multiple paths for activation of naive cd81 t cells: … · multiple paths for activation of naive...

8

Click here to load reader

Upload: duongcong

Post on 18-Aug-2018

212 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Multiple Paths for Activation of Naive CD81 T Cells: … · Multiple Paths for Activation of Naive CD81 T Cells: ... the preactivated DCs can efficiently activate naive ... b1(R&D

of August 18, 2018.This information is current as

T Cells: CD4-Independent Help+Multiple Paths for Activation of Naive CD8

FrelingerJack R. Bennink, Jonathan W. Yewdell and Jeffrey A. Bo Wang, Christopher C. Norbury, Roberta Greenwood,

http://www.jimmunol.org/content/167/3/1283doi: 10.4049/jimmunol.167.3.1283

2001; 167:1283-1289; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/167/3/1283.full#ref-list-1

, 27 of which you can access for free at: cites 56 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2001 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on August 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on August 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Page 2: Multiple Paths for Activation of Naive CD81 T Cells: … · Multiple Paths for Activation of Naive CD81 T Cells: ... the preactivated DCs can efficiently activate naive ... b1(R&D

Multiple Paths for Activation of Naive CD81 T Cells:CD4-Independent Help1

Bo Wang,2* Christopher C. Norbury, † Roberta Greenwood,* Jack R. Bennink,†

Jonathan W. Yewdell,† and Jeffrey A. Frelinger3,4*†

CD81 CTLs play a pivotal role in immune responses against many viruses and tumors. Two models have been proposed. The“three-cell” model focuses on the role of CD41 T cells, proposing that help is only provided to CTLs by CD41 T cells that recognizeAg on the same APC. The sequential “two-cell” model proposes that CD41 T cells can first interact with APCs, which in turnactivate naive CTLs. Although these models provide a general framework for the role of CD41 T cells in mediating help for CTLs,a number of issues are unresolved. We have investigated the induction of CTL responses using dendritic cells (DCs) to immunizemice against defined peptide Ags. We find that help is required for activation of naive CTLs when DCs are used as APCs,regardless of the origin or MHC class I restriction of the peptides we studied in this system. However, CD81 T cells can provideself-help if they are present at a sufficiently high precursor frequency. The important variable is the total number of T cellsresponding, because class II-knockout DCs pulsed with two noncompeting peptides are effective in priming.The Journal ofImmunology,2001, 167: 1283–1289.

T he CD81 T cell plays an important role in eradicatingvirus-infected cells and tumor cells (1, 2). There is in-creasing interest in vaccine development to induce potent

CD81 T cell-mediated cytotoxic responses to viruses and tumorcells. This requires a detailed understanding of how naive CD81 Tcells are activated and differentiate into effector cells. Numerousstudies have demonstrated that activation and differentiation of na-ive CD81 T cells into cytotoxic effector cells requires help fromCD41 Th cells (3–8). Two models have been proposed to explainhow CD41 Th cells might participate in the activation of naiveCD81 T cells. The three-cell model suggests that both CD41 Tcells and naive CD81 T cells must interact simultaneously with acommon APC (9–11). In this model, CD41 Th cells are activatedand provide help via production of cytokines such as IL-2. Thesecond model, named the two-cell model, proposes that CD41 Tcells first interact with dendritic cells (DCs),5 resulting in activa-tion of DCs with up-regulation of costimulatory molecules. Sub-sequently, the preactivated DCs can efficiently activate naiveCD81 T cells (11–13). These models provide a general frameworkfor understanding the requirement of CD41 Th cells for activation

of naive CD81 T cells. In contrast, diverse numbers of studieshave found little or no requirement for CD41 Th cells in inductionof anti-viral CTLs (14–19). Furthermore, naive CD81 T cells ex-pressing a transgenic (tg) TCR can be activated in vitro and in vivowith peptide epitope, apparently without involvement of CD41 Tcells (20, 21). These observations cannot be easily explained bythe above models.

To understand the mechanism of the development of CTL re-sponse, it is crucial to consider DCs as APCs. DCs appear to be theprincipal APC in the initiation of CD81 T cell responses to manyAgs (22–25). DCs present Ags to both MHC class II-restrictedCD41 and class I-restricted CD81 T cells by classical Ag-present-ing pathways (26, 27). DCs can actively acquire exogenous Agsand generate MHC class I-restricted peptides via both TAP-depen-dent and -independent pathways (28–32). Immunization of ani-mals with DCs pulsed with various Ags has been shown to induceprotective cytotoxic responses against viruses (33, 34) and tumorcells (35–37). In this study, we have further dissected the mech-anism of help in the induction of CTL response in vivo using DCsgenerated from spleen. Our data in the present study extend theprevious finding from early studies and indicates that CD41 T cellscan still provide the help essential for activation of naive CD81 Tcells without interacting with the same APC. More important, ourdata show that CD81 T cells can also provide help for their ownactivation.

Materials and MethodsMice

Six-week-old C57BL/6 mice (B6) were obtained from The Jackson Lab-oratory (Bar Harbor, ME), Taconic Farms (Germantown, NY), or the Na-tional Institutes of Health Genetic Stock Center (Taconic Farms). B6 micecarrying a mutation in either the IAb (B6.129-Abbtm1; IAb%) (38, 39) orb2-microglobulin (b2m) gene (B6.129P2(B6)-B2 Mtm1Unc; b2m

%) (40, 41)have previously been described and were further backcrossed to B6 micea minimum of nine times. P14 mice (B6.D2-TgN(TcrLCMV)327Sdz) ex-pressing the H2-Db-restricted tg TCR specific for the determinant (aminoacid 33–41) derived from the lymphocytic choriomeningitis virus (LCMV)gp33 (42) were obtained from The Jackson Laboratory (at N4) and furtherbackcrossed to B6 mice a minimum of five times. B6 mice tg for greenfluorescent protein (GFP) expressed under the

*Department of Microbiology and Immunology, University of North Carolina,Chapel Hill, NC 27599; and†Laboratory for Viral Diseases, National Institute ofAllergy and Infectious Diseases, Bethesda MD 20235

Received for publication March 6, 2001. Accepted for publication May 24, 2001.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by National Institutes of Health Grant AI 20288. C.C.N.was the recipient of a Wellcome Prize Traveling Fellowship.2 Current address: Division of Endocrinology, Children’s Hospital Medical Center,Cincinnati, OH 45229.3 This work was initiated while J.A.F. was on leave at the Laboratory for Viral Dis-eases, National Institute of Allergy and Infectious Diseases (Bethesda, MD).4 Address correspondence and reprint requests to Dr. Jeffrey A. Frelinger, Departmentof Microbiology and Immunology, University of North Carolina, Campus Box 7290,Mary Ellen Jones Building, Chapel Hill, NC 27599-7290. E-mail address:[email protected] Abbreviations used in this paper: DC, dendritic cell;b2m, b2-microglobulin;LCMV, lymphocytic choriomeningitis virus; GFP, green fluorescent protein; VSV,vesicular stomatitis virus; tg, transgenic.

