1
Vol.:(0123456789)
Scien i ic Repo s | (2022) 12:17638 | h ps://doi.o g/10.1038/s41598-022-21576-8
www.na u e.com/scien i ic epo s
Di e en an ibody‑associa ed
au oimmune diseases ha e
dis inc pa e ns o T ollicula cell
dys egula ion
Filipa Ribei o1,2,4, Vasco C. Romão1,3,4, Sa a Rosa1, Ká ia Jesus1, Ana Água‑Doce1,
So ia C. Ba ei a1,3, Pa ícia Ma ins1,3, Susana Lopes da Sil a1,3, Ema Nob e3,
Ma ia João Bugalho3, Vál e R. Fonseca1,5, João Eu ico Fonseca1,3,5 & Luis G aca1,2,5*
Au oan ibodies a e p oduced wi hin ge minal cen e s (GC), in a p ocess egula ed by in e ac ions
be ween B, T ollicula helpe (T h), and T ollicula egula o y (T ) cells. The GC dys egula ion in
human au oimmuni y has been in e ed om ci cula ing cells, albei wi h con lic ing esul s due
o di e se expe imen al app oaches. We applied a consis en app oach o compa e ci cula ing
T and T h subse s in pa ien s wi h di e en au oimmune diseases. We ec ui ed 97 pa icipan s,
including 72 pa ien s wi h Hashimo o’s hy oidi is (HT, n = 18), heuma oid a h i is (RA, n = 16), o
sys emic lupus e y hema osus (SLE, n = 32), and 31 ma ched heal hy dono s (HD). We ound ha he
equency o ci cula ing T ollicula subse s di e ed ac oss diseases. Pa ien s wi h HT had an inc eased
equency o blood T h cells (p = 0.0215) and a educed T /T h a io (p = 0.0338) when compa ed wi h
HD. This was no obse ed in pa ien s wi h sys emic au oimmune heuma ic diseases (RA, SLE),
who had a educ ion in bo h T h (p = 0.0494 and p = 0.0392, espec i ely) and T (p = 0.0003 and
p = 0.0001, espec i ely) cells, esul ing in an unchanged T /T h a io. Ac i a ed PD‑1+ICOS+T h and
CD4+PD‑1+CXCR5–Tph cells we e aised only in pa ien s wi h SLE (p = 0.0022 and p = 0.0054), wi hou
associa ion wi h disease ac i i y. Ou da a sugges ha GC dys egula ion, assessed by T ollicula
subse s, is no uni o m in human au oimmuni y. Speci ic pa e ns o dys egula ion may become
po en ial bioma ke s o disease and pa ien s a i ica ion.
The inding ha T ollicula helpe (T h) and T ollicula egula o y (T ) cells con ol GC eac ions p omp ed
he s udy o ci cula ing T ollicula subse s in human au oimmuni y1–4. T cells de i e om Foxp3+ T eg cells5–7,
ha e a TCR epe oi e di ec ed o sel -an igens and dis inc om he epe oi e o T h cells8, ha e a supp essi e
unc ion9,10, and a e essen ial o p e en he gene a ion o au oimmuni y in mice11–13. I was, he e o e, emp ing
o specula e ha a dys egula ed T /T h a io could be implica ed in an ibody-media ed au oimmuni y. We
p e iously demons a ed ha pa ien s wi h p ima y Sjög en’s synd ome (pSS) had an inc ease in ci cula ing
T cells compa ed o heal hy dono s (HD)4,14. Mo eo e , a subse o pSS pa ien s wi h mo e au oan ibodies
and ec opic lymphoid s uc u es (ELS) wi hin mino sali a y glands had a signi ican ly highe T /T h a io14.
We demons a ed ha he gene a ion o ci cula ing T cells is a no mal p ocess ollowing a humo al esponse,
namely in luenza accina ion4, and can explain he high T /T h a io ollowing ch onic in lamma o y eac ions,
as in pSS4,14. Al hough o he esea che s subsequen ly con i med ou esul s15, he e we e also con lic ing
epo s16. Fu he mo e, epo s ha e also been inconsis en in o he au oimmune diseases, sugges ing an inc ease
o dec ease o he T /T h a io14,17–19.
I has been di icul o compa e hose s udies as he expe imen al app oaches a e no di ec ly compa able. Fo
ins ance, we p e iously ound ha ci cula ing T cells ha e an imma u e pheno ype, gi en he de elopmen al
bi u ca ion o T cells wi hin human lymphoid issue4,20. Howe e , in some s udies, T cells a e selec ed wi hin
OPEN
1Ins i u o de Medicina Molecula João Lobo An unes, Faculdade de Medicina da Uni e sidade de Lisboa, Lisbon
Academic Medical Cen e, A enida P o esso Egas Moniz, 1649-028 Lisboa, Po ugal. 2Ins i u o Gulbenkian
de Ciência, Oei as, Po ugal. 3Rheuma ology Depa men , Hospi al de San a Ma ia, Cen o Hospi ala Lisboa
No e, Lisbon Academic Medical Cen e, Lisboa, Po ugal. 4These au ho s con ibu ed equally: Filipa Ribei o and
Vasco C. Romão. 5
These au ho s join ly supe ised his wo k: Vál e R. Fonseca, João Eu ico Fonseca and Luis
G aca. *email: [email p o ec ed]
2
Vol:.(1234567890)
Scien i ic Repo s | (2022) 12:17638 | h ps://doi.o g/10.1038/s41598-022-21576-8
www.na u e.com/scien i ic epo s/
CD45RO+ o PD-1+ cells17,21, excluding a la ge p opo ion o his cell popula ion. CD25 may also be a poo
ma ke o T cells, especially he mos ma u e wi hin he GC, al hough ci cula ing T cells a e CD25+20,22–25.
