ARTICLE IN PRESS

Immunobiology 215 (2010) 163–171

0171-2985/$ - se

doi:10.1016/j.im

Abbreviations

type hypersensi�CorrespondE-mail addr

www.elsevier.de/imbio

Different mechanisms regulate CD4+

T cell independent induction of oral

and nasal tolerance of CD8+

T cells

Hendrik van den Berg, Mascha Greuter, Georg Kraal, Joke M.M. den Haan�

Department Molecular Cell Biology and Immunology, VU University Medical Center, Amsterdam, The Netherlands

Received 30 October 2008; received in revised form 23 January 2009; accepted 25 January 2009

Abstract

Mucosal administration of antigens is known to induce antigen specific regulatory CD4+ T cells, but less is knownabout the effects on CD8+ T cell function. Using a murine model for mucosal tolerance induction, we show that bothoral and nasal OVA (ovalbumin) application reduced OVA specific CD8+ T effector cell numbers and suppressed in

vivo cytotoxicity in response to subsequent immunisation. To investigate whether CD4+ T cells are essential for oral ornasal CD8+ T cell tolerance, we used MHC class II deficient mice. Normal CD8+ T cell tolerance was observed inMHC class II deficient mice, indicating that CD4+ T cells are not required for both oral and nasal CD8+ T celltolerance induction. To study the direct effects of mucosal antigen application on naive CD8+ T cells, we adoptivelytransferred OVA specific transgenic CD8+ T cells and analysed their fate after mucosal antigen application. Oral OVAapplication reduced the numbers of OVA specific CD8+ T cells, whereas nasal OVA application induced proliferation.However, the expanded population of OVA specific CD8+ T cells from nasally treated mice was unable to proliferateupon re-stimulation. In conclusion, our studies point to suppressive effects of the mucosal immune system directly onCD8+ T cells and indicate multiple mechanisms of tolerance induction. More insight in these mechanisms of toleranceinduction is necessary for the development of mucosal tolerance therapies for autoimmune diseases and allergy.r 2009 Elsevier GmbH. All rights reserved.

Keywords: Mucosal tolerance; Regulatory T cell; CD8+ T cell; CD4+ T cell; Suppression

Introduction

The mucosal immune system is constantly exposed toforeign material and must discriminate between harmfuland innocuous antigens. In healthy individuals, themucosal immune system induces antigen specific downmodulation of immune responses towards antigenspresent in the intestinal tract. Understanding this

e front matter r 2009 Elsevier GmbH. All rights reserved.

bio.2009.01.012

: DC, dendritic cell; Treg, T regulatory; DTH, delayed

tivity reaction; OVA, ovalbumin.

ing author. Tel.: +3120 4448058; fax: +3120 4448081.

ess: j.denhaan@vumc.nl (J.M.M. den Haan).

mechanism of mucosal tolerance induction is important,since defects in mucosal tolerance induction are thoughtto result in inflammatory bowel diseases and allergy tofood antigens.

A number of previous studies have investigated theeffects of oral administration of antigens on CD4+ Tcells. CD103+ dendritic cells (DCs) that are present inthe lamina propria of the intestines take up antigen andmigrate to the mesenteric lymph nodes attracted byCCL19 and CCL21 (Jaensson et al., 2008; Jang et al.,2006; Pabst et al., 2007; Worbs et al., 2006). In themesenteric lymph nodes these DCs present antigen tonaive CD4+ T cells. When high doses of antigen are

ARTICLE IN PRESSH. van den Berg et al. / Immunobiology 215 (2010) 163–171164

administered, the DC-T cell interaction results indeletion of antigen specific CD4+ T cells, whileadministration of low doses induces antigen specificregulatory CD4+ T cells (Tregs) (Weiner, 2001). Thesemucosal CD4+ Tregs have been shown to suppress Th1and Th2 immune responses, which is illustrated by thereduction in delayed type hypersensitivity (DTH) reac-tions, IFNg and IgE production (Chen et al., 1994;van Halteren et al., 1997). Furthermore, antigen specificregulatory CD4+ T cells can transfer their tolerogenicproperties to naive T cells (Chen et al., 1996; vanHalteren et al., 1997; Unger et al., 2003).

Fewer studies have examined the effects of themucosal antigen administration on CD8+ T cellresponses and these have lead to conflicting results.Oral administration of antigen has been shown tosuppress CD8+ T cell responses in some studies (Ke andKapp, 1996), but also stimulation of effector CD8+ Tcells has been demonstrated (Blanas et al., 1996;Hanninen et al., 2001). In addition, the mechanism ofCD8+ T cell modulation by the mucosal immune systemis not clear. One possible mechanism is the inductionof regulatory CD4+ T cells that subsequently suppressthe stimulation of CD8+ T cell responses (Desvigneset al., 1996, 2000; Dubois et al., 2003). Other studiespoint to the direct induction of suppressor CD8+ T cellsor deletion of CD8+ T cells (Lider et al., 1989; Ke andKapp, 1996; Chen et al., 1995; Limmer et al., 2005,2000). However, most of these studies have beenperformed before single cell analyses had becomeavailable, such as MHC-peptide tetramer stainings orintracellular cytokine stainings. The aim of this studywas to characterise the effects of oral and nasal antigenadministration on CD8+ T cells using single cell read-outs and to determine the role of CD4+ T cells in thisprocess. We report that both oral and nasal adminis-tration led to suppression of antigen specific CD8+ Tcell responses after immunisation. This suppression wasnot dependant on the presence of CD4+ T cells.Analysis of antigen specific CD8+ T cells revealed thatoral administration led to deletion of antigen specificCD8+ T cells, whereas nasal administration resulted inan expanded CD8+ T cell population that could notproliferate in vitro. Our studies point to suppressiveeffects of the mucosal immune system that act directlyon CD8+ T cells and indicate multiple mechanisms oftolerance induction.

