% and 18% o di use- and mixed- ype GC
had his pheno ype (Figu e 2a). Gi en hese di e ences, we e-assessed he dis ibu ion o
T cell popula ions conside ing bo h GC his o ype and MSI s a us (Figu e 2b– ). We ound
ha inde e mina e- ype MSS cases ha e mo e helpe , cy o oxic cells and T egs as compa ed
o in es inal- ype MSS cases (p= 0.0261, p= 0.0349 and p= 0.0026, espec i ely); display
inc eased numbe s o helpe T cells and T egs in compa ison o di use- ype MSS (p= 0.0297
and p= 0.0002, espec i ely); and inc eased numbe o T egs in compa ison o mixed- ype
MSS (p= 0.0065; Figu e 2b–d). Al hough, inde e mina e- ype MSS cases had inc eased
numbe o T egs as compa ed o MSI cases (p= 0.0444; Figu e 2d), he T eg/cy o oxic T cell
and T eg/helpe T cell a ios we e no signi ican ly changed (Figu e 2e, ). O e all, hese
esul s sugges ed ha inde e mina e- ype MSS cases elici s onge immune esponses, as
compa ed o he o he his o ypes, and ha he inc eased accumula ion o T egs was likely
he esul o an accumula ion o e ec o T cell popula ions. As o in es inal- ype GC, MSS
cases displayed signi ican ly lowe numbe s o cy o oxic T cells (p= 0.0014), bu no o
T egs, as compa ed o MSI cases (Figu e 2c,d). This led o a highe T eg/cy o oxic T cell
a io in in es inal- ype MSS cases compa ing wi h MSI cases (p= 0.0414; Figu e 2 ).
To u he dissec his unbalanced a io in in es inal- ype GC, we cha ac e ized he
dis ibu ion o T cell popula ions ega ding hei loca ion a he s oma o umo nes a eas
(Figu e 2g–j; Supplemen a y Figu e S1). We obse ed ha he inc eased T eg/cy o oxic
T cell a io in MSS umo s was main ained bo h a he s oma and umo nes egions
(
p= 0.0274
and p= 0.0109, espec i ely; Figu e 2g,h). O no ice, se e al MSS in es inal- ype
umo s showed pa icula ly high numbe s o T egs compa ing o he numbe s o cy o oxic
T cells, a he umo nes (Figu e 2h). Mo eo e , we obse ed ha he densi y o T egs was
compa able a he umo nes o MSS and MSI in es inal- ype GCs, while he densi y o
cy o oxic T cells was only inc eased a he umo nes o MSI umo s (Figu e 2i,j). These
obse a ions may indica e ha he cy o oxic T cell densi y is no being accompanied by
an inc eased T eg densi y a he umo nes , as obse ed o he s oma egion. This da a
u he sugges s ha bo h MSS and MSI cases ha e a popula ion o T egs, in he close
icini y o umo cells, ha does no seem o ha e a pu ely immunosupp ession ole, as
hei densi y does no accompany he inc ease in cy o oxic T cells.
Al oge he , hese obse a ions suppo he hypo hesis ha T egs may ha e a non-
immunosupp essi e ac i i y by di ec ly a ec ing GC umo cells wi h in es inal his ology.
Cance s 2021, 13, x FOR PEER REVIEW 6 o 23
Figu e 2. Dis ibu ion o T cell popula ions in MSS and mic osa elli e ins abili y (MSI) GC umo s. (a) Rela i e and absolu e
equency o MSS and MSI cases o each GC his ological ype. (b– ) Rela i e numbe s o helpe (b), cy o oxic T cells (c),
T egs (d), and T egs no malized o helpe T cells (e) and cy o oxic T cells ( ), acco ding o he MSI s a us and umo
his ology. (g,h) Rela i e numbe o T egs no malized o cy o oxic and helpe T cells wi hin he s oma (g) and umo nes
(h) a eas o in es inal- ype GC cases, acco ding o MSI s a us. Rela i e numbe o T egs, cy o oxic T cells, and helpe T
cells wi hin he s oma (i) and umo nes (j) a eas o in es inal- ype GC cases, acco ding o MSI s a us. See Supplemen a y
Figu e S1 o a isual desc ip ion o s oma and umo nes a eas. Box and whiske s ep esen median ± 10 o 90 pe cen ile.
* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Two-way ANOVA wi h Tukey’s/Sidak’s mul iple compa isons es (b– )
and Mann–Whi ney U es (g–j).
Figu e 2. Con .
Cance s 2021,13, 421 5 o 20
Cance s 2021, 13, x FOR PEER REVIEW 6 o 23
Figu e 2. Dis ibu ion o T cell popula ions in MSS and mic osa elli e ins abili y (MSI) GC umo s. (a) Rela i e and absolu e
equency o MSS and MSI cases o each GC his ological ype. (b– ) Rela i e numbe s o helpe (b), cy o oxic T cells (c),
T egs (d), and T egs no malized o helpe T cells (e) and cy o oxic T cells ( ), acco ding o he MSI s a us and umo
his ology. (g,h) Rela i e numbe o T egs no malized o cy o oxic and helpe T cells wi hin he s oma (g) and umo nes
(h) a eas o in es inal- ype GC cases, acco ding o MSI s a us. Rela i e numbe o T egs, cy o oxic T cells, and helpe T
cells wi hin he s oma (i) and umo nes (j) a eas o in es inal- ype GC cases, acco ding o MSI s a us. See Supplemen a y
Figu e S1 o a isual desc ip ion o s oma and umo nes a eas. Box and whiske s ep esen median ± 10 o 90 pe cen ile.
* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Two-way ANOVA wi h Tukey’s/Sidak’s mul iple compa isons es (b– )
and Mann–Whi ney U es (g–j).
Figu e 2.
Dis ibu ion o T cell popula ions in MSS and mic osa elli e ins abili y (MSI) GC umo s. (
a
) Rela i e and absolu e
equency o MSS and MSI cases o each GC his ological ype. (
b
–
) Rela i e numbe s o helpe (
b
), cy o oxic T cells (
c
),
T egs (
d
), and T egs no malized o helpe T cells (
e
) and cy o oxic T cells (
), acco ding o he MSI s a us and umo his ology.
(
g
,
h
) Rela i e numbe o T egs no malized o cy o oxic and helpe T cells wi hin he s oma (
g
) and umo nes (
h
) a eas o
in es inal- ype GC cases, acco ding o MSI s a us. Rela i e numbe o T egs, cy o oxic T cells, and helpe T cells wi hin he
s oma (
i
) and umo nes (
j
) a eas o in es inal- ype GC cases, acco ding o MSI s a us. See Supplemen a y Figu e S1 o a
isual desc ip ion o s oma and umo nes a eas. Box and whiske s ep esen median
±
10 o 90 pe cen ile.
*p< 0.05
,
** p< 0.01, *** p< 0.001, **** p< 0.0001. Two-way ANOVA wi h Tukey’s/Sidak’s mul iple compa isons es (
b
–
) and
Mann–Whi ney U es (g–j).