Copyright © 2001 by The American Association of Immunologists 0022-1767/01/$02.00

by guest on August 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Page 3: Multiple Paths for Activation of Naive CD81 T Cells: … · Multiple Paths for Activation of Naive CD81 T Cells: ... the preactivated DCs can efficiently activate naive ... b1(R&D

control of the H2-Kb promoter (B6-TgN(KbP-EGFP) were generated in ourlaboratory (43). As expected for a gene under the control of a MHC classI promoter, GFP expression is not down-regulated in T cells followingactivation and is a reliable marker for T cells derived from GFP-expressingmice. All animals used were maintained under specific pathogen-free con-ditions in the American Association of Laboratory Animal Care-accreditedUniversity of North Carolina Department of Laboratory Animal MedicineFacilities (Chapel Hill, NC).

Peptides

The following antigenic peptides were synthesized by the University ofNorth Carolina Microchemical Facility and purified by HPLC: LCMVgp33–41 (restricted by H2-Db), vesicular stomatitis virus (VSV) N52–59

and OVA257–264(restricted by H2-Kb), and peptide 324–332 derived fromthe nuclear protein of Sendai virus (restricted by both Kb and Db). All thepeptides were dissolved in 50% (v/v) DMSO in water at a concentration of10 mg/ml. The amino acid sequences of the peptides were confirmed bymass spectroscopy and shown in Table I.

Establishment of DC cultures

DCs derived from the spleen were used in the experiments for this study.Spleen cell suspensions were prepared from B6 mice, MHC class II-knock-out, or b2m-knockout mice and RBCs lysed. Cells were cultured in 5%CO2 at 37°C in six-well low adherence plates (Costar 3471; Corning Glass,Corning, NY) at 23 106/ml in a volume of 3 ml RPMI 1640 mediumsupplemented with 50mM 2-ME, 10% FBS, glutamine, 10 ng/ml mouseGM-CSF (BD PharMingen, La Jolla CA), and 1 ng/ml human TGF-b1(R&D Systems, Minneapolis, MN). Following culture for 7 days, cellswere harvested, washed, and placed into culture in fresh medium preparedas described above. Cultures were typically split every 2–3 days based onthe density of cells in each well. To assay for cell surface phenotype, DCswere stained with fluorochrome-conjugated mAbs specific for MHC classI (KH95), class II (25-9-17), CD11b (M1/70), CD11c (HL3), CD40(HM40-3), CD80 (16-10A1), and CD86 (GL1) (BD PharMingen) and an-alyzed by flow cytometry. A homogenous population of DCs was obtainedas assessed by their morphology, size (measured by forward scatter), andthe expression of surface Ags CD11b, CD11c, CD40, CD80, CD86, andMHC class I and class II. DCs have been cultured and maintained forseveral months under these conditions with continuous growth and littlechange in cell surface phenotype.

CTL in vivo priming

To prime mice for a CD81 CTL response, DCs were harvested, washedtwice with PBS, and incubated with 10mM peptide in RPMI 1640 con-taining 10% FCS. After incubation in 5% CO2 at 37°C for 2–3 h, cells werewashed three times and resuspended in PBS. The 6- to 10-wk-old micewere primed by i.v. injection (tail vein) of 2–53 105 peptide-pulsed DCsin 200 ml PBS. For coinjection,b2m-deficient DCs were treated as de-scribed above but without peptide and were then mixed together with thesame number (2–53 105) of peptide-pulsed MHC class II-knockout DCsin 200 ml PBS.

For the adoptive transfer experiments, splenocytes from P14 TCR-tgmice were prepared in PBS. B6 mice, 6–8 wk old, were injected i.v. withvarying numbers of P14 splenocytes in 200ml PBS and, 24 h later, re-ceived class II-knockout DCs pulsed with a peptide as described above.Mice that were injected with P14 splenocytes but not primed with peptide-pulsed DCs were used as controls.

In vitro stimulation and CTL assay

Seven days following immunization with peptide-pulsed DCs, splenocyteswere isolated from immunized animals and RBCs lysed with ACK lysisbuffer and resuspended in RPMI 1640 with 10% FCS, antibiotics, glu-tamine, and 50mM 2-ME. Cells (13 106 cells/ml) were stimulated with

1 mM peptide in 75-cm3 flasks (Falcon; BD Biosciences, Franklin Lakes,NJ) in 5% CO2 at 37°C for 5 days.

Cells were then harvested, washed with PBS, resuspended in RPMI1640 with 10% FCS, and used as effector cells in a51Cr-release assay.Syngeneic EL4 cells were labeled with51Cr, pulsed with peptides, and usedas target cells. Standard 4-h51Cr-release assays were performed as previ-ously described (44). The percentage of specific lysis was calculated aspreviously described (45), and each data point represents the mean51Crrelease from triplicate assays.

ResultsDCs generated from spleen can efficiently prime naive mice forcytotoxic responses specific for viral Ags

DCs are potent APCs and have recently been used in a number ofstudies to induce protective immune responses against viruses andtumors (33, 46). We have used DCs generated in vitro to primeCTL responses in vivo so that the function of the components inthe induction of a CTL response can be analyzed. In the presentstudy, our culture conditions were established to generate largenumbers of DCs from mouse spleen. Splenocytes were cultured invitro with mouse GM-CSF and human TGF-b1 (seeMaterials andMethods). After culture for 7 days, most B cells and T cells as wellas other cell types had died, whereas the remaining cells dividedand formed clusters. After expansion for;2 wk, 3–63 107 cellscould be harvested from a single spleen. These cells exhibitedtypical DC morphology and expressed MHC class I and II, CD11b,and CD11c, but low levels of CD80 and CD40 on the cell surfacewere also observed (Fig. 1). No expression of T cell (CD3)- or Bcell (IgM)-specific surface molecules was detected (data notshown). Upon overnight culture with LPS, cell surface expressionof MHC class II, class I, CD11b, CD11c and costimulatory mol-ecules CD80, CD86, and CD40 was up-regulated and the rate ofpinocytosis decreased (data not shown), consistent with maturationof the DCs. DCs derived from class II-knockout orb2m-knockoutmice expressed a similar phenotype but lacked expression of classII or class I, respectively.

To determine whether the DCs established from such cultureswere able to induce CTL responses, B6 mice were primed by i.v.injection of 2–53 105 DCs pulsed with the Db-restricted gp33–41peptide derived from the LCMV glycoprotein. As a control, ani-mals were injected with the same number of DCs that were notpulsed with peptide. Seven days following the priming, spleno-cytes were prepared and restimulated in vitro with the same pep-tide for 5 days, and peptide-specific CTL activity assessed.

