ifn-α-conditioned dendritic cells are highly efficient in inducing cross-priming cd8+ t cells...

IFN-a-conditioned dendritic cells are highlyefficient in inducing cross-priming CD8+ T cells againstexogenous viral antigens

Caterina Lapenta1, Stefano M. Santini1, Massimo Spada1, Simona Donati1,Francesca Urbani1, Daniele Accapezzato2, Debora Franceschini2,Mauro Andreotti1, Vincenzo Barnaba2 and Filippo Belardelli1

1 Department of Cell Biology and Neurosciences, Istituto Superiore di Sanit�, Rome,Italy

2 Department of Internal Medicine, University of Rome 'La Sapienza' Rome, Italy

Dendritic cells (DC) generated after a short-term exposure of monocytes to IFN-a andGM-CSF (IFN-DC) are highly effective in inducing cross-priming of CD8+ Tcells againstviral antigens. We have investigated the mechanisms responsible for the special attitudeof these DC and compared their activity with that of reference DC. Antigen uptake andendosomal processing capabilities were similar for IFN-DC and IL-4-derived DC. BothDC types efficiently cross-presented soluble HCV NS3 protein to the specific CD8+ Tcellclone, even though IFN-DC were superior in cross-presenting low amounts of viralantigens. Moreover, when DC were pulsed with inactivated HIV-1 and injected into hu-PBL-SCID mice, the generation of virus-specific CD8+ T cells was markedly higher inanimals immunized with IFN-DC than in mice immunized with CD40L-matured IL-4-DC. Of interest, in experiments with purified CD8+ T cells, IFN-DC were superior withrespect to CD40L-matured IL-4-DC in inducing in vitro cross-priming of HIV-specificCD8+ T cells. This property correlated with enhanced potential to express the specificsubunits of the IL-23 and IL-27 cytokines. These results suggest that IFN-DC are directlylicensed for an efficient CD8+ T cell priming by mechanisms likely involving enhancedantigen presentation and special attitude to produce IL-12 family cytokines.

Introduction

The priming and expansion of antigen-specific CD8+

T cell response is a complex process involving concertedinteractions between lymphocytes and dendritic cells(DC), the professional antigen-presenting cells (APC)playing a pivotal role in linking innate and adaptiveimmunity [1, 2]. The priming of antigen-specific CD8+

Tcells requires recognition through the Tcell receptor ofpeptide-MHC class I complexes on the surface ofappropriate APC. This event occurs when viral proteinsare synthesized within an infected cell, where cyto-plasmic proteasomes and peptidases degrade them intopeptides, which are then translocated into the endo-plasmic reticulum for the access to newly formed MHCclass I molecules and transport to the cell surface.However, suitable peptides may also be derived fromexogenous antigens intersecting this pathway afterendocytosis by APC, a process named as cross-presenta-tion. Notably, DC are considered as the most efficientcells at cross-presenting exogenous antigens, but theymust undergo a special activation process or “licensing”step in order to cross-prime CD8+ T cells. Underpathological conditions, DC are licensed by engagement

Correspondence: Dr. Filippo Belardelli, Section of Experi-mental Immunotherapy, Department of Cell Biology andNeurosciences, Istituto Superiore di Sanit�, Viale Regina Elena,299, 00161 Rome, ItalyFax: +39-06-4990-2097e-mail: [email protected]

Received 7/10/05Revised 20/2/06

Accepted 31/5/06

[DOI 10.1002/eji.200535579]

Key words:CD8+ T lymphocytes

� Cross-priming� Dendritic cells � HIV

� IFN-a

Abbreviation: AT-2: aldrithiol-2

Caterina Lapenta et al. Eur. J. Immunol. 2006. 36: 2046–20602046

f 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

of surface CD40 by activated CD4+ helper T cells or bymicrobe-derived macromolecules, which can trigger DCmaturation and up-regulate the expression of surface co-stimulatory molecules.

It is generally assumed that only mature DC canefficiently induce cross-priming of CD8+ T cells againstexogenous antigens [3, 4]. In considering the eventsleading to the generation of mature DC frommonocytes,the vision is generally influenced by the widely usedtwo-step culture protocol: (i) immature DC are gener-ated as a result of several days of culture in the presenceof GM-CSF/IL-4 and (ii) a second culture step in thepresence of maturation factors is required to obtainmature DC [3, 5]. We previously demonstrated thathighly active partially mature DC are generated frommonocytes after a single step of 3-day culture withIFN-a/GM-CSF (IFN-DC) [6]. IFN-DC proved to be moreeffective than immature DC generated in the presence ofGM-CSF and IL-4 in inducing a Th-1 type of immuneresponse and CD8+ T cell responses against definedantigens in different models [7–13]. In particular,our results obtained by using IFN-DC pulsed withaldrithiol-2 (AT-2)-inactivated HIV-1 in the hu-PBL-SCID mouse model revealed that IFN-DC were farsuperior with respect to immature IL-4-DC in inducingan in vivo cross-priming of CD8+ T cells againstexogenous HIV antigens [7]. However, the mechanismsunderlying this special attitude of IFN-DC remained tobe determined. In this regard, different and distinctmechanisms could be invoked, including an improvedantigen uptake and processing, as well as an enhancedantigen presentation, possibly due to higher levels ofcostimulatory molecules or, eventually, to the partiallymature phenotype of these cells. In addition, IFN-DChad never been compared with classical mature IL-4-DCand the importance of CD4+ T cells in the IFN-DC-mediated cross-priming of CD8+ T cells remains to beevaluated.

To address all these important issues, we have firstevaluated antigen uptake and the capacity of IFN-DC tocross-present viral antigens and to present viral peptidesto specific CD8+ T cell clones, as compared toconventional IL-4-DC. In the present study, we havealso compared the capability of IFN-DC to induce aCD8+ T cell response against exogenous HIV antigenswith respect to that exhibited by mature DC obtainedafter 6 days of monocyte culture in the presence of GM-CSF/IL-4 and subsequent exposure to sCD40L, which isthought to represent a crucial step for an efficientinduction of cytotoxic CD8+ T cells [14–18].

We report here that low amounts of viral antigens aremore efficiently cross-presented to CD8+ T cells byIFN-DC with respect to IL-4-DC, even though antigenuptake and antigen processing capabilities appear to becomparable. We also report that IFN-DC are more

efficient than CD40L-matured IL-4-DC (mIL-4-DC) ininducing a CD8+ T cell response in hu-PBL-SCID mice.Of interest, IFN-DC were much more efficient than mIL-4-DC in inducing cross-priming of CD8+ T cells againstHIV antigens. Of note, upon CD40-CD40L interaction,IFN-DC up-regulate IL-23 and IL-27 subunit transcriptsto a higher extent than IL-4-DC. Overall, these resultsindicate that IFN-a-conditioned DC, whose differentia-tion/activation pathway may resemble that of DCrapidly generated after in vivo exposure of monocytesto infection-induced cytokines, are directly licensed forefficient CD4-independent CD8+ T cell priming.