Con e sely, T h cells a e o en s udied wi hou exclusion o Foxp3+ cells, o selec ing hem as PD-1+, hus
including bo h T ollicula subse s (T h and T ) as T h cells and/o excluding imma u e PD-1– T h cells26,27.
Gi en ou esul s showing ha pSS pa ien s wi h ELS had a high T /T h a io, we add essed he hypo hesis
ha o he au oimmune diseases wi h a ge o gan in ol emen , au oan ibodies’ p oduc ion and ELS would ha e
a simila p o ile. Fo ha pu pose, we s udied pa ien s wi h Hashimo o’s hy oidi is (HT), an o gan-speci ic
au oimmune disease wi h ELS o ma ion in hy oid gland. In addi ion, we aimed o in es iga e whe he hese
indings could be eplica ed in o he , mo e sys emic, au oimmune diseases. Thus, using a consis en app oach,
we included pa ien s wi h wo di e en au oimmune heuma ic diseases: heuma oid a h i is (RA), a sys emic
disease wi h a clea -cu a ge (i.e., a icula ) in ol emen ; and sys emic lupus e y hema osus (SLE), he
p o o ypical example o an au oan ibody-d i en sys emic au oimmune heuma ic disease.
Pa ien s and me hods
Human samples. We ec ui ed pa ien s wi h HT (n = 18) ollowed in he Endoc inology Depa men o
Hospi al de San a Ma ia, Cen o Hospi ala Uni e si á io Lisboa No e (CHULN), Lisbon Academic Medical
Cen e, Lisbon, Po ugal. Fu he , we included pa ien s wi h RA (n = 16) and SLE (n = 32), who we e ollowed
in he Rheuma ology Depa men o he same hospi al. Only adul , non-p egnan , pa ien s ul illing he clinical
classi ica ion c i e ia o HT (hypo hy oidism wi h he p esence o TPO an ibodies), RA (Ame ican College
o Rheuma ology (ACR)/Eu opean Alliance o Associa ions o Rheuma ology (EULAR) 2010), o SLE (ACR
1997) we e included. We excluded pa ien s unde immunosupp essi e ea men o he han me ho exa e,
hyd oxychlo oquine, o ≤ 7.5mg pe day o p ednisolone o equi alen . Pa ien s diagnosed as ha ing an in ec ious
disease o diabe es, o who had ecei ed any accine in he p e ious mon h we e also excluded. Disease ac i i y
was e alua ed using he Disease Ac i i y Sco e-28 (DAS-28) o RA pa ien s, and Sys emic Lupus E y hema osus
Disease Ac i i y Index (SLEDAI) o SLE pa ien s. Rele an au oan ibodies and se ological ma ke s o disease/
disease ac i i y we e also collec ed. F esh blood samples we e collec ed in EDTA-coa ed ubes and p ocessed
on he same day o low cy ome y analysis. Finally, we collec ed esh blood samples om 31 sex- and age-
ma ched heal hy olun ee s.
E hical app o al. This s udy was app o ed by he Lisbon Academic Medical Cen e E hics Commi ee
( e e ence no. 505/14). In o med consen was ob ained om all pa icipan s. All me hods we e pe o med in
acco dance wi h ele an guidelines and egula ions.
Cell isola ion and low cy ome y. Pe iphe al blood mononuclea cells (PBMCs) we e isola ed by
Ficoll g adien medium (His opaque-1077; Sigma-Ald ich) using SepMa e ubes (S em-Cell Technologies).
Cells we e s ained wi h an i-CD4 (OKT4, BioLegend), an i-CD45RO (UCHL1, BioLegend), an i-CD25
(BC96, eBioscience), an i-CD185/CXCR5 (J252D4, BioLegend), an i-CD279/PD-1 (EH12.2H7, BioLegend),
an i-CD278/ICOS (C398.4A, BioLegend), an i-CCR6 (G034E3, BioLegend), an i-CXCR3 (G025H7, BioLegend)
and an i-Foxp3 (PCH101, eBioscience). In acellula Foxp3 s aining was pe o med using he Foxp3 Fix/Pe m
Ki (eBioscience) acco ding o he manu ac u e ’s ins uc ions. Samples we e acqui ed on a BD LSRFo essa
cy ome e (BD Biosciences) and u he analyzed on FlowJo so wa e (T eeS a ).
S a is ical analysis. The no mali y o da a was e alua ed using he Shapi o–Wilk no mali y es . K uskal–
Wallis one-way analysis o a iance (ANOVA) es wi h Dunn’s compa ison pos - es was applied o pai wise
mul iple compa isons, and wo-way ANOVA wi h Dunne ’s es was pe o med o mul iple compa isons.
P alues less han 0.05 we e conside ed signi ican . S a is ical and g aphic analyzes we e pe o med using
G aphPad P ism so wa e.
Resul s
The ci cula ing T /T h a io is he e ogeneous among dis inc au oimmune diseases. One o he
majo challenges in compa ing s udies o ci cula ing T ollicula cells in au oimmune diseases is he inconsis ency
o he app oaches applied o selec hese cells by low cy ome y, namely he dispa i y in he ma ke s used. To
main ain a consis en s a egy o analyze T ollicula subse s by low cy ome y, allowing compa abili y among
diseases, we in es iga ed he p o ile o ci cula ing T h cells (de ined as CD4+CXCR5+CD45RO+CD25–Foxp3–)
and T cells (de ined as CD4+CXCR5+CD25+ Foxp3+) in dis inc an ibody-media ed au oimmune diseases
(Fig.1A). We included pa ien s wi h HT, ep esen ing an o gan-speci ic au oimmune disease; RA, an au oimmune
heuma ic disease wi h a ge in ol emen ; and SLE, he p o o ypical au oan ibody-d i en sys emic au oimmune
disease (Table1). All pa ien s wi h HT had hypo hy oidism wi h he p esence o hy oid-speci ic au oan ibodies,
and he majo i y we e ea ed wi h le o hy oxine. Pa ien s wi h RA and SLE we e in line wi h he na u e o an
es ablished disease coho , wi h low disease ac i i y and an expec ed equency o au oan ibodies.