Consistent with the data from other studies, CTL responses spe-cific for LCMV gp33 were induced in B6 mice immunized withLCMV gp33-pulsed DCs (Fig. 2A,F). In contrast, no CTL activityspecific for LCMV gp33 was detected when mice were immunizedwith DCs alone (Fig. 2A,E). These results demonstrate that DCs

Table I. Peptides used in immunization

PeptideMHCRestriction Sequence

LCMV gp33 Db KVANFANTMVSV N52–59 Kb RGYVYQGLOVA257–264 Kb SIINFEKLSendai N324–332 Kb/Db FAPGNYPAL

FIGURE 1. Phenotype of DCs grown from spleen. DCs were grownfrom the spleens of B6 mice under the conditions described inMaterial andMethods and stained with fluorochrome-conjugated mAbs specific forMHC class I (Db), class II (Ab), CD11b, CD11c, CD80, and CD40 (dottedline). Surface expression was measured by flow cytometry. Solid line de-picts unstained cells.

1284 BOTH CD4 AND CD8 T CELLS PROVIDE HELP FOR NAIVE CTLp

by guest on August 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Page 4: Multiple Paths for Activation of Naive CD81 T Cells: … · Multiple Paths for Activation of Naive CD81 T Cells: ... the preactivated DCs can efficiently activate naive ... b1(R&D

generated through this protocol, without the need for additionalactivation in vitro, could present exogenous peptide and induce aCD81 CTL response in vivo.

Expression of MHC class II molecules by DCs is essential forthe induction of CTL responses in vivo

Several groups have reported that activation of naive CD81 T cellsby DCs is independent of CD41 T cells (47–49). To determinewhether naive CD81 T cells can be activated and differentiate intocytotoxic effector cells upon specific interaction with DCs butwithout CD41 T cells, MHC class II-deficient DCs were used toprime CTL responses in naive B6 mice. The lack of MHC class IImolecules on the surface prevents these DCs from interacting withCD41 T cells in recipient B6 mice.

Seven days after immunization with LCMV gp33-pulsed classII-negative DCs, the B6 splenocytes were isolated and restimulatedin vitro with the same peptide. CTL activity was measured by astandard 4-h51Cr-release assay. Cytotoxic activity was also ex-amined in mice receiving wild-type DCs pulsed with LCMV gp33peptide. Wild-type B6 DCs pulsed with LCMV gp33 peptide werecapable of priming naive B6 mice (Fig. 2B, filled symbols). Instark contrast, no CTL response was detected in cultures preparedfrom mice receiving DCs lacking MHC class II expression (Fig.2B, open symbols), even after three injections of peptide-pulsedDCs (data not shown). CD81 T cells were purified and restimu-lated with irradiated syngeneic spleen cells pulsed with peptide.CTL activity was detected from mice immunized with peptide-pulsed B6 DCs, but not from mice primed with peptide-pulsedclass II-deficient DCs (data not shown). This indicates that thefailure to induce a CTL response with class II-deficient DCs is dueto the absence of CD41 T cells during priming in vivo.

In addition to CTL activity, restimulation in vitro with LCMVgp33 peptide resulted in proliferation, expansion, and activation(measured by flow cytometry) of CD81 T cells from mice primedwith normal DCs. After culture, there was a significant expansionof CD81 T cells and increased numbers of CD251 and CD691

cells..70% of CD81 T cells in mice primed with B6 DCs wereCD69 positive. In contrast, only 18% of the CD81 T cells wereCD691 in culture prepared from mice primed with MHC classII-deficient DCs.

To ensure that the failure to prime CTLs was not due to someother defect than MHC class II deficiency, we stimulated TCR-tgCD81 T cells with MHC class II-deficient DCs pulsed withLCMV gp33 peptide in vitro. Both B6 DCs (Fig. 2C,F) and MHCclass II-deficient DCs (Fig. 2C,f) could stimulate proliferation ofP14 TCR-tg T cells in vitro to the same extent and with identicalkinetics, suggesting that the latter is not defective in activatingCD81 T cells. We have also used B6 DCs to immunize micelacking CD41 T cells, due to a targeted disruption of the class IIb gene. Transfer of B6 DCs pulsed with LCMV gp33 peptide intoclass II-knockout mice failed to prime a CD81 T cell response(Fig. 3B), whereas the same DCs induced an easily detected CTLresponse in B6 mice (Fig. 3A,F).

Taken together, these data clearly demonstrate that the interac-tion between CD41 T cells and APCs is essential for the activationof naive CD81 T cells in vivo. It is likely that efficient inductionof CTL immunity against virus or tumor cells with DCs in earlystudies may reflect the activation of CD41 T cells specific forpeptides generated during the in vitro culture of DCs in mediumcontaining FBS.

CD41 and CD81 T cells are not required to interact with thesame APC

The two-cell model proposes two sequential events during activa-tion of CD81 T cells. Accordingly, CD41 Th cells first interactwith DCs, and the “conditioned” DCs then encounter and activatenaive CD81 T cells. Indeed, treatment of APCs with anti-CD40Ab in vitro or in vivo can overcome the requirement of CD41 Thcells for the activation of cytotoxic T cells (11–13). These data canexplain the role of DCs in the activation of CD81 T cells in vivo.Indeed, several studies have found that CTL immunity was in-duced in MHC class II-deficient mice by virus infection (16, 17),presumably because virus infection resulted in activation of APCssuch as DCs. However, these studies lack evidence for how, in the

FIGURE 2. A, Peptide-pulsed DCs from B6 mice induce Ag-specificcytotoxic responses in B6 mice. The 6- to 8-wk-old B6 mice were injectedi.v. with either 53 105 DCs pulsed with 10mM LCMV gp33 peptide (F)or B6 DCs alone (E) as a control. Seven days after immunization, spleno-cytes were isolated and stimulated in vitro with 1mM LCMV gp33 peptidefor 5 days. Following stimulation, CTL activity was assessed in a standard4-h 51Cr-release assay. EL4 cells were pulsed with different concentrationsof LCMV gp33 peptide and used as targets. Each data point represents theaverage specific lysis of EL4 cells in triplicates.B, MHC class II-deficientDCs fail to induce a cytotoxic response in B6 mice. B6 mice were immu-nized by injection of 53 105 LCMV gp33 peptide-pulsed B6 DCs or MHCclass II-deficient DCs. The CTL assay was conducted as described in Fig.2A. Percentage of specific lysis of target cells by spleen cells from miceimmunized with LCMV gp33-pulsed B6 DCs (filled symbols) or LCMVgp33-pulsed MHC class II-deficient DCs (open symbols) is shown. Circlesand squares represent individual mice.C, DCs deficient for MHC class IIcan stimulate the proliferation of spleen cells from P14 TCR-tg mice invitro. A total of 2 3 105 splenocytes from P14 TCR-tg mice was culturedwith different numbers of either LCMV gp33 peptide-pulsed B6 DCs (F)or class II-deficient DCs (f). As a control, spleen cells were stimulatedwith either B6 (E) or class II-deficient DCs (M) pulsed with an irrelevantpeptide. Proliferation was evaluated by incorporation of [3H]thymidine.Each data point represents an average of triplicates. When error bars are notvisible, the error bars are smaller than the symbol.