Results

Comparison between IFN-DC and IL-4-DC forcapabilities of antigen uptake and endosomalprocessing

First, we performed a set of experiments aimed atevaluating whether the higher capability of CD8+ T cellcross-priming by the IFN-DC with respect to the IL-4-DC[7] could be associated with an enhanced attitude ofantigen uptake and endosomal processing. Antigenuptake by DC is mediated predominantly by eithermannose receptor-mediated endocytosis or macropino-cytosis, which are modulated during DC differentiation.We have evaluated mannose receptor-mediated endo-cytosis by measuring the uptake of FITC-conjugateddextran polysaccharide, while macropinocytosis andendosomal processing capacity has been evaluated bythe uptake of DQ ovalbumin, which is a self-quenchedconjugate of albumin exhibiting bright green fluores-cence upon endo-lysosomal protease-dependent degra-dation, thus permitting the evaluation of both antigenuptake and processing by live DC. Fig. 1A and B show thephenotype of the two types of DC used in theseexperiments. Consistently with previously publishedresults [6], IFN-DC were characterized by a higherpercentage of cells expressing CD40, CD80, CD86(Fig. 1A). The up-regulation of membrane expressionof these markers (Fig. 1B) was also associated with theappearance of the DC maturation marker CD83+.Notably, IFN-DC exhibited nearly twofold increase ofHLA class I molecule expression intensity as compared toIL-4 DC (Fig. 1B and 4A). As illustrated in Fig. 1C, nomajor difference in the dextran uptake capacity wasdetected between the two DC types (IFN-DC and IL-4-DC) (Fig. 1C). Likewise, both DC types exhibited similarFACS profile after incubation with DQ ovalbumin(Fig. 1D), suggesting that the majority of cells retainedcomparable phagocytic and processing activity. Inparticular, time-course analyses of antigen uptake andprocessing revealed similar kinetics for both DC types

Eur. J. Immunol. 2006. 36: 2046–2060 Cellular immune response 2047


(data not shown). Thus, the finding that both DC typesexhibited a similar capability of antigen uptake andprocessing suggested that other mechanisms wereresponsible for the special attitude of IFN-DC to inducecross-priming of CD8+ T cells against exogenous viralantigens.

Cross-presentation of exogenous solubleantigens to CD8+ cells by monocyte-derived DC

It could be argued that IFN-DC were endowed with anenhanced capability of cross-presenting viral antigens toCD8+ Tcells with respect to IL-4-DC. This hypothesis has

been addressed by experiments aimed at evaluating theefficiency of both DC types loaded with a reference viralsoluble protein to activate a CD8+ T cell clone specificfor the viral antigen. In particular, in a series of threeexperiments with DC from different donors, we havestudied the presentation of exogenous HCV NS3 proteinto a HLA-A2-restricted NS3(1406–1415)-specific CD8+

T cell clone [19]. The response was evaluated byintracellular staining of IFN-c-producing cells followedby flow cytometry. First, by using cells from the samedonors, we performed a series of cross-presentationassays using the same CD8+ Tcell clone and the same DCloaded with the whole recombinant NS3 protein. As

Fig. 1. Phenotype, antigen uptake and antigen processing capacity by IFN-DC and IL-4-DC. (A) Percentage of DC expressing a seriesof selectedmembranemarkers as detected by FACS analysis. (B) Fluorescence intensity of selectedmembranemarkers as detectedby FACS analysis. Bars represent the percentage or the mean fluorescence intensity of cells expressing the selected membranemarker and the standard error. (C and D) Antigen uptake and processing by the IFN-DC and IL-4 DC. Cells were incubated for60 min at 37�C with 50 lg/mL of dextran-FITC conjugate (C) or 100 lg/mL of DQ-ovalbumin (D). After 60 min, cells were washedand analyzed by flow cytometry. DQ ovalbumin is a self-quenched conjugate of albumin that exhibits bright green fluorescenceonly upon proteolytic degradation.



Fig. 2. Presentation assay of theNS3–1406 peptide and cross-presentation of thewholeNS3 protein to the specific CD8+ T cell clone(cloneNS3–1). Cells (3� 104/test) of the CD8+ T cell cloneNS3–1 specific for anHLA-A2 binding peptideHCV1406were incubated, ata stimulator/responder cell ratio of 1:1.5 in a microculture plate, with IFN-DC or IL-4-DC previously loaded with (A) NS3recombinant protein (50 lg, 10 lg, 1 lg or 0) or peptideHCV1406 (100 ng, 10 ng, 1 ng, 0.1 ng, 0.01 ng 0.001 ng/mLor 0) (C). After 18-hincubation at 37�C, cells were assayed for IFN-c production by intracellular immunofluorescence staining followed by flowcytometry (see section "Materials and methods" for details). Each bar represents the mean (� SE) of values of three experiments. (B)Representative dot plot analysis of IFN-c expression by the CD8 clone NS3–1 stimulated with DC loaded with the NS3 protein. (D)Representative dot plot analysis of IFN-c expression by the CD8 cloneNS3–1 stimulatedwithDC loadedwith theNS3–1406 peptide.



shown in Fig. 2A, IFN-DC showed a cross-presentationcapability comparable to that of the IL-4-DC whenloaded with protein concentrations of 50 and 10 lg/mL.However, the IFN-DC proved to be superior in cross-presenting antigen at lower protein concentration(Fig. 2A and B). Consistently, when the DC were loadedwith high concentrations of the corresponding NS3peptide, the activation of the CD8+ Tcell clone by eitherIFN-DC or IL-4-DC proved to be similar (Fig. 2C).

However, at very low peptide concentration (0.01 or0.001 ng/mL), clone activation by IFN-DC was signifi-cantly more efficient (Fig. 2C). This was also supportedby the dot-plot analysis of IFN-c production by thespecific T cell clone when stimulated with DC loadedwith 0.001 ng/mL of the NS3 peptide, which clearlyshowed that IFN-DC were associated with highernumber of IFN-c-producing cells and a strongerflorescence intensity (Fig. 2D).

Fig. 3. Comparative characterization of the expression of HIV-1 receptors and of the DC susceptibility to HIV infection. Membraneexpression of molecules involved in HIVentry and infection (A) Three days after HIV-1 infection, proviral loadwas analyzed in DCby PCR for viral gag sequences (B). The sensitivity of the assay was tested by amplifying serial dilutions of DNA prepared from 8E5cells, which harbor one proviral copy/cell (B). Viral release from infected DCwas assessed bymeasuring the levels of the HIV-1 p24protein in culture supernatants (C), as described in Materials and methods.



