¹Max-Delbrück-Center for Molecular Medicine, MDC, Berlin, Germany

²Institute for Molecular Pathology, IMP, Vienna, Austria

³Department of Immunobiology, Institute for Zoology, University of Bonn, Bonn, Germany

Correspondence to: M Zenke, Max-Delbrück-Center for Molecular Medicine, MDC, Robert-Rössle-Str 10, D-13092 Berlin, Germany

^aPresent address: Imperial Cancer Research Fund, ICRF, 44 Lincoln's Inn Fields, London WC2A 3PX, UK

Dendritic cells (DC) present immunogenic epitopes of antigens in the context of MHC class I and class II molecules in association with costimulatory molecules, and efficiently activate both cytotoxic T cells and T helper cells. Gene modified DC expressing antigen encoding cDNA represent a particularly attractive approach for the immunotherapy of disease. We previously described a gene delivery system for DC based on receptor-mediated endocytosis of ligand/polyethylenimine (PEI) DNA transfer complexes that target cell surface receptors which are abundantly expressed on DC. Employing this gene delivery system, DC were generated that express chicken ovalbumin (OVA) cDNA as a model antigen and introduce antigen into the MHC class I presentation pathway. We demonstrate here that modification of OVA cDNA as transferrin receptor (TfR) or invariant chain (Ii) fusions effectively generate MHC class II specific immune responses in addition to MHC class I responses. TfR-OVA contains the membrane anchoring region of transferrin receptor and represents a membrane-bound form of OVA for access to the MHC class II compartment. Ii-OVA fusions directly target the MHC class II processing pathway. Thus, modification of antigen encoding cDNA represents a convenient and effective means to direct antigens to MHC class II presentation and thus to generate T cell help. Gene Therapy (2001) 8, 487-493.

dendritic cells; MHC class II; invariant chain; ovalbumin; Ad/PEI transfection

Introduction

Dendritic cells (DC) are professional antigen presenting cells (APC) and play a key role in initiating primary immune responses.^1,2,3 DC occur throughout the organism and in particular in tissues such as skin that represent an interface to the environment where they are exposed to and capture antigens. Through antigen uptake and upon activation by inflammatory stimuli, DC are triggered to migrate via lymphatic vessels to the draining lymphoid organs where they present the processed antigens in the context of major histocompatibility complex (MHC) encoded peptide receptors. There are two branches of MHC receptor presentation. MHC class I molecules present endogenously produced peptides that include virus peptides and tumor antigens. In complement to MHC class I molecules that bind their peptides in the endoplasmatic reticulum (ER), MHC class II heterodimers acquire their peptides in endocytic vesicles.^1,2,3 This source of peptides is derived from endocytosed antigens. MHC class II presentation induces T cell help required for antibody production and promotes the generation of cytotoxic T cells.

Peptides are loaded on to MHC class II molecules by different pathways. In the classical pathway, newly synthesized MHC class II molecules associate with invariant chain (Ii) in the ER.⁴ Association with Ii blocks the binding of peptides to the MHC class II peptide binding groove and mediates translocation of the MHC class II/Ii complexes to the endocytic compartment. In the endocytic compartment proteolytic cleavage of Ii chain occurs and the MHC class II peptide binding groove becomes available for binding of peptides derived from extracellular proteins that are taken up into the cell by endocytosis. There are also alternative pathways of peptide loading of MHC class II molecules that are Ii independent.⁵ In addition, exogenous proteins have access to the MHC class I presentation pathway by a so far unknown mechanism.^6,7

Given their unique properties as professional APC, DC are currently being assessed for medical therapy, such as immunotherapy of cancer. Several studies have been initiated mainly employing in vitro generated DC that were loaded with specific tumor antigen peptides.^{2,8,9,10,11,12} Alternative approaches use DC that are transduced with antigen encoding cDNA or RNA (see Refs 13-15). Such gene-modified DC offer several potential advantages over peptide loaded DC in medical therapy. (1) The antigenic peptides are produced by DC themselves, and loaded on to and presented by MHC molecules possibly within multiple MHC alleles. (2) Multiple and/or undefined epitopes are potentially presented. (3) Antigenic peptides are continuously produced and loaded on to MHC molecules, while in peptide pulsed DC only a small proportion of cell surface MHC molecules are loaded with synthetic peptide. (4) cDNA encoding immunomodulators like, for example, cytokines and chemokines can be cotransfected in addition to antigen cDNA to affect DC and T cell functions, and to modulate immune responses.

