Langerhans cells are precommitted to immune tolerance induction

Langerhans cells are precommitted to immunetolerance inductionElena Shklovskayaa, Brendan J. O’Sullivanb, Lai Guan Ngc,d,1, Ben Roedigera,c, Ranjeny Thomasb,Wolfgang Weningerc,d, and Barbara Fazekas de St Grotha,d,2

aT Cell Biology Research Program and cImmune Imaging Program, Centenary Institute of Cancer Medicine and Cell Biology, Newtown NSW 2042, Australia;bUniversity of Queensland Diamantina Institute, Princess Alexandra Hospital, Brisbane QLD 4102, Australia; and dDiscipline of Dermatology, University ofSydney, Sydney NSW 2006, Australia

Edited* by Ralph M. Steinman, The Rockefeller University, New York, NY, and approved September 23, 2011 (received for review June 22, 2011)

Antigen-dependent interactions between T lymphocytes and den-dritic cells (DCs) can produce two distinct outcomes: tolerance andimmunity. It is generally considered that all DC subsets are capableof supporting both tolerogenic and immunogenic responses, de-pending on their exposure to activating signals. Here, we testedwhether epidermal Langerhans cells (LCs) can support immuno-genic responses in vivo in the absence of antigen presentation byother DC subsets. CD4 T cells responding to antigen presentationby activated LCs initially proliferated but then failed to differen-tiate into effector/memory cells or to survive long term. The to-lerogenic function of LCs was maintained after exposure to potentadjuvants and occurred despite up-regulation of the costimulatorymolecules CD80, CD86, and IL-12, but was consistent with theirfailure to translocate the NF-!B family member RelB from the cyto-plasm to the nucleus. Commitment of LCs to tolerogenic functionmay explain why commensal microorganisms expressing Toll-likereceptor (TLR) ligands but con!ned to the skin epithelium are tol-erated, whereas invading pathogens that breach the epithelialbasement membrane and activate dermal DCs stimulate a strongimmune response.

Dendritic cells (DCs) initiate adaptive immune responses bypriming antigen-speci!c T cells in secondary lymphoid

organs. After sampling antigens in peripheral tissues, DCs mi-grate to lymph nodes (LN), where they present antigenic pep-tides bound to major histocompatibility (MHC) molecules (1).Epidermal Langerhans cells (LCs) have long been regarded asprototypic DCs, highly active in antigen uptake and rapidly ac-quiring potent costimulatory capacity after in vitro culture (2).Recently, the immunogenicity of LCs has been questioned on thebasis of !ndings in several in vivo experimental models. Duringherpes viral infection of the skin, migrated LCs isolated fromdraining LN (dLN) were unable to induce proliferation of virus-speci!c CD8 T cells in vitro (3). In LC ablation models, positive(4, 5), negative (6–8), and redundant (9) contributions of LCs tocontact hypersensitivity responses were reported. The currentlack of consensus regarding LC function may relate, at least inpart, to the dif!culties in determining the contribution of a rel-atively small number of LCs to responses driven primarily bynon-LC DC subsets in cutaneous LN (cLN).Here we directly tested the in vivo function of LCs, using

a previously described bone marrow (BM) chimeric mousemodel in which only LCs can present speci!c antigen to CD4 Tcells (10). In this model, all DC subsets express MHC class II IAmolecules but only LCs express MHC class II IE, which is abso-lutely required to present moth cytochrome C peptide (pMCC) to5C.C7 T-cell receptor (TCR) transgenic T cells (11, 12). Theresponse of adoptively transferred 5C.C7 CD4 T cells can thusbe used as a readout for LC function. We compared 5C.C7 T-cellresponses to LCs with those in chimeras expressing IE on non-epidermal DCs or all DC subsets, immunizing with peptide orprotein antigens delivered via multiple routes and with diverseadjuvants. Our results show that LCs displayed tolerogenicfunction under all conditions examined and maintained a tol-erogenic NF-!B signature by failing to translocate RelB to thenucleus (13) even when highly activated.

ResultsRestriction of MHCII-IE Expression to LCs. BM chimeras in which IEexpression is con!ned to LCs have been extensively character-ized previously (10). The chimeras were engineered using twolines of IE"d-transgenic mice on the C57BL/6 (MHCII-IA+IE!)background: 107-1 (here termed IE+), expressing IE with WTdistribution, and 36-2 (here termed IE!), expressing IE only onthymic epithelium and thereby mediating IE-dependent positiveselection and Treg development, as well as tolerance to IE (14).Unlike other DC subsets, LCs are radioresistant (15), such thatin IE!!IE+ chimeras (here termed LC chimeras), only skin LCsand migratory LCs (m-LCs) in cLN expressed IE, whereas theremaining DCs, B cells, and radioresistant stromal cells were IE-negative (Fig. 1 A and B and Fig. S1 A and B) (10). We con-!rmed that migratory dermal DCs (m-DDCs), conventional DCs(cDCs), B cells, and stromal cells from LC chimeras could notpresent the MCC87–103 epitope to IE-restricted 5C.C7 T cellsusing in vitro stimulation with hen egg lysozyme-moth cyto-chrome C protein (HELMCC; a protein antigen containing theMCC87–103 epitope) (10) (Fig. S1C).As controls for the IE!!IE+ LC chimeras, we generated

IE+!IE! chimeras (Fig. 1C) (10). Because all B cells in LCchimeras were IE!, the IE+ BM for control chimeras wasobtained from RAG!/! donors and was mixed with autologousIE! BM to generate an equivalent IE! B-cell compartment. Theproportion of IE+ RAG!/! BM was adjusted to 25% so thatfrequency of skin-derived IE+ migratory DCs (m-DCs) (Fig. 1B)was matched in cLN of LC and 25% control chimeras, to controlfor potential differences in cognate MHCII exposure and pep-tide presentation. Equivalent IE-restricted peptide presentationwas con!rmed by measuring recruitment of 5C.C7 T cells intodivision (Fig. S2 A and B).5C.C7 T cells survived long term in both LC and control

chimeras but rapidly disappeared in IE! mice (Fig. S2C), con-!rming that expression of IE by m-LCs in cLN was suf!cient forsurvival of naïve IE-restricted T cells. Intravenous injection ofpMCC induced equivalent rates of deletion in both chimeras(Fig. S2D), excluding long-term radiation effects as a possiblecause of differential responses in the two chimeras (16).

CD4 T Cells Activated by m-LCs Fail to Differentiate into Effector/Memory Cells and Do Not Survive Long Term. We tested the abilityof LCs to sustain an immunogenic CD4 T-cell response in vivoby transferring carboxy"uorescein diacetate succinimidyl ester

Author contributions: E.S., B.J.O., L.G.N., B.R., R.T., W.W., and B.F.d.S.G. designed re-search; E.S., B.J.O., L.G.N., and B.R. performed research; E.S., B.J.O., L.G.N., and B.R.analyzed data; and E.S. and B.F.d.S.G. wrote the paper.

The authors declare no con!ict of interest.

*This Direct Submission article had a prearranged editor.1Present address: Singapore Immunology Network (SIgN), Agency for Science, Technologyand Research (A*STAR), Biopolis 138648, Singapore.

