ORIGINAL RESEARCH
published: 06 Sep embe 2019
doi: 10.3389/ immu.2019.02116
F on ie s in Immunology | www. on ie sin.o g 1Sep embe 2019 | Volume 10 | A icle 2116
Edi ed by:
Je oen E. J. Guikema,
Academic Medical Cen e
(AMC), Ne he lands
Re iewed by:
Claude-Agnes Reynaud,
Ins i u Na ional de la San é e de la
Reche che Médicale
(INSERM), F ance
Wanli Liu,
Tsinghua Uni e si y, China
*Co espondence:
Michael Meye -He mann
mmh@ heo e ical-biology.de
Special y sec ion:
This a icle was submi ed o
B Cell Biology,
a sec ion o he jou nal
F on ie s in Immunology
Recei ed: 30 No embe 2018
Accep ed: 22 Augus 2019
Published: 06 Sep embe 2019
Ci a ion:
A ul aj T, Binde SC, Robe PA and
Meye -He mann M (2019)
Synch onous Ge minal Cen e Onse
Impac s he E iciency o An ibody
Responses. F on . Immunol. 10:2116.
doi: 10.3389/ immu.2019.02116
Synch onous Ge minal Cen e Onse
Impac s he E iciency o An ibody
Responses
Theinmozhi A ul aj1, Sebas ian C. Binde 1,2, Philippe A. Robe 1and
Michael Meye -He mann1,2,3*
1Depa men o Sys ems Immunology, B aunschweig In eg a ed Cen e o Sys ems Biology, Helmhol z Cen e o In ec ion
Resea ch, B aunschweig, Ge many, 2Cen e o Indi idualized In ec ion Medicine (CIIM), Hano e , Ge many, 3Ins i u e o
Biochemis y, Bio echnology and Bioin o ma ics, Technische Uni e si ä B aunschweig, B aunschweig, Ge many
The ge minal cen e eac ion is an impo an a ge o modula ing an ibody esponses.
An ibody p oduc ion om ge minal cen e s is egula ed by a nega i e eedback
mechanism e med an ibody eedback. By imposing an ibody eedback, ge minal
cen e s can in e ac and egula e he ou pu o o he ge minal cen e s. Using an
agen -based model o he ge minal cen e eac ion, we s udied he impac o an ibody
eedback on kine ics and e iciency o a ge minal cen e . Ou simula ions p edic ha
high eedback o an ibodies om ge minal cen e s educes he p oduc ion o plasma
cells and subsequen ly he e iciency o he ge minal cen e eac ion by p omo ing
ea lie e mina ion. A ini y ma u a ion is only weakly imp o ed by inc eased an ibody
eedback and ul ima ely in e up ed because o p ema u e e mina ion o he eac ion.
The model p edic s ha he asynch onous onse and changes in numbe o ge minal
cen e s could al e he e iciency o an ibody esponse due o changes in eedback by
soluble an ibodies. Consequen ly, la e ini ialized ge minal cen e s ha e a comp omised
ou pu due o highe an ibody eedback om he ge minal cen e s o med ea lie . The
esul s demons a e po en ial e ec s o ge minal cen e in e communica ion and highligh
he impo ance o unde s anding ge minal cen e in e ac ions o op imizing he an ibody
esponse, in pa icula , in he elde ly and in he con ex o accina ion.
Keywo ds: ge minal cen e , an ibody p oduc ion, accina ion, compu e simula ion, ma hema ical modeling
INTRODUCTION
Induc ion o an app op ia e an ibody esponse is c i ical o humo al immuni y and e icien
pa hogen clea ance. Vaccina ion elies p ima ily on modula ing an ibody esponses and gene a ing
immune memo y o boos he immune sys em agains pa hogens (1–4). T cell dependen an ibody
esponses a e media ed by ge minal cen e s (GCs) whe e high-a ini y plasma cells a e o med
s a ing om B cells wi h ela i ely lowe a ini ies. Du ing he GC eac ion, he B cell ecep o
(BCR) is di e si ied by soma ic hype mu a ion, ollowed by selec ion o B cells wi h highe a ini y
BCRs. B cell selec ion in GCs is T-cell media ed, whe e B cells wi h highe a ini y BCR cap u e
highe amoun s o an igen om ollicula dend i ic cells (FDCs) and p esen he p ocessed an igen
o he T cells h ough pMHC (5–7). Al e ing he GC eac ion is a p omising way o modula e
an ibody esponses (8). Injec ion o soluble an igen impac s he apop osis and a ini y ma u a ion in
he GCs (9,10). Ex ended an igen a ailabili y has been shown o enhance he GC esponse (11,12).
A ul aj e al. GC E iciency Impac ed by An ibody Feedback
Fu he , compu a ional simula ions ha e shown ha ex ended
an igen dosing can inc ease he GC esponse by inc easing
an igen cap u e (13).