In previous studies we employed receptor-mediated endocytosis for targeted delivery of DNA into DC.^14,15,16 Adenovirus polyethylenimine DNA (Ad/PEI/DNA) transfer complexes were generated to target the adenovirus receptor on DC. Such DNA transfer complexes consist of adenovirus particles that function as receptor binding and cell entry moiety, and the polycation PEI that binds and condenses plasmid DNA, and following uptake into cells facilitates exit from the endosomal compartment. Ad/PEI/DNA transfer complexes were effective in delivering reporter genes (such as luciferase and green fluorescent protein, GFP) and chicken ovalbumin (OVA) cDNA in DC, and induced antigen specific T cell proliferation in vitro. This gene delivery system, like most DNA transfer systems, introduces antigens into the MHC class I presentation pathway.^14,15 However, several recent studies in experimental tumor models indicated that CD4 T cell help is critical for effective antitumor immunity.^17,18,19,20 Additionally, clinical studies have shown that also in humans CD4 T cell help is required for optimal induction of cytotoxic T cell responses against autologous tumor cells.²¹ This is probably due to the fact that many tumor-specific and tumor-associated antigens are weak antigens.^22,23 Therefore we addressed the question of whether Ad/PEI/DNA transduced DC can be exploited to generate antigen-specific MHC class II T cell responses in addition to MHC class I responses. A gene delivery system that achieves both CD4 and CD8 T cell responses simultaneously would be of considerable advantage over the existing DNA transfer systems.

We describe here two strategies to generate effective MHC class II presentation of transfected antigen cDNA. Chicken ovalbumin (OVA) was employed as a model antigen. First, OVA cDNA was fused to the transmembrane domain of transferrin receptor (TfR) to generate a membrane-bound TfR-OVA fusion protein that has access to the MHC class II compartment. Second, to target the MHC class II processing pathway directly, invariant chain (Ii) OVA fusions were constructed. We demonstrate that such chimeric TfR-OVA and Ii-OVA fusion proteins generate both effective MHC class II and MHC class I restricted T cell responses.

Results

Construction and expression of chimeric OVA antigens

Chicken ovalbumin contains well defined MHC class I and class II binding sequences at positions 257-264 and 323-339, respectively, and was employed as model antigen in this study (Figure 1). Various constructs with one or both epitopes were generated for use in in vitro T cell activation assays with TCR transgenic T cells that are specific for either the MHC class I or class II epitope. First, to obtain a nonsecreted OVA protein, a truncated ovalbumin cDNA lacking sequences that correspond to the first 49 N-terminal amino acids of the leader sequence was constructed (Figure 1). Second, the membrane-bound transferrin receptor-ovalbumin fusion protein (TfR-OVA) version²⁴ was employed to facilitate access to the endocytic compartment. TfR-OVA consists of the first 118 amino acids of human transferrin receptor with its membrane anchoring domain fused in frame to amino acids 139-385 of ovalbumin, containing the MHC class I and class II antigenic epitopes. Finally, for targeting the antigen to the MHC class II pathway two invariant chain (Ii)-OVA fusion proteins were constructed containing either the MHC class I and class II epitope, or the MHC class II epitope only (Ii-OVA.I+II and Ii-OVA.II, respectively, Figure 1).

Expression of the respective proteins was first evaluated by in vitro transcription/translation. All constructs gave rise to proteins of the expected size as shown in Figure 2a. In addition, the identity of the 43 kDa and 38 kDa bands, corresponding to bona fide OVA and OVA, respectively, was verified by Western blotting using polyclonal anti-ovalbumin antibody (Figure 2b). The TfR-OVA fusion protein of 40 kDa size was also detected by this antibody albeit less efficiently. This reduced staining is probably due to the fact that parts of the epitopes recognized by the antibody are located in the first 139 amino acids of the OVA protein that are absent in TfR-OVA. Finally, anti-Ii antibody stained the in vitro translation products of unmodified Ii and of both Ii-OVA fusions but not OVA, OVA and TfR-OVA (Figure 2b and data not shown).

Expression of OVA antigen constructs in Ad/PEI/DNA transfected DC

DC were obtained from mouse bone marrow in the presence of GM-CSF according to standard procedures²⁵ and used for transfection at day 7 of culture when the cells still exhibited an immature phenotype as demonstrated by low DEC205 expression (data not shown). Ad/PEI/DNA transfer complexes containing the expression plasmids for OVA, OVA, TfR-OVA, Ii and the Ii-OVA fusions were generated and transfected into DC. At day 1 after transfection, DC were harvested, RNA was prepared and analyzed by RT-PCR. PCR DNA fragments of the expected size corresponding to OVA, OVA, TfR-OVA, Ii, Ii-OVA.II and Ii-OVA.I+II specific transcripts were readily detected in transfected cells. Empty vector transfected and untransfected control were, as expected, negative (Figure 3). Additionally, OVA and Ii-OVA specific primer pairs amplified comparable amounts of PCR products from the various constructs, indicating equal expression levels in DC.