2To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1110076108/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1110076108 PNAS | November 1, 2011 | vol. 108 | no. 44 | 18049–18054

IMMUN

OLO

GY

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1110076108/-/DCSupplemental/pnas.201110076SI.pdf?targetid=nameddest=SF1





mailto:[email protected]

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1110076108/-/DCSupplemental

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1110076108/-/DCSupplemental

www.pnas.org/cgi/doi/10.1073/pnas.1110076108

(CFSE)-labeled 5C.C7 cells into LC or control chimeras andimmunizing s.c. with pMCC emulsi!ed in complete Freund’sadjuvant (CFA) (Fig. 2). Despite similar recruitment of T cellsinto division (Fig. S2B), an 8.6-fold higher peak in T-cell num-bers was observed in control compared with LC chimeras, andcells survived long term only in control chimeras (Fig. 2 A andB). By day 10, most donor T cells in control but not LC chimerashad acquired a CD44highCD62Llow effector memory (Tem)phenotype (17) (Fig. 2C). To test for effector function, cytokineproduction by 5C.C7 cells was measured following in vitrorestimulation with peptide plus IE+ splenic DCs. Abundantproduction of interleukin (IL)-17 and IFN# was seen in controlbut not LC chimeras (Fig. 2D). The difference in peak effectornumbers was 240-fold for IFN# and 25-fold for IL-17 (Fig. 2E).LC chimeras retained some IL-2-producing CD4 T cells (5.5-folddecreased), accounting for their initial proliferative response. IL-4, IL-5, and IL-10 were never detected, nor was induction offoxp3 expression. Similar results were obtained using HELMCCprotein in CFA as the immunogen, except that cytokine pro-duction during the effector phase was biased toward IL-17 ratherthan IFN# (Fig. S3 A–D).The difference between T-cell survival in LC and control

chimeras could not be explained by the disappearance of IE+

DCs in LC chimeras, because the number of IE+ m-LCs and m-DDCs in dLN of LC and control chimeras, respectively, weresimilar over the course of the response (Fig. S4), consistent withlocal maintenance of LC homeostasis (15).To test for antigen-speci!c memory 60 d (Fig. 2F) and 90 d

(Fig. S5A) post immunization, chimeras were challenged at adifferent site with peptide emulsi!ed in incomplete Freund’sadjuvant (IFA). Sixteen hours after challenge, the frequency of5C.C7 cells in dLN of control chimeras increased by 15-fold (Fig.2F, Left and Fig. S5A, Left). The rapid increase in 5C.C7 cellnumbers was largely due to redistribution to dLN (Fig. S5D).Responding cells underwent blast transformation (Fig. S5C) anddown-regulated CD62L (Fig. 2F, Center and Fig. S5A, Center).Donor 5C.C7 cells produced IFN# after in vitro restimulation(Fig. 2F, Right and Fig. S5A, Right). These responses were notseen in LC chimeras. Interestingly, 5C.C7 cells in dLN of bothchimeras produced IL-2, indicating that the surviving cells in LCchimeras were still capable of responding to TCR stimulation.Similar results were obtained after in vivo challenge of miceprimed with HELMCC protein rather than pMCC (Fig. S3E).

It remained possible that CD4 T cells in LC chimeras failed tomount a memory response to rechallenge because of a speci!cdefect in m-LC antigen presentation. We therefore challengedprimed chimeras with intradermal injection of antigen-pulsedIE+ splenic DCs. Three days after DC injection, the frequency of5C.C7 cells in the dLN of control chimeras had increased !vefold(Fig. 2G) and the cells had become CD62Llow (Fig. S5B, Left).These changes did not occur in LC chimeras. Cytokine pro-duction (mainly IL-2) was higher in control compared with LCchimeras (Fig. S5B, Right). Furthermore, donor T cells werefound at the site of skin challenge in control but not LC chimeras(Fig. 2G), excluding the possibility that sequestration in the skincould account for the disappearance of antigen-speci!c T cells inLC chimeras. Taken together, these experiments indicated thatLCs exposed to s.c. antigen recruited CD4 T cells into an ab-ortive proliferative response that resulted in tolerance ratherthan generation of effector/memory function.

Effect of Activation Status on LC Function. Migrating LCs retainedtheir previously documented CD80/86low phenotype (10) in re-sponse to s.c. immunization, whereas CD80 and CD86 expres-sion on migrating DDCs increased within 4 h and continued toincrease until day 4 postimmunization (Fig. S6 A and B). How-ever, LCs expressed more CD40 than DDCs (10) and further up-regulated their CD40 expression from day 2 onward (Fig. S6 Aand B). We therefore tested whether ligation of CD40 couldconvert LCs to an immunogenic phenotype, as had been de-scribed previously for other tolerogenic DC subsets (18). LCchimeras were treated with agonistic anti-CD40 antibodies ondays 0 and 2 after s.c. immunization. However, even the com-bined pMCC/CFA/anti-CD40 treatment did not support thegeneration of CD4 T-cell memory in LC chimeras (Fig. S6C).Considering that the failure of m-LCs to up-regulate CD80/86

expression after s.c. injection may have indicated inadequateexposure to adjuvant, we switched to an epicutaneous immuni-zation approach in which LCs were directly exposed to proteinantigen/adjuvant via topical application in aqueous cream (19).We did not use tape stripping, which may disturb the integrity ofthe epidermis (20). Even without the addition of adjuvants, ap-plication of cream under an occlusive bandage caused m-LCs toup-regulate both CD80 and CD86, producing a bimodal CD80/86 pro!le as m-LCs !rst reached the dLN 48 h after immuni-zation (Fig. 3 A and B). The tempo of CD69 up-regulation byantigen-speci!c T cells showed a 1–2 d delay after the arrival ofm-LCs from the immunization site (Fig. 3C), suggesting that theT-cell response was driven by migrating antigen-bearing LCsrather than free antigen presented by m-LCs already present inthe LN at the time of immunization. Addition of adjuvants to theepicutaneous cream caused further activation of m-LCs, with2.2- to 3.1-fold increases in CD80 and CD86 expression inresponse to CFA-derived particulate material (heat-killed My-cobacterium tuberculosis H37Ra), TLR1/2 ligand Pam3Cys-Ser-(Lys)4 (Pam3CSK), or the TLR3 ligand polyinosinic acid:poly-cytidylic acid (poly I:C) (Fig. 3D). Epicutaneous immunizationwith cream containing CFA particulates also induced over 20%of m-LCs in dLN of LC chimeras to express IL-12, generatingthree- to fourfold more IL-12-producing IE+ m-DCs than thesame treatment in control chimeras (Fig. 3E). However, despitetheir activated phenotype and IL-12 production, LCs respondingto epicutaneous immunization with a combination of HELMCCand CFA particulates still failed to support the generation ofCD4 T-cell memory, as indicated by the lack of response toin vivo challenge with peptide/IFA (Fig. 3F).

LC-Driven Responses in WT Mice. The experiments described aboveindicated that immunization of LC chimeras rendered themtolerant to speci!c antigen. To test whether LCs also inducedtolerance in unmanipulated animals, we compared responsesto epicutaneous and s.c. immunization in WT mice, reasoningthat if epicutaneous antigen were presented mainly by LCs, then

Fig. 1. Characterization of chimeric mouse models with expression of IErestricted to either LCs or nonepidermal DCs. (A) Schematic representationof LC chimeras. (B) Expression of IE in the skin and skin-draining LNs ofchimeric mice. Representative !ow cytometric plots are gated to show thefrequency of IE+ DCs as a percentage of total DCs. (C) Schematic represen-tation of control chimeras.

18050 | www.pnas.org/cgi/doi/10.1073/pnas.1110076108 Shklovskaya et al.


















epicutaneous responses should recapitulate the tolerogenic re-sponses we had documented in LC chimeras. B10.BR miceadoptively transferred with 5C.C7 cells were either immunizeds.c. with HELMCC/CFA or epicutaneously with HELMCC incream containing a mixture of potent adjuvants (Fig. 3 G and H).Fewer 5C.C7 cells were recovered 6 d after epicutaneous im-munization (down 5.2-fold in dLN and 3.9-fold in spleen com-pared with s.c. immunization), and no donor T cells could berecovered by day 70 (Fig. 3G). The number of effector cells wasalso markedly reduced (down 61-fold for IL-17-, 5.9-fold forIFN#-, and 7.7-fold for IL-2-producing cells;) (Fig. 3H). Ina second experiment comparing epicutaneous immunization ofWT hosts versus LC chimeras, the day 7 response of 5C.C7 cellsin LC chimeras was over 80% of that in WT hosts, indicating thatpresentation of free antigen by resident LN DCs in WT mice isunlikely to account for more than a small proportion of the re-sponse. Thus, the effect of epicutaneous immunization in WTmice mirrored that seen in LC chimeras, con!rming that LCssubserve a tolerogenic function in normal animals.

Activated LCs Fail to Translocate RelB to the Nucleus. The surprisinglack of correlation between costimulatory molecule expression

and LC function in vivo led us to test LCs for further correlatesof DC tolerogenicity. Activation of the NF-!B transcriptionfactor RelB, as indicated by translocation to the nucleus, is oneof the best-established markers of DC immunogenicity in vivo(13, 21). Whereas a proportion of m-DDCs showed clear evi-dence of nuclear translocation of RelB after skin painting witha contact sensitizer, s.c. immunization with CFA, and epicuta-neous immunization with cream/CFA particulates, RelB trans-location was never seen in m-LCs (Fig. 3I and Fig. S7). Thus, theactivation and nuclear translocation of RelB appeared to bea reliable correlate of DC immunogenicity in vivo.