FIGURE 3. B6 DCs fail to induce a cytotoxic response in mice deficientfor MHC class II expression. Age- and sex-matched B6 (A,F) or MHCclass II-deficient mice (B,F) were primed with B6 DCs pulsed withLCMV gp33 peptide as described above. As a control, mice were injectedwith DCs without LCMV gp33 (E). Splenocytes were isolated 7 days afterimmunization and restimulated with 1mM LCMV gp33 peptide for 5 days.A CTL assay was performed as described above. Data in the figure rep-resents specific lysis of EL4 cells pulsed with different concentrations ofLCMV gp33 peptide. Similar results were obtained in three otherexperiments.

1285The Journal of Immunology

by guest on August 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Page 5: Multiple Paths for Activation of Naive CD81 T Cells: … · Multiple Paths for Activation of Naive CD81 T Cells: ... the preactivated DCs can efficiently activate naive ... b1(R&D

two-cell model, CD41 T cells condition DCs. We reasoned that theDCs used in our experiments would have been conditioned be-cause they were cultured in the presence of CD41 T cells, whichshould be able to interact with DCs carrying Ag derived fromculture medium. Therefore, once conditioned, these DCs should beable to prime naive CD81 T cells in vivo without the need forCD41 T cells. However, the results presented in Fig. 3 show thatthis was not the case.

To gain insight into how CD41 T cells provide help, B6 micewere primed by coinjection of LCMV gp33 peptide-pulsed DCsfrom MHC class II-knockout mice andb2m-knockout DCs thatwere not incubated with any specific peptides. Two immune reac-tions could take place in vivo. First, MHC class II-deficient DCspulsed with peptide interact with CD81 T cells, but not CD41 Tcells. In contrast, theb2m-knockout DCs carrying MHC class II-restricted Ags from the culture medium could interact with CD41

T cells. However, these DCs could not present Ags to CD81 Tcells because they lack MHC class I molecules. As describedabove, LCMV gp33 peptide-pulsed-B6 DCs induced a strong CTLresponse (Fig. 4,f), whereas MHC class II-knockout DCs failedto do so (Fig. 4,E). Interestingly, coinjection ofb2m-knockoutDCs with MHC class II-knockout DCs induced a modest CTLresponse (Fig. 4,F).

To examine the extent that the requirement for CD41 Th cellsvaries with determinants used to pulse DCs, we immunized micewith DCs pulsed with CTL epitopes derived from different Ags.VSV N52–59and OVA257–264are restricted by Kb, and Sendai virusN324–332is restricted by Kb and Db (Table I). As described above,B6 mice were primed with DCs prepared from B6 or class II-knockout mice or by coinjection of peptide-pulsed DCs deficientfor class II andb2m-deficient DCs. Regardless of MHC restrictionor origin of the peptides used, CD81 T cell responses specific foreach of the determinants were induced in mice primed with pep-tide-pulsed B6 DCs (Fig. 5,F). The activation of CD81 T cellsspecific for each of the CTL epitopes is CD41 Th cell-dependentbecause peptide-pulsed class II-knockout DCs were not able toprime B6 mice for CTL response (Fig. 5,E). However, coinjectionof b2m-knockout DCs with peptide-pulsed class II-knockout DCsrestored at least partial CTL reactivity to all the tested peptides(Fig. 5,f).

These results demonstrate that, in this system, CD41 Th cellsare generally required for the activation of naive CD81 T cells invivo. Our data indicate that CD41 Th cells need not interact with

a common APC to provide help to CD81 T cells. This is consistentwith the idea that help can be provided by cytokine secretion in thegeneral locale of naive CD81 T cells. The data also suggest thatoptimal help is provided when both CD41 and CD81 T cells in-teract with the same DC, which is consistent with the findings ofBennett et al. (10).

The help essential for CD81 T cell activation can also beprovided by CD81 T cells themselves

We next wanted to determine whether CD81 T cells in high num-bers could substitute for CD41 Th cells. We studied P14 TCR-tgmice bred onto the recombination-activating gene-deficient back-ground, which nearly exclusively produce CD81 T cells express-ing a TCR specific for the LCMV gp33 determinant and have veryfew CD41 T cells. Despite the absence of CD41 T cells, the tgCD81 T cells can be activated by LCMV infection class I tetram-ers (44) or by LCMV gp33 peptide-pulsed DCs, respectively (datanot shown). Consistent with these findings, peptide-pulsed DCsprepared from class II-knockout mice are equally as competent asB6 DCs in activating naive tg CD81 T cells (Fig. 2C).

These findings suggest either that 1) TCR-tg CD81 T cells fromP14 mice differ intrinsically from wild-type B6 mice in their re-quirement for CD41 Th cell-mediated help, or 2) high numbers ofCD81 T cells expressing the Ag-specific TCR in P14 mice gen-erate the help required for their own activation. To distinguishbetween these possibilities, we injected increasing numbers ofsplenocytes from P14 TCR-tg mice into B6 mice, and 1 day later,the recipient mice were either injected with class II-knockout DCspulsed with LCMV gp33 or left untreated. Following immuniza-tion, spleen cells were restimulated in vitro for 5 days, and theCTL response was measured as described above. As shown in Fig.6, peptide-pulsed DCs prepared from class II-knockout mice didnot induce CTL activity in mice without the transfer of P14TCR-tg T cells, consistent with the previous results (Fig. 6,farright). CTL activity was not detected in mice that were injectedwith TCR-tg spleen cells but not primed (E), indicating that acti-vation of CD81 T cells requires immunization with DCs. Ag-spe-cific CTL responses were elicited in B6 mice that had receivedhigher doses of P14 TCR-tg spleen cells, and the CTL inductionwas dependent on the numbers of TCR-tg spleen cells transferred.For example, transfer of 63 104 tg spleen cells (which contained2 3 104 TCR-tg CD81 T cells) resulted in CTL activity. However,no CTL activity was induced in mice that received 33 104 TCR-tgsplenocytes.