Materials and methods

Mice

Specific pathogen free C57Bl/6 mice (8–10 weeks) andLy5.1 mice were purchased from Charles River,

Maastricht, The Netherlands. Specific pathogen freeABBN12 (MHC class II-KO) mice (8 weeks) werepurchased from Taconic, Germantown, NY, USA. OT-Imice have a transgenic Va2Vb5T cell receptor thatrecognise the OVA257–264 (ovalbumin) peptide in thecontext of H2-Kb and were bred at the animal facility ofthe VU Medical Center (Amsterdam, The Netherlands).All mice were kept under routine animal housingconditions and experiments were approved by theanimal experiment committee of the VU UniversityMedical Center, Amsterdam.

Antigen and antibodies

Either intact OVA grade V (Sigma Aldrich, Zwijn-drecht, The Netherlands) or OVA peptide (OVA257–264

for CD8+ T cell stimulation and OVA262–276 for CD4+

T cell stimulation) was used as antigen. For FACSanalysis, FITC conjugated anti-CD11a (M17/4), PEconjugated anti-Va2 (B20.1), PE and APC conjugatedanti-CD4 (GK1.5), PE conjugated anti-CD8 (53-6.7)and APC conjugated anti-IFNg (XMG 1.2) werepurchased from eBioscience. APC conjugated Anti-CD62L (MEL-14) was purchased from BD Bioscience.Anti-Ly5.2 (AL-1) and anti-CD44 (IM781) were affinitypurified from culture supernatant of hybridoma cellswith Protein G–Sepharose and labelled with biotin orFITC according to the manufacturer’s instructions.H-2Kb-OVA257–264 tetramers were a kind gift ofDr. Ton Schumacher (Netherlands Cancer Institute(NKI), Amsterdam, The Netherlands).

Induction of tolerance and DTH

For induction of oral tolerance 25mg of OVA in200 ml saline was administered intragastrically six daysbefore sensitisation. For induction of nasal tolerance,100 mg OVA in 10 ml saline was intranasally adminis-tered on three consecutive days directly before sensitisa-tion. Sensitisation was performed by subcutaneousinjection of 25 ml Incomplete Freund Adjuvant (Difco,Alphen aan den Rijn, The Netherlands) mixed with 25 mlsaline containing 100 mg OVA. Five days later, micewere challenged with 10 mg OVA in 10 ml saline in theauricle of each ear. Directly before challenge, the initialear thickness was determined with an engineer’s micro-metre (Mitutoyo, Tokyo, Japan). Twenty four hourslater, the increase in ear thickness was measured andspleens were isolated for ex vivo re-stimulation.

In vivo cytotoxicity assay

Single cell suspensions were prepared from spleens ofnaive C57Bl/6 and erythrocytes were depleted. Targetcells were incubated with 1 mg/ml OVA257–264 for 1 hour

ARTICLE IN PRESSH. van den Berg et al. / Immunobiology 215 (2010) 163–171 165

at 37 1C and labelled with 0.5 mM CFSE. Control cellswere incubated in medium for 1 hour at 37 1C andlabelled with 5 mM CFSE. Each mouse received 5� 106

OVA257–264 loaded target cells and 5� 106 control cellsin 200 ml saline by i.v. injection. After 16 hours, spleenswere isolated and single cell suspensions were analysedby FACS to determine OVA specific cell killing usingthe following formula, in which OVA represents thenumber of OVA257–264 coated target cells and CTRLrepresents the number of uncoated target cells recoveredfrom either experimental or non-treated (nt) mice:

% specific lysis ¼ 1�OVA exp

CTRL exp�

CTRL nt

OVA nt

� �� 100

Intracellular cytokine detection

For detection of antigen specific IFNg producing Tcells, spleen cells were re-stimulated ex vivo with theappropriate OVA peptide. For CD8+ T cells, spleencells were incubated directly for 5 hours with 0.1 mg/mlOVA257–264 and GolgiPlug (BD). For CD4+ T cells,spleen cells were re-stimulated for 40 hours with 100mg/mlOVA262–276, followed by 5 hours incubation withGolgiPlug. After the re-stimulation with peptide, cellswere stained for CD11a and CD4 or CD8, followed byfixation (2% Para formaldehyde in PBS; pH 7.2) andpermeabilisation (0.5% saponin; 0.5% BSA in PBS).Permeabilised cells were stained for intracellular IFNgand analysed on FACS.