Visualizing Activation and Migration of LCs. In addition to differ-ential activation of RelB, LCs and DDCs show consistent dif-ferences in their migratory behavior, with migrating LCs slowerin reaching dLN than DDCs (9, 10). We used intravital mi-croscopy to monitor the behavior of LCs over the !rst 4 d of theepicutaneous response (Fig. S8). In the steady state, LCs weresessile (mean velocity <1 $m/min), with their dendrites re-maining almost completely immobile as described previously (9,22). Ninety-six hours after application of cream onto ear skin,LCs appeared as round cells with retracted dendrites, deeply

Fig. 2. In vivo response of naïve CD4+ T cells to antigen presented by m-LCs or nonepidermal DCs. Fully reconstituted (>3 mo) LC or control chimeras wereadoptively transferred with 2 ! 105 CFSE-labeled 5C.C7 T cells and s.c. immunized with 10 $g MCC peptide/CFA. (A) Response of donor 5C.C7 T cells in dLNs.Representative !ow cytometric plots of CD4 T cells are gated to show the frequency of donor 5C.C7 cells (Left) and their CFSE-dilution pro"les (Right). (B)Absolute number of donor 5C.C7 T cells in dLN and spleen. Data are from one representative experiment out of three (3–5 animals per group). (C) Acquisitionof effector memory phenotype by donor 5C.C7 T cells. Donor 5C.C7 T cells in dLN were gated for undivided CFSEhigh 5C.C7 cells (gate I) and fully dividedCD62L!CFSE! 5C.C7 cells (gate II). (Right) Mean absolute numbers of cells within the two gates. Values are for one representative experiment. (D and E)Representative !ow cytometric plots (D) and absolute number (E) of cytokine-producing donor 5C.C7 cells in dLN. Numbers in D indicate the frequency of cellsin each of the four quadrants. (F and G) Antigen presentation by LCs does not support differentiation of CD4+ memory cells. Memory response to day 60 s.c.challenge with peptide/IFA (F) or to day 80 intradermal challenge with peptide-pulsed IE+ splenic DCs (G). Draining LNs were collected 16 h (F) or 3 d afterchallenge (G). (F Left) Frequency of 5C.C7 cells expressed as a percentage of total CD4 T cells in unchallenged versus challenged mice. (F Center) Expression ofCD44 and CD62L. (F Right) Cytokine production after challenge. (G) Frequency of donor 5C.C7 cells (expressed as percent of total CD4 T cells) in draining LNsand ear skin after intradermal ear challenge. One representative experiment out of three is shown.

Shklovskaya et al. PNAS | November 1, 2011 | vol. 108 | no. 44 | 18051

IMMUN

OLO

GY



embedded into underlying collagen; these changes were particu-larly apparent with cream/CFA. However, actual crossing of thebasement membrane and entry into dermis were only infrequentlyobserved, consistent with the delayed kinetics of migration.

Migratory LCs Inhibit the Effector Phase of the Immune Response.Although T-cell activation in LC chimeras correlated with thearrival of m-LCs from the immunization site (Fig. 3), it remainedpossible that they arrived too late to rescue a default tolerogenicresponse stimulated by steady-state m-LCs already in the node.To test whether migrating LCs could actively participate in on-going responses initiated by rapidly migrating m-DDCs, wecreated combined radiation chimeras in which both LCs andDDCs expressed IE (Fig. 4A). In these chimeras, IE+ m-DCs incLN comprised a 1:1 mixture of m-LCs and m-DDCs, comparedwith a 1:3 mixture in WT mice (10). The number of donor T cellsin the !rst 10 d post immunization was similar in combined andcontrol chimeras (Fig. 4B), but the number of effector cells incombined chimeras was signi!cantly reduced (down 8.6-fold forIFN# and 5-fold for IL-17) (Fig. 4C). Memory cell numbers wererelatively preserved (Fig. 4B), as was memory function (Fig. 4D).These results indicate that LCs potently regulate the effectorphase of the immune response by limiting T-cell effector func-tion when the ratio of m-LCs to m-DDCs is suf!ciently high. Thislimiting of T-cell effector function appears to be a direct LC-mediated effect, because antigen-speci!c foxp3+ regulatory Tcells did not emerge at any time post immunization.A second possibility is that early presentation of free antigen

by steady-state antigen-presenting m-LCs renders CD4 T cellsunable to respond productively to a subsequent exposure toactivated m-LCs. To test this, we delayed the transfer of 5C.C7 Tcells for 3 d after LC chimera immunization to allow migration ofactivated m-LCs (Fig. S9). T cells transferred into hosts preim-munized with cream/adjuvant/antigen underwent only low-levelCD69 up-regulation and proliferation, suggesting signi!cantcompetition from the endogenous T-cell response (Fig. S9A).When the hosts were treated with cream/adjuvant but adminis-tration of antigen was delayed until the day after 5C.C7 T-celltransfer, signi!cantly more proliferation was observed but noeffector cytokines were detected (Fig. S9B). Thus, primaryantigen presentation by preactivated m-LCs still failed to driveeffector/memory differentiation in naïve CD4 T cells.

DiscussionPrecommitment of DC subsets to specialized functions hasgained acceptance with the demonstration that the ability tocross-present is restricted to CD8+ cDCs and CD103+ DCs (23,24). However, the existence of DC subsets that are precommittedto tolerance induction remains controversial. To test de!ned DCsubsets for tolerogenicity, we have developed a mouse model

Fig. 3. Epicutaneous immunization activates LCs but does not support de-velopment of CD4 memory cells. (A–C) LC or control chimeras were immu-nized with HELMCC in cream applied onto hairless abdominal skin.Representative !ow pro"les (A) and kinetics of CD80 and CD86 expression(B) by m-LCs and m-DDCs in dLN (mean of three per group ± SEM). (C) Ac-tivation of 5C.C7 T cells (three mice per group). MFI, mean !uorescence in-tensity. (D) Expression of CD40, CD80, and CD86 on day 4 after immunizationwith cream containing either CFA particles, imiquimod, Pam3CSK, poly I:C,lipopolysaccharide (LPS), or curdlan, as indicated. (E) Intracellular IL-12p40/

p70 expression by IE+ m-DDCs (control chimeras) and IE+ m-LCs (LC chimeras)in dLN 6 d after epicutaneous immunization with cream containing HELMCCand CFA particles. (Upper) Representative dot plots showing the frequencyof IL-12-positive cells (gated) among IE+ m-DCs. (Lower) Absolute number ofIL-12+ m-DDCs (circles) and IL-12+ m-LCs (triangles) in dLNs. (F) Frequency ofdonor 5C.C7 T cells in dLN after memory recall with peptide/IFA on day 90after epicutaneous immunization of LC chimeras with HELMCC/cream/CFAparticulates. (G and H) Response of adoptively transferred 5C.C7 cells in B10.BR mice immunized either epicutaneously for 5 d with HELMCC/cream con-taining a mixture of CFA particulates, Pam3CSK, poly I:C, imiquimod, andcurdlan (triangles) or s.c. with HELMCC/CFA (circles). Absolute numbers ofdonor 5C.C7 cells (G) and cytokine-producing donor 5C.C7 cells (H) areshown. (I) LCs migrating to draining LNs after immunization fail to trans-locate the NF-!B subunit RelB to the nucleus. Migratory LCs or DDCs were!ow-sorted from dLN of chimeric mice after s.c. (Left) or epicutaneous(Right) immunization, and RelB translocation to the nucleus was analyzed byconfocal microscopy. Graphs show the mean percentage (±SEM) of RelBtranslocations per visual "eld for 6–8 "elds containing >200 DCs per sample.