To determine the contribution of TCR-tg donor cells vs hostCD81 T cells to the CTL activity, B6 mice were injected i.v. withvarious numbers of P14 TCR-tg spleen cells expressing GFP andthen immunized with peptide-pulsed class II-knockout DCs. Thisenables the simple identification of donor CD81 T cells by flowcytometry using GFP fluorescence. After restimulation, cells werestained with mAbs specific for CD69 and CD8 and analyzed byflow cytometry. Following transfer of 23 106 or 5 3 105 GFP1

splenocytes, there was an expansion of CD81 cells, and the vastmajority (.97%) of activated CD81 T cells were derived from therecipient (CD81CD691GFP2) (Fig. 7, f). Because CD69 is con-sidered an early activation marker and might have been down-regulated, we also performed the identical experiment using thecell surface markers CD251, CD44high, and CD62Llow with sim-ilar results (data not shown). In addition, when we tested thesecells for expression of IFN-g by intracellular cytokine staining, allof the IFN-g-producing cells were CD691. Several studies havedemonstrated that TCR-tg T cells were activated and then deletedin vivo following peptide immunization (20, 21). It is possible that,upon DC priming, the transferred TCR-tg T cells were activated in

FIGURE 4. The inability of class II-deficient DCs to induce CTL in B6mice is partially restored by coinjection of DCs fromb2m-knockout mice.B6 mice were immunized with B6 DCs pulsed with LCMV gp33 (f),LCMV gp33-pulsed class II-deficient DCs (E), or peptide-pulsed classII-deficient DCs coinjected with the same numbers ofb2m-deficient DCs(F). As described in Fig. 2C, splenocytes were isolated 7 days after im-munization and stimulated for 5 days in vitro. Cytotoxic activity was mea-sured as described above. Specific lysis of EL4 cells loaded with variousconcentrations of LCMV gp33 is shown. The experiments in Figs. 2Cand5 were performed at the same time and repeated four times.

1286 BOTH CD4 AND CD8 T CELLS PROVIDE HELP FOR NAIVE CTLp

by guest on August 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Page 6: Multiple Paths for Activation of Naive CD81 T Cells: … · Multiple Paths for Activation of Naive CD81 T Cells: ... the preactivated DCs can efficiently activate naive ... b1(R&D

vivo and subsequently deleted after restimulation. Expansion ofCD81 cells or CTLs was not detected in the recipient mice withoutpriming with DCs, although donor cells were found among a smallnumber of CD81 cells expressing CD69 (Fig. 7,M). This effectwas observed when we transferred 10-fold more T cells than weretypically required for priming. It seems unlikely that these fewdonor T cells are responsible for the cytolysis observed in our CTLassay. This is supported by the observation that the fraction ofdonor cells actually decreases in primed mice following in vitroactivation, suggesting that the P14-tg T cells are actually at a com-petitive disadvantage compared with the host gp33-specific T cellpool. These findings demonstrate that the activation of endogenousCD81 T cells is responsible for the observed CTL activity. It isunclear how the transfer of TCR-tg T cells resulted in the activa-tion of endogenous CD81 cells. Because MHC class II-knockoutDCs were used to prime the recipient mice, the involvement ofCD41 T cells was excluded. Therefore, our data suggest that thetransferred CD81 TCR-tg T cells provided help for the activationof host naive CD81 T cells.

To further determine the impact of CD81 T cell frequency onthe induction of CTL response, we reasoned that using two classI-restricted peptides at the same time would increase the frequencyof cells able to respond to any given DC. Therefore, we immunizedB6 mice with MHC class II-deficient DCs pulsed with two pep-tides, LCMV gp33 and VSV N52–59. Because the two peptides are

restricted by different MHC class I molecules, they should notcompete for class I loading. CD81 T cell precursors specific foreither LCMV gp33 or VSV N52–59 will interact with the sameMHC class II-deficient DCs pulsed with the peptides. B6 micewere immunized with MHC class II-deficient DCs pulsed withLCMV gp33 and VSV N52–59. Seven days after immunization,splenocytes were restimulated in vitro with LCMV gp33, and aCTL assay was performed. A CTL response specific for LCMVgp33 was induced in mice primed with class II-knockout DCspulsed with both LCMV gp33 and VSV N52–59 (Fig. 8B, filledsymbols). In contrast, when mice were immunized with the sameDCs but pulsed with only LCMV gp33 peptide, no CTL activitywas detected (Fig. 8A,E). As a control, B6 DCs pulsed with bothLCMV and VSV peptides induced a LCMV gp33-specific CTLresponse (Fig. 8A,F). Our data suggest that increased numbers ofspecific CD81 T cell precursors can help their own activation intoeffector cells. The identical experiment has been performed usingOVA peptide in place of VSV with similar results.

DiscussionThere is intense interest in using DCs propagated in vivo to elicitCTL responses, particularly in the immunotherapy of cancer (50).In this study, we show that CD41 Th cells are generally requiredfor the generation of CD81 T cell-mediated responses to peptide-pulsed DCs. Given that the defined peptides added to the cells werechosen on the basis of binding MHC class I molecules and havenot known to be recognized by CD41 T cells in all the years thatthey have been studied, it is extremely likely that the requirement

FIGURE 6. CD81 cytotoxic T cells can provide help for their own ac-tivation. The 8-wk-old mice were transferred i.v. with different numbers ofGFP-expressing splenocytes from P14 TCR-tg mice as indicated on thetopof each graph. After 24 h, mice were immunized with 53 105 class II-deficient DCs pulsed with 10mM LCMV gp33 (Fandf, each representingindividual mice) or left untreated (E). Mice that did not receive P14TCR-tg spleen cells but were immunized with peptide-pulsed class II-de-ficient DCs were used as a control (far right panel). Seven days afterimmunization, spleen cells were prepared and stimulated in vitro with thesame peptide for 5 days. CTL activity against EL4 cells pulsed with 10mMLCMV gp33 peptide was measured at different E:T ratios in a standard51Cr-release assay.

FIGURE 7. Endogenous CD81 T cells are activated by MHC class II-deficient DCs in the presence of transferred P14 TCR-tg T cells. B6 miceinjected with different numbers of splenocytes (as indicated at thebottom)from P14 TCR-tg mice expressing GFP were primed 24 h later either withpeptide-pulsed class II-deficient DCs (f) or left untreated (M). Seven dayslater, spleen cells were prepared and stimulated as described in Fig. 6.Following the in vitro stimulation, cells were stained with anti-CD8 (Red-670) and anti-CD69 (PE) mAbs and analyzed by flow cytometry. Thehistograms show the percentage of donor cells (CD81CD691GFP1) of thetotal CD81CD691 cells.

FIGURE 5. The requirement for help is a generalrule for the activation of naive cytotoxic CD81 Tcells. Sendai virus N324–332(left panel), VSV N52–59

(middle panel), and OVA257–264 (right panel) pep-tides were used to pulse DCs for immunization. The6- to 8-wk-old B6 mice were injected with peptide-pulsed B6 DCs, peptide-pulsed class II-deficient DCs,or peptide-pulsed class II-deficient DCs together withthe same numbers ofb2m-deficient DCs. Restimula-tion and CTL assay were performed with the sameprocedure as described before. Cytotoxic activities inmice immunized with B6 DCs (F), with class II-de-ficient DCs (E), or with both class II-deficient DCsandb2m-deficient DCs together (f) are presented inthe figure.