2.2. T egs Ac i ely In il a e In es inal-Type GC Sphe oids
To sc u inize he po en ial in e ac ions be ween T egs and umo cells, we es ablished
di ec 3D
in i o
co-cul u es o T cells, isola ed om he pe iphe al blood o heal hy dono s,
wi h GC cell lines (in es inal- ype MKN74, and di use- ype MKN45). Gi en he impossi-
bili y o using in acellula FoxP3 o cell so ing, we isola ed CD3
+
CD4
+
CD127
−
IL2R
α+
T cells, which a e highly en iched in T egs, om he pe iphe al blood o heal hy dono s
(Supplemen a y Figu e S2). As a con ol, CD3
+
CD4
+
CD127
+
IL2R
α−
cells (con en ional
CD4 T cells) we e also collec ed om he same dono s.

Cance s 2021,13, 421 6 o 20
T cells (ei he T egs o con en ional T cells) we e added o GC sphe oids ecapi ula ing
in es inal- o di use- ype GC (Figu e 3a). Co-cul u es, imaged wi h ime-lapse o 48 h, e i-
denced dis inc T eg in il a ion capaci ies depending on GC his ology (
Figu e 3b,c
). Whils
T egs accumula ed inside he in es inal- ype GC sphe oids (Figu e 3b), hey emained
p e e en ially a he pe iphe y o di use- ype sphe oids (Figu e 3c). In bo h si ua ions,
he accumula ion o T egs was p opo ional o he numbe o T egs in he co-cul u e
(
Figu e 3b,c
). As a as con en ional T cells a e conce ned, accumula ion ei he inside o a
he pe iphe y o sphe oids was e y low as compa ed o ha o T egs (
Figu e 3d,e
). These
obse a ions suppo he indings om pa ien s’ umo s and ein o ce speci ic c oss alk be-
ween T egs and in es inal- ype GC cells. Ac i e in il a ion o in es inal- ype GC sphe oids
by T egs was u he alida ed h ough a ou -angle analysis o he sphe oids a dis inc
ime-poin s (Figu e 3 –l; Supplemen a y Figu e S3; Supplemen a y Videos S1–S3). This
analysis e ealed ha only a small ac ion o T egs was ac i ely in il a ing he sphe oids
(Figu e 3 –l), ecapi ula ing he p opo ion o T eg in il a ion obse ed in pa ien s’ in es i-
nal umo s (Figu e 1a).
Cance s 2021, 13, x FOR PEER REVIEW 8 o 23
Figu e 3. T cell in il a ion o GC sphe oids. (a) Schema ic ep esen a ion o he co-cul u e imeline. Sphe oids (g een)
esembling in es inal- and di use- ype GC we e independen ly cul u ed o 7 days. The ea e , T egs and con en ional T
cells ( ed) we e so ed om pe iphe al blood o heal hy dono s, based on he exp ession o CD3, CD4, IL2Rα, and CD127
T cell ma ke s, and added o GC sphe oids. A e 48 h o co-cul u e, sphe oids we e dissocia ed o u he cha ac e iza ion
o bo h T cells and GC cells. (b–e) Moni o iza ion o GC sphe oids-T cell in e ac ions du ing he 48 h o co-cul u e by ime-
lapse mic oscopy. Con ol ep esen s GC sphe oids (g een) ha we e no cul u ed wi h T cells ( ed). 1:1, 1:5 and 1:15 (GC
Figu e 3. Con .
Cance s 2021,13, 421 7 o 20
Cance s 2021, 13, x FOR PEER REVIEW 8 o 23
Figu e 3. T cell in il a ion o GC sphe oids. (a) Schema ic ep esen a ion o he co-cul u e imeline. Sphe oids (g een)
esembling in es inal- and di use- ype GC we e independen ly cul u ed o 7 days. The ea e , T egs and con en ional T
cells ( ed) we e so ed om pe iphe al blood o heal hy dono s, based on he exp ession o CD3, CD4, IL2Rα, and CD127
T cell ma ke s, and added o GC sphe oids. A e 48 h o co-cul u e, sphe oids we e dissocia ed o u he cha ac e iza ion
o bo h T cells and GC cells. (b–e) Moni o iza ion o GC sphe oids-T cell in e ac ions du ing he 48 h o co-cul u e by ime-
lapse mic oscopy. Con ol ep esen s GC sphe oids (g een) ha we e no cul u ed wi h T cells ( ed). 1:1, 1:5 and 1:15 (GC
Figu e 3.
T cell in il a ion o GC sphe oids. (
a
) Schema ic ep esen a ion o he co-cul u e imeline.
Sphe oids (g een) esembling in es inal- and di use- ype GC we e independen ly cul u ed o
7 days
.
The ea e , T egs and con en ional T cells ( ed) we e so ed om pe iphe al blood o heal hy dono s,
based on he exp ession o CD3, CD4, IL2R
α
, and CD127 T cell ma ke s, and added o GC sphe oids.
A e 48 h o co-cul u e, sphe oids we e dissocia ed o u he cha ac e iza ion o bo h T cells and
GC cells. (
b
–
e
) Moni o iza ion o GC sphe oids-T cell in e ac ions du ing he
48 h
o co-cul u e by
ime-lapse mic oscopy. Con ol ep esen s GC sphe oids (g een) ha we e no cul u ed wi h T cells
( ed). 1:1, 1:5 and 1:15 (GC cell:T cell) ep esen he inc easing p opo ions o T cells o GC cells es ed
o (
b
) in es inal- ype sphe oids co-cul u ed wi h T egs, (
c
) di use- ype sphe oids co-cul u ed wi h
T egs, (
d
) in es inal- ype sphe oids co-cul u ed wi h con en ional T cells, (
e
) di use- ype sphe oids
co-cul u ed wi h con en ional T cells. (
,
g
) Ligh -shee mic oscopy o 24 h co-cul u es o in es inal-
ype GC sphe oids (g een) and T egs ( ed). (
) Co-cul u e 3D isualiza ion a e ou -angle usion
o ligh -shee mic oscopy images. (
j
–
l
) La e al iews o he 3D ep esen a ion ob ained om he
ligh -shee imaging da a co esponding o he egions indica ed in (
g
) ~124 nm, (
h
) ~290 nm, and
(i) ~434 nm, o he en i e co-cul u e. Scale ba : 50 µm.