Chimeric TfR-OVA and Ii-OVA expressing DC induce MHC class II restricted T cell activation

Next, DC expressing the various OVA chimeras were examined for their ability to induce antigen specific MHC class I or class II restricted T cell activation. Therefore transduced DC were cocultured with CD4⁺ or CD8⁺ T cells of TCR transgenic OT-I or DO11.10 mice. OT-I T cells recognize OVA 257-264 peptide in the context of H-2-K^b (MHC class I) while DO11.10 T cells detect OVA 323-339 peptide presented by I-A^d (MHC class II). T cell proliferation was assessed after 5 days by thymidine incorporation. OVA, OVA, TfR-OVA and Ii-OVA.I+II expressing DC induced MHC class I restricted T cell proliferation of CD8⁺ OT-I T cells to the same extent as peptide-pulsed DC (Figure 4a). As expected, Ii and Ii-OVA.II transduced DC lacking the MHC class I OVA epitope were inactive in T cell activation, as were -gal transduced and untreated DC (Figure 4a and data not shown). Importantly, DC expressing the membrane-bound TfR-OVA fusion elicited a MHC class II restricted DO11.10 T cell response, while cells expressing unmodified OVA or nonsecreted OVA did not (Figure 4b). Furthermore, the Ii-OVA fusions Ii-OVA.I+II and Ii-OVA.II were also competent in inducing MHC class II restricted proliferation of T cells while unmodified Ii was inactive. This result demonstrates that Ii-OVA fusion constructs effectively direct the OVA antigen to the MHC class II processing pathway and thus represent a simple strategy to provide T cell help.

In further support of these data, IL-2 production in cocultures of transduced DC and TCR transgenic T cells was determined by ELISA assay (Table 1). In accordance with the T cell proliferation data (Figure 4a and b) all OVA constructs expressing the class I epitope yielded IL-2 production by OT-I T cells. Additionally, IL-2 production by MHC class II restricted DO11.10 T cells was achieved with TfR-OVA and the Ii-OVA fusion constructs Ii-OVA.I+II and Ii-OVA.II, in accord with the T cell proliferation data above. Surprisingly, OT-I T cells stimulated by Ii-OVA.I+II expressing DC produce five to seven times more IL-2 than OT-I T cells cocultured with OVA, OVA or TfR-OVA expressing DC (Table 1, day 1). This difference was even more dramatic at day 2 of culture when cocultures with Ii-OVA.I+II transfected DC contained 16 times more IL-2 than cocultures with OVA, OVA or TfR-OVA expressing DC. Ii-OVA.I+II transfected DC were only three times less efficient than peptide-pulsed DC (0.1 M peptide) in inducing IL-2 production in MHC class I restricted T cell activation assays.

Discussion

Several lines of evidence from mouse tumor models suggest that CD4 T cell help is critical for effective antitumor immunity.^17,18,19,20 Thus, when gene-modified DC expressing tumor specific antigens are to be applied in immunotumor therapy in man, it appears to be of pivotal importance that transduced antigens are also presented via the MHC class II antigen processing pathway in order to provide T cell help.²¹ Specific modification of the antigen coding cDNA is a particularly attractive strategy to introduce CD4 T cell epitopes into the MHC class II processing pathway. We employed two membrane proteins, TfR and Ii that contain sorting sequences for transport to endocytic vesicles, for delivery of antigenic sequences to MHC class II processing compartments. The TfR intersects the MHC class II processing pathway in early endosomes, where the receptor delivers iron for cellular consumption. Ii is a MHC class II chaperone that guides trafficking of MHC class II molecules and controls association of antigenic peptides to the MHC class II groove.⁴ Presentation of TfR or Ii antigen fusion proteins requires degradation of the chimeric proteins in the endosomal/lysosomal compartment. Here we demonstrate that DC transduced with OVA cDNA fused to the transmembrane domain of TfR or to the C-terminus of Ii yield potent MHC class II restricted T cell responses in addition to MHC class I responses. This strategy provides a convenient and effective means to ensure antigen entry into the MHC class II processing pathway and thus the generation of T cell help.

TfR-OVA represents a membrane-bound OVA version that directs TfR-OVA to the cell surface.²⁴ The cytosolic domain of TfR contains tyrosine motifs that upon internalization direct trafficking of the receptor and its cargo to early endosomes.^26,27 This sorting signal is present in TfR-OVA and thus we expect the fusion protein to be transported to these endocytic vesicles. Early and late endocytic compartments were shown to contribute peptides for MHC class II presentation.^28,29,30 In addition, it was shown that TfR colocalizes with Ii in early endosomes which contain the proteolytic enzymes cathepsin B and D.³¹ Thus it is very likely that peptide loading of MHC class II molecules occurs in this compartment. Further support for this conclusion stems from a recent report by Fernandes et al,³² who demonstrate that in B cell lines TfR-OVA fusion protein is indeed found in early endosomes. Additionally, in this study more OVA peptides were generated and presented on MHC class II from the TfR-OVA fusion than from an OVA version that was targeted to the secretory pathway.