1287The Journal of Immunology

by guest on August 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Page 7: Multiple Paths for Activation of Naive CD81 T Cells: … · Multiple Paths for Activation of Naive CD81 T Cells: ... the preactivated DCs can efficiently activate naive ... b1(R&D

for CD41 T cells reflects the presentation of immunogenic bovineAgs in association with class II molecules.

The requirement for CD41 Th cells in generating CD81 CTLresponses varies considerably among systems (4–8, 15–18). In ourstudy, activation of CD81 T cells was assessed in the absence ofpreactivated APCs or inflammatory signals normally associatedwith viral or bacterial infections, which may substitute for helprequired for activation of CD81 T cells (51, 52). Indeed, we showthat CD81 T cells themselves can provide help for other respond-ing CD81 T cells if present in sufficient numbers. This providesone mechanism for CD41 Th cell-independent CD81 T cell re-sponses. Possibly, other cells (particularly NK cells) can also sub-stitute for CD41 Th cells under the appropriate conditions. In ret-rospect, it is not surprising that the dependence of CD81 T cellresponses on CD41 Th cells varies considerably with the exactconditions of the experimental system, given the permutations foractivating different cell types in different anatomical locations indifferent ways.

Our findings provide insight into how CD41 Th cells (and byinference, other cell types) help the activation of naive CD81 Tcells. The observation thatb2m-knockout DCs are required for thepartial induction of determinant-specific CTL by peptide-pulsedMHC class II-knockout DCs raises three important points. First,because CD41 T cells can interact with APCs distinct from thoseencountering naive CD81 T cells, the help is likely being providedby cytokine secretion. Second, this cytokine-based help is ineffi-cient relative to both CD41 T cells and CD81 cytotoxic T cellsinteracting with the same DC, either because of a higher concen-tration of cytokines at the site of CD41 Th cell activation or themodification of DCs by their interaction with CD41 T cells. Third,activation of CD81 T cells in these experiments is not due to theinteraction between CD81 T cells and DCs activated by CD41 Tcells, because the DCs that can interact with CD41 T cells lackMHC class I molecules (b2m-knockout DCs). These results do notsupport the two-cell model, which suggests that DCs first activated

by CD41 T cells in turn interact with CD81 T cells, leading to theactivation of CD81 T cells.

Importantly, we found that CD81 T cells can also provide helpwhen tg CD81 T cells were adoptively transferred into mice re-ceiving peptide-pulsed MHC class II-knockout DCs. In support ofthis, a CTL response specific to LCMV gp33 was also inducedwhen mice were immunized with class II-deficient DCs pulsedwith both LCMV gp33 and VSV N52–59. But class II-knockoutDCs pulsed with only LCMV gp33 could not prime CTL in B6mice. Two major possibilities exist for how CD81 T cells mediatehelp under these conditions. First, the simultaneous interaction andactivation of a critical number of CD81 T cells establishes a mi-croenvironment in which cytokine concentration exceeds thethreshold necessary for activating naive CD81 T cells. This couldaccount for the priming of LCMV gp33-specific T cells in P14TCR-tg recombination-activating gene-knockout mice by peptide-pulsed DCs in the absence of CD41 T cells (our unpublished data).This interpretation is supported by our previous data demonstrat-ing that MHC class I tetramers alone were sufficient to activatenaive CD81 T cells from P14 TCR-tg mice (44), accompanied byproduction of large amounts of both IL-2 and IFN-g. Our intra-cellular staining demonstrates that CD81 T cells made more IFN-g(data not shown).

Second, the interaction of a large number of CD81 T cells withDCs may activate the DCs and lead to up-regulation of costimu-latory molecules or secretion of cytokines by DCs. There is in-creasing evidence for this type of mechanism. It has been reportedthat both influenza virus and LCMV induce maturation and acti-vation of DCs, which can then directly activate CD81 T cells (51,53). Recent data from three groups indicates that DCs activated byanti-CD40 mAb either in vitro or in vivo can induce a CTL re-sponse in MHC class II-deficient mice, and MHC class II-deficientDCs activated by anti-CD40 mAb prime CTL responses in theabsence of CD41 T cells (11–13). Similarly, mice treated withanti-CD40 mAb can generate CD81 T cell-mediated CTL re-sponses against tumors in the absence of CD41 T cells (54).

One explanation that could explain part of our data is that thelack of CD4 help has no effect in vivo, but, simply, the lack of CD4priming only impacts the expansion of the CD81 T cells in vitro.Although this explanation could explain some of the results, itcannot explain the fact that increasing the number of CD81 re-sponding cells alone, either by transfer of CD81 TCR-tg cells orby mixing two different peptides, now allows CD8 priming. Underthese conditions, the number of primed CD4 cells would be equalin all cultures (i.e., low), because the priming occurred with classII knockout DCs.

In conclusion, we have demonstrated that the induction ofCD81 CTL responses requires help that can be obtained from mul-tiple sources, including CD41 T cells and CD81 T cells. Thesefindings provide a basis for understanding the variable requirementfor CD41 T cells in generating CD81 T cells responses that hasbeen recorded in the literature, because experimental systems areexpected to vary in the extent to which non-CD4 cells could pro-vide help. The important practical implication is that the inductionof help should be taken into account in the design of vaccinesmeant to induce CD81 CTL responses (55, 56).

AcknowledgmentsWe thank Drs. Roland Tisch, Robert Maile, and Edward Collins for helpfuldiscussions and Katherine Midkiff for technical assistance.

References1. Schmitz, J. E., M. J. Kuroda, S. Santra, V. G. Sasseville, M. A. Simon,

M. A. Lifton, P. Racz, K. Tenner-Racz, M. Dalesandro, B. J. Scallon, et al. 1999.

FIGURE 8. LCMV gp33-specific CTL response was induced in B6mice primed with MHC class II-deficient DCs pulsed with both LCMVgp33 and VSV N52 peptides.A, The 6- to 8-wk-old B6 mice were injectedi.v. with either 53 105 LCMV gp33-pulsed class II-deficient DCs (E) orB6 DCs pulsed with both LCMV gp33 and VSV N52 peptides (F). Asdescribed previously, spleen cells were restimulated with LCMV gp33 pep-tide in vitro for 5 days, and CTL activity specific for LCMV gp33 wasmeasured.B, B6 mice were immunized with 53 105 class II-deficient DCspulsed with both LCMV gp33 and VSV N52 peptides. Similar to Fig. 8A,7 days after immunization, spleen cells were restimulated with LCMVgp33 in vitro for 5 days, and CTL response specific to LCMV was exam-ined (filled symbols). Open symbols represent CTL activity against EL4cells without peptide.