2.3. T egs Induce IL2RαExp ession a he Memb ane o In es inal-Type GC Cells
Nex , we in es iga ed whe he GC cells could a ec he pheno ype o T cells and
ice- e sa. We s a ed by analyzing he exp ession o CD3, CD4, IL2R
α
, and FoxP3 T cell
ma ke s in cells collec ed om he co-cul u e condi ioned media, and a e dissocia ion
o GC sphe oids in o single-cell suspensions. To assess he exp ession o hese ma ke s
bo h in T cells (T egs and con en ional, s ained wi h CTV) and in GC cells, we ollowed he
ga ing s a egy showed on Supplemen a y Figu e S4. Bo h T egs and con en ional T cells
main ained hei o iginal pheno ype a e 48 h o co-cul u e (Figu e 4a), indica ing ha GC
cells unlikely impac he exp ession o ma ke s o T eg ac i a ion and immunosupp ession,
such as IL2R
α
and FoxP3. Su p isingly, when co-cul u ing in es inal- ype GC cells wi h
T egs, bu no wi h con en ional T cells, GC cells show de no o exp ession o memb anous
IL2R
α
(Figu e 4b,c), he
α
-chain o he IL2 ecep o . This phenomenon was speci ic o
in es inal- ype GC cells, and he ex en o exp ession was p opo ional o he numbe o
T egs in he co-cul u e (Figu e 4b–e). When in es inal- ype GC sphe oids we e cul u ed in
media supplemen ed wi h IL2 and inc easing concen a ions o an i-CD3/an i-CD28 beads,
no IL2R
α
induc ion was obse ed (Supplemen a y Figu e S5A). Mo eo e , condi ioned
media om co-cul u es did no elici IL2R
α
exp ession in umo cells om sphe oids ha
ha e no been p e iously exposed o T cells (Supplemen a y Figu e S5B,C).
Cance s 2021,13, 421 8 o 20
Cance s 2021, 13, x FOR PEER REVIEW 10 o 23
Figu e 4. Cha ac e iza ion o GC and T cells a e 48 h o co-cul u e. (a) Rep esen a i e pseudocolo plo s wi h he exp es-
sion o CD3, CD4, IL2Rα, and FoxP3 T cells ma ke s by low cy ome y show ha T egs and con en ional T cells main ain
hei cha ac e is ic pheno ype a e 48 h o co-cul u e. (b–e) Quan i a i e exp ession o IL2Rα in GC cells. (b,d) Plo s show-
ing pe cen age o IL2Rα posi i e in es inal- (b) o di use- ype GC (d) cells a e 48 h co-cul u e wi h T egs (black) o
con en ional T cells (g ey). Da a a e shown as mean ± SD o co-cul u es ea ed wi h T cells isola ed om a leas six
heal hy dono s. ** p < 0.01, **** p < 0.0001. Two-way ANOVA wi h Dunne ’s mul iple compa isons es . (c,e) Rep esen a-
i e his og ams o IL2Rα exp ession in in es inal- (c) o di use- ype GC (e) cells a e co-cul u ed a 1:1 ( ed) o 1:5 (blue)
p opo ions wi h T cells. ( ,g) De ec ion o memb anous IL2Rα exp ession by imaging low cy ome y. ( ) Rep esen a i e
images o IL2Rα exp ession in IL2Rα posi i e ( op le ) and nega i e ( op igh ) in es inal GC cells, as well as in T egs
(bo om le ) and con en ional T cells (bo om igh ). Scale ba : 10 μm. (g) Rep esen a i e his og am o IL2Rα exp ession
a he cell memb ane (le g aph) o in es inal- ype GC cells co-cul u ed wi h con en ional T cells (1-da k line) o T egs (2-
pink line) a 1:5 p opo ion and a e 48 h o co-cul u e. Quan i ica ion o he median IL2Rα in ensi y a he cell memb ane
( igh g aph). G aphs ep esen da a om a leas h ee independen expe imen s.
Figu e 4.
Cha ac e iza ion o GC and T cells a e 48 h o co-cul u e. (
a
) Rep esen a i e pseudocolo plo s wi h he exp ession
o CD3, CD4, IL2R
α
, and FoxP3 T cells ma ke s by low cy ome y show ha T egs and con en ional T cells main ain
hei cha ac e is ic pheno ype a e 48 h o co-cul u e. (
b
–
e
) Quan i a i e exp ession o IL2R
α
in GC cells. (
b
,
d
) Plo s
showing pe cen age o IL2R
α
posi i e in es inal- (
b
) o di use- ype GC (
d
) cells a e 48 h co-cul u e wi h T egs (black)
o con en ional T cells (g ey). Da a a e shown as mean
±
SD o co-cul u es ea ed wi h T cells isola ed om a leas six
heal hy dono s. ** p< 0.01, **** p< 0.0001. Two-way ANOVA wi h Dunne ’s mul iple compa isons es . (
c
,
e
) Rep esen a i e
his og ams o IL2R
α
exp ession in in es inal- (
c
) o di use- ype GC (
e
) cells a e co-cul u ed a 1:1 ( ed) o 1:5 (blue)
p opo ions wi h T cells. (
,
g
) De ec ion o memb anous IL2R
α
exp ession by imaging low cy ome y. (
) Rep esen a i e
images o IL2R
α
exp ession in IL2R
α
posi i e ( op le ) and nega i e ( op igh ) in es inal GC cells, as well as in T egs
(bo om le ) and con en ional T cells (bo om igh ). Scale ba : 10
µ
m. (
g
) Rep esen a i e his og am o IL2R
α
exp ession a
he cell memb ane (le g aph) o in es inal- ype GC cells co-cul u ed wi h con en ional T cells (1-da k line) o T egs (2-pink
line) a 1:5 p opo ion and a e 48 h o co-cul u e. Quan i ica ion o he median IL2R
α
in ensi y a he cell memb ane ( igh
g aph). G aphs ep esen da a om a leas h ee independen expe imen s.
Cance s 2021,13, 421 9 o 20
We u he con i med by imaging low cy ome y, ha a e co-cul u e wi h T egs,
in es inal- ype GC cells displayed a 1.8- old inc ease o he median memb anous IL2R
α
luo escence in ensi y (3312
±
72 a.u.), in compa ison o GC cells cul u ed wi h con en ional
T cells (1839 ±71 a.u.; Figu e 4 ,g).
Al oge he , hese esul s sugges ha while GC cells do no seem o a ec he phe-
no ype o T egs, hei di ec in e ac ion induces exp ession o IL2R
α
a he memb ane o
in es inal- ype GC cells.
2.4. IL2RαExp ession in In es inal-Type GC Cells Associa es wi h MAPK Signalling Pa hway
Ac i a ion and Sphe oid G ow h
IL2 is a key cy okine in he egula ion o immune cell ac i a ion and p oli e a ion,
pa icula ly in T egs and e ec o T cells. In immune cells, he high-a ini y IL2 ecep o ,
comp ising he
α
-chain (CD25),
β
-chain (CD122) and
γ
c-chain (CD132), ini ia es signal
ansduc ion ia JAK1/3, leading o he ac i a ion o MAPK, PI-3K and STAT signaling
pa hways [
23
,
24
]. We es ed whe he hese pa hways we e ac i a ed in in es inal- ype
GC cells exp essing IL2R
α
, since, o he bes o ou knowledge IL2R
α
exp ession has
ne e been desc ibed in epi helial umo cells. We obse ed ha so ed IL2R
α+
in es inal-
ype GC cells o e exp essed o al and phospho-ERK1/2 (Figu e 5a,b; Supplemen a y
Figu e S6A), bu STAT3 and Ak exp ession/ac i a ion we e no de ec ed (Supplemen a y
Figu e S6B). Acco dingly, we obse ed ha in es inal- ype GC sphe oids co-cul u ed
wi h T egs had inc eased g ow h, as a consequence o highe p oli e a ion (
p= 0.0021
;
Figu e 5c,d
; Supplemen a y Figu e S6C). Indeed, du ing he i s 24 h o co-cul u e, GC:T eg
sphe oid g ow h a e pa icula ly inc eased in he 1:5 condi ion in compa ison o con ol
sphe oids, and o GC:con en ional T cell co-cul u es (Figu e 5c). This obse a ion indica es
an ea ly p oli e a i e e ec o T egs o e in es inal- ype GC cells. In con as , T egs, bu
no con en ional T cells, induced a educ ion o di use- ype GC sphe oid g ow h a e
24 h
o co-cul u e as compa ed o con ol sphe oids (p< 0.05; Supplemen a y Figu e S6D).