An ibodies enhance o supp ess an ibody esponses and
mechanisms go e ning his a e ac i ely being s udied
(14,15). Masking o an igen epi ope by soluble an ibodies
is one o he supp essi e mechanisms ha has long
been ecognized (16). Be gs öm e al. ha e shown ha
adminis a ion o IgG supp esses ex a ollicula an ibody
sec e ing cells, GC B cells, long- e m plasma cells, IgG esponses,
and induc ion o memo y esponse and demons a ed
ha an igen clea ance is unlikely o be he mechanism
unde lying he obse ed e ec s (17). Epi ope speci ici y o
an ibody esponse supp essed by injec ed an ibodies a e also
obse ed (17,18).
Injec ed an ibodies a e ound o be deposi ed on FDCs
and al e he apop osis and a ini y ma u a ion o B cells
in he GCs (19). Hence, a mechanism o sel - egula ion
o GCs by he an ibodies p oduced om plasma cells is
p oposed. An ibody eedback modula es an igen a ailabili y
indi ec ly by masking an igen on FDCs and hus compe e
wi h he B cells o he an igen displayed on he su ace o
FDCs wi h dynamics de e mined by GC ou pu (19). In silico
analysis and simula ions p edic ed ha he injec ion o soluble
an ibodies p omo es p ope shu down o he GC eac ions
and quicke a ini y ma u a ion due o inc eased selec ion
e iciency (19). This also sugges s ha selec ion o B cells
in he GCs could be in luenced by he in e communica ion
be ween GCs due o soluble an ibodies (19). E ec s o al e ing
an igen a ailabili y and T h help a e also being s udied
ex ensi ely in he con ex o de eloping b oadly neu alizing
an ibodies (12,20).
Ma hema ical models a e being de eloped and employed o
iden i ying and unde s anding he mechanisms o many non-
in ui i e biological p ocesses (21,22). In silico simula ions ha e
acili a ed a be e unde s anding o he B cell-T cell in e ac ions
in spleen (23), GC eac ion and in e p e a ion o expe imen al
esul s conce ning GC kine ics, a ini y ma u a ion and an ibody
p oduc ion (13,20,24–30).
Unde s anding he mechanisms ha egula e an ibody
esponses is impo an o de ising speci ic op imiza ion
s a egies o imp o e he accina ion esponse. The e ec s
o GC-GC in e ac ions due o soluble an ibodies on
indi idual GC eac ions a e no known. He e, we ocus
on unde s anding he con ibu ion o in e ac ion be ween
GCs in a ec ing he an ibody esponses by ex ending a
p e iously de eloped agen -based model o he GC eac ion
o co e in e -GC in e ac ion and ela ed ead-ou s (see
Ma e ials and me hods). We s udy he impac o an ibody
eedback on shu down and a ini y ma u a ion o GC
eac ions by a ying he s eng h o an ibody eedback.
We also in es iga e he e ec o an ibody eedback when
he GC onse is delayed a e he onse o ea lie GCs
al eady p oducing an ibody. We p opose ha a change
in he numbe o ge minal cen e s and asynch onous GC
ini ia ion could ha e implica ions in GC unc ion due o al e ed
an ibody eedback.
MATERIALS AND METHODS
Cells a e ep esen ed as agen s on a h ee-dimensional la ice
wi h a la ice cons an o 5 µm. The GC eac ion olume is
a sphe e o adius 160 µm wi hin he la ice and is di ided
equally in o da k and ligh zones. Founde B cells en e he da k
zone a a a e o 2 cells/h and di ide six imes (30). Di iding
cen oblas s mu a e a a p obabili y o 0.5 s a ing om day 1
o he eac ion. Cen oblas s di e en ia ing o cen ocy es sea ch
o an igen on FDCs and hei success ul con ac depends on he
BCR a ini y. A ou -dimensional shape space is used o a ini y
ep esen a ion (31). B cells ha ailed o collec an igen wi hin
he collec ion pe iod unde go apop osis. Fu he , T h signaling is
pola ized owa d he B cell ha collec ed he maximum amoun
o an igen, hus, p e e en ially selec ing high a ini y B cells
which we e mo e e icien in collec ing and p ocessing an igen.
Selec ed cells a he bo de o he eac ion olume exi owa d
he T zone and di e en ia e in o an ibody p oducing plasma
cells a a a e o ln 2
24 h−1. An agen -based ep esen a ion was
chosen in o de o ep esen he beha io o indi idual cells and
hei in e ac ions wi h a high deg ee o accu acy and he model
has been adap ed (26,29) o be consis en wi h expe imen al
indings on spa ial dynamics, selec ion (32), and e olu ion o
clonal dominance (33). A de ailed desc ip ion o he model is
p o ided in he Supplemen a y Ma e ial.
An ibody P oduc ion and Feedback
An ibodies a e esol ed in o 11 bins (i=0, 1, ..., 10) e lec ing
hei a ini ies. Change in he concen a ion o an ibody A(i)in
each bin i ollows he equa ion:
dA (i)
d =k1np(i)−k2A(i)(1)
The concen a ion o an ibody in each bin is inc eased a a a e
k1, e lec ing he p oduc ion o an ibodies wi h 10−17 mol/h
om each plasma cell (np(i) is he numbe o plasma cells wi h
a ini y co esponding o bin i) and hei dilu ion o e a olume
o 10 ml. Hence, he o al an ibody p oduced is dilu ed o e
he whole o ganism and he concen a ion is assumed o be he
same e e ywhe e. An ibodies ha e a hal -li e o 30 days (k2=
ln 2
30 day−1).