To achieve a direct targeting of OVA to the MHC class II compartment, Ii-OVA fusion genes were generated, thereby further extending our previous studies and similar approaches by others.^4,33,34,35 It is demonstrated that DC expressing Ii-OVA fusion cDNA with the OVA-specific MHC class I and II, or the MHC class II epitope only, were effective in eliciting potent MHC class II-restricted T cell responses in DC/T cell cocultures in vitro. Interestingly, efficient MHC class I presentation was also achieved although the Ii fusion targets the antigen to the MHC class II processing pathway. Furthermore, by measuring IL-2 production in such DC/T cell cocultures Ii-OVA.I+II-expressing DC were found to be only three-fold less potent in MHC class I-restricted T cell activation as peptide pulsed DC.

Surprisingly, the MHC class I-restricted T cell stimulation by Ii-OVA.I+II expressing DC was more effective than that by DC expressing OVA, OVA or TfR-OVA protein as assessed by IL-2 production in DC/T cell cocultures. Delivery of Ii-OVA to the MHC class I pathway was reported before and appears to follow the classical TAP-dependent pathway.³⁶ Access of Ii-OVA to the MHC class I pathway could be due to a deficiency in translocation of the translated protein into the ER or because protein is retrieved from the ER to the cytosol for degradation. The insertion of 111 amino acids of OVA sequence at the C-terminus of Ii might lead to the formation of misfolded Ii-OVA protein. Proteins undergo a quality control in the ER and misfolded molecules are shuttled back to the cytoplasm by a retrograde transport system.³⁷ Retrograde transport and ubiquitination appear to be coupled processes that target misfolded proteins to degradation by the proteasome. Such a mechanism might well explain the more efficient MHC class I presentation of the Ii-OVA.I+II fusion protein.

In previous studies retrovirus-transduced DC expressing unmodified bona fide OVA cDNA were demonstrated to yield both MHC class I and II responses.¹⁹ Additionally, a truncated OVA devoid of its signal peptide sequence for translocation into the ER was not presented in association with MHC class II. This observation is in apparent contrast to the results presented here, where only MHC class I-restricted T cell responses were found for both unmodified OVA cDNA and OVA cDNA devoid of leader sequence. The lack of MHC class II responses is not due to failure of OVA cDNA expression, since OVA protein was readily detected after in vitro transcription/translation and also in transient expression experiments (Figure 2 and data not shown), and yielded MHC class I restricted immune responses (Figure 4 and Table 1). Therefore, the reason for these apparent disparate findings might very well be in the different transduction systems used, retroviral versus receptor targeted gene delivery. Infection of DC with recombinant retroviruses requires proliferating cell populations and frequently multiple rounds of infection are employed for efficient transduction.^19,38 Such retrovirus-transduced DC might produce OVA protein in sufficient amounts to achieve high OVA concentrations in the vicinity of the cell to allow antigen uptake into the endosomal compartment. In our study postmitotic differentiated DC were employed that are effectively transduced by Ad/PEI/DNA transfer complexes,^14,15,16 yet the production of OVA protein might be too low to yield sufficiently high OVA concentrations at the cell surface for efficient endocytic uptake. However, as demonstrated here, this apparent limitation of our gene delivery system can be readily overcome by fusion of antigen encoding cDNA to TfR transmembrane domain or to Ii chain. Such fusion constructs provide a simple and effective way of targeting the MHC class II processing pathway and achieving CD4 T cell responses. The data reported here show that such strategies can be applied for primary immunocompetent DC and should pave the way for future clinical studies. Current efforts aim at determining the potential of such chimeric molecules in antigen specific T cell activation in vivo.

Materials and methods

Cells and cell culture

For generation of mouse DC, bone marrow suspensions were prepared from 8-12-week-old C57BL/6 and BALB/c mice (Charles River, Sulzfeld, Germany) as described.²⁵ Cells were cultured in RPMI 1640 medium (GIBCO-BRL, Karlsruhe, Germany) containing 10% inactivated FCS (GIBCO-BRL), 50 M -mercaptoethanol, 100 U/ml penicillin and streptomycin in the presence of recombinant mouse GM-CSF.^14,15,16 The non-adherent cell fraction consisting of DC was harvested at day 6 of culture and was used for transfection at day 7.