Su p isingly, he equency o T h and T cells wi hin he CD4+ T cell popula ion a ied subs an ially among
he di e en diseases (Fig.1A,B). Pa ien s wi h HT showed an inc eased equency o T h cells compa ed o
HD (p = 0.0215). In con as , he equency o ci cula ing T h cells was dec eased in pa ien s wi h RA and SLE
(p = 0.0494 and p = 0.0392, espec i ely). Fu he mo e, RA and SLE pa ien s also displayed educed equency
o T cells (p = 0.0003 and p = 0.0001, espec i ely), which was no obse ed in HT pa ien s. Simila esul s we e
obse ed wi h he absolu e cell numbe s (Supplemen a y Fig.S1). These indings con as wi h ou p e ious
esul s in pa ien s wi h pSS, who had simila T h and highe T equencies han HD14.
3
Vol.:(0123456789)
Scien i ic Repo s | (2022) 12:17638 | h ps://doi.o g/10.1038/s41598-022-21576-8
www.na u e.com/scien i ic epo s/
Con a y o wha we had hypo hesized, pa ien s wi h HT had a signi ican educ ion in T /T h a io
(p = 0.0338) (Fig.1C), a he han an inc ease, as epo ed o pSS, an au oimmune disease which also has
p edominan a ge -o gan (i.e., sali a y/lac imal gland) in ol emen 14. O no e, he low T /T h a io in HT
pa ien s was mainly d i en by an inc ease in he equency o blood T h cells (Fig.1A–C), which we e p e iously
shown o be unchanged in pa ien s wi h pSS14. No co ela ion was obse ed be ween he equency and he
numbe o T h cells wi h he se um an ibody i e s o HT pa ien s (Supplemen a y Fig.S2). Fu he mo e, he
educ ions in ci cula ing T h and T cells obse ed in RA and SLE esul ed in T /T h a ios simila o heal hy
con ols. The esul s conce ning RA a e o e all in acco dance wi h wha we ha e p e iously demons a ed o
pa ien s wi h es ablished RA in emission/low disease ac i i y28.
No ably, he T /T h a io om RA and SLE pa ien s showed a signi ican in e indi idual a ia ion
(Fig.1C). We could no ind any clinical pa ame e s, such as disease du a ion, disease ac i i y, au oan ibody
s a us, o ea men ha could explain he in e indi idual a iabili y o he T /T h a io among hese pa ien s
(Supplemen a y Figs.S3 and S4). Ne e heless, he T /T h a io in pa ien s wi h RA in emission (DAS28 ≤ 2.6)
ends o be highe when compa ed o pa ien s who a e no in emission, as p e iously epo ed28. As an icipa ed,
we did no ind signi ican di e ences ega ding he equency o CD4+ T cells (Fig.1D).
O e all, hese esul s show ha di e en an ibody-media ed au oimmune diseases ha e dis inc pa e ns o
ci cula ing T h and T cells in ol ed in he egula ion o GCs, sugges ing ha hese diseases may ha e dis inc
o ms o GC dys egula ion.
Ac i a ed PD‑1+ICOS+ T h cells a e inc eased in some, bu no all, au oimmune diseases. A e
assessing he o e all T h and T popula ions, we in es iga ed he equency o T h subse s. Fi s , we quan i ied he
equency o ac i a ed T h cells, de ined as PD-1+ICOS+ cells wi hin he T h ga e (Fig.2A). The PD-1+ICOS+ cells
ep esen he popula ion o T h cells ha ansien ly inc eases in ci cula ion sho ly a e immune s imula ion,
such as ha p o ided by in luenza accina ion29–31. We ound a end o a highe equency o PD-1+ICOS+
T h cells in pa ien s wi h o gan-speci ic and sys emic au oimmune disease, in compa ison o heal hy con ols
(Fig.2A and Supplemen a y Fig.S1). This inc ease was, howe e , s a is ically signi ican only o pa ien s wi h
SLE (p = 0.0022) (Fig.2A and Supplemen a y Fig.S1). No co ela ion be ween PD-1+ICOS+ T h cells o RA and
SLE pa ien s wi h disease ac i i y was obse ed, possibly due o he low numbe o pa ien s wi h high disease
ac i i y, as hese a e es ablished disease coho s (Supplemen a y Fig.S5). Mo eo e , no co ela ion o disease
ac i i y wi h o al T h cells, bo h in numbe and pe cen age, was also obse ed (Supplemen a y Fig.S5).
Figu e1. The ci cula ing T /T h a io is he e ogeneous among di e en au oimmune diseases. (A)
Rep esen a i e plo s and (B) pooled da a o he equency o CD4+CXCR5+CD25–Foxp3– T h cells and
CD4+CXCR5+CD25+Foxp3+ T cells in pe iphe al blood o pa ien s wi h Hashimo o’s hy oidi is (HT, n = 18),
heuma oid a h i is (RA, n = 16), sys emic lupus e y hema osus (SLE, n = 32), and sex- and age-ma ched heal hy
dono s (HD, n = 31). (C) Blood T /T h a io and (D) equency o CD4+ T cells in HT, RA, SLE, and sex- and
age-ma ched HD. Each da a poin ep esen s one indi idual subjec ; ba s ep esen mean ± SEM; *p < 0.05,
**p < 0.01, ***p < 0.001, K uskal–Wallis one-way analysis o a iance (ANOVA) es wi h Dunn’s compa ison
pos - es .