An ibodies a ibu ed o he di e en bins ha e he same
associa ion a e cons an kon (106M−1.s−1) (34), while he ko
a ies such ha hei dissocia ion cons an s a e be ween 10−5.5
and 10−9.5 M. These an ibodies o m immune complex (CFDC)
wi h an igen displayed on each FDC si e (GFDC) ollowing he
dynamic equa ion:
dCFDC (i)
d =konGFDC NA (i)+Aea ly (i)−ko (i)CFDC (i)(2)
He e, A(i)is he an ibodies p oduced by he simula ed GC
and Aea ly (i)is he an ibodies p oduced by ea ly GC used in
he simula ions o delayed ini ializa ion o GC (see me hods–
Simula ion o delayed ini ializa ion o GCs). Nis a scaling ac o
con olling he s eng h o an ibody eedback on o he simula ed
GC. We assume ha GCs h oughou he o ganism concu en ly
F on ie s in Immunology | www. on ie sin.o g 2Sep embe 2019 | Volume 10 | A icle 2116
A ul aj e al. GC E iciency Impac ed by An ibody Feedback
p oduce an ibodies which a e homogeneously dis ibu ed on he
whole o ganism and in pa icula appea in he simula ed GC.
We assume ha he soluble an ibodies e e sibly bind o
FDC an igen and bound an igen is no a ailable o up ake by
B cells. This esul s in compe i ion be ween B cells and soluble
an ibodies o bind o he an igen. Wi h each success ul con ac
wi h FDCs, B cells consume an an igen po ion equi alen o
10−8M. The amoun o ee an igen dec eases o inc eases due o
immune complex o ma ion o dissocia ion. The dec ease in he
concen a ion o an ibodies due o immune complex o ma ion
is neglec ed.
Immune Powe Calcula ion
A measu e o he e iciency o a GC eac ion e med immune
powe (IP) is in oduced (35), which e lec s a combina ion
o a ini y ma u a ion and he amoun o p oduced an ibody
o ming plasma cells. The IP es ima es he abili y o an ibodies
p oduced by a GC in binding he an igen. Fo his, we use an
an igen concen a ion (G) o 10−6M and es ima e he p opo ion
o an igen bound o he an ibodies o a pa icula a ini y. We
assume a high concen a ion o an igen when compa ed o he
concen a ion o an ibodies and use he ollowing s eady s a e
app oxima ion o calcula e he concen a ion o bound an igen:
Gbound (i)=A(i)G
K(i)+G(3)
whe e iis he bin numbe , A(i) is he concen a ion o an ibody
p oduced by he simula ed GC and K(i)=10−5.5−0.4iis he
dissocia ion cons an . IP is calcula ed as he a io o an igen
bound o he soluble an ibodies o he o al an igen:
IP =PiGbound (i)
G(4)
IP combines he e ec s o changes in quali y (a ini y) as well as
quan i y (concen a ion) o he an ibodies p oduced in binding
he an igen.
Simula ion o Delayed Ini ializa ion o GCs
To simula e a GC ini ialized wi h a ce ain delay (wi h espec
o ea lie ini ialized GC), we simula e his GC eac ion (N=
1) along wi h ex e nally added an ibodies Aea ly (see Equa ion
2) a e e y ime s ep. The added an ibodies co espond o
soluble an ibodies p oduced om ea lie ini ialized GC. Fo
his, we gene a e an an ibody concen a ion p o ile o a GC
ini ialized i s and escale i wi h a scaling ac o o 300 o
e lec an an ibody concen a ion o s ong eedback. Delay
o ini ializa ion o he simula ed GC is achie ed by adding
he ex e nally added an ibodies a e shi ing he an ibody
concen a ion p o ile depending on he ime o delay. Hence, he
GC ini ialized la e is unde he in luence o an ibody eedback
due o ea lie ini ialized GCs. Feedback o an ibodies p oduced
om he la e GC on ea ly GC is igno ed.
RESULTS
We use he p e iously desc ibed agen -based model o es
he in luence o an ibody eedback on he ou pu o he GC
eac ion wi h a ocus on he e mina ion o he GC eac ion
and a ini y ma u a ion o B cells. We a y he scaling ac o
(N) o an ibody p oduc ion as a p oxy o a ying he s eng h
o an ibody eedback. Inc easing N is simila o inc easing he
numbe o synch onous GCs p oducing an ibodies al hough we
a e simula ing a ep esen a i e GC. He e, we use he scaling
ac o s 1, 10, 30, and 300.
To moni o he binding o soluble an ibodies o an igen
displayed on FDCs, we plo he ac ion o o al an igen on FDCs
ha is a pa o immune complex o med wi h soluble an ibodies.
Wi h scaling ac o N=1, app oxima ely 65% o he o al an igen
is bound o soluble an ibodies on day 21 (Figu e 1A). Howe e ,
as he eedback s eng h is inc eased (N=10, 30, and 300),
he p opo ion o immune complex inc eases mo e quickly and
almos all an igen on FDCs is co e ed wi h an ibodies a an
ea lie ime poin (Figu e 1A). The p opo ion o an igen no
bound o soluble an ibodies is a ailable o up ake by B cells
sea ching o an igen on FDCs. Consequen ly, he concen a ion
o ee an igen a ailable o B cells d ops mo e quickly wi h
inc easing an ibody eedback (Figu e 1B), which inc eases he
selec ion p essu e o B cells.