Plasmid vectors

The following plasmid constructs were used: pcDNA3-OVA contains chicken ovalbumin cDNA in pcDNA3 (Invitrogen, Groningen, The Netherlands).^14,15 pcDNA3-OVA represents an N-terminally truncated OVA version (amino acids 49-386) that is devoid of the signal sequence for translocation into the ER. For construction of pcDNA3-OVA, PCR primers were used that generate a consensus ATG³⁹ at amino acid 49 of OVA cDNA, and HpaI and XbaI restriction endonuclease cleavage sites 5' and 3' of the translational start and stop sites, respectively. OVA cDNA was then introduced as a HpaI-XbaI fragment into EcoRV/XbaI of pcDNA3. The integrity of OVA cDNA was verified by DNA sequencing. pcDNA3-TfR-OVA contains the first 118 amino acids of human transferrin receptor fused in-frame to amino acids 139-386 of OVA. pcDNA3-TfR-OVA was obtained by cloning the HindIII/XbaI fragment of pBlueRIP-TfR-OVA²⁴ into pcDNA3 vector. OVA cDNA fragments containing the MHC class II epitope (amino acids 315-353) and class I and class II epitopes (amino acids 242-353) were amplified by PCR from pcDNA-OVA. PCR primers were designed to introduce DraIII restriction sites at either end of the fragments. PCR fragments were inserted into the DraIII site of pEXV3-Ii31³⁴ and Ii-OVA fusions were then cloned into EcoRI of pcDNA3 thereby generating pcDNA3-Ii-OVA.II and pcDNA3-Ii-OVA.I+II plasmid, respectively. The integrity of the Ii-OVA fusions was confirmed by DNA sequencing. Plasmid DNA was prepared using the Qiagen EndoFree Plasmid Maxi Kit (Qiagen, Hilden, Germany).

Ad/PEI transfection

For formation of Ad/PEI/DNA complexes, adenovirus particles of E4-defective Ad5 strain dl1014 (E4^-Ad)⁴⁰ and low molecular weight PEI (MW 2000, Aldrich, Deisenhofen, Germany) were used. Transfection complexes were prepared as described^14,41,42 and mouse DC were transfected under serum-free conditions with 250 l transfection mix per 0.5 ´ 10⁶ cells in 24-well plates with 3000 adenovirus particles per cell. For evaluation of T cell stimulatory activity DC were harvested on day 1 after transfection, irradiated (5000 rad) and used for in vitro coculture with T cell receptor (TCR) transgenic T cells of OT-I or DO11.10 mice.

In vitro translation and detection of antigen constructs

Plasmids were linearized by PvuI (pcDNA-3, pcDNA3-OVA, pcDNA3-TfR-OVA and pcDNA3-OVA) or SmaI (pcDNA3-Ii, pcDNA3-Ii-OVA.II and pcDNA3-Ii-OVA.I+II) and subjected to in vitro trans- cription/translation (TnT Coupled Wheat Germ Extract System; Promega, Madison, WI, USA) in the presence of [³⁵S]methionine (Amersham Pharmacia Biotech, Freiburg, Germany). Radiolabeled proteins were separated by SDS-PAGE and blotted on to nitrocellulose membranes (BA85, Schleicher and Schuell, Dassel, Germany) using a semidry blotting apparatus (Pharmacia). Membranes were exposed to film for 16 h. OVA protein was detected by Western blot analysis using rabbit polyclonal anti-OVA (ICN Biomedicals, Eschwege, Germany) and monoclonal anti-CD74 antibody (clone In-1, PharMingen, San Diego, CA, USA), respectively, as described.⁴³

RT-PCR

RNA was prepared from DC at day 1 after transfection using RNeasy Mini Kit (Qiagen) including DNase treatment of samples. Subsequent reverse transcription of RNA samples was performed with SuperScript Preamplification System using random hexamers (GIBCO-BRL). The synthesized cDNA was then amplified by PCR with OVA-, Ii- or Ii-OVA fusion specific primer sets using standard settings (OVA sense primer: AGA AATGTCCTTCAGCCAAGCTC and anti-sense primer: GCCCATAGCCATTAAGACAGATGTG; Ii sense primer: TTCCGAAATCTGCCAAACCTG and anti-sense primer: TCATCTCAAACAAGAGCCACTGC; Ii-OVA sense primer was Ii sense primer and anti-sense primer: GGACGATGAAACAGACACCA) yielding 431 bp (OVA), 288 bp (Ii), 729 bp (Ii-OVA.I+II) and 510 bp (Ii-OVA.II) amplification products, respectively. Equal amounts of cDNA were used as template for PCR reactions and PCR conditions were chosen that ensured a semiquantitative assessment of the amplified PCR products. PCR products were loaded onto 1% agarose gel and visualized by ethidium bromide staining.