Adoptive transfer of T cells into Ly5.1 mice

Lymph nodes and spleens were isolated from C57Bl/6-Ly5.2 mice and OT-I-Ly5.2 mice and single cellsuspensions were prepared. After depletion of erythro-cytes, WT CD4+ and OT-I CD8+ T cells were enrichedusing Dynal negative isolation kits following manufac-turers recommendations (Invitrogen). The remainingcells, 80–90% pure CD4+ or CD8+ T cells, were re-suspended in PBS. Each acceptor Ly5.1 mouse received4� 106 purified naive OVA specific CD8+ T cells fromOT-I (Ly5.2) mice and a control population of 4� 106

purified naive CD4+ T cells from C57Bl/6 (Ly5.2) mice.These mice received OVA for 5 weeks either once a weekorally or on three consecutive days each week intrana-sally. Control mice were left untreated. Ten days afterthe last treatment, spleens and lymph nodes werecollected and OT-I cell survival was analysed by FACSstaining for Ly5.2, CD8 and CD4. The number of OT-Icells was normalised using WT Ly5.2 CD4+ T cells andrepresented as percentage of OT-I cells in non treated(nt) animals using the following formula. OT-I repre-sents the number of CD8+ OT-I cells and CTRL

represents the number of CD4+ control cells recoveredfrom either experimental or nt mice:

% OT-I cells ¼OT-I exp

CTRL exp�

CTRL nt

OT-I nt� 100

For re-stimulation of OT-I cells, 2� 106 spleen cellswere incubated with 200 mg/ml OVA for 3 days at 37 1Cand analysed by FACS staining for Ly5.2, CD4, CD8and IFNg. The percentage of OT-I cells in theexperimental groups as compared to those in nt groupswas calculated as described above.

Results

Oral administration of antigen suppresses antigen

specific cytotoxic T cells

Oral administration of antigens induces regulatoryCD4+ T cells, which results in suppression of delayedhypersensitivity (DTH) responses. To investigate whethercytotoxic T cell responses are suppressed after oralantigen application, we analysed in vivo cytotoxicactivity in addition to the DTH response. To inducetolerance, mice received OVA orally six days before theywere subcutaneously immunised with OVA in IFA. Fivedays after immunisation, mice were challenged by aninjection of OVA in the auricles of both ears and theincrease in ear thickness was measured 24 hours later tomeasure the DTH response. Animals that had beentreated orally with OVA became tolerant for OVA, asear swelling responses were reduced compared tocontrol mice (Fig. 1A). To determine in vivo cytotoxicactivity, spleen cells coated with H-2Kb restrictedOVA257–264 peptide and non-coated control cells werelabelled with low or high doses of CFSE, respectively,and together adoptively transferred to tolerised andimmunised mice. After 16 hours, OVA specific cyto-toxicity was determined by FACS. OVA specificcytotoxicity was significantly suppressed in tolerisedanimals when compared to non-tolerised animals(Fig. 1B).

The suppression of cytotoxicity could be caused by areduction in numbers or by a suppressed function ofOVA specific CD8+ T cells. To quantify the numbersof OVA specific CD8+ T cells in tolerised mice versusimmunised mice, we isolated spleen cells 6 days afterimmunisation and stained with H-2Kb-OVA257–264

tetramers. Immunised mice showed a clear CD8+ Tcell response to OVA, in which 0.5% of total CD8+ Tcells was OVA specific. Mice that were previouslytolerised showed significantly reduced numbers ofOVA-specific CD8+ T cells compared to immunisedmice (Fig. 1C). In addition, spleen cells were re-stimulated ex vivo with MHC class I and II restricted

ARTICLE IN PRESS

OVAOVA

-OVA

--

i.g.s.c.

- OVA

100Δ

ear-

thic

knes

s (μ

m)

50

150

0- CD11a

0

% ly

sis

20

40

60

800.47 0.21

104

101

103

102

100

100 101 102 103 104

104

101

103

102

100

100 101 102 103 104

OVA-

H2-

Kb O

VA25

7-26

4 Tet

ram

er

OVA

0.4

0.2

0.6

0.0

CD

11a+

Tetra

mer

+

of C

D8+ T

cel

ls

OVAOVA

-OVA

--

i.g.s.c.

0.0

0.2

0.4

0.6

0.8

CD

11a+

IFN

γ+

of C

D8+ T

cel

ls

OVAOVA

-OVA

--

i.g.s.c.

CD

11a+

IFN

γ+

of C

D4+ T

cel

ls

0.2

0.1

0.3

0.0

*** *****

******

Fig. 1. Oral administration of OVA results in reduced antigen specific cytotoxicity and number of antigen specific CD8+ T cells.Mice received OVA or saline intragastrically and six days later the mice were sensitised s.c. with 100mg OVA in IFA. At day 11 micewere challenged with 10mg OVA in 10 ml saline in the auricle of both ears and 24 h later the DTH response was determined by

measuring the increase in ear-thickness (A). After measuring the DTH these mice received CFSE labelled OVA257–264-peptide coatedspleen cells together with non-coated control cells i.v. After 16 hours, the spleens of the recipient animals were analysed for thepresence of OVA257–264 peptide coated cells and cytotoxicity was calculated using the control cells as internal standard (B). Six days

after immunisation, antigen specific CD8+ T cells were stained with H-2Kb-OVA257–264 tetramers (C and D) or re-stimulated withMHC class I restricted OVA peptide for 5 h, or MHC class II restricted OVA peptide for 2 days, and OVA specific INFg producingCD8+ (E) and CD4+ (F) T cells were analysed. Black bars represent immunised control mice, grey bars represent OVA mice treated

with oral OVA followed by immunisation, open bars represent untreated mice. Experiments were repeated 14 times for (A) and (E),7 times for (F), and 2 times for (B) each with minimally 7 mice per group. Error bars indicate SEM of 8 treated or 3 non-treated miceper group (**po0.01, ***po0.001).