1288 BOTH CD4 AND CD8 T CELLS PROVIDE HELP FOR NAIVE CTLp

by guest on August 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Page 8: Multiple Paths for Activation of Naive CD81 T Cells: … · Multiple Paths for Activation of Naive CD81 T Cells: ... the preactivated DCs can efficiently activate naive ... b1(R&D

Control of viremia in simian immunodeficiency virus infection by CD81 lym-phocytes.Science 283:857.

2. Doherty, P. C., W. Allan, M. Eichelberger, and S. R. Carding. 1992. Roles ofabandgd T cell subsets in viral immunity.Annu. Rev. Immunol. 10:123.

3. Cardin, R. D., J. W. Brooks, S. R. Sarawar, and P. C. Doherty. 1996. Progressiveloss of CD81 T cell-mediated control of a gamma-herpesvirus in the absence ofCD41 T cells.J. Exp. Med. 184:863.

4. Flano, E., D. L. Woodland, and M. A. Blackman. 1999. Requirement for CD41

T cells in Vb41CD81 T cell activation associated with latent murine gamma-herpesvirus infection.J. Immunol. 163:3403.

5. von Herrath, M. G., M. Yokoyama, J. Dockter, M. B. Oldstone, and J. L. Whitton.1996. CD4-deficient mice have reduced levels of memory cytotoxic T lympho-cytes after immunization and show diminished resistance to subsequent viruschallenge.J. Virol. 70:1072.

6. Stohlman, S. A., C. C. Bergmann, M. T. Lin, D. J. Cua, and D. R. Hinton. 1998.CTL effector function within the central nervous system requires CD41 T cells.J. Immunol. 160:2896.

7. Zajac, A. J., J. N. Blattman, K. Murali-Krishna, D. J. Sourdive, M. Suresh,J. D. Altman, and R. Ahmed. 1998. Viral immune evasion due to persistence ofactivated T cells without effector function.J. Exp. Med. 188:2205.

8. Wild, J., M. J. Grusby, R. Schirmbeck, and J. Reimann. 1999. Priming MHC-I-restricted cytotoxic T lymphocyte responses to exogenous hepatitis B surfaceantigen is CD41 T cell dependent.J. Immunol. 163:1880.

9. Mitchison, N. A., and C. O’Malley. 1987. Three-cell-type clusters of T cells withantigen-presenting cells best explain the epitope linkage and noncognate require-ments of the in vivo cytolytic response.Eur. J. Immunol. 17:1579.

10. Bennett, S. R., F. R. Carbone, F. Karamalis, J. F. Miller, and W. R. Heath. 1997.Induction of a CD81 cytotoxic T lymphocyte response by cross-priming requirescognate CD41 T cell help.J. Exp. Med. 186:65.

11. Ridge, J. P., F. Di Rosa, and P. Matzinger. 1998. A conditioned dendritic cell canbe a temporal bridge between a CD41 T-helper and a T-killer cell.Nature 393:474.

12. Bennett, S. R., F. R. Carbone, F. Karamalis, R. A. Flavell, J. F. Miller, andW. R. Heath. 1998. Help for cytotoxic-T-cell responses is mediated by CD40signalling.Nature 393:478.

13. Schoenberger, S. P., R. E. Toes, E. I. van der Voort, R. Offringa, and C. J. Melief.1998. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40Linteractions.Nature 393:480.

14. Johnson, A. J., M. K. Njenga, M. J. Hansen, S. T. Kuhns, L. Chen, M. Rodriguez,and L. R. Pease. 1999. Prevalent class I-restricted T-cell response to the Theiler’svirus epitope Db:VP2121–130 in the absence of endogenous CD4 help, tumornecrosis factora, g interferon, perforin, or costimulation through CD28.J. Virol.73:3702.

15. Buller, R. M., K. L. Holmes, A. Hugin, T. N. Frederickson, and H. C. Morse, III.1987. Induction of cytotoxic T-cell responses in vivo in the absence of CD4helper cells.Nature 328:77.

16. Tripp, R. A., S. R. Sarawar, and P. C. Doherty. 1995. Characteristics of theinfluenza virus-specific CD81 T cell response in mice homozygous for disruptionof the H-2lAb gene.J. Immunol. 155:2955.

17. Bodmer, H., G. Obert, S. Chan, C. Benoist, and D. Mathis. 1993. Environmentalmodulation of the autonomy of cytotoxic T lymphocytes.Eur. J. Immunol. 23:1649.

18. Stevenson, P. G., G. T. Belz, J. D. Altman, and P. C. Doherty. 1998. Virus-specific CD81 T cell numbers are maintained during gamma- herpesvirus reac-tivation in CD4-deficient mice.Proc. Natl. Acad. Sci. USA 95:15565.

19. Chen, Y., R. G. Webster, and D. L. Woodland. 1998. Induction of CD81 T cellresponses to dominant and subdominant epitopes and protective immunity toSendai virus infection by DNA vaccination.J. Immunol. 160:2425.

20. Garza, K. M., S. M. Chan, R. Suri, L. T. Nguyen, B. Odermatt,S. P. Schoenberger, and P. S. Ohashi. 2000. Role of antigen-presenting cells inmediating tolerance and autoimmunity.J. Exp. Med. 191:2021.

21. Kyburz, D., P. Aichele, D. E. Speiser, H. Hengartner, R. M. Zinkernagel, andH. Pircher. 1993. T cell immunity after a viral infection versus T cell toleranceinduced by soluble viral peptides.Eur. J. Immunol. 23:1956.

22. Banchereau, J., F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu,B. Pulendran, and K. Palucka. 2000. Immunobiology of dendritic cells.Annu.Rev. Immunol. 18:767.

23. Inaba, K., J. W. Young, and R. M. Steinman. 1987. Direct activation of CD81

cytotoxic T lymphocytes by dendritic cells.J. Exp. Med. 166:182.24. Kast, W. M., C. J. Boog, B. O. Roep, A. C. Voordouw, and C. J. Melief. 1988.

Failure or success in the restoration of virus-specific cytotoxic T lymphocyteresponse defects by dendritic cells.J. Immunol. 140:3186.

25. Condon, C., S. C. Watkins, C. M. Celluzzi, K. Thompson, and L. D. Falo, Jr.1996. DNA-based immunization by in vivo transfection of dendritic cells.Nat.Med. 2:1122.

26. Steinman, R. M. 1991. The dendritic cell system and its role in immunogenicity.Annu. Rev. Immunol. 9:271.

27. Banchereau, J., and R. M. Steinman. 1998. Dendritic cells and the control ofimmunity. Nature 392:245.

28. Yewdell, J. W., C. C. Norbury, and J. R. Bennink. 1999. Mechanisms of exog-enous antigen presentation by MHC class I molecules in vitro and in vivo: im-plications for generating CD81 T cell responses to infectious agents, tumors,transplants, and vaccines.Adv. Immunol. 73:1.