4
Vol:.(1234567890)
Scien i ic Repo s | (2022) 12:17638 | h ps://doi.o g/10.1038/s41598-022-21576-8
www.na u e.com/scien i ic epo s/
I has also been epo ed ha ci cula ing T h cells can be subse ed in Th1-like, Th2-like, and Th17-like T h
cells based on CCR6 and CXCR3 exp ession2. We did no ind signi ican di e ences ega ding he equency o
T h1, T h2, and T h17 subse s in pa ien s wi h HT o SLE, compa ed o HD. Howe e , we obse ed a dec ease o
T h17 cells in RA pa ien s (p = 0.0109), wi h a con e se inc ease o he T h2 cells (p = 0.0105) (Fig.2B).
Ci cula ing CD4+PD‑1+CXCR5– Tph cells and CXCR5+CD25–Foxp3+ T ollicula cells a e inc eased
in pa ien s wi h SLE. We nex assessed he equency o ci cula ing T pe iphe al helpe (Tph) cells, de ined
as CD4+PD-1+CXCR5–, which accumula e in in lamed issues and also ha e he abili y o p o ide help o B
cells32,33. We ound ha his popula ion o cells is especially inc eased among SLE pa ien s (p = 0.0054) (Fig.3A),
as p e iously desc ibed by o he g oups34–36. Al hough he Tph equency among SLE pa ien s ollows a bimodal
dis ibu ion, wi h a small subg oup o pa ien s displaying highe le els o ci cula ing Tph cells, we did no ind an
associa ion wi h clinical pa ame e s, such as disease du a ion, disease ac i i y, au oan ibody s a us, o ea men
(Supplemen a y Fig.S6).
Finally, we assessed he equency o ci cula ing CD4+CXCR5+CD25–Foxp3+ cells. We called hese cells
CD25– T cells, gi en he simila i ies wi h he pheno ype o bona ide T cells. We ound ha ci cula ing
CD25– T cells we e inc eased only in pa ien s wi h SLE (p = 0.0096), wi h no di e ences obse ed o HT and
RA (Fig.3B). Gi en he exclusi e associa ion o his cell popula ion wi h SLE, we pe o med a ROC analysis ha
iden i ied he equency o CD25– T cells as a signi ican p edic o o SLE compa ed o HD (a ea unde he
cu e (AUC): 0.744; p = 0.0009; Supplemen a y Fig.S7). Namely, alues o CD25– T cells equal o abo e 0.61%
(o o al CD4+ T cells) ha e a sensi i i y o 62.5% and a speci ici y o 71% o he diagnosis o SLE, in compa ison
wi h HD (Supplemen a y Fig.S7).
Fu he , SLE pa ien s also had an inc ease in he o al popula ion o ci cula ing CD25– T eg cells (p = 0.0050)
(Fig.3B). This inding sugges s ha he inc ease o he CD25– T cells does no explain he g ea e equency o
o al CD25– T eg cells (independen ly o CXCR5 exp ession), which has been p e iously poin ed ou as a possible
bioma ke o he s a i ica ion o SLE pa ien s ega ding enal in ol emen 37.
Discussion
In ecen yea s we wi nessed majo ad ances ega ding he unde s anding o he egula ion o humo al esponses
leading o an ibody and au oan ibody p oduc ion. While T h cells p omo e GC o ma ion, a ini y ma u a ion,
and p oduc ion o high-a ini y an ibodies, T cells egula e he magni ude o he GC esponse and p e en
au oimmuni y in mice23,38,39. E en hough i is emp ing o assume ha au oimmuni y a ises when he T /T h
balance is shi ed by he educ ion o T cells o inc eased T h cells, he e is e idence ha human au oimmuni y
Table 1. Demog aphic and clinical cha ac e is ics o s udy pa icipan s. HD heal hy dono s, HT hashimo o’s
hy oidi is, RA heuma oid a h i is, SLE sys emic lupus e y hema osus, DAS28 disease ac i i y sco e-28,
SLEDAI sys emic lupus e y hema osus disease ac i i y index, An i-TG an i- hy oglobulin an ibodies, An i-TPO
an i- hy oid pe oxidase an ibodies, An i-CCP an i-cyclic ci ullina ed pep ides, RF heuma oid ac o , ANA
an inuclea an ibody. *The e we e no signi ican di e ences agains HD con ols. **An i-TG and an i-TPO
i e s una ailable o 2 pa ien s.