In o de o unde s and he p og ess and du a ion o he GC
eac ion, we plo he GC olume de ined as he o al numbe s o
cen oblas s and cen ocy es p io o he o ma ion o ou pu cells
ha a e in u n capable o di e en ia ing o an ibody p oducing
plasma cells. Wi h inc easing eedback, he GC olume eaches
ze o mo e quickly, sugges ing ha he e mina ion is accele a ed
due o an ibody eedback (Figu e 1C). This is consis en wi h
he esul s o p e ious s udies (19). The numbe o plasma cells
di e en ia ed om ou pu cells is highe wi h lowe an ibody
eedback and i dec eases ma kedly in he case o highe eedback
s eng hs (Figu e 1D). This obse a ion is a consequence o he
low an igen a ailabili y o B cells ha inc eases i s selec ion
p essu e esul ing in he selec ion o ewe B cells.
We u he in es iga e he a ini y ma u a ion p ocess, as his
could also be a a ge due o he selec ion p essu e induced
by he dec eased an igen a ailabili y o B cells. App oxima ely
a e 10 days, he a ini y in he case o low an ibody eedback
is highe (Figu e 2) bu he e ec is weak compa ed o ha
on he numbe o plasma cells p oduced. The lowe mean
a ini y obse ed a la e ime poin s is a consequence o ea lie
shu down. Hence, he ea lie shu down has dec eased he ime
o e ec i e a ini y ma u a ion and, hus, supp essing a u he
inc ease in a ini y. This sugges s ha p ema u e e mina ion o
he GC eac ion would no only dec ease he amoun o selec ed
ou pu cells bu could also dec ease he e ec i eness o he a ini y
ma u a ion p ocess.
In o de o quan i a i ely cap u e he p oduc o bo h e ec s o
inc eased an ibody eedback, namely he in e media e change in
a ini y and he educed p oduc ion o ou pu cells, we calcula e
he immune powe IP (see Equa ion 4) by es ima ing he ac ion
o an igen bound o he an ibodies p oduced om he plasma
cells (Figu e 3A). As expec ed om he ela i e s eng h o bo h
e ec s, he immune powe is educed wi h inc easing eedback
s eng h, e lec ing a dec ease in he e iciency o he GC eac ion
(Figu e 3B).
Nex , we pe o med simula ions o GC (N=1) ini ialized
wi h a delay o 0, 72, and 120 h a e he ini ializa ion o GCs
F on ie s in Immunology | www. on ie sin.o g 3Sep embe 2019 | Volume 10 | A icle 2116
A ul aj e al. GC E iciency Impac ed by An ibody Feedback
FIGURE 1 | Ge minal cen e eac ion kine ics wi h a ying an ibody eedback s eng h. The scaling ac o Nis p opo ional o he s eng h o he an ibody eedback
on o he simula ed GC. (A) F ac ion o FDC an igen bound in immune complexes. (B) Concen a ion o ee an igen. (C) GC olume kine ics measu ed as numbe o
GC-BCs. (D) To al numbe o plasma cells gene a ed in he simula ed GC eac ion o e ime.
FIGURE 2 | E ec o a ying an ibody eedback on a ini y ma u a ion: mean
a ini y o all plasma cells de i ed om he simula ed GC eac ion.
(N=300) inducing an ibody eedback. The an ibody om
he ea ly GCs is used as an inpu o he simula ion o he
la e GC. Wi hou delay, he ac ion o immune complexes on
FDCs eaches 1 app oxima ely in 10 days (Figu e 4A). The
concen a ion o ee an igen a ailable o B cells d ops mo e
quickly wi h inc easing delay (Figu e 4B). GC olume kine ics
show a dec ease in he maximum olume and ea lie shu down
wi h inc easing delay (Figu e 4C). A GC ini ialized 120 h la e
is e mina ed e en be o e 10 days and he maximum olume is
educed by 75% (Figu e 4C). The numbe o plasma cells o med
is d ama ically educed wi h inc easing delay (Figu e 4D). The e
is also a dec ease in he mean a ini y wi h inc easing delay
(Figu e 4E), al hough he e is a small inc ease obse ed a ea lie
ime poin s p io o 10 days. The s ong an ibody eedback
pushes a ini y ma u a ion o a ew B cells, bu mos ly supp esses
selec ion. Co espondingly, he immune powe is educed wi h
delayed ini ializa ion, showing a dec ease in he e iciency o la e
ini ialized GCs (Figu e 4F).