H. van den Berg et al. / Immunobiology 215 (2010) 163–171166

OVA peptides and the numbers of IFNg producingCD8+ and CD4+ T cells were measured. Oral admin-istration of OVA significantly reduced the numbers ofantigen specific IFNg producing CD8+ and CD4+ Tcells compared to non-tolerised animals (Fig. 1D and E).

These results indicate that oral administration ofantigen significantly reduces the number of antigen specificCD8+ and CD4+ T cells and DTH and cytotoxicityresponses elicited by subsequent immunisation.

CD8+

T cell suppression in the absence of CD4+

T

cells

Mucosal administration has been shown to induceantigen specific regulatory CD4+ T cells and regulatoryCD4+ T cells are known to suppress antigen specificCD8+ T cell responses in a number of models. One ofthe possible mechanisms of CD8+ T cell toleranceinduction would be indirect via the induction of antigenspecific CD4+ regulatory T cells. To investigate thispossibility, we analysed both oral and nasal toleranceinduction in MHC class II deficient mice that lackregulatory CD4+ T cells. As depicted in Fig. 2A and B,

MHC class II deficiency has no effect on the ability toinduce strong OVA-specific CD8+ T cell responses afterimmunisation. In fact, these responses were higher thanin wild-type C57Bl/6 mice, probably due to the absenceof thymus derived CD25+CD4+ regulatory T cells.Interestingly, oral administration of OVA before sensi-tisation led to an antigen specific suppression of theCD8+ T cell response in MHC class II deficient mice,comparable to the situation in wild-type C57Bl/6 mice(Fig. 2A). Also mice that were depleted of CD4+ T cellsusing a monoclonal antibody could efficiently betolerised by oral administration (data not shown).Similarly nasal administration, which has been shownto activate CD4+ Tregs (Unger et al., 2003), resulted inCD8+ T cell tolerance in MHC class II deficient mice(Fig. 2B).

These results indicate that CD4+ T cells are notessential for the induction of CD8+ T cell tolerance.However, the deficiency in CD4+ T cells could clearlybe demonstrated when next to CD8+ T cell responses,DTH responses were determined in MHC class IIdeficient mice after oral and nasal tolerance induction.In contrast to wild type animals, no specific DTHresponses could be measured in MHC class II deficient

ARTICLE IN PRESS

0.0

0.3

0.6

0.9

1.2C

D11

a+ IF

Nγ+

of C

D8+ T

cel

ls

C57Bl/6

MHC class II-KO

OVAOVA

-OVA

--

i.g.s.c.

OVAOVA

-OVA

--

OVAOVA

-OVA

--

i.g.s.c.

OVAOVA

-OVA

--

OVAOVA

-OVA

--

i.n.s.c.

OVAOVA

-OVA

--

OVAOVA

-OVA

--

i.n.s.c.

OVAOVA

-OVA

--

CD

11a+

IFN

γ+

of C

D8+ T

cel

ls

0.0

0.6

1.2

1.8

Δ ea

r-th

ickn

ess

(μm

)

Δ ea

r-th

ickn

ess

(μm

)

0.0

0.3

0.6

0.9

1.2

0.0

0.6

1.2

1.8

Oral tolerance Nasal tolerance

***

***

***

***

** ***

Fig. 2. Both oral and nasal OVA administration leads to CD8+ T cell suppression in a CD4+ T cell independent way. WT andMHC class II deficient mice were orally (A and C) or intranasally (B and D) tolerised and sensitised as described in Fig. 1. OVAspecific CD8+ T cells were analysed on day 12 as described in Fig. 1 (A and B). Mice were challenged for DTH measurement on day

11. After 24 h, increase in ear thickness was determined as measure for DTH response (C and D). Black bars represent WT mice,grey bars represent MHC class II deficient mice. Mice were treated as indicated. Error bars indicate SEM of 8 treated or 3 non-treated mice per group (***po0.001). Oral tolerance induction was also reproducibly observed in CD4+ T cell depleted mice (2

experiments each with 7 mice per group).

H. van den Berg et al. / Immunobiology 215 (2010) 163–171 167

mice, illustrating the important role of CD4+ T effectorcells in DTH responses (Fig. 2C and D).

Together these data show that both oral and nasalCD8+ T cell tolerance induction can occur in theabsence of CD4+ T cells, suggesting that CD8+ T celltolerance induction is mediated by direct interactionof tolerising mucosal DCs and CD8+ T cells. Theseinteractions could potentially lead to deletion of CD8+

T cells or to the generation of anergic or suppressorCD8+ T cells incapable of proliferation upon stimula-tion by regular stimulating antigen presenting DCspresent upon immunisation (Walker and Abbas, 2002).