HD (n = 31) HT (n = 18) RA (n = 16) SLE (n = 32)
Age, mean ± SD yea s* 48.6 ± 17.4 49.0 ± 14.0 53.9 ± 15.7 52.8 ± 16.6
Female, no. (%)* 26 (81) 17 (94) 14 (88) 26 (81)
Disease du a ion, mean ± SD yea s – 7.8 ± 12.7 12.1 ± 10.1 12.3 ± 8.4
DAS28, mean ± SD – – 2.9 ± 1.4 –
SLEDAI, mean ± SD – – – 1.8 ± 1.8
An i-TG posi i e (> 40 U/mL), no. (%)** 12 (75)
An i-TPO posi i e (> 60 U/mL, no. (%)** – 14 (88) – –
An i-CCP posi i e, no. (%) – – 13 (81) –
RF posi i e, no. (%) – – 15 (94) –
ANA ≥ 1/160, no. (%) – – – 26 (81)
An i-SSA posi i e, no. (%) – – – 6 (19)
An i-SSB posi i e, no. (%) – – – 2 (6)
An i-Sm posi i e, no. (%) – – – 5 (16)
An i-RNP posi i e, no. (%) – – – 9 (28)
An i-dsDNA posi i e, no. (%) – – – 14 (44)
Me ho exa e, no. (%) – – 17 (100) 4 (13)
P ednisolone, no. (%) – – 12 (75) 25 (78)
Hyd oxychlo oquine, no. (%) – – 5 (31) 29 (91)
Sul asalazine, no. (%) – – 4 (25) –
Aza hiop ine, no. (%) – – – 6 (19)
Le o hy oxin, no. (%) – 17 (94) – –
5
Vol.:(0123456789)
Scien i ic Repo s | (2022) 12:17638 | h ps://doi.o g/10.1038/s41598-022-21576-8
www.na u e.com/scien i ic epo s/
Figu e2. Ac i a ed PD-1+ICOS+ T h cells a e expanded in SLE pa ien s. (A) Rep esen a i e plo s (le ) and
pooled da a ( igh ) o he equency o ac i a ed PD-1+ICOS+ T h cells in he pe iphe al blood o HT (n = 18),
RA (n = 16), and SLE (n = 32) pa ien s, and sex- and age-ma ched HD (n = 18). Each da a poin ep esen s
an indi idual subjec ; ba s ep esen mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001, K uskal–Wallis one-way
analysis o a iance (ANOVA) es wi h Dunn’s compa ison pos - es (B) Rep esen a i e plo s (le ) and pooled
da a ( igh ) o he dis ibu ion o Th1-like (CXCR3+CCR6–), Th2-like (CXCR3–CCR6–), and Th17-like (CXCR3–
CCR6+) T h cell subse s in HT (n = 18), RA (n = 16), and SLE (n = 32) pa ien s, and sex- and age-ma ched HD
(n = 19). *p < 0.05, **p < 0.01, ***p < 0.001, wo-way ANOVA wi h Dunne ’s mul iple compa isons es was
pe o med.
Figu e3. SLE pa ien s show an inc eased equency o ci cula ing Tph and CXCR5+CD25–Foxp3+ T ollicula
cells. Rep esen a i e plo s (le ) and pooled da a ( igh ) o he equency o (A) Tph cells and (B) CXCR5+CD25–
Foxp3+ T cells in he pe iphe al blood o HT (n = 18), RA (n = 16), and SLE (n = 32) pa ien s, and sex- and age-
ma ched HD (A: n = 16; B: n = 31). Each da a poin ep esen s an indi idual subjec ; ba s ep esen mean ± SEM;
*p < 0.05, **p < 0.01, ***p < 0.001, K uskal–Wallis one-way analysis o a iance (ANOVA) es wi h Dunn’s
compa ison pos - es .

6
Vol:.(1234567890)
Scien i ic Repo s | (2022) 12:17638 | h ps://doi.o g/10.1038/s41598-022-21576-8
www.na u e.com/scien i ic epo s/
can coexis wi h a blood T /T h a io shi ed in he opposi e di ec ion14. Se e al explana ions may accoun o
his appa en ly coun e -in ui i e inding. One possibili y is ha he equency o T ollicula cells in blood may
no di ec ly e lec hei incidence in he seconda y lymphoid o gans o ELS. S udies showing T ollicula cells in
he a ec ed o gans in human au oimmuni y a e s ill a e, meaning ha his co ela ion be ween blood and issue
has no been ye ully es ablished40,41. Ano he possibili y s ems om he obse a ion ha ch onic s imula ion
by au oan igens, and he subsequen humo al ac i i y, can lead o pe sis en p oduc ion o ci cula ing T cells
(as i occu s ollowing in luenza accina ion)4.
Howe e , he e a e se e al epo s wi h inconsis en esul s, ei he o di e en diseases o ocused on he
same disease, as di e en me hods a e used o de ine T and T h subse s. The same designa ion (T h o T )
has been used wi h cell popula ions wi h dispa a e pheno ypic cha ac e is ics. Fo ins ance, T h cells a e o en
conside ed as CD4+CXCR5+ cells, wi hou exclusion o Foxp3+ T cells26,27. These inconsis encies make i di icul
o compa e he di e en s udies, hus no allowing o in e whe he he e is a uni o m T /T h dys egula ion ha
explains he pa hogenesis o humo al imbalance in human au oimmuni y.
We s udied a g oup o o gan-speci ic and sys emic au oimmune diseases, using a consis en analysis
app oach, o in es iga e whe he common cha ac e is ics ega ding he equency o T and T h subse s
could be es ablished. We ound ha he e is a ma ked he e ogenei y be ween di e en au oan ibody-media ed
au oimmune diseases ega ding he equency o ci cula ing T and T h subse s. I emains unclea whe he he
obse ed changes in T ollicula cells a e he cause o he consequence o he humo al immune dys egula ion
unde lying hese diseases. As a esul , i is possible ha some a e ac ual au oan ibody-media ed condi ions,
whe eas o he s a e diseases ha ha e de ec able au oan ibodies, bu a e no necessa ily media ed by hem, hus
con ibu ing o he he e ogenei y we obse ed. This could be o e come by accessing he dys egula ion o he
ge minal cen e esponse di ec ly in he a ec ed issues, bo h in he a ge o gan and in d aining lymphoid issue
whe e au oan ibody p oduc ion occu s. Un o una ely, in he p esen s udy we compa ed di e en diseases wi h
dis inc a ge issues. The e o e, a issue-based app oach would isk inc easing he he e ogenei y o he analysis,
u he complica ing he in e p e a ion o ou and p e ious indings.
