Tumors Protective Immune Responses against Delivery of Antigen to ...

The Journal of Immunology

Delivery of Antigen to CD40 Induces Protective Immune Responses against Tumors1 Karoline W. Schjetne,2 Agnete B. Fredriksen,2,3 and Bjarne Bogen3 Ligation of CD40 induces maturation of dendritic cells (DC) and could be a useful target for vaccines. In this study, we have constructed two types of Ab-based vaccine constructs that target mouse CD40. One type is a recombinant Ab with V regions specific for CD40 and has defined T cell epitopes inserted into its C region. The other type is a homodimer, each chain of which is composed of a targeting unit (single-chain fragment variable targeting CD40), a dimerization motif, and an antigenic unit. Such proteins bound CD40, stimulated maturation of DC, and enhanced primary and memory T cell responses. When delivered i.m. as naked DNA followed by electroporation, the vaccines induced T cell responses against MHC class IIrestricted epitopes, Ab responses, and protection in two tumor models (myeloma and lymphoma). Two factors apparently contributed to these results: 1) agonistic ligation of CD40 and induction of DC maturation, and 2) delivery of Ag to APC and presentation on MHC class II molecules. These results highlight the importance of agonistic targeting of Ag to CD40 for induction of long-lasting and protective immune responses. The Journal of Immunology, 2007, 178: 4169 – 4176.

V

accines should not only deliver Ag to APC, but should also induce maturation of the APC so that they can potently stimulate T cells. Dendritic cells (DC)4 are increasingly considered to be the most important APC (1). Maturation of DC initiates up-regulation of MHC class II and costimulatory molecules, and migration to secondary lymphoid tissues where they activate T cells. In the absence of maturation, as under steady-state conditions, DC may rather induce T cell tolerance than activation (2). Consistent with such a model, T cell tolerance was replaced by memory T cell responses when Ag-loaded DC received an additional maturation signal such as ligation of CD40 on DC (3, 4). CD40 is a member of the TNF family and is expressed by a range of cells, including DC, macrophages, B cells, fibroblasts, epithelial cells, and endothelial cells (5). During a normal T cell response, CD40 on DC is engaged by CD40L, which is transiently expressed on activated CD4⫹ Th cells. CD40 molecules are thereby cross-linked, which induces DC maturation and enable them to efficiently present Ag to T cells (6 – 8). The changes in the DC are not fully understood, but probably involve a combination of improved Ag processing, increased expression of costimulatory

and adhesion molecules, and up-regulation of cytokine production. Consistent with a crucial importance of CD40L-CD40 interaction in induction of immune responses, mice and humans that lack CD40 or CD40L (CD154) genes have reduced Ab production and Ig-class switching, and are unable to mount effective responses against infectious agents. Recently, the importance of activating DC via CD40 ligation was demonstrated in a series of studies, where DEC205-targeted Ab-Ag fusion proteins induced tolerance to the Ag unless an agonistic anti-CD40 mAb was coadministered (4, 9, 10). Thus, agonistic mAbs against CD40 appears to be an effective substitute for Th cells in maturing DC. In this study, we have exploited this finding and designed two types of Ig-like vaccine constructs that agonistically target CD40 on APC for delivery of Ag. One type resembles Ab molecules and has Ag introduced as defined T cell epitopes in the C region. The other type carries larger parts of Ag expressing both B and T cell antigenic determinants. These CD40specific molecules elicited Ag-specific immune responses and induced protection in two different tumor models (myeloma and lymphoma).

Materials and Methods Institute of Immunology, University of Oslo and Rikshospitalet-Radiumhospitalet Medical Center, Oslo, Norway Received for publication April 21, 2006. Accepted for publication January 22, 2007. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was funded by the Research Council of Norway, Norwegian Cancer Society, and Multiple Myeloma Research Foundation. 2

K.W.S. and A.B.F. contributed equally to this work.

3

Address correspondence and reprint requests to Dr. Agnete B. Fredriksen, Institute of Immunology, University of Oslo and Rikshospitalet-Radiumhospitalet Medical Center, N-0027 Oslo, Norway; E-mail address: [email protected] or Dr. Bjarne Bogen, Institute of Immunology, University of Oslo and Rikshospitalet-Radiumhospitalet Medical Center, N-0027 Oslo, Norway; E-mail address: bjarne.bogen@ medisin.uio.no Abbreviations used in this paper: DC, dendritic cell; h␥3, human ␥3; NIP, 4-hydroxy-3-iodo-5-nitrophenylacetic acid; scFv, single-chain fragment variable; 3 [ H]Thd, [3H]thymidine; wt, wild type.