using transgenic expression of MHCII-IE to target speci!c an-tigen presentation to individual DC subsets, enabling directfunctional measurement in vivo. Using this approach, we showhere that LCs maintain tolerogenic function under a range ofconditions that are commonly believed to induce immunoge-nicity in all DC subsets.Function of IE+ LCs was measured by comparing IE-

restricted CD4 T-cell responses under three different conditions:when LCs were the only DC subtype capable of processing andpresenting speci!c antigen (LC chimeras) (Fig. 1A); when all DCsubsets with the exception of LCs could present antigen (controlchimeras) (Fig. 1C); and when both LCs and non-LC DCs couldpresent antigen (combined chimeras) (Fig. 4A). The full comple-ment of MHCII-IA-expressing DCs was present in all three mod-els, the only differences being due to DC subset-speci!c expressionof the additional MHCII-IE allele required for speci!c antigenpresentation. We chose this approach to avoid the dif!culties in-herent in interpreting the data from MHCII knockout mice andchimeras, in which adoptively transferred CD4 T cells are rapidlydesensitized due to lack of baseline TCR engagement (25, 26).Our results indicate that naïve CD4 T cells initially prolif-

erated strongly in response to antigen presented by LCs but thengradually disappeared without effector/memory cell differentia-tion, rendering the animal tolerant to subsequent challenge withspeci!c antigen. This response was independent of whether pe-ptide or protein antigens were used, whether they were deliveredsubcutaneously or epicutaneously, and whether potent adjuvantsincluding CFA, agonistic anti-CD40 mAb, and TLR ligands wereincluded in the immunization. Thus, LCs appear to possess aninherent commitment to tolerogenic function, even when dis-

playing a CD80/86high phenotype associated with immunogenic-ity in other DC subtypes.Although this !nding may be considered surprising in the light

of the currently accepted two-signal model of T-cell activation(27), it is consistent with the well-established phenomenon of astrong CD28-dependent proliferative burst preceding i.v. pep-tide-mediated tolerance induction in vivo (12). Our results in-dicate that costimulatory molecule expression by DCs may benecessary but not suf!cient for immunogenicity in vivo. One ofthe additional biochemical requirements for immunogenicity isbelieved to be activation of the NF-!B subunit RelB (13). DCsderived from RelB!/! mice or RelB!/! chimeras, or treated withan NF-!B inhibitor (RelBlow DCs), can induce antigen-speci!ctolerance (13) and suppress in"ammatory arthritis (28). Our!nding that LCs fail both to activate RelB (Fig. 3I and Fig. S7)and to generate an effector/memory CD4 T-cell response addssupport to the notion that RelB may serve as a master regulatorof DC function.The ability of LCs to drive proliferation of naïve 5C.C7 CD4 T

cells in vivo is consistent with the potent ability of LCs to drivein vitro responses (2), but differs from published results obtainedwith OTII CD4 T cells in MHCII knockout chimeras (26). Thedifference may be due to the relatively low af!nity of OTII cellsfor speci!c antigen-MHC and/or to the MHCII!/! DC milieu,which would have led to TCR desensitization via TCR% chaindephosphorylation (25, 29). Indeed, we established that in vivoT-cell responses proceeded under essentially physiological con-ditions in our models. Thus, the IE-expressing LCs and m-LCs inthe chimeras fully supported survival of naïve T cells (Fig. S2C),which have the most stringent requirements for cognate MHCIIcontact (30). We also demonstrated identical kinetics of dele-tional tolerance in fully reconstituted LC and control chimeras(Fig. S2D), to exclude quantitative differences in antigen pre-sentation as a cause of differential cell fate in our chimeras.Stromal effects were excluded by showing that LN stroma couldnot present IE-restricted antigen to CD4 T cells (Fig. S1C). Thisis in sharp contrast to recently reported results for CD8 T cells,which can survive by means of contact with MHCI expressed byeither hematopoietic or stromal compartments (31), and can berendered tolerant by speci!c antigen presented by radioresistantLN stroma (32).LC-dependent presentation of antigen could potently suppress

generation of IL-17- and IFN#-secreting effector cells in com-bined chimeras in which the ratio of antigen-presenting m-LCs tom-DDCs was made arti!cially high to provide an unequivocalresult (Fig. 4). In unmanipulated mice in which the ratio of m-LCs to m-DDCs in cLN is 1:3 rather than 1:1, the effect of LCswould be smaller, which may explain the lack of effect in some(9, 33, 34), but not all (7, 8, 35), models of contact sensitivity.The ability of LCs to suppress the response to antigen presen-tation by other DC subsets argues against the possibility thattolerance in our models is a default response to presentation offree antigen without active involvement of migrated skin DCs.Understanding the in vivo function of LCs may provide clues

as to how DCs can mediate tolerance to TLR-expressing com-mensal organisms colonizing epithelial surfaces such as skin andbowel, whilst retaining the ability to prime a strong immuneresponse to pathogens. We propose that LCs mediate toleranceto skin commensals under steady-state conditions when thestructural integrity of the basement membrane that usuallyprovides an epidermal/dermal barrier is intact. In contrast, in-vading pathogens that breach the barrier will generate a strongresponse overwhelmingly mediated by rapidly migrating DDCs,whereas minor disturbances will be subject to a combinationof immunogenic DDC signals and LC modulation of effectorfunction but not memory generation.Finally, our !ndings provide direct evidence of a DC subset

committed to tolerance induction while responding to immunogenicsignals and displaying what is currently considered to be an immu-nogenic surface phenotype. The four recently described skin DC

Fig. 4. LCs inhibit CD4 T-cell effector responses initiated and maintained bynonepidermal DC subsets. (A) Schematic representation of combined chi-meras. (B–D) Combined, control, and LC chimeras were adoptively trans-ferred with 2 ! 105 CFSE-labeled 5C.C7 T cells and s.c. immunized with pMCC/CFA. (B) Absolute number of donor 5C.C7 T cells in dLN. Data are from onerepresentative experiment of two, with three or four animals per group. (C)Representative !ow cytometric plots (Left) and absolute number (Right) ofcytokine-producing donor 5C.C7 cells 10 d postimmunization. Each symbolrepresents an individual mouse. (D) Memory response of combined chimerasto intradermal challenge with peptide-loaded IE+ DCs 80 d after priming.(Left) Frequency of 5C.C7 T cells in draining LN and skin of challenged versusunchallenged mice. (Right) Expression of CD44 and CD62L.

Shklovskaya et al. PNAS | November 1, 2011 | vol. 108 | no. 44 | 18053

IMMUN

OLO

GY





subsets (10) thus include those specialized for negative regulation ofCD4 T cells in addition to those specialized for cross-presentation toCD8 T cells (23). On the basis of these !ndings, we predict that DCsubsets precommitted to induction of tolerance or immunity in CD4T cells will coexist with cross-presenting DCs in many organs,allowing the full range of differential T-cell responses to be gener-ated as CD4 T cells integrate a range of tolerogenic and immuno-genic signals from DCs and, in turn, regulate tolerance andimmunity within the CD8 T-cell compartment.

Materials and MethodsMice. IE"d transgenic mouse lines 107-1 and 36-2 and 5C.C7 RAG1!/! TCRtransgenic mice are described in ref. 10. CD11c-YFP transgenic mice (36) wereobtained from M. Nussenzweig (The Rockefeller University, New York, NY).More details in SI Materials and Methods. Approval for all animal experi-mentation was obtained from the Animal Ethics Committees at the Universityof Sydney and the Wistar Institute.

BM Chimeras. LC chimeras and control chimeras are described in ref. 10. Moredetails in SI Materials and Methods.

Adoptive Transfer of T Cells and Immunizations. T-cell adoptive transfer and s.cimmunization were performed essentially as described in ref. 10. For epi-cutaneous immunization, 10 $g HELMCC was mixed with adjuvants in 150mg aqueous cream (Sorbolene; Kenkay) applied onto hairless skin and se-cured with an occlusive bandage. More details in SI Materials and Methods.

Flow Cytometry. The analysis and antibodies used are described in detail in SIMaterials and Methods.

T-Cell Effector and Memory Assays. For effector restimulation, lymph nodeand spleen cell suspensions were cultured with 10 $M pMCC for 10 h(effectors) or 16 h (memory cells) in the presence of magnetically isolated(Miltenyi Biotech) IE+ splenic DCs and Brefeldin A. After culture, cells werestained as for !ow cytometry, "xed, permeabilized, and stained using anti-IFN#, anti-IL2, and anti-IL17 antibodies. For memory recall, mice were chal-lenged s.c. into front footpads with 10 $g pMCC in IFA or intradermally intothe ear pinna with MCC-pulsed IE+ splenic DCs. Culture and staining forcytokine detection were as described for effector cells. More details areavailable in SI Materials and Methods.