Different effects of oral and nasal OVA

administration on antigen specific CD8+ T cells

To elucidate the direct effects of oral and nasalantigen administration on the fate of naive antigenspecific CD8+ T cells, we transferred Ly5.1 mice with4� 106 purified naive OVA specific CD8+ T cells fromOT-I (Ly5.2) mice. During 5 weeks mice received aweekly dose of OVA via the oral or nasal route. Controlmice were left untreated. Ten days after the lastadministration, spleens and lymph nodes were collected

and the numbers of OT-I cells and their function wereanalysed. Although both oral and nasal administrationresulted in tolerance and suppression of CD8+ T cellresponses after subsequent immunisation (Fig. 1), thedirect effects of oral and nasal antigen administration onnaive antigen specific CD8+ T cells were very different.Whereas oral antigen administration led to a significantreduction of antigen specific CD8+ T cells, expansion ofthese cells was observed after intranasal application(Fig. 3A). In both cases OT-I cells were activated in vivo

as was shown by down-regulation of CD62L (Fig. 3B)and up-regulation of CD44 (Fig. 3C). In addition, highpercentages of OT-I cells from both orally and nasallytolerised mice produced IFNg in response to OVA257–264

peptide, whereas OT-I cells from control mice showed anaive phenotype without IFNg production (Fig. 3D).These results suggest that the DC-CD8+ T cellinteractions leading to oral and nasal tolerance mustbe different, since they result in different effects on theCD8+ T cells.

Next we determined the capacity of OT-1 cells fromanimals that had been treated orally and nasallyto proliferate in vitro upon re-stimulation with OVA.As expected, OT-I cells from naive, non-treatedanimals proliferated upon re-stimulation. Similarly, the

ARTICLE IN PRESS

oralOVA

nasalOVA-

% O

T-I c

ells

0

100

200

300

oralOVA

nasalOVA-

0

125

250

375

500CD62L

mea

n flu

ores

cenc

e

oralOVA

nasalOVA-

% IF

Nγ+

OT-

I cel

ls

0

25

50

75

100Medium

OVA

oral OVA nasal OVA-0

100

200

300

400%

OT-

I cel

ls

***

0

50

100

150

200

oralOVA

nasalOVA-

CD44

mea

n flu

ores

cenc

e ***

*****

**

**** **

**

*

Fig. 3. Different effects of oral and nasal OVA application on OVA specific CD8+ T cells. C57Bl/6 mice were transferred with4� 106 CD8+ OT-I cells and WT CD4+ control cells and left untreated for 40 days, or treated with weekly oral or nasal OVA

applications for 5 weeks. On day 40, the numbers of OT-I cells in the spleen were determined by FACS. Using the WT CD4+ T cellsas an internal control, the percentage of OT-I cells was calculated and depicted relative to the number of OT-I cells present in non-treated animals (A) and the mean fluorescence intensity of CD62L (B) and CD44 (C) and percentage of IFNg producing cells from

the surviving OT-I cells was determined using FACS analysis (D). Black bars represent non-treated mice, grey bars represent orallytreated mice, and open bars represent nasally treated mice. Spleen cells were cultured in the absence (black bars) or presence (greybars) of OVA for 3 days and the number of OT-I cells was analysed using FACS (E). Mice were treated as indicated. Experiments

were performed with 5–7 mice per group. Error bars indicate SEM (*po0.05, **po0.01, ***po0.001).

H. van den Berg et al. / Immunobiology 215 (2010) 163–171168

remaining OT-I cells from the orally tolerised mice werestill able to proliferate. In contrast, the OT-I cells fromnasally treated animals did not proliferate upon ex vivo

re-stimulation (Fig. 3E).Together, these data indicate that different mucosal

routes of tolerance induction results in differentmechanisms of CD8+ T cell tolerance. Whereas oraltolerance leads to a direct deletion of antigen specificCD8+ T cells, nasal tolerance induction results inexpansion of CD8+ T cells that are lacking proliferativecapacity upon secondary stimulation.

Discussion

In this study we investigated the mechanism of CD8+

T cell suppression upon both oral and nasal antigenapplication. We report that in mice, applicationof antigen via the oral and nasal mucosa leads tosuppression of CD8+ T cell responses upon secondary

immunisation. This suppression was detected as adecrease in antigen specific CD8+ T cell numbers aswell as a decrease in cytotoxicity. Suppression of CD8+

T cells could still be established in MCH class IIdeficient mice by both oral and nasal antigen applica-tion, indicating that regulatory CD4+ T cells werenot directly mediating CD8+ T cell tolerance. Theregulatory CD4+ T cell independence is an unexpectedobservation, because we and others have shown thatmucosal antigen administration induces regulatory Tcells (Faria and Weiner, 2006; Unger et al., 2003; Hauet-Broere et al., 2003), which have been demonstrated tosuppress CD8+ T cells in different models (Piccirillo andShevach, 2001; Sutmuller et al., 2001; Chen et al., 2005;Boettler et al., 2005; Antony et al., 2005; Mempel et al.,2006). Instead, our study suggests that mucosal antigenadministration causes CD8+ T cell tolerance via directinteraction between the antigen presenting cell and theCD8+ T cell, as has also been indicated in a hapten-specific CD8+ T cell tolerance model (Goubier et al.,2008). However, this does not exclude that antigen

ARTICLE IN PRESSH. van den Berg et al. / Immunobiology 215 (2010) 163–171 169

specific regulatory CD4+ T cells are induced, which canprovide an additional pathway of CD8+ T cell toleranceby suppressing CD4+ T cell help for CD8+ T cellactivation. In the absence of CD4+ T cell help, CD8+ Tcells can be activated, but undergo an abortive responsethat results in CD8+ T cell apoptosis upon secondaryencounter with antigen (Janssen et al., 2003).