OVA specific T cell activation assay

CD8⁺ T cells from OT-I mice express a transgenic TCR that recognizes OVA_257-264 peptide on H-2K^b.^44,45 Splenocytes of OT-I mice were prepared and CD8⁺ T cells were obtained by immunomagnetic bead purification using MACS anti-CD8 Microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer's instructions. CD8⁺ T cells were then cocultured with irradiated mouse DC (5000 rad) in 96-well microtiter plates. CD4⁺ T cells from DO11.10 mice expressing a transgenic TCR that recognizes OVA_323-339 peptide on I-A^d were prepared accordingly using anti-CD4 Microbeads (Miltenyi Biotec) for purification.⁴⁶

Transfected DC were used at day 1 after transfection; untreated DC and DC pulsed with 0.1 M OVA_257-264 peptide (SIINFEKL) or 0.1 M OVA_323-339 peptide (ISQAVHAAHAEINEAGR) were employed as controls for MHC class I and MHC class II restricted T cell stimulation using OT-I and DO11.10 T cells, respectively. As further controls, T cells were stimulated by phorbol 12-myristate 13 acetate (PMA, 25 ng/ml, Sigma) and ionomycin (1 g/ml, Sigma, Traufkirchen, Germany). After 1 and 2 days of culture supernatant was harvested and tested by ELISA for IL-2 production (PharMingen). At day 4 of coculture 1 Ci of ³H-thymidine (Amersham) was added per well, cells were harvested 6 h later and ³H-thymidine incorporation was measured in a Microbeta counter (Wallac, Turku, Finland). All values of ³H-thymidine incorporation represent means of triplicates.

Acknowledgements

We thank FR Carbone and WR Heath for OT-I mice and plasmid DNA, SM Kurz for recombinant mouse GM-CSF, T Blankenstein for helpful discussions, S Knespel for technical assistance and I Gallagher for expert secretarial assistance. This work was in part funded by grants from the Deutsche Forschungsgemeinschaft (DFG, SFB506 and Ze231 and Ko 810/4-4) to MZ and NK, respectively. SD was supported by the MDC Gene Therapy Program.

1 Hart DN. Dendritic cells: unique leukocyte populations which control the primary immune response. Blood 1997; 90: 3245-3287, MEDLINE

2 Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature 1998; 392: 245-252, Article MEDLINE

3 Austyn JM. Dendritic cells. Curr Opin Hematol 1998; 5: 3-15, MEDLINE

4 Koch N, van Driel IR, Gleeson PA. Highjacking a chaperone: manipulation of the MHC class II presentation pathway. Immunol Today 2000; 21: 546-550, MEDLINE

5 Rovere P et al. Dendritic cell maturation and antigen presentation in the absence of invariant chain. Proc Natl Acad Sci USA 1998; 95: 1067-1072, MEDLINE

6 Norbury CC et al. Class I MHC presentation of exogenous soluble antigen via macropinocytosis in bone marrow macrophages. Immunity 1995; 3: 783-791, MEDLINE

7 Albert ML, Sauter B, Bhardwaj N. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 1998; 392: 86-89, Article MEDLINE

8 Hsu FJ et al. Vaccination of patients with B-cell lymphoma using autologous antigen-pulsed dendritic cells. Nat Med 1996; 2: 52-58, MEDLINE

9 Murphy G et al. Phase I clinical trial: T-cell therapy for prostate cancer using autologous dendritic cells pulsed with HLA-A0201-specific peptides from prostate-specific membrane antigen. Prostate 1996; 29: 371-380, Article MEDLINE

10 Nestle FO et al. Vaccination of melanoma patients with peptide- or tumor lysate-pulsed dendritic cells. Nat Med 1998; 4: 328-332, MEDLINE

11 Thurner B et al. Vaccination with mage-3A1 peptide-pulsed mature, monocyte-derived dendritic cells expands specific cytotoxic T cells and induces regression of some metastases in advanced stage IV melanoma. J Exp Med 1999; 190: 1669-1678, MEDLINE

12 Bellone M, Iezzi G, Imro MA, Protti MP. Cancer immunotherapy: synthetic and natural peptides in the balance. Immunol Today 1999; 20: 457-462, MEDLINE

13 Boczkowski D, Nair SK, Snyder D, Gilboa E. Dendritic cells pulsed with RNA are potent antigen-presenting cells in vitro and in vivo. J Exp Med 1996; 184: 465-472, MEDLINE

14 Diebold SS et al. Efficient gene delivery into human dendritic cells by adenovirus polyethylenimine and mannose polyethylenimine transfection. Hum Gene Ther 1999; 10: 775-786, MEDLINE

15 Diebold SS et al. Mannose polyethylenimine conjugates for targeted DNA delivery into dendritic cells. J Biol Chem 1999; 274: 19087-19094, Article MEDLINE