Comparison of 3-day IFN-DC versus CD40L-activated IL-4-DC for their capability to inducehumoral response and cross-priming in hu-PBL-SCID mice

In a previous study based on the use of DC pulsed withinactivated HIV-1 as antigen model, we had shown thatvirus-pulsed IFN-DC were superior with respect toimmature IL-4-DC in inducing a potentially protectivehumoral and cellular immune response against HIVantigens when tested in hu-PBL-SCID mice [7]. How-ever, it remained to be evaluated whether IFN-DC couldcompare favorably with reference mature DC (mIL-4-DC), as those obtained after in vitro maturation of IL-4-DC by exposure to CD40L. Before addressing this issue, itwas also important to evaluate whether the IFN-DC andIL-4-DC could exhibit any differential property ininteracting with HIV-1. In our previous study [7], thevirus inactivation was achieved by using aldrithiol-2(AT-2), which selectively disrupts the p7 nucleocapsid(NC) protein, thus resulting in inactivation withoutaffecting the conformation and fusogenic activity of thegp120. We have now analyzed the two DC types by flowcytometry for the expression of selected membranemolecules involved in viral entry. The phenotypicanalysis showed lower levels of expression of membraneCD4, CXCR4, CCR5 and DC-SIGN in IFN-DC ascompared to IL-4-DC (Fig. 3A), consistently with resultsfrom other groups [12, 13]. Similar proviral load wasdetected in both IFN-DC and IL-4-DC previously exposedto HIV (Fig. 3B). IL-4-DC proved to be capable ofreleasing higher amounts of HIV with respect to the IFN-DC. Overall, these results suggested that the superiorcapability of the HIV-pulsed IFN-DC to induce a humanhumoral and cellular immune response in hu-PBL-SCIDmice was not due to an enhanced susceptibility of theseDC to virus entry and infection.

Fig. 4 illustrates the phenotype (Fig. 4A) andcytokine secretion patterns (Fig. 4B), before and afterCD40L stimulation, of the DC types utilized in thesubsequent studies. As expected, only a small fraction ofthe IFN-DC expressed the CD83 maturation marker,while the large majority of both mIL-4-DC and mIFN-DCwere CD83+ (Fig. 4A). Both mIFN-DC and mIL-4-DCexpressed comparable levels of the costimulatorymolecules CD80 and CD86, higher than the correspond-ing immature DC. As illustrated in Fig. 4B, IFN-DCsecreted higher amounts of TNF-a, PGE2 and IL-6 thanIL-4-DC. Interestingly, after CD40L-inducedmaturation,the levels of the secreted IL-12 and TNF-a were higherfor IFN-DC than for IL-4-DC. In contrast, no or very lowlevels of secretion of IL-15, IL-18, IL-1b, IL-7, TGF-b1and IL-2 were detected in the different DC cultures (datanot shown).

The immune priming activity of IFN-DC and mIL-4-DC pulsed with AT-2-HIV-1 was tested in hu-PBL-SCIDmice, by measuring their in vivo capability to induce thegeneration of human antibodies and, more importantly,of CD8+ T cells against HIV-1 antigens. Fig. 5 shows theantibody response to HIV-1 gp41 immunodominantpeptides obtained in hu-PBL-SCIDmice immunized witheither IFN-DC or mIL-4-DC loadedwith AT-2-inactivatedHIV-1. At 1 week after primary and boost immunization,comparable levels of anti-HIV antibodies were detectedin mouse sera, indicating that both DC types exhibitedsimilar efficacy in the elicitation of a human antibodyresponse. When CD40L-treated IFN-DC (mIFN-DC)were compared with IFN-DC, no major difference inthe antibody production was observed (data notshown), suggesting that the subsequent maturationstep did not result in any significant enhancement of theDC functional activity. Interestingly, however, IFN-DCwere more efficient than mIL-4-DC in inducing thegeneration of HIV-1-specific CD8+ T cells in theimmunized hu-PBL-SCID mice, as revealed by IFN-cELISPOT assay (Fig. 6, Exp.1). Notably, treatment ofIFN-DC with sCD40L did not significantly enhance thegeneration of HIV-specific CD8+ T cells (Fig. 6, Exp. 2),suggesting that IFN-DC are fully committed to theefficient cross-priming of CD8+ T cells without therequirement of additional maturation steps provided byCD4+ T cells.

Efficient CD4+ T cell-independent generation ofeffector CD8+ T cells against HIV-antigens by IFN-DC in vitro

The in vivo studies illustrated above suggested that IFN-DC are especially effective in inducing the cross-primingof virus specific CD8+ T cells in vivo. Thus, we haveperformed in vitro experiments to characterize thecapability of IFN-DC of inducing antigen-specificeffector CD8+ T cells against exogenous HIV antigensin the presence or absence of CD4+ T cell help. Inparticular, we compared the in vitro cross-priming ofhighly purified CD8+ Tcells using the two types of AT-2-HIV-1-pulsed DC: IFN-DC and mIL-4-DC. Positivelyselected CD8+ T cells represented >97% of the cellpopulation as assessed by flow cytometry (Fig. 7A). IFN-DC were far superior in the induction of specific CD8+

T cell response in absence of CD4+ T cell help asevaluated by both ELISPOT enumeration of IFN-c andgranzyme-B-releasing cells after restimulationwith HIV-1 antigens (Fig. 7B and C). Comparable numbers of IFN-c-producing T cells were detected when the total PBL,instead of purified CD8+ T cells, were co-cultured witheither IFN-DC or mIL-4-DC (Fig. 7B). The generation ofgranzyme-B-releasing cells was more efficiently inducedby IFN-DC both in the presence and in absence of CD4+



Fig. 4. Phenotype and cytokine production by the differentimmature and mature DC types. (A) Representative dothistogram FACSJ profiles of four types of DC used in the invivo experiments in the hu-PBL-SCID mouse model. (B) PGE2and cytokine levels in the culture supernatants of IFN-DC,IL-4-DC, mIFN-DC and mIL4-DC after their culture for 24 h infresh medium. Each bar represents the mean concentrationvalues (� SE) of three experiments.



T cells (Fig. 7C). Fig. 7D illustrates the production ofIL-6, IL-10, IL-12, TNF-a and PGE2 in supernatants fromthe last DC restimulation. Of interest, high levels of IL-12were detected in supernatants from co-cultures ofpurified CD8+ T cells with antigen-pulsed-IFN-DC,suggesting that IFN-DC had acquired the full capacityto release this cytokine during co-culture. However, thedifferential capability of the two DC populations toinduce a CD8+-specific T cell response (Fig. 7B and C)did not correlate with major differences in the pattern ofcytokine production (Fig. 7D).