The la e obse a ions sugges ha T egs may ha e he opposi e e ec o e di use- ype
GC sphe oids, and likely h ough a dis inc mechanism. Al oge he , hese esul s show
ha T egs po en ia e ac i a ion o he MAPK signaling pa hway and in es inal- ype GC
sphe oid g ow h.
Cance s 2021, 13, x FOR PEER REVIEW 12 o 23
Figu e 5. Sphe oid g ow h a e co-cul u e wi h T cells. (a,b) Wes e n blo analysis o phospho yla ed (p-ERK1/2) and
o al ERK1/2 p o ein le els in in es inal- ype IL2Rα posi i e and nega i e cells a e 48 h co-cul u e wi h T egs (1:5 p o-
po ion). (a) WB scans and (b) no malized exp ession ep esen h ee independen expe imen s. (c) Plo s o g ow h a ea
o in es inal GC sphe oids co-cul u ed wi h T egs (le g aph) o con en ional T cells ( igh g aph) o 48 h. Con ol sphe-
oids (da k) ep esen in es inal- ype GC sphe oids wi hou T cells in co-cul u e. Co-cul u es a 1:1, 1:5, and 1:15 p opo -
ions a e ep esen ed in ed, blue, and o ange, espec i ely. Da a shown mean ± SD o h ee independen co-cul u es. ** p
< 0.01. Two-way ANOVA wi h Dunne ’s mul iple compa isons es . (d) Rep esen a i e images o Ki-67 nuclea exp ession
(da k b own nuclei) in in es inal- ype GC sphe oids a e 48 h o co-cul u e wi h T egs o con en ional T cells, a di e en
1:0 (con ol), 1:1, 1:5, o 1:15 p opo ions.
Figu e 5. Con .
Cance s 2021,13, 421 16 o 20
cells (CD3
+
CD4
+
CD127
+
IL2R
α−
) was pe o med on FACSA iaII low cy ome e (BD Bio-
sciences). Fu he de ails on he an ibodies used can be ound in Supplemen a y Table S2.
4.5. Co-Cul u e o GC Sphe oids and T Cells
Fo he co-cul u e o GC sphe oids and T cells, MKN74 (in es inal- ype) and MKN45
(di use- ype) GC sphe oids we e ans e ed o ound-bo om 96-well pla es, and
T egs/con en ional T cells we e added o he cul u e media a 1:1, 1:5, and 1:15 (
GC cell:T cell
)
a ios. Co-cul u es we e main ained in RPMI 1640 wi h 25 mM HEPES (Lonza, Basel,
Swi ze land) supplemen ed wi h 10% FBS, 1% PS, 1% Sodium py u a e (Sigma-Ald ich)
and ecombinan human IL-2 (10 ng/mL; Pep oTech, London, UK). T eg cul u es we e
u he supplemen ed wi h an i-CD3/an i-CD28 MACSiBead (0.5 beads/T cell; T Cell Ac i-
a ion Ki , Mil enyi Bio ec, Be gisch Gladbach, Ge many). As a con ol, MKN74/MKN45
single cul u es (1:0; GC cell:T cell) we e main ained in he same condi ions as co-cul u es.
Fo all co-cul u e expe imen s (Figu es 3–5), a minimal numbe o 6 independen T cells
dono s, collec ed and analyzed in a leas 3 independen days a e ep esen ed.
4.6. Li e Cell Imaging
To moni o T cell in il a ion in o umo sphe oids, MKN74 and MKN45 cell lines
s ably exp essing mEme ald we e used o o m he GC sphe oids, and so ed T egs and
con en ional T cells we e s ained wi h CellT ace
™
Fa Red (CTFR) Cell P oli e a ion
Ki (20 min, 37
◦
C; The mo Fishe Scien i ic) be o e co-cul u e. Fi s , co-cul u es we e
imaged using an au oma ed luo escence wide ield HCS mic oscope (IN Cell Analyze
2000; GE Heal hca e, Chicago, IL, USA), equipped wi h a Plan-Fluo Nikon 20
×
/0.45
objec i e lens and a la ge-chip CCD came a (CoolSNAP K4). Eme ald-GC sphe oids we e
acqui ed in he FITC channel (Exci a ion/Emission: 490/ 525 nm; Exposu e: 50 ms) and
T cells in he Cy5 channel (Exci a ion/Emission: 645/705 nm; Exposu e: 30 ms). Images
we e acqui ed a e e y 2 h o 60 h, unde empe a u e-con olled condi ions.
To u he con i m T eg (CTFR) in il a ion in o in es inal- ype GC sphe oids (MKN74-
Eme ald), sphe oids we e ans e ed o a luo ina ed e hylene p opylene (FEP) mic o ube
and imaged wi h a cus om-buil Digi al Scanne Lase Ligh Shee mic oscope (LSFM)
equipped wi h a Nikon Plan-Fluo 10x/0.3 wa e -imme sion objec i e lens in illumina ion
and de ec ion plans. The luo escence signals o he Eme ald-MKN74 sphe oids and CTFR-
T egs we e eco ded using a Hamama su O ca-Flash 4.0 V3 came a, a e sample exci a ion
wi h a 488 (1.0 mW) and 640 nm (1.0 mW) lase lines, espec i ely, wi h a 525/50 nm
bandpass (BP) emission il e o eme ald-sphe oids and a 700/75 nm BP emission il e
o CTFR-T egs. The co-cul u e was imaged a h ee imepoin s (16 h, 24 h, and 48 h)
om ou angles (90
◦
o a ion) e e y 16 min. Each iew consis s o mul iple slices 1
µ
m
apa co e ing he en i e sphe oid. Image econs uc ion was pe o med using he a i is
Vision4D 3.1.4 (a i is AG, Ros ock, Ge many).