The functional outcome of both oral and nasalantigen application is suppression of CD8+ T cellactivation upon subsequent immunisation. However, weprovide clear evidence that different mechanisms ac-count for this CD8+ T cell suppression. Oral antigenapplication leads to deletion of naive antigen specificCD8+ T cells, thereby reducing the numbers of antigenspecific CD8+ T cells that can be activated uponimmunisation. In contrast, nasal antigen applicationleads to proliferation of effector CD8+ T cells thatcannot proliferate upon re-stimulation. These nasallytolerised CD8+ T cells behave similar to the ‘‘helpless’’CD8+ T cells that are primed in the absence of CD4+ Tcell help. These ‘‘helpless’’ CD8+ T cells upregulate thedeath receptor TRAIL which activates caspase-depen-dant apoptosis pathways upon secondary stimulation(Janssen et al., 2005, 2003). However, caspase inhibitorsdid not relieve the block in proliferation of the nasallytolerised CD8+ T cells (data not shown), which suggeststhat a different mechanism is responsible.

What causes oral administration to delete antigenspecific CD8+ T cells, while nasal administration leadsto expansion of the CD8+ T cell population? In bothtreatments antigens are taken up by DCs present in themucosal tissues, which subsequently migrate to thedraining lymph node. Indeed, antigen specific CD8+ Tcell proliferation can be observed in the draining cervicaland mesenteric lymph nodes early after nasal and oraladministration (Hanninen et al., 2001). However,cervical and mesenteric lymph node DCs may differ intheir capacity to induce deletion of CD8+ T cells. Wehave previously shown that peripheral and cervicallymph nodes exhibit profoundly different characteristicsthat influence tolerance induction (Wolvers et al., 1999).Also upregulation of homing receptors on CD8+ T cellsis differentially regulated by DCs present in differenttypes of lymph nodes (Agace, 2006). Similarly, cervicaland mesenteric lymph node DCs may differ in CD8+ Tcell tolerance induction as they are influenced by thestromal cell compartment and the peripheral tissues thatdrain to the lymph nodes.

In addition, one of the major differences between oraland nasal administration is the uptake of oral antigensin the blood leading to the liver via the portal vein.Oral administered antigens can be detected in the blood(Wakabayashi et al., 2006) and are taken up by liversinusoidal endothelial cells, that can cross-presentthem to CD8+ T cells and induce deletion (Knolleand Limmer, 2003; Limmer et al., 2005, 2000). More

recently, liver plasmacytoid DCs have also been shownto cross-present orally administered antigens and toinduce deletion of CD8+ T cells (Goubier et al., 2008).This indicates that release of food antigens in the bloodfollowed by uptake by liver resident cells form animportant suppressive mechanism for oral antigensand may account for the deletion of CD8+ T cell afteroral administration. In addition, this suggests thatoral antigen administration forms a more potent path-way for immune suppression than nasal antigen admin-istration.

Even though oral administration of antigen resultedin deletion of antigen specific CD8+ T cells, theremaining cells still exhibited an activated phenotypewith the capacity to produce IFNg, which is consistentwith mucosal CD8+ T cell priming observed inexperimental autoimmune diabetes studies (Blanaset al., 1996; Hanninen et al., 2001). In these studiestransgenic RIP-OVA mice were adoptively transferredwith OVA specific CD8+ T cells and tolerised with OVAvia the oral and nasal routes. Both treatments inducedCTL activation and diabetes in these mice. Interestingly,nasal administration was significantly more potent ininducing diabetes than oral administration (Hanninen etal., 2001), which is in agreement with our observationthat CD8+ T cells expand after nasal administration.Since these studies have been performed with highnumbers of adoptively transferred antigen specificCD8+ T cells, that have been shown to behavedifferently compared to endogenously present lowCD8+ T cell frequencies (Mintern et al., 2002; Marzoet al., 2005), the possible consequences of mucosaladministration of auto antigens in humans is unclear.However, our study in combination with these studiesdoes warrant precaution in the use of mucosal admin-istration of antigens for the prevention or treatment ofautoimmune diseases. In this light, oral administrationwould be expected to result in more beneficial resultsthan nasal administration. In addition, mucosal toler-ance induction should only be attempted in those casesin which low precursor frequencies are present.

In conclusion, we show that both oral as well as nasaladministration results in suppression of CD8+ T cellfrequencies and functional responses upon subsequentimmunisation. This suppression is not dependent onregulatory CD4+ T cells and is induced via differentmechanisms. Insight in the mechanisms of mucosaltolerance induction is essential to develop strategies toprevent and treat autoimmune diseases and allergies.

Acknowledgements

This work was supported by Danone Researchand SenterNovem. J.d.H. was supported by GrantNWO917.46.311 from the Dutch Scientific Research

ARTICLE IN PRESSH. van den Berg et al. / Immunobiology 215 (2010) 163–171170

Programme. We thank Dr. Ton Schumacher(Netherlands Cancer Institute (NKI), Amsterdam, TheNetherlands) for providing us with tetramers and thestaff of our animal facility for the care of the animalsused in this study.

References

Agace, W.W., 2006. Tissue-tropic effector T cells: genera-tion and targeting opportunities. Nat. Rev. Immunol. 6,682–692.

Antony, P.A., Piccirillo, C.A., Akpinarli, A., Finkelstein, S.E.,Speiss, P.J., Surman, D.R., Palmer, D.C., Chan, C.C.,Klebanoff, C.A., Overwijk, W.W., Rosenberg, S.A.,

Restifo, N.P., 2005. CD8+ T cell immunity against atumor/self-antigen is augmented by CD4+ T helper cellsand hindered by naturally occurring T regulatory cells.