IFN-DC exhibit a high capability to express theIL-12 family cytokines IL-23 and IL-27 uponsCD40L-induced maturation

The results reported above showed that IFN-DC werecapable of efficiently generating an effective CD8+ Tcellresponse, including the production of high levels ofIFN-c. It was reasonable to suppose that this specialproperty of IFN-DC could be due to their capability toexpress certain cytokines involved in the amplification ofthe action of IL-12 and in the generation and expansionof a cytotoxic CD8+ T cells. In this regard, Th1 and CTLresponses have been demonstrated to be promoted bythe IL-12 family cytokines IL-23 and IL-27 [20–22].Thus, we have measured the mRNA levels of IL-23 p19/p40 and IL-27 EBI-3/p28 subunits in the two types

Fig. 6. Generation of HIV-specific human CD8+ T cells in hu-PBL-SCID mice immunized with AT2-HIV-1-pulsed DC. ELISPOTanalysis of anti-HIV-1 CD8+ T cell response. Human cells recovered from three spleens of hu-PBL-SCIDmice from each groupwerepooled. The assay was performed using as stimulators autologous AT2-HIV-1-pulsed or unpulsed DC. Bars represent the CD8+

T cell response from hu-PBL-SCID mice immunized with either IFN-DC or mIL-4-DC (Exp. 1) and IFN-DC or mIFN-DC (Exp. 2), ascompared to the basal CD8+ T cell response in non-immunized hu-PBL-SCID mice (CTR). Control cultures were incubated withunpulsed autologous DC. The panel shows the results of one representative experiment out of three. Hu-PBL-SCID mice wereimmunized as described and Materials and methods.

Fig. 5. Generation of anti-HIV-1 antibodies in hu-PBL-SCIDmice immunizedwithAT2-HIV-1-pulsedDC. ELISAdetection ofantibodies to the HIV-1 gp41 ectodomain epitope AVERY in thesera from hu-PBL-SCID mice immunized with virus-pulsedIFN-DC or CD40L-matured IL-4-DC (mIL4-DC) as comparedto the basal response in non-immunized hu-PBL-SCID mice(CTR). Three tenfold serumdilutions (1:10&; 1:100&; 1:1000&)from threemice in each groupwere tested. Each bar representsthe mean (� SE) of values of three serum samples fromindividual mice.



Fig. 7. In vitro cross-priming of CD8+ T cells against exogenous HIV-1 antigens by DC co-cultivatedwith either total PBL or purifiedCD8+ T cells. Purified CD8+ or total PBL were stimulated on day 0 and restimulated on day 7 with the autologous IFN-DC ormIL-4-DC pulsed with AT-2-inactivated HIV-1 (stimulator/responder ratio of 1:4). Panel (A) shows the light scatter and dot plotanalyses of the purified CD8+ T cell population used in the experiment illustrated in panels (B) and (C). Control cultures wereincubated with unpulsed autologous DC. Exogenous IL-2 (25 U/mL) was added every 4 days. At day 14, the cultures wererestimulated with DC pulsed with AT2-HIV-1, before performing ELISPOT IFN-c (B) and ELISPOT granzyme-B (C) assays, asdescribed in Materials and methods. (D) Cytokine production in the supernatants of primary cultures stimulated three times withautologous DC. Cytokines were measured as described inMaterials and methods. Data are representative of three experiments. Nomeasurable levels of IL-2, IL-1b, IL-7, IL-18, IL-15 and TGFb1 were detected.



Fig. 8. Evaluation of the levels of mRNA expression of the subunits of the IL-23 and IL-27 cytokines by TaqMan real-time RT-PCRanalysis and of IL-12/IL-23 cytokine production by ELISA. (A) DCwere obtained from bloodmonocytes as described inMaterials andmethods. ImmatureDCwere then induced todifferentiatebyovernight exposure to sCD40L.Tomeasure cytokinemRNAexpression,TaqManreal-timeRT-PCRanalysiswasused (AppliedBiosystems, FosterCity, CA). Total RNAwasextracted frommonocytesandDCat different time points, and reverse transcription was carried out as previously described. TaqMan assays were performedaccording to themanufacturer's instructionswith anABI 7700 thermocycler (Applied Biosystems). PCRwas performed, amplifyingthe target cDNA (p40, and p19 transcripts for IL-23. EBI-3 and p28 for IL-27), with b-actin cDNAas an endogenous control. DatawereanalyzedwiththePERelativeQuantificationsoftwareofAppliedBiosystems.SpecificmRNAtranscript levelswereexpressedas foldincrease over the basal condition (untreated monocytes). (B) IL-12 and IL-23 protein release in culture supernatant was tested byusing a commercially available ELISA Kit. Each bar represents the mean concentration values (� SE) of three experiments.



immature DC and their corresponding mature counter-parts. As shown in Fig. 8A, p40 subunit mRNA, which isshared by IL-12 and IL-23 heterodimers, was up-regulated in both IL-4-DC and IFN-DC at comparablelevels upon maturation, while the p19 subunit wasspecifically up-regulated in IFN-DC more than 1000-fold, as confirmed by the higher levels of secreted IL-23detected in supernatants frommatured IFN-DC by ELISA(Fig. 8B). Likewise, the IL-27 EBI-3/p28 subunit mRNAlevels proved to be strongly up-regulated in the IFN-DC.Thus, these results suggested that IFN-DC exhibited agreater attitude to produce IL-12 family cytokinescapable of supporting IL-12 activity and promotingT cell IFN-c production.

Discussion

In the present study, we have shown that one single stepculture ofmonocytes in the presence of IFN-a/GM-CSF issufficient to generateDC endowedwith a special attitudefor cross-priming of CD8+ T cells against exogenousantigens in vivo and in vitro, even in the absence of CD4+

T cell help. This special attitude to induce cross-primingof CD8+ T cells against exogenous antigens was notexplained by increased antigen uptake and antigenprocessing capabilities, since these functions weresomehow comparable between the IFN-DC and theimmature IL-4-DC (Fig. 1C, D). However, the IFN-DCretained a superior attitude in cross-presenting low orlimiting amounts of viral antigens to CD8+ T cells. Sincesimilar resultswere obtainedwith peptide pulsedDC, it islikely that the higher levels of costimulatory and HLAclass I molecules expressed on IFN-DC may explain thissuperior function, although we cannot rule out thepossibility that the capacity of targeting antigens ontoclass I processing pathway is more efficient in IFN-DCthan in IL-4-DC. However, this differencemay, at least inpart, be responsible for the enhanced capability of theIFN-DCwith respect to IL-4-DC to induce an in vivo cross-priming of CD8+ T cells against HIV antigens in the hu-PBL-SCID model [7], even though other mechanisms,such as production of a special set of cytokines/chemokines by IFN-DC could also play an important role.