4.7. Immunopheno ype by Con en ional and Imaging Flow Cy ome y Analysis
A e 48 h o co-cul u e, condi ioned media and sphe oids we e collec ed o analyze
he pheno ype o T egs and con en ional T cells. Be o e co-cul u e wi h GC sphe oids, T egs
and con en ional T cells we e s ained using CellT ace
™
Viole (CTV) Cell P oli e a ion Ki
(20 min, 37
◦
C; The mo Fishe Scien i ic), o allow u he e alua ion o T cell p oli e a ion
and o sepa a e T cells (CTV+) om GC cells (CTV-) in he low cy ome y analysis. To
p epa e single-cell suspensions, sphe oids we e dissocia ed using 0.5% T ypsin-EDTA
(5 min, 37
◦
C; The mo Fishe Scien i ic). Bo h T cells and GC cells we e s ained wi h
an i-CD3-PE, an i-CD4-FITC, and an i-IL2R
α
-PE/Cy7, as p e iously men ioned, and wi h
Fixable Viabili y Dye eFluo
™
780 (1:1000 dilu ion; The mo Fishe Scien i ic) o assess cell
iabili y. In acellula s aining o FoxP3 was pe o med using he Foxp3/T ansc ip ion
Fac o S aining Bu e Se (The mo Fishe Scien i ic), ollowing manu ac u e ’s ins uc ions.
B ie ly, cells we e ixed wi h Fixa ion/Pe meabiliza ion Bu e (30 min; The mo Fishe
Scien i ic), blocked wi h human IgG (1 mg/mL, 15 min; Sigma-Ald ich) and s ained wi h

Cance s 2021,13, 421 17 o 20
an i-FoxP3-APC (1:15, 30 min; The mo Fishe Scien i ic). Da a acquisi ion was pe o med
on an LSRFo essa cy ome e (Bec on, Dickinson & Company, F anklin Lakes, NJ, USA)
and analyzed using FlowJo 10 so wa e (Bec on, Dickinson & Company) ollowing he
ga ing s a egy explained on Supplemen a y Figu e S3A.
Imaging low cy ome y was pe o med on single-cell suspensions de i ed om
co-cul u es o MKN74-GC sphe oids wi h T egs/con en ional T cells and s ained wi h an i-
IL2R
α
-PE/Cy7 an ibody. Da a was acqui ed on an imaging low cy ome e (ImageS eamX;
Amnis Co po a ion, Sea le, WA, USA) equipped wi h INSPIRE so wa e. Samples we e
exci ed wi h a 488 nm a gon lase , and o each e en , b igh - ield cell images we e ac-
qui ed (40
×
magni ica ion) on channel 1, while IL2R
α
signal was de ec ed on channel 6
(Exci a ion/Emission: 496/774 nm). Lase powe was no modi ied h oughou he sample
acquisi ion. Da a analysis was pe o med using he IDEAS so wa e (Amnis Co po a ion).
Fu he de ails on he an ibodies used can be ound in Supplemen a y Table S2.
4.8. Reco e y o IL2Rα+GC Cells and Wes e n Blo Analysis
To analyze he e ec o IL2R
α
exp ession speci ically on in es inal- ype GC cells,
MKN74-GC cells we e s ained wi h CellT ace
™
Yellow (CTY) Cell P oli e a ion Ki (The mo
Fishe Scien i ic) be o e g own as sphe oids, while T egs and con en ional T cells we e
s ained wi h CellT ace
™
Fa Red (CTFR) Cell P oli e a ion Ki (The mo Fishe Scien i ic)
be o e co-cul u e. As p e iously, a e co-cul u e, sphe oids we e dissocia ed, and he e-
sul ing single-cell suspension was s ained wi h an i-IL2R
α
-PE/Cy7 an ibody. MKN74-GC
cells (CTY+) posi i e o nega i e o IL2R
α
we e so ed on he FACSA ia II low cy ome-
e (Bec on, Dickinson & Company) and collec ed o FBS. Acco ding o he numbe o
so ed cells, samples we e esuspended in p opo ional olumes o adioimmunop ecipi a-
ion assay bu e (RIPA bu e ) supplemen ed wi h p o ease and phospha ase inhibi o s
(The mo Fishe Scien i ic) and cell lysis was allowed o 30 min, on ice. To enhance p o ein
eco e y, lysa es we e sonica ed wice (10 s, 20
×
, 50%) be o e cen i uga ion a 14,000 pm
(15 min, 4
◦
C) o eco e clea ed lysa es. Equal amoun s o p o ein lysa e (25
µ
L) om
each biological eplica e we e subjec ed o Wes e n blo ing. P ima y an ibodies a ge ing
phospho-S a 3 (Ty 705, D3A7; Cell Signalling Technology), S a 3 (124 h6; Cell Signalling
Technology), phospho-Ak (Se 473, D9E; Cell Signalling Technology), Ak (Cell Signalling
Technology), phospho-p44/42 MAPK (phospho-ERK1/2, Th 202/Ty 204; Cell Signalling
Technology), p44/42 MAPK (ERK1/2; Cell Signalling Technology), and GAPDH (1E6D9;
P o einTech, Rosemon , IL, USA) we e used, as well as co esponding ECLTM an i-mouse
and an i- abbi IgG ho se adish pe oxidase-conjuga ed seconda y an ibodies (GE Heal h-
ca e). De ec ed signals we e quan i ied using Quan i y One
®
Basic so wa e (Bio-Rad,
He cules, CA, USA). Unp ocessed scans o Wes e n blo s a e p o ided in Supplemen a y
Figu e S9. Addi ional in o ma ion on he an ibodies used is de ailed in Supplemen a y
Table S2.
4.9. Ki-67 S aining on GC-Sphe oids
To in es iga e GC-sphe oid p oli e a ion a e co-cul u e, sphe oids we e ixed wi h
2% pa a o maldehyde (o e nigh ; Me ck, Da ms ad , Ge many), s ained wi h Gill’s hema-
oxylin (10 min; Bio-Op ica, Milan, I aly) and injec ed wi hin a d op o 2.4% low mel ing
poin aga ose (50
◦
C; Lonza). A e gelling (10 min a oom empe a u e ollowed by
20 min
on ice), he aga ose s uc u es we e included in pa a in blocks and hen sec ioned
in o 3-
µ
m slides. An igen e ie al was pe o med using ci a e bu e (10 mM, pH 6.0, a
98 ◦C
, 40 min; Abcam). Endogenous pe oxidase ac i i y was blocked using 0.3% hyd ogen
pe oxidase solu ion (20 min; Sigma-Ald ich) ollowed by incuba ion wi h an i-Ki-67 an i-
body (SP6, 1:200 dilu ion, 90 min; The mo Fishe Scien i ic). A e washing, slides we e
incuba ed wi h REAL EnVision De ec ion Sys em (Dako) subs a e bu e (30 min) and
wi h DAB Ch omogen (10 min).
Cance s 2021,13, 421 18 o 20
5. Conclusions
O e all, ou s udy sugges s ha du ing he ea ly s eps o in es inal- ype gas ic
ca cinogenesis, T egs accumula e wi hin he umo mic oen i onmen and, likely h ough
a con ac -dependen mechanism, p omo e IL2R
α
exp ession and s imula ion o g ow h
signaling pa hways, such as MAPK pa hway, in umo cells. Ou indings hold p omising
and ele an implica ions o u u e IO he apies and pa ien s a i ica ion in GC, which
may po en ially be expanded o o he immune cold umo s.