J. Immunol. 174, 2591–2601.Blanas, E., Carbone, F.R., Allison, J., Miller, J.F., Heath,

W.R., 1996. Induction of autoimmune diabetes by oraladministration of autoantigen. Science 274, 1707–1709.

Boettler, T., Spangenberg, H.C., Neumann-Haefelin, C.,Panther, E., Urbani, S., Ferrari, C., Blum, H.E., von,W.F., Thimme, R., 2005. T cells with a CD4+CD25+

regulatory phenotype suppress in vitro proliferation ofvirus-specific CD8+ T cells during chronic hepatitis C virusinfection. J. Virol. 79, 7860–7867.

Chen, M.L., Pittet, M.J., Gorelik, L., Flavell, R.A., Weissle-der, R., von, B.H., Khazaie, K., 2005. Regulatory T cellssuppress tumor-specific CD8T cell cytotoxicity through

TGF-beta signals in vivo. Proc. Natl. Acad. Sci. USA 102,419–424.

Chen, Y., Kuchroo, V.K., Inobe, J., Hafler, D.A., Weiner,H.L., 1994. Regulatory T cell clones induced by oral

tolerance: suppression of autoimmune encephalomyelitis.Science 265, 1237–1240.

Chen, Y., Inobe, J., Weiner, H.L., 1995. Induction of oral

tolerance to myelin basic protein in CD8-depleted mice:both CD4+ and CD8+ cells mediate active suppression.J. Immunol. 155, 910–916.

Chen, Y., Inobe, J., Kuchroo, V.K., Baron, J.L., Janeway Jr.,C.A., Weiner, H.L., 1996. Oral tolerance in myelin basicprotein T-cell receptor transgenic mice: suppression ofautoimmune encephalomyelitis and dose-dependent induc-

tion of regulatory cells. Proc. Natl. Acad. Sci. USA 93,388–391.

Desvignes, C., Bour, H., Nicolas, J.F., Kaiserlian, D., 1996.

Lack of oral tolerance but oral priming for contactsensitivity to dinitrofluorobenzene in major histocompat-ibility complex class II-deficient mice and in CD4+ T cell-

depleted mice. Eur. J. Immunol. 26, 1756–1761.Desvignes, C., Etchart, N., Kehren, J., Akiba, I., Nicolas, J.F.,

Kaiserlian, D., 2000. Oral administration of hapten inhibits

in vivo induction of specific cytotoxic CD8+ T cellsmediating tissue inflammation: a role for regulatory CD4+T cells. J. Immunol. 164, 2515–2522.

Dubois, B., Chapat, L., Goubier, A., Papiernik, M., Nicolas,

J.F., Kaiserlian, D., 2003. Innate CD4+CD25+ regula-tory T cells are required for oral tolerance and inhibition of

CD8+ T cells mediating skin inflammation. Blood 102,

3295–3301.

Faria, A.M., Weiner, H.L., 2006. Oral tolerance and TGF-

beta-producing cells. Inflamm. Allergy Drug Targets 5,

179–190.

Goubier, A., Dubois, B., Gheit, H., Joubert, G., Villard-Truc,

F., sselin-Paturel, C., Trinchieri, G., Kaiserlian, D., 2008.

Plasmacytoid dendritic cells mediate oral tolerance. Im-

munity 29, 464–475.

Hanninen, A., Braakhuis, A., Heath, W.R., Harrison, L.C.,

2001. Mucosal antigen primes diabetogenic cytotoxic T-

lymphocytes regardless of dose or delivery route. Diabetes

50, 771–775.

Hauet-Broere, F., Unger, W.W., Garssen, J., Hoijer, M.A.,

Kraal, G., Samsom, J.N., 2003. Functional CD25� and

CD25+ mucosal regulatory T cells are induced in gut-

draining lymphoid tissue within 48 h after oral antigen

application. Eur. J. Immunol. 33, 2801–2810.

Jaensson, E., Uronen-Hansson, H., Pabst, O., Eksteen, B.,

Tian, J., Coombes, J.L., Berg, P.L., Davidsson, T., Powrie,

F., Johansson-Lindbom, B., Agace, W.W., 2008. Small

intestinal CD103+ dendritic cells display unique functional

properties that are conserved between mice and humans.

J. Exp. Med. 205, 2139–2149.

Jang, M.H., Sougawa, N., Tanaka, T., Hirata, T., Hiroi, T.,

Tohya, K., Guo, Z., Umemoto, E., Ebisuno, Y., Yang,

B.G., Seoh, J.Y., Lipp, M., Kiyono, H., Miyasaka, M.,

2006. CCR7 is critically important for migration of

dendritic cells in intestinal lamina propria to mesenteric

lymph nodes. J. Immunol. 176, 803–810.

Janssen, E.M., Lemmens, E.E., Wolfe, T., Christen, U., von

Herrath, M.G., Schoenberger, S.P., 2003. CD4+ T cells are

required for secondary expansion and memory in CD8+ T

lymphocytes. Nature 421, 852–856.

Janssen, E.M., Droin, N.M., Lemmens, E.E., Pinkoski, M.J.,

Bensinger, S.J., Ehst, B.D., Griffith, T.S., Green, D.R.,

Schoenberger, S.P., 2005. CD4+ T-cell help controls

CD8+ T-cell memory via TRAIL-mediated activation-

induced cell death. Nature 434, 88–93.