Ou indings suppo he hypo hesis ha T ollicula cell dys egula ion di e s ac oss au oimmune diseases,
bo h o gan-speci ic and sys emic. Howe e , we should unde line ha his s udy was no designed o iden i y
blood bioma ke s o speci ic au oimmune diseases, based on T h/T cell pa e ns. Such a s udy would ha e
limi ed applicabili y as HT, RA and SLE a e e y di e en diseases om a clinical poin o iew and, he e o e,
he use o a bioma ke would likely no be clinically use ul. Howe e , he e may be some po en ial o hese ini ial
da a o be u he explo ed wi hin each disease, aiming a pa ien s a i ica ion o e en pe sonalized ea men .
Indeed, he di e se pa e ns obse ed wi hin a pa icula disease, aise he hypo hesis ha di e en subg oups o
pa ien s ha e dis inc pa e ns o humo al dys egula ion, which possibly also a y o e ime. I is well es ablished
ha some clinical ea u es, such as posi i i y o speci ic au oan ibodies o he p esence o ELS can occu in only
a subg oup o pa ien s wi h he same disease, such as in RA o pSS14,42–44.
This ype o he e ogenei y, linked o dis inc immunopa hogenesis, may p o ide an impo an explana ion
o di e en esponses obse ed o he same ea men egimens. Thus, s a egies o u he s a i y pa ien s
acco ding o he unde lying pa e n o immune dys egula ion, using T /T h as a p oxy, may con ibu e o
de elop ailo ed indi idualized ea men s. In ac , we and o he s ha e p e iously shown ha he equency o
ci cula ing PD-1+ICOS+ T h cells could be used o iden i y pSS pa ien s wi h g ea e disease ac i i y14,15, whe eas
he T /T h a io could iden i y hose pa ien s wi h ELS wi hin sali a y glands14. This suppo s ou hypo hesis
ha , in speci ic diseases, subpopula ions o T and T h cells may allow pa ien s a i ica ion. In addi ion, o he
componen s o he in lamma o y esponse, namely se um cy okine p o iles, can po en ially con ibu e o a be e
cha ac e iza ion o unique ea u es o dis inc au oimmune diseases and subg oups o pa ien s wi h he same
disease, p o iding addi ional ools o a mo e e ined pa ien s a i ica ion.
Ou s udy has se e al limi a ions. Pa ien s wi h dis inc diseases could no be ully ma ched o demog aphic
o clinical cha ac e is ics, such as ea men o disease du a ion. Thus, i is possible ha speci ic d ugs o disease
s age (i.e., ea ly/la e) may a ec T and T h subse s di e en ly. Ne e heless, we we e unable o ind an associa ion
be ween he obse ed he e ogenei y in T and T h subse s and speci ic ea men s o o he clinical pa ame e s,
such as disease du a ion o se ological ma ke s. In addi ion, he numbe o pa ien s included in each g oup may
be ela i ely small, conside ing he desc ibed disease he e ogenei y, especially in RA and SLE. Analyzing a la ge
numbe o pa ien s wi h such condi ions may o e come his issue, especially wi h ega d o pa ien s a i ica ion,
hus allowing o mo e obus conclusions. Also, he he e ogenei y ound in T and T h subse s in pa ien s unde
di e en ea men s (Supplemen a y Fig.S8) ein o ces he need o la ge s udies. I is also possible ha he
pa e n o dys egula ion in issues wi h humo al esponses d i ing au oan ibody p oduc ion may di e om he
con en o T ollicula subse s in ci cula ion. Ne e heless, he pe iphe al blood compa men sha es common
ea u es ac oss diseases, namely he p esence o ci cula ing au oan ibodies and dys egula ed humo al esponses.
In addi ion, i is easie o access and mo e ho oughly s udied, wi h equen claims o ci cula ing T h/T
dys egula ion as a uni e sal hallma k au oan ibody p oduc ion—an assump ion we ha e shown canno be made.
In conclusion, humo al dys egula ion in au oimmuni y canno be asc ibed o a common mechanism ela ed o
T ollicula cells. Howe e , he he e ogenei y o T and T h subse s ound ac oss di e en au oimmune diseases,
and wi hin speci ic pa ien subg oups, o e s an oppo uni y o explo e po en ial bioma ke s o he di e en
diseases and o u he s udies on pa ien s a i ica ion. This may po en ially con ibu e o de eloping ailo ed
he apeu ic s a egies and imp o ing pa ien ou comes.
Da a a ailabili y
The da a unde lying his a icle will be sha ed on easonable eques o he co esponding au ho L.G.
7
Vol.:(0123456789)
Scien i ic Repo s | (2022) 12:17638 | h ps://doi.o g/10.1038/s41598-022-21576-8
www.na u e.com/scien i ic epo s/
Recei ed: 21 July 2022; Accep ed: 29 Sep embe 2022
Re e ences
1. Sage, P. T., Al a ez, D., Godec, J., Von And ian, U. H. & Sha pe, A. H. Ci cula ing T ollicula egula o y and helpe cells ha e
memo y-like p ope ies. J. Clin. In es . 124, 5191–5204 (2014).
2. Schmi , N., Ben ebibel, S. E. & Ueno, H. Pheno ype and unc ions o memo y T h cells in human blood. T ends Immunol. 35,
436–442 (2014).
3. Mo i a, R. e al. Human blood CXCR5+CD4+ T cells a e coun e pa s o T ollicula cells and con ain speci ic subse s ha
di e en ially suppo an ibody sec e ion. Immuni y 34, 108–121 (2011).