RelB Staining. Chimeric mice were skin-painted with !uorescein iso-thiocyanate as described (10) or immunized s.c. or epicutaneously. m-LCs andm-DDCs were isolated from draining LNs by !ow sorting, cytospun ontoglass slides, "xed, and stained for RelB and nuclear DNA and analyzed byconfocal microscopy. Details of sorting and staining procedures are in SIMaterials and Methods.

Two-Photon Intravital Microscopy. Two-photon intravital microscopy of LCsand DDCs was performed on ear skin of anesthetized CD11c-YFPmice. Detailsof imaging and image analysis are described in SI Materials and Methods.

ACKNOWLEDGMENTS. We thank C. Zhu and T. Hartkopf for technicalassistance, the staffs of the Centenary Institute Flow Cytometry and AnimalFacilities for excellent technical support, and A. Smith and members of ourlaboratories for stimulating discussion. This work was supported by theAustralian National Health and Medical Research Council (E.S., B.R., R.T.,W.W., and B.F.d.S.G.), the Queensland Government (B.J.O.), Arthritis Queens-land (R.T.), and the New South Wales Government (W.W.).

1. Steinman RM, Banchereau J (2007) Taking dendritic cells into medicine. Nature 449:419–426.

2. Romani N, Clausen BE, Stoitzner P (2010) Langerhans cells and more: Langerin-expressing dendritic cell subsets in the skin. Immunol Rev 234:120–141.

3. Allan RS, et al. (2003) Epidermal viral immunity induced by CD8"+ dendritic cells butnot by Langerhans cells. Science 301:1925–1928.

4. Bennett CL, et al. (2005) Inducible ablation of mouse Langerhans cells diminishes butfails to abrogate contact hypersensitivity. J Cell Biol 169:569–576.

5. Bennett CL, Noordegraaf M, Martina CA, Clausen BE (2007) Langerhans cells are re-quired for ef"cient presentation of topically applied hapten to T cells. J Immunol 179:6830–6835.

6. Grabbe S, Steinbrink K, Steinert M, Luger TA, Schwarz T (1995) Removal of the ma-jority of epidermal Langerhans cells by topical or systemic steroid application en-hances the effector phase of murine contact hypersensitivity. J Immunol 155:4207–4217.

7. Kaplan DH, Jenison MC, Saeland S, Shlomchik WD, Shlomchik MJ (2005) EpidermalLangerhans cell-de"cient mice develop enhanced contact hypersensitivity. Immunity23:611–620.

8. Igyarto BZ, et al. (2009) Langerhans cells suppress contact hypersensitivity responsesvia cognate CD4 interaction and Langerhans cell-derived IL-10. J Immunol 183:5085–5093.

9. Kissenpfennig A, et al. (2005) Dynamics and function of Langerhans cells in vivo:Dermal dendritic cells colonize lymph node areas distinct from slower migratingLangerhans cells. Immunity 22:643–654.

10. Shklovskaya E, Roediger B, Fazekas de St Groth B (2008) Epidermal and dermaldendritic cells display differential activation and migratory behavior while sharing theability to stimulate CD4+ T cell proliferation in vivo. J Immunol 181:418–430.

11. Seder RA, Paul WE, Davis MM, Fazekas de St Groth B (1992) The presence of in-terleukin 4 during in vitro priming determines the lymphokine-producing potential ofCD4+ T cells from T cell receptor transgenic mice. J Exp Med 176:1091–1098.

12. Smith AL, Wikstrom ME, Fazekas de St Groth B (2000) Visualizing T cell competitionfor peptide/MHC complexes: A speci"c mechanism to minimize the effect of precursorfrequency. Immunity 13:783–794.

13. Martin E, O’Sullivan B, Low P, Thomas R (2003) Antigen-speci"c suppression ofa primed immune response by dendritic cells mediated by regulatory T cells secretinginterleukin-10. Immunity 18:155–167.

14. Widera G, et al. (1987) Transgenic mice selectively lacking MHC class II (I-E) antigenexpression on B cells: An in vivo approach to investigate Ia gene function. Cell 51:175–187.

15. Merad M, et al. (2002) Langerhans cells renew in the skin throughout life understeady-state conditions. Nat Immunol 3:1135–1141.

16. Shklovskaya E, Fazekas de St Groth B (2006) Severely impaired clonal deletion of CD4+

T cells in low-dose irradiated mice: Role of T cell antigen receptor and IL-7 receptorsignals. J Immunol 177:8320–8330.

17. Sallusto F, Geginat J, Lanzavecchia A (2004) Central memory and effector memoryT cell subsets: Function, generation, and maintenance. Annu Rev Immunol 22:745–763.

18. Hawiger D, et al. (2001) Dendritic cells induce peripheral T cell unresponsivenessunder steady state conditions in vivo. J Exp Med 194:769–779.

19. Stoitzner P, Tripp CH, Douillard P, Saeland S, Romani N (2005) Migratory Langerhanscells in mouse lymph nodes in steady state and in!ammation. J Invest Dermatol 125:116–125.

20. Holzmann S, et al. (2004) A model system using tape stripping for characterization ofLangerhans cell-precursors in vivo. J Invest Dermatol 122:1165–1174.

21. O’Sullivan BJ, Thomas R (2002) CD40 ligation conditions dendritic cell antigen-presenting function through sustained activation of NF-!B. J Immunol 168:5491–5498.

22. Ng LG, et al. (2008) Migratory dermal dendritic cells act as rapid sensors of protozoanparasites. PLoS Pathog 4:e1000222.

23. Bedoui S, et al. (2009) Cross-presentation of viral and self antigens by skin-derivedCD103+ dendritic cells. Nat Immunol 10:488–495.

24. Edelson BT, et al. (2010) Peripheral CD103+ dendritic cells form a uni"ed subset de-velopmentally related to CD8"+ conventional dendritic cells. J Exp Med 207:823–836.

25. Dorfman JR, Stefanová I, Yasutomo K, Germain RN (2000) CD4+ T cell survival is notdirectly linked to self-MHC-induced TCR signaling. Nat Immunol 1:329–335.

26. Allenspach EJ, Lemos MP, Porrett PM, Turka LA, Laufer TM (2008) Migratory andlymphoid-resident dendritic cells cooperate to ef"ciently prime naive CD4 T cells.Immunity 29:795–806.

27. Watts TH (2010) Staying alive: T cell costimulation, CD28, and Bcl-xL. J Immunol 185:3785–3787.

28. Martin E, et al. (2007) Antigen-speci"c suppression of established arthritis in mice bydendritic cells de"cient in NF-!B. Arthritis Rheum 56:2255–2266.

29. Hochweller K, et al. (2010) Dendritic cells control T cell tonic signaling required forresponsiveness to foreign antigen. Proc Natl Acad Sci USA 107:5931–5936.

30. Takeda S, Rodewald HR, Arakawa H, Bluethmann H, Shimizu T (1996) MHC class IImolecules are not required for survival of newly generated CD4+ T cells, but affecttheir long-term life span. Immunity 5:217–228.

31. Markiewicz MA, Brown I, Gajewski TF (2003) Death of peripheral CD8+ T cells in theabsence of MHC class I is Fas-dependent and not blocked by Bcl-xL. Eur J Immunol 33:2917–2926.

32. Lee JW, et al. (2007) Peripheral antigen display by lymph node stroma promotes T celltolerance to intestinal self. Nat Immunol 8:181–190.

33. Stoecklinger A, et al. (2007) Epidermal Langerhans cells are dispensable for humoraland cell-mediated immunity elicited by gene gun immunization. J Immunol 179:886–893.

34. Bursch LS, Rich BE, Hogquist KA (2009) Langerhans cells are not required for the CD8 Tcell response to epidermal self-antigens. J Immunol 182:4657–4664.