4

Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00 www.jimmunol.org

Construction of CD40-specific vaccine constructs Troybodies. CD40-specific troybodies were constructed essentially as described previously (11, 12). Rearranged V(D)J region genes of the H and L chain were PCR amplified from cDNA synthesized from mRNA extracted from FGK-45 cells (13), cloned, sequenced, and inserted into expression vectors encoding either a complete H chain (pLNOH2) or L chain (pLNO␬). Primers used were as follows (restriction enzyme sites are underlined): 5⬘VHCD40, ggtgtgcattcc gag gtg cag ctg gtg gag; 3⬘VHCD40, gacg tacgactcacc tga gga gac tgt gac cat gac; 5⬘VLCD40, ggtgtgcattcc gac act gta ctg acc cag tct cc; 3⬘VLCD40, gacgtacgttctactcacg ttt caa ttc cag ctt gg. The T cell epitope ␭2315 (91–101) from L chain of M315 (14) was introduced into the CH gene of pLNOH2 by site-specific mutagenesis (11, 15) exchanging loop 6 (loop F-G) of CH1 in the C region of human ␥3 (h␥3). The vector encoding the 4-hydroxy-3-iodo-5-nitrophenylacetic acid (NIP)-specific control Ab was as described previously (11). Vaccibodies. The CD40-specific V(D)J genes from the FGK-45 hybridoma were modified by PCR to provide linker sequences (bold). The primers were as follows: 5⬘VLCD40, as described above; 3⬘VLCD40, gcc aga gcc acc tcc gcc aga tcc gcc tcc acc ttt caa ttc cag ctt gg; 5⬘VHCD40, ggc gga ggt ggc tct ggc ggt ggc gga tcg gag gtg cag ctg gtg gag; 3⬘VHCD40, as described above. The VL and VH PCR products were combined to a single-chain

4170

TARGETING CD40 INDUCES PROTECTIVE IMMUNE RESPONSES

fragment variable (scFv) format by PCR SOEing and inserted into the targeting unit of the pLNOH2-vaccibody expression vectors encoding a dimerization unit (hinge and CH3 from hIgG3) and an antigenic unit (scFv315 or scFvA20) as described previously (16). The vectors encoding the NIP-specific controls were as described previously (16).

Production of vaccine protein Transient and stable transfections were performed in HEK293E cells essentially as described previously (16). All recombinant Abs were affinity purified from cell supernatant by use of anti-human C␬ Ab (A8B5)-conjugated, NP-lysin conjugated, or by DNP-lysin-conjugated Sepharose columns.

Mice, cells, and mAbs BALB/cABom (H-2d) mice were obtained from Taconic Farms. The ␭2315specific TCR-transgenic mice on a BALB/c background (17) and the ␭2315specific TCR-transgenic mice on a C.B-17 scid⫺/⫺ background have been described previously (18). The studies have been reviewed and approved by the National Animal Research Authority. NS0, HEK293E, and MOPC315 cells (19) were obtained from American Type Culture Collection (ATCC). The A20 B lymphoma cell line (20) was a gift from S. Buus (Institute for Medical Microbiology and Immunology, Copenhagen, Denmark). The ␭2315-specific CD4⫹ T cell clone 7A10B2 (21) has been described previously. Rat anti-mouse CD40 FGK-45 hybridoma (13) was a gift from J. Andersson (Basel Institute of Immunology, Basel, Switzerland), and the immature DC cell line (D2SC/1) (22) was provided by P. Ricciardi-Castagnoli (University of Milan-Bicocca, Milan, Italy).

Immediately following injection, electroporation was performed using a caliper electrode as described previously (23).

In vivo detection of primed APC in draining lymph nodes BALB/c mice were injected i.m. with 50 ␮g of plasmids and electroporated. Eight days later, draining (lumbar and sacral) and nondraining (mesenteric) lymph node cells were treated with collagenase and DNase, irradiated (20 Gy), and 5 ⫻ 105 cells/well were incubated with polarized ␭2315-specific Th2 cells (2 ⫻ 104). After 48 h, the cultures were pulsed with 1 ␮Ci of [3H]Thd. Incorporated [3H]Thd was measured after 16 h.