In o de o compa e he impo ance o he s eng h o
an ibody eedback s. GC delay on a ini y ma u a ion and
ou pu p oduc ion, we sys ema ically a ied bo h pa ame e s (see
Figu e 5). Inc easing he alue o ei he pa ame e inc eases he
s eng h o he an ibody eedback and can eplace he e ec
o he espec i e o he . The mean a ini y o plasma cells on
day 21 dec eased om ∼0.7–0.35 in he ange o pa ame e
alues es ed (Figu e 5B). The inc ease in mean a ini y obse ed
on day 5 is e y low (Figu e 5A) compa ed o he dec ease
F on ie s in Immunology | www. on ie sin.o g 4Sep embe 2019 | Volume 10 | A icle 2116
A ul aj e al. GC E iciency Impac ed by An ibody Feedback
FIGURE 3 | To al an ibody p oduced by simula ed GC (A) and e iciency o he GC eac ion in e ms o immune powe (IP) (de ined in Equa ion 4) (B).
FIGURE 4 | Delayed ini ializa ion o GC eac ion (N=1) wi h an ibody eedback om ea ly GCs (N=300). (A) F ac ion o FDC an igen bound o soluble an ibodies.
(B) Concen a ion o ee an igen. (C) GC olume as numbe o GC B cells. (D) Numbe o all plasma cells p oduced in he delayed GC eac ion. (E) Mean a ini y o all
plasma cells gene a ed in he delayed GC eac ion. (F) Immune powe (IP) (de ined in Equa ion 4) o he delayed GC eac ion.
obse ed on day 21. Hence, a ini y ma u a ion is imp o ed by
an ibody eedback ea ly in he GC eac ion bu is no con inued
in he long e m. In con as , he immune powe exhibi s a
consis en dec ease wi h inc easing an ibody eedback s eng h
o GC delay a any ime o he GC eac ion (Figu es 5C,D).
Hence, he e iciency o he GC eac ion is domina ed by he
nega i e e ec o an ibody eedback on he numbe o plasma
cells a he han he in e media e posi i e e ec on he a ini y
o plasma cells.
DISCUSSION
Using an agen -based model app oach, we in es iga ed he ole
o soluble an ibodies o med as a esul o he GC eac ion in
F on ie s in Immunology | www. on ie sin.o g 5Sep embe 2019 | Volume 10 | A icle 2116

A ul aj e al. GC E iciency Impac ed by An ibody Feedback
FIGURE 5 | Ini ializa ion o a GC (N=1) wi h a ying delay wi h espec o ea ly GCs simula ed wi h di e en scaling ac o s N. Mean a ini y o plasma cells on day 5
(A) and day 21 (B) a e GC ini ializa ion. Immune powe (IP) (de ined in Equa ion 4) o he delayed GC on day 5 (C) and day 21 (D).
modula ing he ou pu o he GCs. Ou simula ions wi h a ying
s eng h o an ibody eedback show ha he inc eased an ibody
eedback esul s in a dec eased p oduc ion o plasma cells. GC
eac ions e mina e ea lie wi h inc eased an ibody eedback.
Immune complex o ma ion o FDC-bound an igen wi h soluble
an ibodies esul s in a dec eased a ailabili y o an igen o he B
cells unde going selec ion. The aised selec ion p essu e induces
a dec eased p oduc ion o plasma cells and ea lie e mina ion o
he GC.
Al hough he inc eased selec ion p essu e is expec ed o
imp o e o accele a e a ini y ma u a ion, we ind ha he
imp o emen in a ini y ma u a ion by an ibody eedback is
weak. The e is a small imp o emen in he a ini y a ea lie
ime poin s. Howe e , because o he educ ion o GC size
kine ics, he e is also a e mina ion o he a ini y ma u a ion
p ocess. Wi h highe an ibody eedback, he a ini y ma u a ion
p ocess is p ema u ely e mina ed and, hence, he mean a ini ies
o plasma cells do no imp o e u he in la e s ages o he
GC eac ion. The e o e, he mean plasma cell a ini y emains
low when compa ed o lowe an ibody eedback, which allows
o be e a ini y ma u a ion. These obse a ions sugges ha
kine ics o he GC eac ion is he key p ocess a ge ed by an ibody
eedback and he e ec on a ini y ma u a ion is domina ed by he
ea lie e mina ion o he GC eac ion.
Calcula ion o he e iciency o GC eac ions (immune
powe ) shows an o e all dec ease in he e iciency due o
he dec eased amoun o an ibodies (due o lowe numbe
o plasma cells) and a educ ion in he mean a ini y. This
sugges s ha he inc eased selec ion p essu e due o an ibody
eedback could ha e a de imen al e ec on he unc ioning o
he GC. Mo eo e , i sugges s he ole o soluble an ibodies
in egula ing he GC eac ion, p e en ing ch onic con inua ion
o GC eac ions, and p e en ing he excess p oduc ion o
plasma cells unde no mal condi ions. As he soluble an ibodies
a e p oduced as he esul o he GC eac ion, an ibody
eedback migh be conside ed as a na u al mechanism o sel -
egula ion o he p oduc ion o an ibodies and changes in
hei a ini ies.
Conside ing he in e ac ions be ween mul iple GCs, i is
likely ha simila scaling o an ibody eedback migh occu
wi h inc easing he numbe o GCs p oducing plasma cells.
Consequen ly, he ou pu o indi idual GCs migh be al e ed
depending on he numbe o GCs o med in esponse
o immuniza ion.