35. Bobr A, et al. (2010) Acute ablation of Langerhans cells enhances skin immuneresponses. J Immunol 185:4724–4728.

36. Lindquist RL, et al. (2004) Visualizing dendritic cell networks in vivo. Nat Immunol 5:1243–1250.


http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1110076108/-/DCSupplemental/pnas.201110076SI.pdf?targetid=nameddest=STXT










Supporting InformationShklovskaya et al. 10.1073/pnas.1110076108SI Materials and MethodsMice. All mice were housed under speci!c pathogen-free con-ditions in the Centenary Institute (CI) Animal Facility. MHCII-IE!d transgenic mouse lines 107-1 and 36-2 were bred on aCD45.1 C57BL/6 background. IE+ mice were also bred ontoa CD45.2 RAG1!/! C57BL/6 background. WT B10.BR (H-2k)mice were bred on a CD45.1 background. 5C.C7 transgenic miceexpressing the 5C.C7 T-cell receptor (TCR) (V!11+V"3+) (1, 2)were on either a C57BL/6 IE+ RAG1!/! or B10.BR RAG!/!

background (both CD45.2).

Bone Marrow Chimeric Mice. IE!!IE+ [Langerhans cell (LC)chimeras] and IE+!IE! (control chimeras) are described indetail in ref. 3. Brie"y, for LC chimeras, IE+CD45.1 hosts weretreated with 1,200 cGy split-dose irradiation (2 ! 600 cGy, 3 hapart) and i.v. injected with 10 ! 106 IE!CD45.1 bone marrow(BM) cells. For control chimeras, IE!CD45.1 hosts were irra-diated with 600 cGy and grafted with IE+CD45.2 RAG1!/! BMmixed with host-type BM (10 ! 106 cells per mouse). Whenmixed at a 1:3 ratio, the resulting IE+ chimerism in the dendriticcell (DC) lineage was on average 22%. For combined chimeras,IE+CD45.1 hosts were treated with 1,200 cGy split-dose irradi-ation (as above) and received the same BM graft as controlchimeras. All chimeric mice were allowed to rest for at least3 mo before experimental use.

Adoptive Transfer of T Cells and Immunizations.Naïve CD4+ T cellsfrom 5C.C7 RAG1!/! mice were labeled with 5 #M carboxy-"uorescein diacetate succinimidyl ester (CFSE; Invitrogen) asdescribed (4). Cells (2 ! 105) were injected i.v. into chimeras andmice were immunized 24 h later.Subcutaneous immunization. Mice were immunized s.c. with 10 #gof moth cytochrome C peptide (pMCC) 87–103 KANER-ADLIAYLKQATK (Auspep) or with recombinant hen egg ly-sozyme-moth cytochrome C protein (HELMCC) containing theMCC87–103 epitope between residues 64 and 76 of the matureHEL protein. HELMCC was produced in a yeast expressionsystem (Pichia pastoris; Invitrogen) and af!nity-puri!ed as de-scribed (3). The dose of HELMCC per mouse was equivalent to1 #g of MCC peptide, as calculated from in vivo dose compar-isons of the response of 5C.C7 T cells to s.c. immunization incomplete Freund’s adjuvant (CFA). Peptide or protein antigenwas diluted in PBS, emulsi!ed 1:1 in CFA (Sigma), and injecteds.c. into both hind footpads (50 #L) and the base of the tail (100#L). In some experiments, mice received an s.c. injection ofpMCC in CFA, as described above, and were additionally im-munized with 50 #g of agonistic anti-CD40 antibodies (cloneFGK45) intraperitoneally on days 0 and 2 postimmunization.Epicutaneous immunization. After abdominal hair removal with hairremoval cream, 10 #g HELMCC in 150 mg aqueous cream(Sorbolene; Kenkay) was applied onto the abdominal skin andsecured with an occlusive bandage. The bandage was removedafter 5 d. In some experiments, adjuvants were mixed into thecream before immunization (dose per mouse): 25 #g Pam3Cys-Ser-(Lys)4 (Pam3CSK), 50 #g polyinosinic acid:polycytidylic acid(poly I:C), 1.25 mg lipopolysaccharide (LPS), 150 #g curdlan (allfrom InvivoGen), or particles obtained from CFA (equivalent of150 #L of CFA pelleted and extensively washed in PBS). Imi-quimod was used as 5% Aldara cream (iNova Pharmaceuticals)at 150 mg per mouse. Pooled draining lymph nodes (dLNs)(inguinal, brachial, and axillary) were analyzed for expression ofCD80/86 by "ow cytometry.

Flow Cytometry. Staining with mAbs was performed in PBScontaining 5% FCS, 10 mM EDTA, and 0.02% sodium azide.All mAbs were "uorochrome- or biotin-conjugated and werefrom either BD Pharmingen, eBioscience, or custom-made (CI).Nonspeci!c staining was blocked with anti-CD16/32 (2.4G2).Dead cells were stained with 4’,6-diamidino-2-phenylindole(DAPI; Invitrogen). Acquisition was performed on an LSRIIdigital "ow cytometer equipped with blue, red, violet, and UVlasers (BD Biosciences). FlowJo software (Tree Star) was usedfor data analysis.T cells. Draining LN and spleen cells were stained for CD62L(clone MEL14), V!11 (RR8), CD4 (RM4-5), V"3 (KJ25), CD44(IM78.1), and CD45.1 (A20).DCs. Spleens and LNs were digested with collagenase/DNaseI(both from Sigma-Aldrich) as described (5). Ear skin was sepa-rated into epidermal and dermal sheets with 5 U/mL dispase(Sigma-Aldrich), followed by digestion with 2 mg/mL collage-nase IV (Sigma-Aldrich). Cells were stained for MHCII-IE(clone 14.4.4s), pan-MHCII (M5/114), CD40 (3/23), CD80 (16-10A1), CD86 (GL1), CD11c (HL3), and pan-CD45 (30-F11).Intracellular staining with anti-IL12p40/p70 (clone C15.6) wasperformed after overnight culture in the presence of Brefeldin A(Sigma-Aldrich).Analysis. DAPI-negative events were gated for forward scatter(FSC) height vs. area to exclude DC-DC and T-DC clusters.Hematopoietic cells in epidermal and dermal samples wereidenti!ed with pan-CD45. DC subsets were identi!ed usingmAbs against CD11c, B220, pan-MHCII, and MHCII-IE. DonorT cells were identi!ed as CD4+CD45.1!TCRV!11+ orTCRV"3+; CFSE pro!les were analyzed for cell-division patternas described (6, 7).Cell sorting. Lymph nodes from LC chimeras and IE+ micewere collected, digested, stained with mAbs, and "ow-sorted to>90% purity for the following subsets: B cells (MHCII+B220+CD11c!), LN stromal cells (CD45!MHCI+), conventional DCs(CD11chighB220!MHCIIint), migratory (m-)LCs (CD11cintB220!MHCIIhighIE+), and migratory dermal DCs (m-DDCs) (CD11cintB220!MHCIIhighIE!). Sorted cells (1 ! 105 per well) were culturedwith CFSE-labeled naïve 5C.C7 T cells (3 ! 105 per well) and 1 #g/mL HEL/MCC. CFSE dilution was analyzed by "ow cytometryafter 4 d of culture.

T-Cell Effector and Memory Assays. Effector restimulation. Ten millionLN or spleen cells were cultured with 10 #M MCC87–103 for 10 h(effectors) or 16 h (memory cells) in the presence of 0.5 ! 106per well freshly isolated IE+ DCs. DCs were obtained fromspleens of IE+ donors using a CD11c magnetic selection kit(Miltenyi Biotech). Brefeldin A was added after 2 h of culture to5 #g/mL !nal concentration. After culture, cells were stained forCD4, CD45.1, and either V!11 or V"3, !xed with 4% para-formaldehyde, and permeabilized with 0.1% BSA and 0.5%saponin in PBS. Intracellular staining was performed usingmonoclonal antibodies against IFN$ (XMG1.2), IL-2 (JES6-5H4), and IL-17A (eBioTC11-18H10.1). The background stain-ing of host CD4+ cells was below 0.5% for all cytokines.Memory recall. Mice were challenged s.c. into front footpads with10 #g MCC87–103 dissolved in PBS and emulsi!ed 1:1 in incom-plete Freund’s adjuvant (IFA; Sigma-Aldrich). Alternatively,mice were injected intradermally into the ear pinna with antigen-pulsed IE+ DCs (5 ! 105 per ear); DCs were puri!ed fromspleens of IE+ donors using CD11c magnetic beads (MiltenyiBiotech), pulsed with 10 #M pMCC for 30 min at 37 °C, and

Shklovskaya et al. www.pnas.org/cgi/content/short/1110076108 1 of 11

www.pnas.org/cgi/content/short/1110076108

extensively washed before injection. Draining LNs and spleenswere isolated 16 h after s.c. challenge or 3 d after intradermalchallenge, respectively. Culture and staining for cytokine dete-ction was as described for effector cells.