16 Diebold SS, Zenke M. Gene modified antigen presenting dendritic cells for T cell mediated immunity. In: Motman K, Michiels L and Walker JM (eds). Methods in Molecular Medicine, Cancer Immunotherapy. Humana Press: Totowa, NJ (in press),

17 Hung K et al. The central role of CD4(+) T cells in the antitumor immune response. J Exp Med 1998; 188: 2357-2368, MEDLINE

18 Pardoll DM, Topalian SL. The role of CD4+ T cell responses in antitumor immunity. Curr Opin Immunol 1998; 10: 588-594, MEDLINE

19 De Veerman M et al. Retrovirally transduced bone marrow-derived dendritic cells require CD4+ T cell help to elicit protective and therapeutic antitumor immunity. J Immunol 1999; 162: 144-151, MEDLINE

20 Schnell S, Young JW, Houghton AN, Sadelain M. Retrovirally transduced mouse dendritic cells require CD4+ T cell help to elicit antitumor immunity: implications for the clinical use of dendritic cells. J Immunol 2000; 164: 1243-1250, MEDLINE

21 Baxevanis CN et al. Tumor-specific CD4+ T lymphocytes from cancer patients are required for optimal induction of cytotoxic T cells against the autologous tumor. J Immunol 2000; 164: 3902-3912, MEDLINE

22 Boon T, Coulie PG, Van den Eynde B. Tumor antigens recognized by T cells. Immunol Today 1997; 18: 267-268, Article MEDLINE

23 Schuler G, Steinman RM. Dendritic cells as adjuvants for immune-mediated resistance to tumors. J Exp Med 1997; 186: 1183-1187, MEDLINE

24 Teasdale RD, D'Agostaro G, Gleeson PA. The signal for Golgi retention of bovine beta 1,4-galactosyltransferase is in the transmembrane domain. J Biol Chem 1992; 267: 4084-4096, MEDLINE

25 Inaba K et al. Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. J Exp Med 1992; 176: 1693-1702, MEDLINE

26 Jing SQ et al. Role of the human transferrin receptor cytoplasmic domain in endocytosis: localization of a specific signal sequence for internalization. J Cell Biol 1990; 110: 283-294, MEDLINE

27 Girones N et al. Mutational analysis of the cytoplasmic tail of the human transferrin receptor. Identification of a sub-domain that is required for rapid endocytosis. J Biol Chem 1991; 266: 19006-19012, MEDLINE

28 Barnes KA, Mitchell RN. Detection of functional class II-associated antigen: role of a low density endosomal compartment in antigen processing. J Exp Med 1995; 181: 1715-1727, MEDLINE

29 Pinet VM, Long EO. Peptide loading onto recycling HLA-DR molecules occurs in early endosomes. Eur J Immunol 1998; 28: 799-804, MEDLINE

30 Neefjes J. CIIV, MIIC and other compartments for MHC class II loading. Eur J Immunol 1999; 29: 1421-1425, MEDLINE

31 Guagliardi LE et al. Co-localization of molecules involved in antigen processing and presentation in an early endocytic compartment. Nature 1990; 343: 133-139, MEDLINE

32 Fernandes DM, Vidard L, Rock KL. Characterization of MHC class II-presented peptides generated from an antigen targeted to different endocytic compartments. Eur J Immunol 2000; 30: 2333-2343, MEDLINE

33 Sanderson S, Frauwirth K, Shastri N. Expression of endogenous peptide-major histocompatibility complex class II complexes derived from invariant chain-antigen fusion proteins. Proc Natl Acad Sci USA 1995; 92: 7217-7221, MEDLINE

34 Sponaas A, Carstens C, Koch N. C-terminal extension of the MHC class II-associated invariant chain by an antigenic sequence triggers activation of naive T cells. Gene Therapy 1999; 6: 1826-1834, MEDLINE

35 van Bergen J et al. Get into the groove! Targeting antigens to MHC class II. Immunol Rev 1999; 172: 87-96, MEDLINE

36 Paz P, Brouwenstijn N, Perry R, Shastri N. Discrete proteolytic intermediates in the MHC class I antigen processing pathway and MHC I-dependent peptide trimming in the ER. Immunity 1999; 11: 241-251, MEDLINE

37 Mori K. Tripartite management of unfolded proteins in the endoplasmic reticulum. Cell 2000; 101: 451-454, MEDLINE

38 Westermann J et al. Retroviral interleukin-7 gene transfer into human dendritic cells enhances T cell activation. Gene Therapy 1998; 5: 264-271, MEDLINE

39 Kozak M. Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 1986; 44: 283-292, MEDLINE

40 Bridge E, Ketner G. Redundant control of adenovirus late gene expression by early region 4. J Virol 1989; 63: 631-638, MEDLINE