29. Albert, M. L., B. Sauter, and N. Bhardwaj. 1998. Dendritic cells acquire antigenfrom apoptotic cells and induce class I-restricted CTLs.Nature 392:86.

30. Sigal, L. J., S. Crotty, R. Andino, and K. L. Rock. 1999. Cytotoxic T-cell im-munity to virus-infected non-haematopoietic cells requires presentation of exog-enous antigen.Nature 398:77.

31. Moore, M. W., F. R. Carbone, and M. J. Bevan. 1988. Introduction of solubleprotein into the class I pathway of antigen processing and presentation.Cell54:777.

32. Norbury, C. C., B. J. Chambers, A. R. Prescott, H. G. Ljunggren, and C. Watts.1997. Constitutive macropinocytosis allows TAP-dependent major histocompat-ibility complex class I presentation of exogenous soluble antigen by bone mar-row-derived dendritic cells.Eur. J. Immunol. 27:280.

33. Reis e Sousa, C., A. Sher, and P. Kaye. 1999. The role of dendritic cells in theinduction and regulation of immunity to microbial infection.Curr. Opin. Immu-nol. 11:392.

34. Nair, S., J. S. Babu, R. G. Dunham, P. Kanda, R. L. Burke, and B. T. Rouse.1993. Induction of primary, antiviral cytotoxic, and proliferative responses withantigens administered via dendritic cells.J. Virol. 67:4062.

35. Mayordomo, J. I., T. Zorina, W. J. Storkus, L. Zitvogel, C. Celluzzi, L. D. Falo,C. J. Melief, S. T. Ildstad, W. M. Kast, A. B. Deleo, et al. 1995. Bone marrow-derived dendritic cells pulsed with synthetic tumour peptides elicit protective andtherapeutic antitumour immunity.Nat. Med. 1:1297.

36. Paglia, P., C. Chiodoni, M. Rodolfo, and M. P. Colombo. 1996. Murine dendriticcells loaded in vitro with soluble protein prime cytotoxic T lymphocytes againsttumor antigen in vivo.J. Exp. Med. 183:317.

37. Nestle, F. O., S. Alijagic, M. Gilliet, Y. Sun, S. Grabbe, R. Dummer, G. Burg, andD. Schadendorf. 1998. Vaccination of melanoma patients with peptide- or tumorlysate-pulsed dendritic cells.Nat. Med. 4:328.

38. Grusby, M. J., R. S. Johnson, V. E. Papaioannou, and L. H. Glimcher. 1991.Depletion of CD41 T cells in major histocompatibility complex class II-deficientmice.Science 253:1417.

39. Gosgrove, D., D. Gray, A. Dierich, J. Kaufman, M. Lemeur, C. Benoist, andD. Mathis. 1991. Mice lacking MHC class II molecules.Cell 66:1051.

40. Zijlstra, M., M. Bix, N. E. Simister, J. M. Loring, D. H. Raulet, and R. Jaenisch.1990. b2-microglobulin deficient mice lack CD4281 cytolytic T cells. Nature344:742.

41. Koller, B. H., P. Marrack, J. W. Kappler, and O. Smithies. 1990. Normal devel-opment of mice deficient inb2M, MHC class I proteins, and CD81 T cells.Science 248:1227.

42. Pircher, H., K. Burki, R. Lang, H. Hengartner, and R. M. Zinkernagel. 1989.Tolerance induction in double specific T-cell receptor transgenic mice varies withantigen.Nature 342:559.

43. Wu, Y. P., E. McMahon, M. R. Kraine, R. Tisch, A. Meyers, J. Frelinger,G. K. Matsushima, and K. Suzuki. 2000. Distribution and characterization ofGFP1 donor hematogenous cells in twitcher mice after bone marrow transplan-tation.Am. J. Pathol. 156:1849.

44. Wang, B., R. Maile, R. Greenwood, E. J. Collins, and J. A. Frelinger. 2000. NaiveCD81 T cells do not require costimulation for proliferation and differentiationinto cytotoxic effector cells.J. Immunol. 164:1216.

45. Quinn, D. G., A. J. Zajac, C. E. Hioe, and J. A. Frelinger. 1997. Virus-specific,CD81 major histocompatibility complex class I-restricted cytotoxic T lympho-cytes in lymphocytic choriomeningitis virus-infectedb2-microglobulin-deficientmice.J. Virol. 71:8392.

46. Lane, P. J., and T. Brocker. 1999. Developmental regulation of dendritic cellfunction.Curr. Opin. Immunol. 11:308.

47. Young, J. W., and R. M. Steinman. 1990. Dendritic cells stimulate primary hu-man cytolytic lymphocyte responses in the absence of CD41 helper T cells.J. Exp. Med. 171:1315.

48. McCoy, K. D., I. F. Hermans, J. H. Fraser, G. Le Gros, and F. Ronchese. 1999.Cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) can regulate dendriticcell-induced activation and cytotoxicity of CD81 T cells independently of CD41

T cell help.J. Exp. Med. 189:1157.49. Mehta-Damani, A., S. Markowicz, and E. G. Engleman. 1994. Generation of

antigen-specific CD81 CTLs from naive precursors.J. Immunol. 153:996.50. Fong, L., and E. G. Engleman. 2000. Dendritic cells in cancer immunotherapy.

Annu. Rev. Immunol. 18:245.51. Cella, M., M. Salio, Y. Sakakibara, H. Langen, I. Julkunen, and A. Lanzavecchia.

1999. Maturation, activation, and protection of dendritic cells induced by double-stranded RNA.J. Exp. Med. 189:821.

52. Rescigno, M., F. Granucci, S. Citterio, M. Foti, and P. Ricciardi-Castagnoli.1999. Coordinated events during bacteria-induced DC maturation.Immunol. To-day 20:200.

53. Ruedl, C., M. Kopf, and M. F. Bachmann. 1999. CD81 T cells mediate CD40-independent maturation of dendritic cells in vivo.J. Exp. Med. 189:1875.

54. French, R. R., H. T. Chan, A. L. Tutt, and M. J. Glennie. 1999. CD40 antibodyevokes a cytotoxic T-cell response that eradicates lymphoma and bypasses T-cellhelp.Nat. Med. 5:548.

55. Zajac, A. J., K. Murali-Krishna, J. N. Blattman, and R. Ahmed. 1998. Therapeuticvaccination against chronic viral infection: the importance of cooperation be-tween CD41 and CD81 T cells.Curr. Opin. Immunol. 10:444.

56. Maecker, H. T., D. T. Umetsu, R. H. DeKruyff, and S. Levy. 1998. Cytotoxic Tcell responses to DNA vaccination: dependence on antigen presentation via classII MHC. J. Immunol. 161:6532.

1289The Journal of Immunology

by guest on August 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from