It is generally believed that DC ability to activate andexpand Ag-specific CD8+ T cells depends on the DCmaturation stage and that DC need to receive a licensingsignal, associated with IL-12 production, in order toelicit cytolytic immune response. In particular, theprovision of signals through CD40 ligand-CD40 inter-actions on CD4+ T cells and DC, respectively, isconsidered important for the DC licensing and inductionof cytotoxic CD8+ T cells [14–16]. Although differentstimuli can activate DC, in our comparative studies withthe IFN-DC we have utilized mature DC activated by

CD40 ligation, which sustains prolonged NF-jB activa-tion, high levels of IL-12 and effective CTL induction[14–18, 23, 24]. The finding that IFN-DC were moreeffective than mIL-4-DC in inducing cross-priming ofCD8+ T cells against exogenous HIV antigens suggeststhat DC licensing for CD8+ T cell cross-priming canefficiently occur after a single-step short-term culture ofmonocytes in the presence of infection-induced cyto-kines (i.e. IFN-a). In our experiments, the capability ofthe different DC to induce cross-priming of CD8+ Tcellsagainst HIV antigens did not correlate with IL-12production, at the time of DC injection into hu-PBL-SCID mice or before their co-culture with autologousT cells. In fact, as expected, the mIL-4-DC utilized in ourexperiments produced large quantities of IL-12, whileonly low IL-12 levels were detected in supernatants fromIFN-DC (Fig. 4B). However, the finding that largeamounts of IL-12 were secreted by IFN-DC after contactwith autologous lymphocytes indicates that these DC arealready committed to undergo terminal activation/maturation. Notably, a further stimulation of IFN-DCwith sCD40L before in vivo immunization did notsignificantly enhance the capacity to stimulate CD8+

Tcells, suggesting that the single IFN-a conditioning stepwas fully sufficient to directly generate licensed DC. Inour in vitro studies, we measured two types of cellresponse: (i) CD8+ cells producing IFN-c; (ii) CD8+ cellsreleasing granzyme-B, which may represent cytotoxiceffector cells. Of note, in the absence of CD4+ T helpercells, IFN-DC were superior with respect to mIL-4-DC ininducing both types of CD8+ T cell responses. Interest-ingly, when the response of total PBL stimulated withAT-2-HIV-pulsed DC was studied, both IFN-DC and mIL-4-DC were equally capable to efficiently stimulate theexpansion of IFN-c-producing CD8+ T cells, while IFN-DC were superior in the induction of granzyme-B-producing cells. In this regard, it is worth mentioningthat CTL effector diversity in terms of dissociatedexpression of granzyme-B and IFN-c has been described[25]. Most assays describe specificity and frequencyof antigen-specific CD8+ cells rather than directantiviral-effect [26], and IFN-c-producing cells aremainly involved in macrophage activation and inflam-mation, while direct killing activity is associated withgranzyme-B release. Our data showing that thegranzyme-B-releasing CD8+ T cells are more effectivelyinduced by IFN-DC further emphasize the concept thatdistinct DC types can preferentially induce differentCD8+ T cell subsets, which may differentially affect thequality of response.

There are several mechanisms by which IFN-a caninfluence the licensing of DC, including the enhance-ment of expression of the peptide transporter TAP-1[27], up-regulation of MHC class I antigens andinduction of factors sustaining generation and activity



of CD8+ T cells [28]. In particular, our results suggest arole of IL-23 and IL-27 in the Th1-promoting activity ofIFN-DC, as these cytokines appear to be produced athigher levels by the IFN-DC and are important inenhancing IL-12-mediated CD8+ T cell responses [20,21]. Noteworthy, the adjuvant activity of these cytokineshas been demonstrated by a recent paper reporting anincrease in the number of IFN-c-producing specificCD8+ cells upon administration of IL-23 and IL-27 [22].Moreover, IL-23 has been shown to sustain CTL and Th1immune responses to DNA immunization by increasingthe rate of survival and proliferation [29].

Notably, IL-23 activity appears to be preferentiallyrestricted to memory T cells, although it has also beendemonstrated that IL-23 can synergize with IL-12 inpromoting cytokine production by DC themselves [30].High levels of this cytokine could explain, at least in part,the better T cells response of IFN-DC in terms of T cellIFN-c production (Fig. 6 and 7B). On the other hand,IL-27 synergizes with IL-12 to induce IFN-c productionby na�ve T cells and regulates IL-12 responsiveness ofna�ve CD4+ T cells through IL-12Rb2 chain up-regulation [31, 32]. We may envisage that an up-regulation of IL-27 production by the IFN-DC couldresult in a higher response of na�ve T cells to IL-12action, thus leading to high levels of T cell IFN-csecretion. We postulate that early exposure to IL-27,produced by IFN-DC would commit naive T cells towardTh1 phenotype while exposure to IL-12 would favorsubsequent expansion and stabilization of Th1 responsewith the contribution of IL-23, which had been shown tosustain the proliferation of memory T cells.

Recent studies have revealed the important role oftype I IFN in linking innate and adaptive immunity [28,33]. In particular, A. Le Bon and co-workers [34] haveshown that type I IFN-a can efficiently promote thecross-priming of CD8+ T cells in mouse models. Theseresults have led to the suggestion that virus-induced IFNcan act as a major stimulus for vigorous generation ofCD8+ T cell response, often observed in the course ofsome viral infections, by multiple mechanisms, includ-ing the promotion of cross-priming of CD8+ T cellsagainst exogenous antigens [35]. Our results suggestthat mechanisms similar to those described in mice [34,35] can be operative in humans, supporting the conceptthat the in vivo generation of IFN-a-conditioned DCrepresent a natural event required for an efficient in vivocross-priming of CD8+ T cells against exogenousantigens in the course of infections. Lastly, our datamay be relevant for the development of DC-basedvaccines, which has recently emerged as an attractivestrategy of therapeutic vaccination in patients withcancer and infectious diseases [3, 36, 37]. In fact, onecritical issue for optimization of DC-based vaccines is theidentification of DC endowed with functional features

“optimal” for the induction of a protective anti-tumorresponse. While our results lead to a general attention toconsider IFN-DC as candidates for development of DC-based vaccines, our data also underline a specificinterest for using IFN-a-conditioned DC and AT-2-HIVas immunotherapy of HIV-1 infection. The interest inthis strategy is, in fact, enhanced by a recently publishedreport showing the efficacy of an AT-2-HIV-DC vaccinein lowering HIV viremia in HIV-infected patients [38]. Inview of this and of the several studies showing a specialanti-HIV activity of IFN-DC with respect to conventionalDC [6, 7, 13], the use of IFN-DC in clinical trials oftherapeutic vaccination of HIV-1-infected patientsrepresents the natural direct extension of the presentwork.