Fu he , ou simula ions show ha delayed ini ializa ion
o a GC eac ion would dec ease he p oduc ion o plasma
cells by impai ing he GC size kine ics and dec ease he ex en
o a ini y ma u a ion, esul ing in a dec eased e iciency
due o an ibody eedback om GCs ini ialized ea lie .
Asynch onous onse o GCs is e lec ed in expe imen al
indings as he numbe o GCs seem o inc ease o e a
pe iod o ime a e immuniza ion (36,37). This sugges s
ha synch oniza ion o GC onse can al e he e iciency
o GCs.
The combina ion o he numbe o GCs and he ex en o
synch oniza ion o hei onse migh play a ole in de e mining
he e iciency o o e all an ibody esponses due o GC-GC
in e ac ions. The numbe o GCs o med in he elde ly in
F on ie s in Immunology | www. on ie sin.o g 6Sep embe 2019 | Volume 10 | A icle 2116
A ul aj e al. GC E iciency Impac ed by An ibody Feedback
esponse o immuniza ion is educed (36,38). Such changes
migh esul in al e a ions in an ibody eedback in addi ion o
o he de ec s.
Fu u e s udies on unde s anding he numbe o GCs induced
and asynch onous onse migh allow p ecise p edic ion o
changes in an ibody esponses and ways o modula e hem.
In his s udy, we ocus mainly on he con ibu ion o plasma
cells o he an ibody eedback. Hence, he e ec s o low a ini y
an ibodies p oduced by ex a ollicula an ibody sec e ing cells
on GC onse and i s con ibu ion o he an ibody eedback
a e neglec ed. In he case o a seconda y immune esponse,
he an ibody eedback on GCs migh be inc eased due o a
highe ex a ollicula esponse gene a ed by memo y B cells. A
la ge numbe o ac o s in luence accina ion success including
cy okine p oduc ion, B and T epe oi e di e si y and p opo ion
o T egula o y cells (39,40). Se e al s a egies ha e been
sugges ed o imp o e accine e icacy including ex ended an igen
a ailabili y, a ge ing pa e n ecogni ion ecep o s, dend i ic
cells, and T cells (11,12,41–44). In addi ion o inc eased T h help
(11,12) he enhanced GC esponse obse ed upon p olonged
an igen deli e y could also be an ou come o o e coming
an ibody eedback by inc eased an igen deposi ion on FDCs.
Ou esul s show ha changes in an ibody eedback due o
GC-GC in e ac ions migh also play a ole in he op imiza ion
o accina ion.
AUTHOR CONTRIBUTIONS
PR, SB, and MM-H designed he s udy. MM-H de eloped he
model. TA pe o med he simula ions. TA, SB, and MM-H
analyzed he esul s and w o e he manusc ip .
FUNDING
TA was suppo ed by he Eu opean Union’s Ho izon
2020 esea ch and inno a ion p og amme unde he
Ma ie Skłodowska-Cu ie g an ag eemen no. 765158.
PR was suppo ed by he Human F on ie Science
P og am (RGP0033/2015).
ACKNOWLEDGMENTS
The au ho s would like o hank Alexey U a o skii o inspi ing
discussions. We hank Michelle Lin e man and Alyssa Sil a
Caye ano o insigh ul discussions on GCs in aged o ganisms.
SUPPLEMENTARY MATERIAL
The Supplemen a y Ma e ial o his a icle can be ound
online a : h ps://www. on ie sin.o g/a icles/10.3389/ immu.
2019.02116/ ull#supplemen a y-ma e ial
REFERENCES
1. Sa kande J, Hojyo S, Tokoyoda K. Vaccina ion o gain humo al immune
memo y. Clin T ansl Immunol. (2016) 5:e120. doi: 10.1038/c i.2016.81
2. Pulend an B, Ahmed R. Immunological mechanisms o accina ion. Na
Immunol. (2011) 12:509–17. doi: 10.1038/ni.2039
3. Sallus o F, Lanza ecchia A, A aki K, Ahmed R. F om accines o memo y and
back. Immuni y. (2010) 33:451–63. doi: 10.1016/j.immuni.2010.10.008
4. Kas u i SP, Skoun zou I, Alb ech RA, Kou sonanos D, Hua T, Nakaya HI,
e al. P og amming he magni ude and pe sis ence o an ibody esponses wi h
inna e immuni y. Na u e. (2011) 470:543–7. doi: 10.1038/na u e09737
5. Vic o a GD, Nussenzweig MC. Ge minal cen e s. Annu Re Immunol. (2012)
30:429–57. doi: 10.1146/annu e -immunol-020711-075032
6. MacLennan ICM. Ge minal cen e s. Annu Re Immunol. (1994) 12:117–39.
doi: 10.1146/annu e .iy.12.040194.001001
7. S ebegg M, Kuma SD, Sil a-Caye ano A, Fonseca VR, Lin e man MA, G aca
L. Regula ion o he ge minal cen e esponse. F on Immunol. (2018) 9:2469.
doi: 10.3389/ immu.2018.02469
8. DeF anco AL. The ge minal cen e an ibody esponse in heal h and disease.
F1000Resea ch. (2016) 5:F1000. doi: 10.12688/ 1000 esea ch.7717.1
9. Pulend an B, Kannou akis G, Nou i S, Smi h KGC, Nossal GJV. Soluble
an igen can cause enhanced apop osis o ge minal-cen e B cells. Na u e.