Supplemen a y Ma e ials:
The ollowing a e a ailable online a h ps://www.mdpi.com/2072-669
4/13/3/421/s1, Figu e S1: Dis ibu ion o T cell popula ions in MSS and MSI GC umou s, Figu e S2:
Ga ing s a egy o T egs and con en ional T cell so ing om pe iphe al blood o heal hy dono s,
Figu e S3: T cell in il a ion o GC sphe oids by ligh -shee mic oscopy, Figu e S4: Cha ac e iza ion o
T cells be o e and a e co-cul u e, Figu e S5: E ec o IL2, an i-CD3/an i-CD28 beads, and co-cul u e
condi ioned medium in he exp ession o IL2R
α
in GC cells., Figu e S6: GC cell pheno ype a e
co-cul u e wi h T cells, Figu e S7: Dis ibu ion o T cell popula ions in in es inal- ype GC pa ien s and
associa ion wi h clinicopa hological ea u es, Figu e S8: S ep-wise mul ispec al imaging analysis
o he immune T cell landscape in GC issue sec ions, Figu e S9: Wes e n blo quan i ica ion o
phospho yla ed (p-ERK1/2) and o al ERK1/2 p o ein le els in in es inal- ype IL2R
α
- and IL2R
α
+
GC cells a e 48 h co-cul u e wi h T egs, Table S1: Desc ip ion o pa ien in o ma ion included in he
analysed GC coho , Table S2: De ailed desc ip ion o an ibodies, Videos S1–S3: T egs ( ed) in il a ion
o in es inal- ype GC sphe oids (g een) a e 16 h (Supplemen a y Video S1), 24 h (Supplemen a y
Video S2) and 48 h (Supplemen a y Video S3) o co-cul u e a 1:5 (GC:T) cell p opo ion, by ligh -shee
mic oscopy, File S1: Tumou p esen s lymphoid s oma (his ological analysis).
Au ho Con ibu ions:
Concep ualiza ion, S.R., J.C., and C.O.; Collec ion o ma e ial and clini-
copa hological da a om GC pa ien s me hodology, S.R., G.G., I.G., G.M.A., J.C., F.C., and C.O.;
Immunopheno ype analysis, S.R., M.E.I., and N.F.C.C.d.M.; Es ablishmen and cha ac e iza ion o
co-cul u es, S.R., A.P.B., S.P.T., M.M.A., J.J.M., M.I.O., J.T.B., and L.G.; w i ing—o iginal d a p epa-
a ion, S.R.; w i ing— e iew and edi ing, S.R., J.C., and C.O.; p ojec adminis a ion and unding
acquisi ion, C.O. All au ho s ha e ead and ag eed o he published e sion o he manusc ip .
Funding:
This wo k was suppo ed by g an s: (1) “Ins i u e o Resea ch and Inno a ion in Heal h
Sciences” (POCI-01-0145-FEDER-007274), by FEDER—Fundo Eu opeu de Desen ol imen o Regional
unds h ough he COMPETE 2020—Ope acional P og am o Compe i i eness and In e na ion-
alisa ion (POCI), PORTUGAL 2020, Po uguese unds h ough Po uguese Science & Technology
Founda ion (FCT)/Minis é io da Ciência, Tecnologia e Ino ação; (2) P ojec CANCER Re . NORTE-01-
0145-FEDER-000029 and DOCne Re . NORTE-01-0145-FEDER-000003), by No e Po ugal Regional
P og am (NORTE 2020), unde he PORTUGAL 2020 Pa ne ship Ag eemen , h ough he Eu o-
pean Regional De elopmen Fund (ERDF); (3) P ojec Re . PTDC/BBB-ECT/2518/2014, by FCT;
(4) FCT PhD P og ams and by P og ama Ope acional Po encial Humano (POCH), speci ically by he
Bio echHeal h P og ame (Doc o al P og am on Cellula and Molecula Bio echnology Applied o
Heal h Sciences); (5) “The ole o gas ic cance cell-de i ed ex acellula esicles”, by IPATIMUP
Boa d o Di ec o s; (6) GenomePT p ojec (POCI-01-0145-FEDER-022184), by COMPETE 2020—POCI,
Lisboa Po ugal Regional Ope a ional P og am (Lisboa2020), Alga e Po ugal Regional Ope a ional
P og am (CRESC Alga e2020), unde he PORTUGAL 2020 Pa ne ship Ag eemen , h ough ERDF,
and by FCT; (7) Ma ie Skłodowska-Cu ie g an ag eemen No. 722148 o T ain-EV p ojec by Eu o-
pean Union’s Ho izon 2020 esea ch and inno a ion p og am; 8) P ojec CANCERSTEM unded by
ERDF, POCI and FCT; 9) FCT ellowship PD/BD/128406/2017 o SR, and Junio Resea ch con ac
o JC ( egula ed by he dec ee-law 57/2016 emended by law 57/2017). IPATIMUP in eg a es he
i3S Resea ch Uni , which is pa ially suppo ed by FCT. NFCCdM has ecei ed unding om he
Eu opean Resea ch Council (ERC) unde he Eu opean Union’s Ho izon 2020 esea ch and inno a ion
p og am (g an ag eemen No. 852832).
Ins i u ional Re iew Boa d S a emen :
The s udy was conduc ed acco ding o he guidelines o he
Decla a ion o Helsinki, and app o ed by he E hics Commi ee o Cen o Hospi ala Uni e si á io
de São João (p o ocol code 78/13).
In o med Consen S a emen :
In o med consen was ob ained om all subjec s in ol ed in he s udy.
Cance s 2021,13, 421 19 o 20
Da a A ailabili y S a emen :
Da a is con ained wi hin he a icle and/o supplemen a y ma e ial.
Addi ional da a p esen ed in his s udy a e a ailable on eques om he co esponding au ho .
Acknowledgmen s:
The au ho s would like o hank Se iço de Imunohemo e apia o Cen o
Hospi ala Uni e si á io de São João (CHUSJ), o kindly dona ing Bu y Coa s and o Ma ia José
Oli ei a and Susana San os o enabling his collabo a ion p o ocol. The au ho s acknowledge
he suppo o he Ad anced Ligh Mic oscopy, Bioimaging, BioSciences Sc eening and His ology
and Elec on Mic oscopy i3S Scien i ic Pla o ms membe s o he na ional in as uc u e PPBI—
Po uguese Pla o m o Bioimaging (PPBI-POCI-01-0145-FEDER-022122), and he suppo p o ided
by he T ansla ional Cy ome y i3S Scien i ic Pla o ms. The au ho s also acknowledge he excellen
assis ance o Ca a ina Mei eles, Emília Ca doso, Paula Sampaio, Ma ia Láza o, And éMaia, Rossana
Co eia and Nuno Mendes.