Measure T cell proliferation in vivo Vaccibodies. DNA-vaccinated BALB/c mice were adoptively transferred with 1.5 ⫻ 107 lymph node cells from ␭2315-specific TCR transgenic mice, and BrdU incorporation was performed as described previously (16). Troybodies. Adoptively transferred BALB/c mice (1.5 ⫻ 107 lymph node cells) were protein vaccinated (s.c. injection on right flank, 100 ␮g/mouse), and BrdU incorporation was performed as described previously (12).

IFN-␥ ELISPOT

Spleen cells were stained with CD40-specific vaccine proteins, or NIPspecific control proteins, and detected with anti-human IgG3 (troybodies) (Southern Biotechnology Associates) or HP6017 (vaccibodies) (ATCC). allophycocyanin-conjugated CD19 (BD Pharmingen) was used to stain B cells.

Adoptively transferred BALB/c mice (3 ⫻ 106 TCR transgenic CD4⫹ T cells/mouse) were protein vaccinated (s.c. injection on right flank, 100 ␮g/mouse) or DNA vaccinated. After 3 wk, spleens were removed and the number of IFN-␥ secreting, peptide-specific T cells in fresh splenocyte preparations was determined by ELISPOT essentially as described previously (23). Briefly, 96-well nitrocellulose plates (MultiScreen; Millipore) were coated with mAbs specific for mouse IFN-␥ (AN-18). Splenocytes were plated in triplicates (1 ⫻ 106 cells/well) and synthetic peptide (␭2315 aa 89 –107) was added. Biotinylated anti-mouse IFN-␥ mAb (XMG1.2) and streptavidin alkaline phosphatase (Sigma Chemical) was used for detection. Spots were developed by adding alkaline phosphatase conjugate substrate solution (50 ␮l) (Bio-Rad). Spots were counted electronically with a Zeiss KS-ELISPOT-401 instrument. The number of spots per 105 splenocytes is shown. Each point represents two mice.

CD69 expression

Tumor challenge experiments

The ␭2315-specific TCR transgenic mice on a SCID background were injected i.m. with 50 ␮g of naked DNA and electroporated. At day 10 after vaccination, draining and nondraining lymph node CD4⫹ T cells were analyzed for CD69 expression. Cells were analyzed on a FACSCalibur (BD Biosciences).

BALB/c mice were immunized by naked DNA as described above. Two weeks later, mice were injected s.c. on the right flank with MOPC315 myeloma cells (1.6 ⫻ 105) or A20 B cell lymphoma cells (1.2 ⫻ 105). Ten mice were included in each group and tumor growth was observed biweekly. Tumor avoidance curves and statistical analyses were generated with GraphPad Prism 4.0 software.

Flow cytometry

Agonistic ligation of CD40 on splenic B cells Splenocytes (1 ⫻ 105) were cultured with CD40-specific troybodies or vaccibodies, the corresponding nontargeted controls, or FGK-45 (all 10 ␮g/ml) in medium containing IL-4 (20 U/ml) for 3 days. Incorporation of [3H]thymidine ([3H]Thd) was measured for the last 16 h.

Maturation of DC and IL-12 secretion The immature splenic DC line D2SC/1 cells (22) or immature bone marrow-derived DC cultured with IL-4 and GM-CSF were incubated with CD40-specific troybodies, the corresponding nontargeted troybody (10 ␮g/ ml), LPS (10 ng/ml), or medium alone for 48 h before the cells were stained with Abs specific for MHC class II, CD86, or CD54 and CD11c (BD Pharmingen). Secretion of IL-12p40 from bone marrow DC incubated for 24 h with CD40-specific troybodies, the corresponding nontargeted troybody (10 ␮g/ml), LPS (10 ng/ml), or medium alone was measured by ELISA (OptEIA; BD Biosciences).

ELISA Wells were coated with NIP-BSA or recombinant mouse CD40/Fc Ig chimeric protein (R&D Systems) and detected with biotinylated HP6017 (antihuman IgG, Fc-region) (Zymed Laboratories). Anti-Id315 Ab was detected in ELISA as described previously (16).

In vitro T cell proliferation T cell proliferation assays were performed essentially as described (12) using either polarized ␭2315-specific Th2 cells or cloned ␭2315-specific CD4⫹ T cells (7A10B2) (both 2 ⫻ 104 cells/well) as responder cells and irradiated BALB/c splenocytes (5 ⫻ 105/well) as APC.

DNA immunization Naked DNA immunization was performed essentially as described previously (16). As a negative control, mice were injected with NaCl alone.