41 Baker A, Cotten M. Delivery of bacterial artificial chromosomes into mammalian cells with psoralen-inactivated adenovirus carrier. Nucleic Acids Res 1997; 25: 1950-1956, MEDLINE

42 Baker A et al. Polyethylenimine (PEI) is a simple, inexpensive and effective reagent for condensing and linking plasmid DNA to adenovirus for gene delivery. Gene Therapy 1997; 4: 773-782, MEDLINE

43 Madruga J et al. Polarised expression pattern of focal contact proteins in highly motile antigen presenting dendritic cells. J Cell Sci 1999; 112: 1685-1696, MEDLINE

44 Hogquist KA et al. T cell receptor antagonist peptides induce positive selection. Cell 1994; 76: 17-27, MEDLINE

45 Kurts C et al. Constitutive class I-restricted exogenous presentation of self antigens in vivo. J Exp Med 1996; 184: 923-930, MEDLINE

46 Murphy KM, Heimberger AB, Loh DY. Induction by antigen of intrathymic apoptosis of CD4+CD8+TCRlo thymocytes in vivo. Science 1990; 250: 1720-1723, MEDLINE

Figure 1 Schematic representation of chimeric OVA antigen constructs. Ovalbumin cDNA (OVA) contains at its C-terminus defined immunogenic MHC class I and class II epitopes (amino acids 257-264 and 323-339, respectively). In OVA, a non-secreted form of OVA protein, the first 48 amino acids containing the signal peptide sequence are deleted. TfR-OVA contains the first 118 amino acids of human transferrin receptor (TfR) fused to amino acids 139-386 of ovalbumin, resulting in a membrane-bound form of OVA. The invariant chain (Ii)-OVA fusion constructs Ii-OVA.I+II and Ii-OVA.II contain amino acids 242-353 and 315-353, respectively, of ovalbumin fused to the C-terminal end of Ii, yielding constructs with the MHC class I and II epitopes, or with the MHC class II epitope only, as indicated.

Figure 2 In vitro translation of chimeric OVA antigen constructs and Western blot analysis. (a) OVA, OVA, TfR-OVA, Ii, Ii-OVA.I+II and Ii-OVA.II cDNA were subjected to in vitro transcription/translation and reaction products were resolved by SDS-PAGE, blotted on to nitrocellulose membrane and exposed to film. M, size marker. (b) OVA, OVA and TfR-OVA proteins were detected by OVA specific antibody and ECL (lanes 1-3). Similarly Ii, Ii-OVA.I+II and Ii-OVA.II proteins are detected with Ii specific antibody (lanes 4-6). M, size marker.

Figure 3 RT-PCR analysis of Ad/PEI/DNA transfected mouse DC. Ad/PEI/DNA transfer complexes containing OVA, OVA, TfR-OVA, Ii, Ii-OVA.I+II and Ii-OVA.II cDNA (lanes 3-8, respectively) and empty vector (lane 2) were generated and used for transfection of mouse DC. Transgene expression was analyzed by RT-PCR at day 1 after transfection. As controls untreated DC (lane 1) and plasmid templates (OVA plasmid, Ii-OVA.I+II and Ii plasmid in a, b and c, respectively; lane 10) were used. H₂O control, lane 9. OVA (a), Ii-OVA.I+II and Ii-OVA.II fusion (Ii/OVA, b) and Ii (c) cDNA was amplified by specific PCR primer pairs and ethidium bromide staining of PCR fragments resolved in 1% agarose gel is shown.

Figure 4 TfR-OVA and Ii-OVA transduced DC induce MHC class I and class II restricted T cell proliferation. (a) OVA cDNA and chimeric TfR-OVA and Ii-OVA cDNAs were introduced into DC by Ad/PEI transfection and MHC class I restricted T cell activation was determined in cocultures with OT-I TCR transgenic T cells by ³H-thymidine incorporation. Control, untreated DC; OVA, OVA, TfR-OVA, Ii, Ii-OVA.I+II and Ii-OVA.II cDNA constructs were the same as in Figure 1; T cells only, SIINFEKL peptide-pulsed DC and PMA plus ionomycin treated T cells were used as additional controls. Means of triplicate values are shown. One representative experiment of five experiments is depicted. (b) OVA cDNA and various variants thereof were introduced into DC as in (a) and MHC class II restricted T cell activation was determined in cocultures with DO11.10 TCR transgenic T cells by ³H-thymidine incorporation. Controls, OVA and various OVA variants were the same as in (a); the ISQAVHAAHAEINEAGR peptide was used for loading of peptide-pulsed DC. Means of triplicate values are shown as in (a).

Table 1 IL-2 production in cocultures of T cells with DC expressing TfR-OVA and Ii-OVA variants