Materials and methods

Cell separation and culture

Peripheral blood mononuclear cells were obtained fromheparinized blood of healthy donors by Ficoll density gradientcentrifugation (Seromed). Monocytes were isolated by im-munomagnetic selection (MACS Cell Isolation Kits; MiltenyiBiotec). Positively selected CD14+ monocytes were plated atthe concentration of 2 � 106 cells/mL in AIM-V medium(GIBCO BRL), supplemented with 2% autologous plasma,500 U/mL GM-CSF and either 250 U/mL IL-4 (R&D Systems)for 6 days or 10 000 U/mL natural IFN-a (Alfaferone;AlfaWasserman) for 3 days. DC were matured by treatmentwith CD40L (1 lg/mL) + 0.1 lg/mL enhancer (AlexisBiochimicals) for one additional day. CD40L and enhancerkit Human rhsCD40L FLAGJ Set is a commercial kit from Alexiscorporation. The extracellular domain of human CD40L(CD154) (aa 116–261) is fused at the N terminus to a linkerpeptide (6 aa) and a FLAGJ-tag. FLAG is a registered trademarkof Sigma-Aldrich. The “Enhancer” for Ligands (Prod. No. ALX-804–034) increases the biological activity of rhsCD40L at least1000-fold by ligand cross-linking. Human CD8+ T cells wereisolated by positive immunomagnetic selection (MACS CellIsolation Kits; Miltenyi Biotec).

Immunophenotypic analysis

Cells were washed and resuspended in PBS containing 1%human serum and incubated with a anti-CD80 (BectonDickinson), CD40, CD86, CD83, HLA-ABC, DC-SIGN, CCR5,CXCR4 and CD4 (BD PharMingen). Cells were analyzed byflow cytometry by using a FACSortTM (Becton Dickinson) flowcytometer.

Phagocytosis

DC (0.5 � 106 cells) were incubated for 60 min at 37�C witheither 50 lg/mL of dextran-FITC conjugate or 10 lg/mL ofDQ-Ovalbumin (Molecular Probes). Cells were washed andresuspended in 500 lL of PBS. DC incubated with either



dextran-FITC or DQ-Ovalbumin at 4�C were used as control.Cells were analyzed by flow cytometry.

Antigen presentation assay of the HCV-NS3 protein tothe specific CD8+ T cell clone

Cells of a CD8+ T cell clone specific for the HLA-A2 bindingpeptide NS31406–1415 (KLVALGINAV) of the HCV-NS3 c33crecombinant protein [39] were stimulated with the relevantpeptide or with protein-loaded DC (either IFN-DC or IL-4-DC)in U-bottom microculture wells at 2 � 104 DC/3 � 104 CD8+

T cell/well in 0.2 mL of RPMI 1640 medium supplementedwith 10% fetal calf serum (RPMI-1640–10%). DC loaded witheither the peptide or the NS3 protein for 18 h at 37�C werethen washed with RPMI-1640–10% and added to the culture atthe DC/T cell ratio of 1:1.5. After a 2-h culture, the cells werefurther treated with brefeldin-A (10 lg/mL, Sigma-Aldrich) at37�C for 18 h. Cells were washed and stained with anti-CD8tricolor (TC) (Caltag Laboratories, Burlingame, CA) for 20 minat 4�C, fixed, permeabilized using Cytofix/Cytoperm solution(BD PharMingen) at 4�C for 20 min, rewashed with PermWash Buffer (BD PharMingen), intracellularly stained withFITC-labeled anti-IFN-c antibody (BD PharMingen) for 30 minat 4�C and finally subjected to flow cytometry.

Detection of HIV-1 infection in DC cultures

DC (IFN-DC and IL4-DC) were washed and infected with HIV-1SF162 strain for 2 h at 37�C. After extensive washing, DC werecultured in RPMI containing 10% FCS at the concentration of106 cells/mL. Culture mediumwas harvested at day 3. For PCRdetection of HIV-1 proviral sequences, DNAwas extracted fromDC. The presence of human sequences was determined byDNA-PCR using specific primers for the HLA-DQa gene: GH2650 GTGCTGCAGGTGTAAACTTGTACCAG, GH27 30CACGGAT-CCGGTAGCAGCGGTAGAGTTG. HIV-1 proviral DNA wasdetected by specific amplification of HIV-1 gag sequences:GAG 881 50 GGTACATCAGGCCATATCACC, GAG 882 30

ACCGGTCTACATAGTCTC. The sensitivity of the assay wastested by amplifying serial dilutions of DNA prepared from 8E5cells (which harbor one proviral copy/cell). The 8E5 DNA wasserially diluted into human cell DNA. Virus replication wasdetermined after 3 days of culture by detection of p24 gagantigen in culture supernatant using a commercial ELISA kit(Dupont, Bruxelles, Belgium).

Immunization of hu-PBL-SCID mice

CB17 scid/scid female mice (Charles River Laboratories) wereused at 4 weeks of age. Three or four mice for each group wereinjected i.p. with 30–40�106 PBL resuspended in 0.5 mL AIM-V medium. To prepare the inactivated HIV-1, different SF-162HIV-1 stocks were inactivated by treatment for 1 h at 37�Cwith2,20-dithiodipyridine (AT-2) as described elsewhere [6]. Fouror seven days after reconstitution, hu-PBL-SCID mice wereinjected i.p. with 2.5 � 106 autologous DC pulsed for 2 h at37�C with AT-2-inactivated HIV-1 (100 ng p24). Mature DCwere loaded with antigens prior to the induction of maturationby sCD40L treatment. The vaccinated mice received one boostimmunization at day 7 and were sacrificed after additional7 days.

ELISA for human immunoglobulins

Sera from control and vaccinated hu-PBL-SCID mice, col-lected at 7 and 14 days after the first immunization, weretested for the presence of antibodies to HIV-1 by an ELISAsystem for quantifying human immunoglobulins to the AVERYHIV-1 gp41 epitope, based on the use of anti-human total IgGor IgM (Cappel-Cooper Biomedical), as described in detailelsewhere [7].

Recovery of cells from hu-PBL-SCID mice and ELISPOTassay

Hu-PBL-SCID mice were sacrificed 7–10 days after the lastimmunization. Cells were collected from the peritoneal cavityand spleen. Human cells frommouse spleens were enriched byFicoll density gradient centrifugation and pooled (three to fourmice per group). Autologous DC were pulsed for 2 h at 37�Cwith AT-2-inactivated HIV-1 (100 ng p24), washed and used asAPC for stimulation of human cells recovered from hu-PBL-SCID mice. Control uninfected DC were used as stimulators forthe calculation of background spots. PBMC cultures treatedwith 2 lg/mL PHA served as positive controls. The cells wereadded at 106 per well and incubated at 37�C overnight in a finalvolume of 2 mL of AIM-V medium (GIBCO) supplementedwith 2 mM L-glutamine and 2% heat-inactivated autologousplasma. After incubation with autologous DC at a responder/stimulator ratio of 4:1, CD8+ Tcells were positively selected byMACS Micro Beads (Miltenyi Biotec) and tested 105/well in anELISPOT assay for the production of IFN-c (Euroclone) [7].