(1995) 375:331–4. doi: 10.1038/375331a0
10. Pulend an B, Smi h KG, Nossal GJ. Soluble an igen can impede a ini y
ma u a ion and he ge minal cen e eac ion bu enhance ex a ollicula
immunoglobulin p oduc ion. J Immunol. (1995) 155:1141–50.
11. Ci elli KM, C o y S. Ge minal cen e enhancemen by ex ended
an igen a ailabili y. Cu Opin Immunol. (2017) 47:64–9.
doi: 10.1016/j.coi.2017.06.008
12. Ci elli KM, Ca na han DG, Nogal B, Ma in JT, Rod iguez OL, Upadhyay AA,
e al. Slow deli e y immuniza ion enhances HIV neu alizing an ibody and
ge minal cen e esponses ia modula ion o immunodominance. Cell. (2019)
177:1153–71.e28. doi: 10.1016/j.cell.2019.04.012
13. Tam HH, Melo MB, Kang M, Pele JM, Ruda VM, Foley MH, e al. Sus ained
an igen a ailabili y du ing ge minal cen e ini ia ion enhances an ibody
esponses o accina ion. P oc Na l Acad Sci USA. (2016) 113:E6639–48.
doi: 10.1073/pnas.1606050113
14. Heyman B. Regula ion o an ibody esponses ia an ibodies,
complemen , and Fc ecep o s. Annu Re Immunol. (2000) 18:709–37.
doi: 10.1146/annu e .immunol.18.1.709
15. Wang X-Y, Wang B, Wen Y-M. F om he apeu ic an ibodies o immune
complex accines. NPJ Vaccines. (2019) 4:2. doi: 10.1038/s41541-018-0095-z
16. Ka lsson MC, We ne sson S, Diaz de S åhl T, Gus a sson S, Heyman
B. E icien IgG-media ed supp ession o p ima y an ibody esponses in
Fcgamma ecep o -de icien mice. P oc Na l Acad Sci USA. (1999) 96:2244–9.
doi: 10.1073/pnas.96.5.2244
17. Be gs öm JJE, Xu H, Heyman B. Epi ope-speci ic supp ession o IgG
esponses by passi ely adminis e ed speci ic igg: e idence o epi ope masking.
F on Immunol. (2017) 8:238. doi: 10.3389/ immu.2017.00238
18. Fo sell MNE, K as ad L, Sedimbi SK, Ande sson J, Ka lsson MCI. Regula ion
o subuni -speci ic ge minal cen e B cell esponses o he HIV-1 en elope
glycop o eins by an ibody-media ed eedback. F on Immunol. (2017) 8:738.
doi: 10.3389/ immu.2017.00738
19. Zhang Y, Meye -He mann M, Geo ge LA, Figge MT, Khan M, Goodall M,
e al. Ge minal cen e B cells go e n hei own a e ia an ibody eedback. J
Exp Med. (2013) 210:457–464. doi: 10.1084/jem.20120150
20. De Boe RJ, Pe elson AS. How Ge minal cen e s e ol e b oadly neu alizing
an ibodies: he b ead h o he ollicula helpe T cell esponse. J Vi ol. (2017)
91:e00983–17. doi: 10.1128/JVI.00983-17
21. Mackey MC, Maini PK. Wha has ma hema ics done o biology? Bull Ma h
Biol. (2015) 77:735–8. doi: 10.1007/s11538-015-0065-9
22. Ma io C, G an L, Ca men M-P, Ribei o RM. Ma hema ics in mode n
immunology. In e ace Focus. (2016) 6:20150093. doi: 10.1098/ s s.2015.0093
23. Wayne S eilein J, Read CB. A ma hema ical model in immuno egula ion
whe e he spleen has a pi o al ole. J Theo Biol. (1976) 61:363–76.
doi: 10.1016/0022-5193(76)90024-2
24. Wang P, Shih C, Qi H, Lan Y. A s ochas ic model o he ge minal
cen e in eg a ing local an igen compe i ion, indi idualis ic T–B
In e ac ions, and B cell ecep o signaling. J Immunol. (2016) 197:1169–82.
doi: 10.4049/jimmunol.1600411
F on ie s in Immunology | www. on ie sin.o g 7Sep embe 2019 | Volume 10 | A icle 2116
A ul aj e al. GC E iciency Impac ed by An ibody Feedback
25. Thomas MJ, Klein U, Lyge os J, Rod íguez Ma ínez M. A p obabilis ic
model o he ge minal cen e eac ion. F on Immunol. (2019) 10:689.
doi: 10.3389/ immu.2019.00689
26. Meye -He mann M, Binde SC, Mesin L, Vic o a GD. Compu e simula ion
o mul i-colo b ainbow s aining and clonal e olu ion o B cells in ge minal
cen e s. F on Immunol. (2018) 9:2020. doi: 10.3389/ immu.2018.02020
27. Robe PA, Ras ogi A, Binde SC, Meye -He mann M. How o
simula e a ge minal cen e . Me hods Mol Biol. (2017) 1623:303–34.
doi: 10.1007/978-1-4939-7095-7_22
28. Ke¸smi C, De Boe RJ. A ma hema ical model on ge minal cen e kine ics and
e mina ion. J Immunol. (1999) 163:2463–9.