RelB Staining. Chimeric mice were skin-painted with "uoresceinisothiocyanate (FITC) in acetone:dibutyl phthalate as described (3)or immunized subcutaneously or epicutaneously. On days 2 and 3after FITC painting, day 2 after s.c. immunization, and day 4 afterepicutaneous immunization, draining LNs were digested and cellswere stained for "ow cytometry. IE+ m-LCs and IE! m-DDCsfrom LC chimeras, and IE+ m-DDCs from control chimeras, were"ow-sorted, cytospun onto glass slides, air-dried, and !xed with4% paraformaldehyde. Slides were blocked with 5% BSA in 0.1%Tween-20/Tris (Fluka) and stained overnight at 4 °C for RelB (C-20; Santa Cruz Biotechnology) followed by goat anti-rabbit Alexa-555 (Invitrogen). Nuclei were counterstained with DAPI. Imageswere acquired on an LSM510 Meta confocal microscope.

Two-Photon Intravital Microscopy and Image Analysis. CD11c-YFPmice were anesthetized and ear hair was removed. Themouse was

placed on a custom-built stage maintained at 36 °C. The ear waspositioned on a small metal platform, immersed in PBS/glycerin(70:30, vol:vol), and covered with a coverslip (8). Imaging wasperformed on a LaVision BioTec TriM Scope attached to anOlympus BX-51 !xed-stage microscope equipped with 20! (NA0.95) and 40! (NA 0.8) water-immersion objectives. The setupincluded six external nondescanned dual-channel re"ection/"uorescence detectors and a diode-pumped, wide-band mode-locked Ti:sapphire fs laser (MaiTai HP; Spectra-Physics; 720–1,050 nm, <140 fs, 90 MHz). Three-dimensional images of earskin were acquired (1- to 6-#m spacing in z axis over a totaldistance of 30–40 #m) every 30–60 s for a period of up to 4 h.Three-dimensional image stacks were analyzed using Volocitysoftware (Improvision). Migration parameters were assessed asdescribed (9).

Statistical Analysis. One-way ANOVA with Newman–Keuls mul-tiple-comparison posttest (GraphPad) was used to analyze thedifferences in effector cell numbers between combined andcontrol chimeras.

1. Fazekas de St. Groth B, Patten PA, Ho WY, Rock EP, Davis MM (1992) An analysis ofT cell receptor-ligand interaction using a transgenic antigen model for T cell toleranceand T cell receptor mutagenesis. Molecular Mechanisms of Immunological Self-Recognition, eds Alt FW, Vogel HJ (Academic, San Diego), pp 123e127.

2. Seder RA, Paul WE, Davis MM, Fazekas de St Groth B (1992) The presence of interleukin4 during in vitro priming determines the lymphokine-producing potential of CD4+

T cells from T cell receptor transgenic mice. J Exp Med 176:1091e1098.3. Shklovskaya E, Roediger B, Fazekas de St Groth B (2008) Epidermal and dermal

dendritic cells display differential activation and migratory behavior while sharing theability to stimulate CD4+ T cell proliferation in vivo. J Immunol 181:418e430.

4. Smith AL, Wikstrom ME, Fazekas de St Groth B (2000) Visualizing T cell competition forpeptide/MHC complexes: A speci!c mechanism to minimize the effect of precursorfrequency. Immunity 13:783e794.

5. Smith AL, Fazekas de St. Groth B (1999) Antigen-pulsed CD8!+ dendritic cells generatean immune response after subcutaneous injection without homing to the draininglymph node. J Exp Med 189:593e598.

6. Fazekas de St Groth B, et al. (1999) Carboxy"uorescein diacetate succinimidyl esterand the virgin lymphocyte: A marriage made in heaven. Immunol Cell Biol 77:530e538.

7. Shklovskaya E, Fazekas de St Groth B (2006) Severely impaired clonal deletion of CD4+

T cells in low-dose irradiated mice: Role of T cell antigen receptor and IL-7 receptorsignals. J Immunol 177:8320e8330.

8. Ng LG, et al. (2008) Migratory dermal dendritic cells act as rapid sensors of protozoanparasites. PLoS Pathog 4:e1000222.

9. Mrass P, et al. (2006) Random migration precedes stable target cell interactions oftumor-in!ltrating T cells. J Exp Med 203:2749e2761.



Fig. S1. Migratory LCs but not LN stromal cells express IE and activate IE-restricted CD4 T cells in vitro. (A) Flow cytometric analysis of skin-draining LNs from LCchimera. IE+ and IE! cell subsets (gated as shown; Left) were further analyzed for MHCII and CD11c expression (Right). (B) LN stromal cells do not express MHCII-IE. Cutaneous LNs (Upper Right) or thymuses (Lower Right) from !ve mouse strains, as indicated, were analyzed by "ow cytometry for expression of IE andMHCI after gating for CD45+ and CD45! subsets (Left). Images are representative overlays of CD45+ hematopoietic cells (blue) and CD45! stromal cells (red).Numbers indicate the mean "uorescence intensity (MFI) value of IE expression within the respective color-coded gates. (C) Only IE+ m-LCs in LC chimeras processand present protein antigen to 5C.C7 T cells in vitro. Naïve CFSE-labeled 5C.C7 T cells (3 ! 105) were cocultured with 1 ! 105 of antigen-presenting cells (APCs)"ow-sorted from pooled cutaneous LNs of LC chimeras (Upper) or spleens of IE+ mice (Lower) in the presence of 1 #g/mL HELMCC protein. The following cellsubsets were sorted: B cells, CD45!MHCI+ LN stromal cells, MHCIIintCD11chigh conventional DCs, MHCIIhighIE+ m-LCs, and MHCIIhighIE! m-DDCs. n.d., not done.Proliferation of T cells was measured 4 d later by "ow cytometry.



Fig. S2. LCs expressing IE support survival and activation of IE-restricted naïve CD4 T cells in vivo. (A) IE-negative mice, IE+ mice, LC chimeras, and two groupsof control chimeras reconstituted with either a lower or higher dose of IE+ RAG!/! BM (25% or 40% BM inoculum, respectively) received 2 ! 105 IE-restrictednaïve 5C.C7 CD4+ T cells and were immunized either i.v. with MCC87–103 peptide (Left) or s.c. with MCC87–103 peptide emulsi!ed in CFA (Right). Mice were killedon day 3 postimmunization. Filled histograms show proliferation of donor 5C.C7 cells as measured by CFSE dilution, whereas unimmunized controls are in-dicated in solid bold lines. Numbers indicate the frequency of donor T cells recruited into division in response to antigen. (B) 5C.C7 T-cell recruitment into celldivision after s.c. immunization was similar for LC chimeras and 25% control chimeras. Each dot represents a single animal, with bars indicating the mean. (Cand D) IE+ migratory (m-)DCs in LC and control chimeras support long-term survival (C) and peptide-mediated deletion (D) of 5C.C7 T cells in vivo. Absolutenumber of donor 5C.C7 T cells per mouse (mean of 3 ± SEM) was estimated after adoptive transfer of 2 ! 105 naïve 5C.C7 T cells (C) followed by i.v. injection of10 #g MCC peptide in D. Control chimeras, closed circles; LC chimeras, open triangles; IE-negative mice, open circles.