In vitro induction of cross-priming of CD8+ T cellsagainst HIV-1 antigens by using either purified CD8+

T cells or total PBL

CD8+ T cells and PBL (4 � 106) were stimulated with 106

autologous IFN-DC or mIL-4-DC, pulsed with AT-2-inactivatedHIV-1 (100 ng of p24) for 2 h at 37�C. In the case of IL4-DC,cells were first loaded with antigens and subsequently inducedtomaturation by sCD40L treatment. CD8+ Tcells and PBLwererestimulated 7 days later with HIV-pulsed DC. Seven dayslater, the frequency of HIV-1-specific T cells was evaluated byELISPOT assays for IFN-c (Euroclone) or granzyme-B (BectonDickinson) according to the manufacturer's instructions.Tenfold dilutions (from 105 to 102) of DC-stimulated CD8+

T cells and PBL from primary cultures were restimulatedovernight with DC pulsed with inactivated HIV-1 (E/S ratio of1:1), added to duplicate wells, and incubated for 18 h. Controluninfected DC were used as stimulators/targets for thecalculation of background spots to be subtracted for theevaluation of the specific number of IFN- c or granzyme-B-spot-forming cells. PBMC cultures treated with 2 lg/mL PHAserved as positive controls. IFN- c or granzyme-B-producingcells was evaluated by enumerating single spots using anautomatic analyzer.

Detection of cytokine production

Commercial ELISA were used to quantitate in the cell culturesupernatants the following cytokines: IL-6, IL-2, IL-1b IL-12



and TNF-a (Endogen), IL-23 (Bender MedSystem), IL-7(D.R.G.), IL-10 and IL-15 and TGF-b1 (R&D Systems), IL-18(M.B.L.) and for measuring PGE2 (Assay, Designs.). Assaysensitivity was as follows: IL-6 (10.24 pg/mL), IL-7 (15.6 pg/mL), IL-10 (3.6 pg/mL), IL-12 (25.6 pg/mL), IL-23 (78 pg/mL), IL-15 (3.9 pg/mL), IL-18 (25.6 pg/mL), TNF-a (15.6 pg/mL), IL-2 (38.4 pg/mL), IL-1b (10.24 pg/mL TGFb1 (31.2 pg/mL) and PGE2 (39.1 pg/mL). ELISA were performed intriplicate and laboratory standards were included on eachplate.

Evaluation of IL-23 and IL-27 subunitmRNAexpressionby real-time RT-PCR analysis

DC were obtained from blood monocytes as described aboveand then induced to differentiate by overnight exposure tosCD40L. To measure cytokine mRNA expression, TaqMan real-time RT-PCR analysis was used (Applied Biosystems). TotalRNA was extracted from monocytes and DC at different timepoints, and reverse transcribed. TaqMan assays were per-formed according to the manufacturer's instructions with anABI 7700 thermocycler (Applied Biosystems). PCR wasperformed, amplifying the target cDNA (p40, and p19transcripts for IL-23. EBI-3 and p28 for IL-27), with b-actincDNA as an endogenous control. Data were analyzed with thePE Relative Quantification software of Applied Biosystems. Attime zero, mRNA levels, normalized to b-actin, weredetermined for each individual cytokine chain and wereexpressed relative to b-actin mRNA. Specific mRNA transcriptlevels were expressed as fold increase over the basal condition(untreated monocytes).

Acknowledgements: We are grateful to Cinzia Gaspar-rini and Anna Ferrigno for excellent secretarial assis-tance and Dr. M. Ferrantini for useful comments on themanuscript.

References

1 Banchereau, J. and Steinman, R. M., Dendritic cells and the control ofimmunity. Nature 1998. 392: 245–252.

2 Banchereau, J., Briere, F., Caux, C., Davoust, J., Lebecque, S., Liu, Y. J.,Pulendran, B. et al., Immunobiology of dendritic cells. Annu. Rev. Immunol2000. 18: 767–811.

3 Schuler, G., Schuler-Thurner, B. and Steinman, R. M., The use of dendriticcells in cancer immunotherapy. Curr. Opin. Immunol 2003. 15: 138–147.

4 Kaiser, A., Bercovici, N., Abastado, J. P. and Nardin A., Naive CD8+ T cellrecruitment and proliferation are dependent on stage of dendritic cellmaturation. Eur. J. Immunol 2003. 33: 162–171.

5 Sallusto, F. and Lanzavecchia, A., Efficient presentation of soluble antigenby cultured human dendritic cells is maintained by granulocyte/macrophagecolony-stimulating factor plus interleukin 4 and downregulated by tumornecrosis factor alpha. J. Exp. Med. 1994. 179: 1109–1118.

6 Santini, S. M., Lapenta, C., Logozzi, M., Parlato, S., Spada, M., DiPucchio, T. and Belardelli, F., Type I interferon as a powerful adjuvant formonocyte-derived dendritic cell development and activity in vitro and in Hu-PBL-SCID mice. J. Exp. Med 2000. 191: 1777–1788.

7 Lapenta, C., Santini, S. M., Logozzi, M., Spada, M., Andreotti, M., DiPucchio, T., Parlato, S. et al., Potent immune response against HIV-1 andprotection from virus challenge in hu-PBL-SCID mice immunized with

inactivated-virus-pulsed dendritic cells generated in the presence of IFN-a. J.Exp. Med. 2003. 198: 361–367.

8 Santodonato, L., D'Agostino, G., Nisini, R., Mariotti, S., Monque, D. M.,Spada, M., Lattanzi, L. et al., Monocyte-derived dendritic cells generatedafter a short-term culture with IFN-a and granulocyte-macrophage colony-stimulating factor stimulate a potent Epstein-Barr virus-specific CD8+ T cellresponse. J. Immunol 2003. 170: 5195–5202.

9 Gabriele, L., Borghi, P., Rozera, C., Sestili, P., Andreotti, M., Guarini, A.,Montefusco, E. et al., IFN-a promotes the rapid differentiation ofmonocytes from patients with chronic myeloid leukemia into activateddendritic cells tuned to undergo full maturation after LPS treatment. Blood2004. 103: 980–987.

10 Buelens, C., Bartholome, E. J., Amraoui, Z., Boutriaux, M., Salmon, I.,Thielemans, K., Willems, F. et al., Interleukin-3 and interferon betacooperate to induce differentiation of monocytes into dendritic cells withpotent helper T-cell stimulatory properties. Blood 2002. 99: 993–998.

11 Blanco, P., Palucka, A. K., Gill, M., Pascual, V. and Banchereau, J.,Induction of dendritic cell differentiation by IFN-a in systemic lupuserythematosus. Science 2001. 294: 1540–1543.