Ke, Y., Kapp, J.A., 1996. Oral antigen inhibits priming

of CD8+ CTL, CD4+ T cells, and antibody responses

while activating CD8+ suppressor T cells. J. Immunol.

156, 916–921.

Knolle, P.A., Limmer, A., 2003. Control of immune responses

by scavenger liver endothelial cells. Swiss. Med. Wkly. 133,

501–506.

Lider, O., Santos, L.M., Lee, C.S., Higgins, P.J., Weiner, H.L.,

1989. Suppression of experimental autoimmune encepha-

lomyelitis by oral administration of myelin basic protein: II.

Suppression of disease and in vitro immune responses is

mediated by antigen-specific CD8+ T lymphocytes.

J. Immunol. 142, 748–752.

Limmer, A., Ohl, J., Kurts, C., Ljunggren, H.G.,

Reiss, Y., Groettrup, M., Momburg, F., Arnold, B.,

Knolle, P.A., 2000. Efficient presentation of exogenous

antigen by liver endothelial cells to CD8+ T cells

results in antigen-specific T-cell tolerance. Nat. Med. 6,

1348–1354.

Limmer, A., Ohl, J., Wingender, G., Berg, M., Jungerkes, F.,

Schumak, B., Djandji, D., Scholz, K., Klevenz, A.,

ARTICLE IN PRESSH. van den Berg et al. / Immunobiology 215 (2010) 163–171 171

Hegenbarth, S., Momburg, F., Hammerling, G.J., Arnold,B., Knolle, P.A., 2005. Cross-presentation of oral antigensby liver sinusoidal endothelial cells leads to CD8T cell

tolerance. Eur. J. Immunol. 35, 2970–2981.Marzo, A.L., Klonowski, K.D., Le, B.A., Borrow, P., Tough,

D.F., Lefrancois, L., 2005. Initial T cell frequency dictates

memory CD8+ T cell lineage commitment. Nat. Immunol.6, 793–799.

Mempel, T.R., Pittet, M.J., Khazaie, K., Weninger, W.,

Weissleder, R., von, B.H., von Andrian, U.H., 2006.Regulatory T cells reversibly suppress cytotoxic T cellfunction independent of effector differentiation. Immunity

25, 129–141.Mintern, J.D., Davey, G.M., Belz, G.T., Carbone, F.R.,

Heath, W.R., 2002. Cutting edge: precursor frequencyaffects the helper dependence of cytotoxic T cells.

J. Immunol. 168, 977–980.Pabst, O., Bernhardt, G., Forster, R., 2007. The impact of cell-

bound antigen transport on mucosal tolerance induction.

J. Leukoc. Biol. 82, 795–800.Piccirillo, C.A., Shevach, E.M., 2001. Cutting edge: control of

CD8+ T cell activation by CD4+CD25+ immunoregu-

latory cells. J. Immunol. 167, 1137–1140.Sutmuller, R.P., van Duivenvoorde, L.M., van, E.A., Schu-

macher, T.N., Wildenberg, M.E., Allison, J.P., Toes, R.E.,

Offringa, R., Melief, C.J., 2001. Synergism of cytotoxic Tlymphocyte-associated antigen 4 blockade and depletion ofCD25(+) regulatory T cells in antitumor therapy revealsalternative pathways for suppression of autoreactive

cytotoxic T lymphocyte responses. J. Exp. Med. 194,823–832.

Unger, W.W.J., Jansen, W., Wolvers, D.A.W., van Halteren,A.G.S., Kraal, G., Samsom, J.N., 2003. Nasal toleranceinduces antigen-specific CD4(+)CD25(–) regulatory T cells

that can transfer their regulatory capacity to naive CD4(+)T cells. International Immunology 15, 731–739.

van Halteren, A.G., van der Cammen, M.J., Cooper, D.,

Savelkoul, H.F., Kraal, G., Holt, P.G., 1997. Regulationof antigen-specific IgE, IgG1, and mast cell responsesto ingested allergen by mucosal tolerance induction.

J. Immunol. 159, 3009–3015.Wakabayashi, A., Kumagai, Y., Watari, E., Shimizu, M.,

Utsuyama, M., Hirokawa, K., Takahashi, H., 2006.

Importance of gastrointestinal ingestion and macromole-cular antigens in the vein for oral tolerance induction.Immunology 119, 167–177.

Walker, L.S., Abbas, A.K., 2002. The enemy within: keeping

self-reactive T cells at bay in the periphery. Nat. Rev.Immunol. 2, 11–19.

Weiner, H.L., 2001. Oral tolerance: immune mechanisms and

the generation of Th3-type TGF-beta-secreting regulatorycells. Microbes Infect. 3, 947–954.

Wolvers, D.A., Coenen-de Roo, C.J., Mebius, R.E., van der

Cammen, M.J., Tirion, F., Miltenburg, A.M., Kraal, G.,1999. Intranasally induced immunological tolerance isdetermined by characteristics of the draining lymph nodes:

studies with OVA and human cartilage gp-39. J. Immunol.162, 1994–1998.

Worbs, T., Bode, U., Yan, S., Hoffmann, M.W., Hintzen, G.,Bernhardt, G., Forster, R., Pabst, O., 2006. Oral tolerance

originates in the intestinal immune system and relies onantigen carriage by dendritic cells. J. Exp. Med. 203, 519–527.