Fig. S3. In vivo response of naïve CD4+ T cells to protein antigen presented by m-LCs or nonepidermal DCs. Established LC or control chimeras (Fig. 1 A and C)were adoptively transferred with 2 ! 105 CFSE-labeled 5C.C7 T cells and s.c. immunized in hind footpads and the base of the tail with 10 #g HELMCC protein inCFA. (A) Absolute number (mean ± SEM) of donor 5C.C7 T cells in draining LNs and spleens of control chimeras (circles) and LC chimeras (triangles). Data arefrom one representative experiment with 3–5 animals per group. (B) Acquisition of effector memory phenotype by antigen-speci!c 5C.C7 T cells after s.c.immunization of LC and control chimeras with protein/CFA. (Left) Representative "ow cytometric analyses of expression of CD62L by 5C.C7 T cells in drainingLNs of control chimeras (Upper) and LC chimeras (Lower). Cells were gated as shown (Left). (Right) Mean absolute numbers of undivided CFSEhigh 5C.C7 cells(gate I) and fully divided CD62L!CFSE! 5C.C7 cells (gate II) in control chimeras (Upper) and LC chimeras (Lower). Values for one representative experiment areshown. (C and D) Representative "ow cytometric plots (C) and absolute number (D) of cytokine-producing donor 5C.C7 cells in draining LNs. Numbers in Cindicate the frequency of cells in each of the four quadrants. Control chimeras, circles; LC chimeras, triangles. (E) Lack of memory in LC chimeras. Memory recallwas performed with MCC peptide in IFA in front footpads 60 d after primary immunization, as described for Fig. 2F. Draining LNs (pooled brachial and axillary)were collected 16 h after challenge and analyzed by "ow cytometry. (Left) Frequency of 5C.C7 cells expressed as a percentage of total CD4 T cells in un-challenged versus challenged mice. (Center) Expression of CD44 and CD62L. (Right) Cytokine expression after challenge. (Upper) Control chimeras. (Lower) LCchimeras. Data are for one representative experiment out of three.



Fig. S4. Absolute numbers of IE+ m-LCs and IE+ m-DDCs in draining LNs of s.c. immunized chimeric mice. In the experiment described in Fig. 2, migration of IE+

LCs and IE+ DDCs to the draining popliteal and inguinal LNs of LC chimeras (! ) and control chimeras (•) was analyzed by "ow cytometry. Mean absolutenumbers per mouse (±SEM) of IE+ m-DCs and total CD11cintB220!MHCIIhigh m-DCs (crosses) are shown. One experiment is shown (3–5 animals per group).



Fig. S5. Antigen presentation by LCs does not support differentiation of CD4+ memory cells. LC or control chimeras received naïve 5C.C7 T cells and were s.c.immunized with peptide/CFA, as described for Fig. 2. Antigen-speci!c memory was assessed 80 or 90 d later. (A) Memory response to s.c. challenge withpeptide/IFA in front footpads (this panel is identical to Fig. 2F, except that challenge was performed on day 90 rather than day 60 postimmunization). DrainingLNs (pooled brachial and axillary) were collected 16 h after challenge and analyzed by "ow cytometry. (Left) Frequency of 5C.C7 cells expressed as a percentageof total CD4 T cells in unchallenged versus challenged mice. (Center) Expression of CD44 and CD62L. (Right) Cytokine expression after challenge. (Upper)Control chimeras. (Lower) LC chimeras. One representative experiment out of three is shown. (B) Memory response to intradermal challenge with peptide-pulsed IE+ splenic DCs 80 d postimmunization (this panel complements Fig. 2G). Lymphoid tissues and skin from unchallenged and challenged mice wereharvested 72 h after challenge and analyzed by "ow cytometry or cultured in vitro for cytokine expression, as indicated. Shown are representative analyses ofdonor T-cell CD44 and CD62L expression pro!les (Left) and cytokine expression after in vitro restimulation (Right). Numbers indicate the frequency of cellswithin the gates. (Upper) Control chimeras. (Lower) LC chimeras. (C and D) Response of 5C.C7 memory cells to s.c. challenge with antigen in the experimentdescribed in Fig. 2F. LNs draining the site of primary immunization in nonchallenged mice (1° dLN) or the site of challenge (2° dLN) in control chimeras wereanalyzed by "ow cytometry 16 h after challenge. (C) Blast transformation of 5C.C7 T cells in 2° dLN. (Left) CD4 T cells were gated for donor 5C.C7 cells (red) andhost CD4 T cells (blue), and forward scatter was used to compare cell size. Red histogram, donor 5C.C7 cells; blue histogram, host CD4 T cells. (Right) FCS-A value(mean ± SEM) is plotted for each group of four or !ve mice. (D) T-cell redistribution in vivo 16 h after challenge. Spleens, 1° dLN, and 2° dLN were analyzed forfrequency (Left) and absolute number (Right) of donor 5C.C7 cells.



Fig. S6. Differential activation of LCs and DDCs migrating to draining LNs after s.c. immunization. (A) Activation of migrated MHCIIhighCD11cintIE+ LCs andMHCIIhighCD11cintIE! DDCs was assessed by "ow cytometry at the indicated times after s.c. immunization of LC chimeras with peptide/CFA. Representativepro!les of the expression of CD80 (Left) and CD86 (Right) are shown as gray-!lled histograms for IE+ m-LCs and solid bold lines for IE! m-DDCs. (B) MFI of CD40,CD80, and CD86 expression at the indicated times after immunization (mean of 3 per group ± SEM). m-LCs are shown as triangles and m-DDCs as circles. Theresult is representative of at least three independent experiments. (C) Stimulation of LCs via agonistic anti-CD40 antibody fails to rescue antigen-speci!cmemory in LC chimeras. Fifty micrograms of anti-CD40 was injected intraperitoneally on days 0 and 2 after immunization with peptide/CFA. Sixty days later,memory was assessed by s.c. challenge with peptide/IFA, as described for Fig. 2F. (Left) Donor 5C.C7 cells as the frequency of total CD4 T cells in draining LNs.(Center) Expression of CD44 and CD62L. (Right) Cytokine expression after challenge.



Fig. S7. LCs migrating to draining LNs after immunization fail to translocate the NF-%B subunit RelB to the nucleus. IE+ m-DCs were sorted from draining LNsof LC or control chimeras, as appropriate, after (A) contact sensitization (days 2 and 3 after sensitization), (B) s.c. immunization (day 2 after immunization), or(C) epicutaneous immunization (day 4 after immunization). Cells were stained for nuclear DNA (DAPI; blue) and RelB (red), and analyzed by confocal mi-croscopy. White arrows indicate a translocation event. B and C complement Fig. 3I.



Fig. S8. Two-photon intravital microscopic analysis of skin LCs in the steady state and after epicutaneous immunization. (A) Representative time-lapse (min:s)images of steady-state LCs in the ear skin of CD11c-YFP mice (Left). LCs are shown in yellow whereas collagen !bers are in blue. (Scale bars, 25 #m.) (Right)Mean velocity and displacement plots of LCs (open bars) and DDCs (gray-!lled bars) from 15-min tracks (mean ± SEM). (B and C) Ear skin of CD11c-YFP mice wastreated with hair removal cream, and mice were epicutaneously immunized with cream (left ear) or cream/CFA (right ear) applied to ear skin and secured witha bandage for the duration of immunization. Bandages were removed just before acquisition of still images 96 h postimmunization. (B) Representative 3Dreconstructions of ear skin of CD11c-YFP mice showing the morphology and distribution of LCs (yellow) in relation to collagen !bers (blue) after immunizationwith cream or cream/CFA, as indicated. (C) Three-dimensional sectioning images showing the localization of an LC in relation to collagen !bers after cream orcream/CFA immunization. Blue lines indicate the position in x-y-z planes.



Fig. S9. Response of antigen-speci!c T cells to antigen presented by preactivated LCs. (A) Response of 5C.C7 T cells in LC chimeras epicutaneously immunized4 d before T-cell transfer with protein antigen in cream containing CFA particulates, Pam3CSK, poly I:C, imiquimod, and curdlan. (Upper) Experimental setup.(Lower) Flow cytometric analysis of 5C.C7 T-cell activation in pooled draining axillary, inguinal, and brachial LNs. Row 1, frequency of donor TCRV"3+CD45.1!

5C.C7 T cells (shown as a percentage of total CD4 T cells).; row 2, expression of early activation marker CD69 on donor T cells; row 3, histogram of CFSEexpression. (B) Response of 5C.C7 T cells in LC chimeras treated on day !4 with skin application of cream plus adjuvants as above, transferred with T cells onday !1 and s.c. immunized with soluble peptide antigen on day 0. (Upper) Experimental setup. (Lower) Flow cytometric analysis of 5C.C7 T-cell activation indraining axillary, inguinal, and brachial LNs. Row 1, frequency of donor TCRV"3+CD45.1! 5C.C7 T cells (shown as a percentage of total CD4 T cells); row 2,histogram of CFSE expression. (Right) Analysis of cytokine production 30 d postimmunization after in vitro restimulation with peptide and IE+ DCs.



Documents

Langerhans cells are precommitted to immune tolerance induction