29. Meye -He mann M, Moh E, Pelle ie N, Zhang Y, Vic o a GD, Toellne K-M.
A heo y o ge minal cen e B cell selec ion, di ision, and exi . Cell Rep. (2012)
2:162–174. doi: 10.1016/j.cel ep.2012.05.010
30. Binde SC, Meye -He mann M. Implica ions o in a i al imaging o mu ine
ge minal cen e s on he con ol o B cell selec ion and di ision. F on
Immunol. (2016) 7:593. doi: 10.3389/ immu.2016.00593
31. Pe elson AS, Os e GF. Theo e ical s udies o clonal selec ion: minimal
an ibody epe oi e size and eliabili y o sel -non-sel disc imina ion. J Theo
Biol. (1979) 81:645–70.
32. Vic o a GD, Schwicke TA, Fooksman DR, Kampho s AO, Meye -He mann
M, Dus in ML, e al. Ge minal cen e dynamics e ealed by mul ipho on
mic oscopy wi h a pho oac i a able luo escen epo e . Cell. (2010) 143:592–
605. doi: 10.1016/j.cell.2010.10.032
33. Tas JMJ, Mesin L, Pasqual G, Ta g S, Jacobsen JT, Mano YM, e al.
Visualizing an ibody a ini y ma u a ion in ge minal cen e s. Science (80-).
(2016) 351:1048–54. doi: 10.1126/science.aad3439
34. Ba is a FD, Neube ge MS. A ini y dependence o he B cell esponse o
an igen: a h eshold, a ceiling, and he impo ance o o - a e. Immuni y.
(1998) 8:751–9.
35. U a o skii A. Ma hema ical Modelling o Lymphocy e-Media ed Immune
Responses. (2017).
36. Szakal AK, Taylo JK, Smi h JP, Kosco MH, Bu on GF, Tew JJ. Kine ics o
ge minal cen e de elopmen in lymph nodes o young and aging immune
mice. Ana Rec. (1990) 227:475–85.
37. Rao SP, Vo a KA, Manse T. Di e en ial exp ession o he inhibi o y
IgG Fc ecep o FcγRIIB on ge minal cen e cells: implica ions o
selec ion o high-a ini y B cells. J Immunol. (2002) 169:1859–68.
doi: 10.4049/jimmunol.169.4.1859
38. Ayda Y, Balogh P, Tew JG, Szakal AK. Al e ed egula ion o FcγRII on
aged ollicula dend i ic cells co ela es wi h immuno ecep o y osine-
based inhibi ion mo i signaling in B cells and educed ge minal cen e
o ma ion. J Immunol. (2003) 171:5975–87. doi: 10.4049/jimmunol.171.
11.5975
39. Fink K. Can we imp o e accine e icacy by a ge ing T and B cell epe oi e
con e gence?. F on Immunol. (2019) 10:110. doi: 10.3389/ immu.2019.00110
40. Wiede mann U, Ga ne -Spi ze E, Wagne A. P ima y accine ailu e o
ou ine accines: why and wha o do? Hum Vaccin Immuno he . (2016)
12:239–43. doi: 10.1080/21645515.2015.1093263
41. Vasou A, Sul anoglu N, Goodbou n S, Randall RE, Kos ikis LG. Ta ge ing
Pa e n Recogni ion Recep o s (PRR) o accine adju an a ion: om
syn he ic PRR agonis s o he po en ial o de ec i e in e e ing pa icles o
i uses. Vi uses. (2017) 9:186. doi: 10.3390/ 9070186
42. Panagio i E, Klene man P, Lee LN, an de Bu g SH, A ens R. Fea u es
o e ec i e T cell-inducing accines agains ch onic i al in ec ions. F on
Immunol. (2018) 9:276. doi: 10.3389/ immu.2018.00276
43. Sho man K, Lahoud MH, Caminschi I. Imp o ing accines by
a ge ing an igens o dend i ic cells. Exp Mol Med. (2009) 41:61.
doi: 10.3858/emm.2009.41.2.008
44. Lin e man MA, Hill DL. Can ollicula helpe T cells be a ge ed
o imp o e accine e icacy? F1000Resea ch. (2016) 5:F1000.
doi: 10.12688/ 1000 esea ch.7388.1
Con lic o In e es S a emen : The au ho s decla e ha he esea ch was
conduc ed in he absence o any comme cial o inancial ela ionships ha could
be cons ued as a po en ial con lic o in e es .
Copy igh © 2019 A ul aj, Binde , Robe and Meye -He mann. This is an open-
access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion
License (CC BY). The use, dis ibu ion o ep oduc ion in o he o ums is pe mi ed,
p o ided he o iginal au ho (s) and he copy igh owne (s) a e c edi ed and ha he
o iginal publica ion in his jou nal is ci ed, in acco dance wi h accep ed academic
p ac ice. No use, dis ibu ion o ep oduc ion is pe mi ed which does no comply
wi h hese e ms.
F on ie s in Immunology | www. on ie sin.o g 8Sep embe 2019 | Volume 10 | A icle 2116