Con lic s o In e es : The au ho s decla e no con lic o in e es .
Re e ences
1.
Pellino, A.; Riello, E.; Nappo, F.; B ignola, S.; Mu gioni, S.; Djaballah, S.A.; Lona di, S.; Zagonel, V.; Rugge, M.;
Loupakis, F.; e al
.
Ta ge ed he apies in me as a ic gas ic cance : Cu en knowledge and u u e pe spec i es. Wo ld J. Gas oen e ol.
2019
,
25, 5773–5788. [C ossRe ] [PubMed]
2.
B ay, F.; Fe lay, J.; Soe joma a am, I.; Siegel, R.L.; To e, L.A.; Jemal, A. Global cance s a is ics 2018: GLOBOCAN es ima es o
incidence and mo ali y wo ldwide o 36 cance s in 185 coun ies. CA Cance J. Clin. 2018,68, 394–424. [C ossRe ] [PubMed]
3.
Kawazoe, A.; Kuwa a, T.; Kuboki, Y.; Shi a a, K.; Naga suma, A.K.; Aizawa, M.; Yoshino, T.; Doi, T.; Oh su, A.; Ochiai, A.
Clinicopa hological ea u es o p og am dea h ligand 1 exp ession wi h umo -in il a ing lymphocy e, misma ch epai , and
Eps ein-Ba i us s a us in a la ge coho o gas ic cance pa ien s. Gas ic Cance 2017,20, 407–415. [C ossRe ] [PubMed]
4.
Böge , C.; Beh ens, H.-M.; Ma hiak, M.; K üge , S.; Kal ho , H.; Röcken, C. PD-L1 is an independen p ognos ic p edic o in
gas ic cance o Wes e n pa ien s. Onco a ge 2016,7, 24269–24283. [C ossRe ]
5.
Wing, K.; Sakaguchi, S. Regula o y T cells exe checks and balances on sel ole ance and au oimmuni y. Na . Immunol.
2009
,
11, 7–13. [C ossRe ]
6.
Togashi, Y.; Shi a a, K.; Nishikawa, H. Regula o y T cells in cance immunosupp ession—implica ions o an icance he apy.
Na . Re . Clin. Oncol. 2019,16, 356–371. [C ossRe ]
7.
Facciabene, A.; Mo z, G.T.; Coukos, G. T- egula o y cells: Key playe s in umo immune escape and angiogenesis. Cance Res.
2012,72, 2162–2171. [C ossRe ]
8.
Ichiha a, F.; Kono, K.; Takahashi, A.; Kawaida, H.; Sugai, H.; Fujii, H. Inc eased popula ions o egula o y T cells in pe iphe al
blood and umo -in il a ing lymphocy es in pa ien s wi h gas ic and esophageal cance s. Clin. Cance Res. 2003,9, 4404–4408.
9.
Sasada, T.; Kimu a, M.; Yoshida, Y.; Kanai, M.; Takabayashi, A. CD4+CD25+ egula o y T cells in pa ien s wi h gas oin es inal
malignancies: Possible in ol emen o egula o y T cells in disease p og ession. Cance 2003,98, 1089–1099. [C ossRe ]
10.
Shen, L.; Wang, J.; Shen, D.-F.; Yuan, X.; Dong, P.; Li, M.-X.; Xue, J.; Zhang, F.-M.; Ge, H.-L.; Xu, D. CD4+CD25+CD127low/
−
egu-
la o y T cells exp ess Foxp3 and supp ess e ec o T cell p oli e a ion and con ibu e o gas ic cance s p og ession. Clin. Immunol.
2009,131, 109–118. [C ossRe ]
11.
Mao, F.-Y.; Kong, H.; Zhao, Y.-L.; Peng, L.-S.; Chen, W.; Zhang, J.-Y.; Cheng, P.; Wang, T.-T.; L , Y.-P.; Teng, Y.-S.; e al. Inc eased
umo -in il a ing CD45RA
−
CCR7
−
egula o y T-cell subse wi h immunosupp essi e p ope ies os e gas ic cance p og ess.
Cell Dea h Dis. 2017,8, e3002. [C ossRe ] [PubMed]
12.
Yuan, L.; Xu, B.; Yuan, P.; Zhou, J.; Qin, P.; Han, L.; Chen, G.; Wang, Z.; Run, Z.; Zhao, P.; e al. Tumo -in il a ing CD4+ T cells in
pa ien s wi h gas ic cance . Cance Cell In . 2017,17, 114. [C ossRe ] [PubMed]
13.
Kono, K.; Kawaida, H.; Takahashi, A.; Sugai, H.; Mimu a, K.; Miyagawa, N.; Oma a, H.; Fujii, H. CD4(+)CD25high egula o y T
cells inc ease wi h umo s age in pa ien s wi h gas ic and esophageal cance s. Cance Immunol. Immuno he .
2005
,55, 1064–1071.
[C ossRe ] [PubMed]
14.
Pe one, G.; Ru ini, P.A.; Ca alano, V.; Spino, C.; San ini, D.; Mu e o, P.; Spo o, C.; Zinga e i, C.; Sis i, V.;
Alessand oni, P.; e al
.
In a umo al FOXP3-posi i e egula o y T cells a e associa ed wi h ad e se p ognosis in adically esec ed gas ic cance .
Eu . J. Cance 2008,44, 1875–1882. [C ossRe ] [PubMed]
15.
Shen, Z.; Zhou, S.; Wang, Y.; Li, R.-L.; Zhong, C.; Liang, C.; Sun, Y. Highe in a umo al in il a ed Foxp3+ T eg numbe s
and Foxp3+/CD8+ a io a e associa ed wi h ad e se p ognosis in esec able gas ic cance . J. Cance Res. Clin. Oncol.
2010
,
136, 1585–1595. [C ossRe ] [PubMed]
16.
Kashimu a, S.; Saze, Z.; Te ashima, M.; Soe a, N.; Oh ani, S.; Osuka, F.; Kogu e, M.; Go oh, M. CD83+ dend i ic cells and
Foxp3+ egula o y T cells in p ima y lesions and egional lymph nodes a e in e sely co ela ed wi h p ognosis o gas ic cance .
Gas ic Cance 2011,15, 144–153. [C ossRe ]
17.
Zhou, S.; Xu, S.; Tao, H.; Zhen, Z.; Chen, G.; Zhang, Z.; Yang, Y. CCR7 exp ession and in a umo al FOXP3+ egula o y T cells a e
co ela ed wi h o e all su i al and lymph node me as asis in gas ic cance . PLoS ONE 2013,8, e74430. [C ossRe ]
18.
Haas, M.; Dimmle , A.; Hohenbe ge , W.; G abenbaue , G.G.; Niedobi ek, G.; Dis el, L. S omal egula o y T-cells a e associa ed
wi h a a ou able p ognosis in gas ic cance o he ca dia. BMC Gas oen e ol. 2009,9, 65. [C ossRe ]
Cance s 2021,13, 421 20 o 20
19.