4. Fonseca, V. R. e al. Human blood T cells a e indica o s o ongoing humo al ac i i y no ully licensed wi h supp essi e unc ion.
Sci. Immunol. 1487, 1–13 (2017).
5. Wollenbe g, I. e al. Regula ion o he ge minal cen e eac ion by Foxp3+ ollicula egula o y T cells. J. Immunol. 187, 4553–4560
(2011).
6. Chung, Y. e al. Follicula egula o y T cells exp essing Foxp3 and Bcl-6 supp ess ge minal cen e eac ions. Na . Med. 17, 983–988
(2011).
7. Lin e man, M. A. e al. Foxp3+ ollicula egula o y T cells con ol he ge minal cen e esponse. Na . Med. 17, 975–982 (2011).
8. Macei as, A. R. e al. T ollicula helpe and T ollicula egula o y cells ha e di e en TCR speci ici y. Na . Commun. 8, 1–12
(2017).
9. Sage, P. T. e al. Supp ession by T cells leads o du able and selec i e inhibi ion o B cell e ec o unc ion. Na . Immunol. 17,
1436–1446 (2016).
10. Wing, J. B., Ise, W., Ku osaki, T. & Sakaguchi, S. Regula o y T cells con ol an igen-speci ic expansion o T h cell numbe and
humo al immune esponses ia he co ecep o CTLA-4. Immuni y 41, 1013–1025 (2014).
11. Vae h, M. e al. Follicula egula o y T cells con ol humo al au oimmuni y ia NFAT2- egula ed CXCR5 exp ession. J. Exp. Med.
211, 545–561 (2014).
12. Fu, W. e al. De iciency in T ollicula egula o y cells p omo es au oimmuni y. J. Exp. Med. 215, 815–825 (2018).
13. Sage, P. T. & Sha pe, A. H. T ollicula egula o y cells in he egula ion o B cell esponses. T ends Immunol. 36, 410–418 (2015).
14. Fonseca, V. R. e al. The a io o blood T ollicula egula o y cells o T ollicula helpe cells ma ks ec opic lymphoid s uc u e
o ma ion while ac i a ed ollicula helpe T cells indica e disease ac i i y in p ima y Sjög en’s Synd ome. A h i is Rheuma ol.
70, 774–784 (2018).
15. Ve s appen, G. M. e al. Is he T ollicula egula o y: ollicula helpe T cell a io in blood a bioma ke o ec opic lymphoid
s uc u e o ma ion in Sjög en’s Synd ome? Commen on he a icle by Fonseca e al.. A h i is Rheuma ol. 70, 1354–1355. h ps://
doi. o g/ 10. 1002/ a . 40488 (2018).
16. Aq awi, L. A. e al. Diminished CXCR5 exp ession in pe iphe al blood o pa ien s wi h Sjög en’s synd ome may ela e o bo h
geno ype and sali a y gland homing. Clin. Exp. Immunol. 192, 259–270 (2018).
17. Dhaeze, T. e al. Ci cula ing ollicula egula o y T cells a e de ec i e in mul iple scle osis. J. Immunol. 195, 832–840 (2015).
18. Xu, B. e al. The a io o ci cula ing ollicula T helpe cell o ollicula T egula o y cell is co ela ed wi h disease ac i i y in sys emic
lupus e y hema osus. Clin. Immunol. 183, 46–53 (2017).
19. Shan, Y. e al. Highe equency o pe iphe al blood ollicula egula o y T cells in pa ien s wi h new onse ankylosing spondyli is.
Clin. Exp. Pha macol. Physiol. 42, 154–161 (2015).
20. Kuma , S. e al. De elopmen al bi u ca ion o human T ollicula egula o y cells. Sci. Immunol. 6, 1–15 (2021).
21. Macei as, A. R., Fonseca, V. R., Agua-Doce, A. & G aca, L. T ollicula egula o y cells in mice and men. Immunology 152, 25–35
(2017).
22. Bo a, D. e al. Dynamic egula ion o T ollicula egula o y cell esponses by in e leukin 2 du ing in luenza in ec ion. Na .
Immunol. 18, 1249–1260 (2017).
23. Fonseca, V. R., Ribei o, F. & G aca, L. T ollicula egula o y (T ) cells: Dissec ing he complexi y o cell compa men s. Immunol.
Re . 288, 112–127 (2019).
24. Wing, J. B. e al. A dis inc subpopula ion o CD25(-) T ollicula egula o y cells localizes in he ge minal cen e s. P oc. Na l. Acad.
Sci. U S A 114, E6400–E6409 (2017).
25. Ri o, P.-G.G. e al. T cells lack IL-2Rα bu exp ess decoy IL-1R2 and IL-1Ra and supp ess he IL-1–dependen ac i a ion o T h
cells. Sci. Immunol. 2, eaan0368 (2017).
26. Wang, X. e al. Imbalance o ci cula ing T /T h a io in pa ien s wi h heuma oid a h i is. Clin. Exp. Med. 19, 55–64 (2019).
27. Deng, J. e al. Signal ansduce and ac i a o o ansc ip ion 3 hype ac i a ion associa es wi h ollicula helpe T cell di e en ia ion
and disease ac i i y in heuma oid a h i is. F on . Immunol. 9, 1–11 (2018).
28. Romão, V. C., Fonseca, J. E., Agua-Doce, A. & G aca, L. T ollicula egula o y cells a e dec eased in pa ien s wi h es ablished
ea ed heuma oid a h i is wi h ac i e disease: Commen on he a icle by Liu e al.. A h i is Rheuma ol. 70, 1893–1895 (2018).
29. Ben ebibel, S. E. e al. Induc ion o ICOS+CXCR3+CXCR5+ T H cells co ela es wi h an ibody esponses o in luenza accina ion.