12 Mohty, M., Vialle-Castellano, A., Nunes, J. A., Isnardon, D., Olive, D. andGaugler, B., IFN-alpha skews monocyte differentiation into Toll-likereceptor 7-expressing dendritic cells with potent functional activities. J.Immunol 2003. 171: 3385–3393.

13 Carbonneil, C., Aouba, A., Burgard, M., Cardinaud, S., Rouzioux, C.,Langlade-Demoyen, P. and Weiss L., Dendritic cells generated in thepresence of granulocyte-macrophage colony-stimulating factor and IFN-alpha are potent inducers of HIV-specific CD8 T cells. AIDS 2003. 17:1731–1740.

14 Ridge, J. P., Di Rosa, F. and Matzinger, P., A conditioned dendritic cell canbe a temporal bridge between a CD4+ T-helper and T-killer cell.Nature 1998.393: 474–478.

15 Schoenberger, S. P., Toes, R. E., van der Voort, E. I., Offringa, R. andMelief, C. J., T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions. Nature 1998. 393: 480–483.

16 Bennett, S. R., Carbone, F. R., Karamalis, F., Flavell, R. A.,Miller, J. F. andHeath, W. R., Help for cytotoxic-T-cell responses is mediated by CD40signalling. Nature 1998. 393: 478–480.

17 Fujii, S., Liu, K., Smith, C., Bonito, A. J. and Steinman, R. M., The linkageof innate to adaptive immunity via maturing dendritic cells in vivo requiresCD40 ligation in addition to antigen presentation and CD80/86 costimula-tion. J. Exp. Med. 2004. 199: 1607–1618.

18 O'Sullivan, B. J. and Thomas, R., CD40 ligation conditions dendritic cellantigen-presenting function through sustained activation of NF-kappaB. J.Immunol 2002. 168: 5491–5498.

19 Accapezzato, D., Visco, V., Francavilla, V., Molette, C., Donato, T., Paroli,M., Mondelli, M. U. et al., Chloroquine enhances human CD8+ T cellresponses against soluble antigens in vivo. J. Exp. Med 2005. 202: 817–28.

20 Trinchieri, G., Pflanz, S. and Kastelein, R. A., The IL-12 family ofheterodimeric cytokines: new players in the regulation of T cell responses.Immunity 2003. 19: 641–644.

21 Trinchieri, G.Interleukin-12 and the regulation of innate resistance andadaptive immunity. Nat. Rev. Immunol 2003. 3: 133–146.

22 Matsui, M., Moriya, O., Belladonna, M. L., Kamiya, S., Lemonnier, F. A.,Yoshimoto, T. and Akatsuka, T., Adjuvant activities of novel cytokines,interleukin-23 (IL-23) and IL-27, for induction of hepatitis C virus-specificcytotoxic T lymphocytes in HLA-A*0201 transgenic mice. J. Virol. 2004. 78:9093–9104.

23 O'Sullivan, B. J. and Thomas, R., CD40 and dendritic cell function. Crit.Rev. Immunol 2003. 23: 83–107.

24 Valenzuela, J., Schmidt, C. and Mescher, M., The roles of IL-12 inproviding a third signal for clonal expansion of na�ve CD8 Tcells. J. Immunol2002. 169: 6842–6849.

25 Kleen, T. O., Asaad, R., Landry, S. J., Boehm, B. O. and Tary-Lehmann,M., Tc1 effector diversity shows dissociated expression of granzyme B andinterferon-gamma in HIV infection. AIDS 2004. 1: 383–392.

26 Yang, O. O., Will we be able to 'spot' an effective HIV-1 vaccine? TrendsImmunol 2003. 24: 67–72.



27 Cho, H. J., Hayashi, T., Datta, S. K., Takabayashi, K., Van Uden, J. H.,Horner, A., Corr, M. and Raz, E., IFN-alpha beta promote priming ofantigen-specific CD8+ and CD4+ T lymphocytes by immunostimulatoryDNA-based vaccines. J. Immunol. 2002. 168: 4907–4913.

28 Belardelli, F. and Ferrantini, M., Cytokines as a link between innate andadaptive antitumor immunity. Trends Immunol 2002. 23: 201–208.

29 Ha, S. J., Kim, D. J., Baek, K. H., Yun, Y. D. and Sung, Y. C., IL-23 inducesstronger sustained CTL and Th1 immune responses than IL-12 in hepatitis Cvirus envelope protein 2 DNA immunization. J. Immunol. 2004. 172:525–531.

30 Belladonna, M. L., Renauld, J. C., Bianchi, R., Vacca, C., Fallarono, F.,Orabona, C., Fioretti, M. C. et al., IL-23 and IL-12 have overlapping, butdistinct, effects on murine dendritic cells. J. Immunol 2002. 168:5448–5454.

31 Lucas, S., Ghilardi, N., Li, J. and de Sauvage, F. J., IL-27 regulates IL-12responsiveness of naive CD4+ T cells through Stat1-dependent and -independent mechanisms. Proc. Natl. Acad. Sci. USA 2003. 100:15047–15052.

32 Hibbert, L., Pflanz, S., DeWaal Malefyt, R. and Kastelein, R. A., IL-27 andIFN-alpha signal via Stat1 and Stat3 and induce T-Bet and IL-12Rbeta2 innaive T cells. J. Interferon Cytokine Res 2003. 23: 513–522.

33 Le Bon, A. and Tough, D. F., Links between innate and adaptive immunityvia type I interferon. Curr. Opin. Immunol 2002. 14: 432–436.

34 Le Bon, A., Etchart, N., Rossmann, C., Ashton, M., Hou, S., Gewert, D.,Borrow, P. et al., Cross-priming of CD8+ Tcells stimulated by virus-inducedtype I interferon. Nat. Immunol 2003. 4: 1009–1015.

35 Beignon, A. S., Skoberne, M. and Bhardwaj, N., Type I interferonspromotes cross-priming: more functions for old cytokines. Nat. Immunol2003. 4: 939–941.

36 Figdor, C. G., de Vries, I. J., Lesterhuis, W. J. and Melief, C. J., Dendriticcell immunotherapy: mapping the way. Nat. Med 2004. 10: 475–480.

37 O'Neill, D. W., Adams, S. and Bhardwaj, N., Manipulating dendritic cellbiology for the active immunotherapy of cancer. Blood 2004. 104:2235–2246.

38 Lu, W., Arraes, L. C., Ferreira, W. T. and Andrieu, J. M., Therapeuticdendritic-cell vaccine for chronic HIV-1 infection. Nat. Med 2004. 10:1359–1365.

39 Barnaba, V., Franco, A., Alberti, A., Benvenuto, R. and Balsano, F.,Selective killing of hepatitis B envelope antigen-specific B cells by class I-restricted, exogenous antigen-specific T lymphocytes. Nature 1990. 345:258–260.



ifn-α-conditioned dendritic cells are highly efficient in inducing cross-priming cd8+ t cells...

Documents