Kim, K.-J.; Lee, K.S.; Cho, H.J.; Kim, Y.H.; Yang, H.K.; Kim, W.H.; Kang, G.H. P ognos ic implica ions o umo -in il a ing FoxP3+
egula o y T cells and CD8+ cy o oxic T cells in mic osa elli e-uns able gas ic cance s. Hum. Pa hol.
2014
,45, 285–293. [C ossRe ]
20.
Choi, H.S.; Ha, S.Y.; Kim, H.-M.; Ahn, S.M.; Kang, M.-S.; Kim, K.-M.; Choi, M.G.; Lee, J.H.; Sohn, T.S.; Bae, J.M.; e al. The
p ognos ic e ec s o umo in il a ing egula o y T cells and myeloid de i ed supp esso cells assessed by mul icolo low
cy ome y in gas ic cance pa ien s. Onco a ge 2016,7, 7940–7951. [C ossRe ]
21.
Tan, W.; Zhang, W.; S asne , A.; G i enniko , S.; Cheng, J.Q.; Ho man, R.M.; Ka in, M. Tumo -in il a ing egula o y T cells
s imula e mamma y cance me as asis h ough RANKL-RANK signalling. Na u e 2011,470, 548–553. [C ossRe ] [PubMed]
22.
Facciabene, A.; Peng, X.; Hagemann, I.S.; Balin , K.; Ba che i, A.; Wang, L.P.; Gimo y, P.A.; Gilks, C.B.; Lal, P.; Zhang, L.; e al.
Tumo hypoxia p omo es ole ance and angiogenesis ia CCL28 and T( eg) cells. Na u e
2011
,475, 226–230. [C ossRe ] [PubMed]
23.
A enas-Rami ez, N.; Woy schak, J.; Boyman, O. In e leukin-2: Biology, Design and Applica ion. T ends Immunol.
2015
,
36, 763–777. [C ossRe ]
24.
Malek, T.R.; Cas o, I. In e leukin-2 Recep o Signaling: A he In e ace be ween Tole ance and Immuni y. Immuni y
2010
,
33, 153–165. [C ossRe ] [PubMed]
25.
Cheng, H.-H.; Tseng, G.-Y.; Yang, H.-B.; Wang, H.-J.; Lin, H.-J.; Wang, W.-C. Inc eased numbe s o Foxp3-posi i e egula o y T
cells in gas i is, pep ic ulce and gas ic adenoca cinoma. Wo ld J. Gas oen e ol. 2012,18, 34–43. [C ossRe ]
26.
Jang, T.J. The numbe o Foxp3-posi i e egula o y T cells is inc eased in Helicobac e pylo i gas i is and gas ic cance .
Pa hol. Res. P ac . 2010,206, 34–38. [C ossRe ]
27.
Ying, L.; Yan, F.; Meng, Q.; Yuan, X.; Yu, L.; Williams, B.R.G.; Chan, D.W.; Shi, L.; Tu, Y.; Ni, P.; e al. Unde s anding immune
pheno ypes in human gas ic disease issues by mul iplexed immunohis ochemis y. J. T ansl. Med. 2017,15, 1–11. [C ossRe ]
28.
Gullo, I.; Oli ei a, P.; A helogou, M.; Gonçal es, G.; Pin o, M.L.; Ca alho, J.; Valen e, A.; Pinhei o, H.; And ade, S.;
Almeida, G.M.; e al
. New insigh s in o he in lamed umo immune mic oen i onmen o gas ic cance wi h lymphoid s oma:
F om mo phology and digi al analysis o gene exp ession. Gas ic Cance 2018,22, 77–90. [C ossRe ]
29.
Mizukami, Y.; Kono, K.; Kawaguchi, Y.; Akaike, H.; Kamimu a, K.; Sugai, H.; Fujii, H. CCL17 and CCL22 chemokines wi hin
umo mic oen i onmen a e ela ed o accumula ion o Foxp3+ egula o y T cells in gas ic cance . In . J. Cance
2008
,
122, 2286–2293. [C ossRe ]
30.
Yuan, X.-L.; Chen, L.; Zhang, T.-T.; Ma, Y.-H.; Zhou, Y.-L.; Zhao, Y.; Wang, W.-W.; Dong, P.; Yu, L.; Zhang, Y.Y.; e al. Gas ic
cance cells induce human CD4+Foxp3+ egula o y T cells h ough he p oduc ion o TGF-be a1. Wo ld J. Gas oen e ol.
2011
,
17, 2019–2027. [C ossRe ]
31.
Du, L.; Xiao, X.; Wang, C.; Zhang, X.; Zheng, N.; Wang, L.; Zhang, X.; Li, W.; Wang, S.; Dong, Z. Human leukocy e an igen-G
is closely associa ed wi h umo immune escape in gas ic cance by inc easing local egula o y T cells. Cance Sci.
2011
,
102, 1272–1280. [C ossRe ] [PubMed]
32.
Kim, S.Y.; Pa k, C.; Kim, H.-J.; Pa k, J.; Hwang, J.; Kim, J.-I.; Choi, M.G.; Kim, S.; Kim, K.-M.; Kang, M.-S. De egula ion o
Immune Response Genes in Pa ien s wi h Eps ein-Ba Vi us-Associa ed Gas ic Cance and Ou comes. Gas oen e ology
2015
,
148, 137–147.e9. [C ossRe ] [PubMed]
33.
Rocha, S.; Ca alho, J.; Oli ei a, P.; Vogls ae e , M.; Sch a z, D.; Thomsen, A.R.; Wal e , N.; Khandu i, R.; Sanchez, J.C.;
Kelle , A.; e al
. 3D Cellula A chi ec u e A ec s Mic oRNA and P o ein Ca go o Ex acellula Vesicles. Ad . Sci.
2019
,6, 1800948.
[C ossRe ] [PubMed]
34.
Choudh y, H.; Helmi, N.; Abdulaal, W.H.; Zeyadi, M.; Zamzami, M.A.; Wu, W.; Mahmoud, M.M.; Wa si, M.K.; Rasool, M.;
Jamal, M.S. P ospec s o IL-2 in Cance Immuno he apy. BioMed Res. In . 2018,2018, 1–7. [C ossRe ] [PubMed]
35.
Flynn, M.J.; Ha ley, J.A. The eme ging ole o an i-CD25 di ec ed he apies as bo h immune modula o s and a ge ed agen s in
cance . B . J. Haema ol. 2017,179, 20–35. [C ossRe ] [PubMed]
36.
Ijssels eijn, M.E.; B ouwe , T.P.; Abdul ahman, Z.; Reidy, E.; Ramalhei o, A.; Hee en, A.M.; Vah meije , A.; Jo dano a, E.S.;
De Mi anda, N.F.C.C.
Cance immunopheno yping by se en-colou mul ispec al imaging wi hou y amide signal ampli ica ion.
J. Pa hol. Clin. Res. 2018,5, 3–11. [C ossRe ]