Sci. T ansl. Med. 5, 1–11 (2013).
30. Pilkin on, M. A. e al. G ea e ac i a ion o pe iphe al T ollicula helpe cells ollowing high dose in luenza accine in olde adul s
o ecas s se ocon e sion. Vaccine 35, 329–336 (2017).
31. He a i, R. S. e al. Successi e annual in luenza accina ion induces a ecu en oligoclono ypic memo y esponse in ci cula ing T
ollicula helpe cells. Sci. Immunol. 2, eaag2152 (2017).
32. Rao, D. A. T cells ha help B cells in ch onically in lamed issues. F on . Immunol. 9, 1924 (2018).
33. Yoshi omi, H. & Ueno, H. Sha ed and dis inc oles o T pe iphe al helpe and T ollicula helpe cells in human diseases. Cell.
Mol. Immunol. 18, 523–527 (2021).
34. Bocha niko , A. V. e al. PD-1hiCXCR5- T pe iphe al helpe cells p omo e B cell esponses in lupus ia MAF and IL-21. JCI Insigh
4, e130062 (2019).
35. Lin, J., Yu, Y., Ma, J., Ren, C. & Chen, W. PD-1+CXCR5-CD4+T cells a e co ela ed wi h he se e i y o sys emic lupus
e y hema osus. Rheuma ology 58, 2188–2192 (2019).
36. Makiyama, A. e al. Expanded ci cula ing pe iphe al helpe T cells in sys emic lupus e y hema osus: Associa ion wi h disease
ac i i y and B cell di e en ia ion. Rheuma ology 58, 1861–1869 (2019).
37. Bonelli, M. e al. CD4+CD25-Foxp3+ T cells: A ma ke o lupus neph i is?. A h i is Res. The . 16, R104 (2014).
38. Wing, J. B., Tekgüç, M. & Sakaguchi, S. Con ol o ge minal cen e esponses by T ollicula egula o y cells. F on . Immunol. 9,
1910 (2018).
39. Vinuesa, C. G., Lin e man, M. A., Yu, D. & MacLennan, I. C. M. Follicula helpe T cells. Annu. Re . Immunol. 34, 335–368 (2016).
40. Chu, Y., Wang, F., Zhou, M., Chen, L. & Lu, Y. A p elimina y s udy on he cha ac e iza ion o ollicula helpe T (T h) cells in
heuma oid a h i is syno ium. Ac a His ochem. 116, 539–543 (2014).
41. Rao, D. A. e al. Pa hologically expanded pe iphe al T helpe cell subse d i es B cells in heuma oid a h i is. Na u e 542, 110–114
(2017).
8
Vol:.(1234567890)
Scien i ic Repo s | (2022) 12:17638 | h ps://doi.o g/10.1038/s41598-022-21576-8
www.na u e.com/scien i ic epo s/
42. Co sie o, E., Ne iani, A., Bomba die i, M. & Pi zalis, C. Ec opic lymphoid s uc u es: Powe house o au oimmuni y. F on .
Immunol. 7, 430 (2016).
43. Dua e, J. H. Hi ing he b akes on ec opic lymphoid s uc u e o ma ion. Na . Re . Rheuma ol. 11, 621–621 (2015).
44. Bomba die i, M. & Pi zalis, C. Ec opic lymphoid neogenesis and lymphoid chemokines in Sjog en’s Synd ome: a he in e play
be ween ch onic in lamma ion, au oimmuni y and lymphomagenesis. Cu . Pha m. Bio echnol. 13, 1989–1996 (2012).
Acknowledgemen s
We would like o hank he Biobank o Ins i u o de Medicina Molecula João Lobo An unes, Faculdade de
Medicina, Uni e sidade de Lisboa, Lisbon Academic Medical Cen e, o hei help in sample collec ion. We
also hank o he Flow Cy ome y Facili y o Ins i u o de Medicina Molecula João Lobo An unes, Faculdade
de Medicina, Uni e sidade de Lisboa, Lisbon Academic Medical Cen e, o hei se ices and assis ance. We
hank o he Endoc inology and Rheuma ology Depa men s o Hospi al de San a Ma ia, Cen o Hospi ala
Uni e si á io Lisboa No e, Lisbon Academic Medical Cen e, o hei collabo a ions. We g a e ully hank he
pa icipa ion o all olun ee s who dona ed blood samples o his s udy.
Au ho con ibu ions
L.G. and V.F. p oposed and designed he s udy. L.G., F.R. and V.C.R. w o e he manusc ip . F.R., S.R., K.J.
and A.A.D. pe o med he expe imen s. V.C.R., S.C.B, P.M., S.L.S., E.M., M.J.B. and J.E.F. con ibu ed o he
ec ui men o he pa ien s. All au ho s we e esponsible o he c i ical e ision o he manusc ip .
Funding
This wo k was suppo ed by he Fundação pa a a Ciência e Tecnologia, Po ugal (EJPRD/0003/2019).
Compe ing in e es s
The au ho s decla e no compe ing in e es s.
Addi ional in o ma ion
Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps:// doi. o g/
10. 1038/ s41598- 022- 21576-8.
Co espondence and eques s o ma e ials should be add essed o L.G.
Rep in s and pe missions in o ma ion is a ailable a www.na u e.com/ ep in s.
Publishe ’s no e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and
ins i u ional a ilia ions.
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional
License, which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o
o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he
C ea i e Commons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his
a icle a e included in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he
ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no
pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om
he copy igh holde . To iew a copy o his licence, isi h p:// c ea i eco mmons. o g/ licen ses/ by/4. 0/.
© The Au ho (s) 2022