Results Mouse CD40-specific vaccine constructs bind agonistically to CD40 We engineered two different vaccine constructs, both based on Ab structure. troybodies (11) are recombinant Ab molecules with V regions that target surface molecules on APC (herein CD40). In addition, troybodies carry T cell epitopes introduced into a loop connecting ␤-strands in Ig-C domains (Fig. 1A and Table I). As negative controls we used nontargeted troybodies with V regions specific for the hapten NIP. These should not be able to bind to surface molecules on APC, but still harbor the ␭2315 T cell epitope in their C regions. In addition, CD40-specific Ab molecules lacking the epitope (wild-type (wt) h␥3) were used (Table I). Vaccibodies (16) are homodimeric fusion proteins, each chain consisting of a scFv that target APC (herein CD40), a dimerization motif (shortened hinge and CH3 of h␥3), and an antigenic unit (herein scFv from MOPC315 plasmacytoma or A20 B cell lymphoma) (Fig. 1A and Table I). Nontargeted vaccibodies with specificity for the hapten NIP, and with the corresponding antigenic units as described above (Table I), served as negative controls. We first investigated whether the cloned V regions from the FGK-45 hybridoma were able to bind to CD40 after they had been cloned and produced as proteins in the troybody and vaccibody formats. Splenocytes were stained with CD40-specific proteins or the nontargeted NIP-specific controls. Indeed, the CD40-specific


4171 specific vaccine constructs, the corresponding nontargeted NIPspecific controls, FGK-45 mAb, or isotype-matched control Ab (rat IgG1). CD40-specific proteins induced a strong proliferation of splenocytes that was comparable to that of the FGK-45 mAb, whereas nontargeted constructs and isotype-matched rat IgG1 failed to do so (Fig. 2A). Ability of CD40-specific vaccine proteins to induce DC maturation was examined by using an immature DC line (D2SC/1) as well as bone marrow-derived DC. CD40-specific troybodies triggered up-regulation of MHC class II, CD86, and CD54 to a level similar to that achieved by LPS (Fig. 2B). By contrast, the nontargeted NIP-specific proteins had no effect compared with medium alone. Furthermore, incubation of bone marrow-derived DC with the CD40-specific protein induced secretion of IL-12 (Fig. 2C). These results indicate that the agonistic activity previously reported for the FGK-45 mAb was maintained when transposing the V regions to the CD40-specific vaccine proteins. Targeting Ag to CD40 enhances in vitro proliferation of Ag-specific CD4⫹ T cells and requires a physical link between V regions and Ag

FIGURE 1. A, Schematic figure of mouse CD40-specific vaccine molecules; troybodies (left) and vaccibodies (right). Both vaccine constructs are based on Ig structure and target CD40, either via V regions on H and L chains (troybodies) or scFv (vaccibodies) (light shade). The Ag is in troybodies, a defined T cell epitope (dark star), whereas vaccibodies carry complete scFv derived from Ig produced by B cell tumors (dark shade). troybodies include h␥3 and human C␬, whereas vaccibodies include a shortened hinge and CH3 domain from h␥3 (open) that cause disulphide bond formation and homodimerization. See Table I for details. B, Staining of CD19⫹ splenocytes with CD40-specific troybodies (left) and vaccibodies (right) (filled histograms), or corresponding nontargeted NIP-specific vaccine proteins (open diagrams). C, Binding of CD40-specific or NIPspecific troybodies and vaccibodies to CD40-Ig fusion protein (left) or NIP-BSA (right) in ELISA.

molecules bound much better to gated CD19⫹ B cells expressing CD40 (Fig. 1B) compared with the nontargeted controls. Furthermore, the CD40-specific proteins bound to CD40-Ig fusion protein but not NIP-BSA in ELISA, whereas the converse was true for the NIP-specific controls (Fig. 1C). Ligation of CD40 has been reported to induce maturation of DC (3). Moreover, the anti-CD40 FGK-45 mAb has previously been reported in several different studies to be agonistic, i.e., induce proliferation of spleen cells (13), up-regulate CD86 and CD40 expression on ex vivo DC from lymph nodes (4), and secretion of IL-12 and IFN-␥ from lymph node cells (24). To examine whether the cloned CD40-specific V regions had maintained their ability to agonistically bind to CD40, we incubated splenocytes with CD40-

In the troybody molecules, the CD4⫹ ␭2315 T cell epitope was introduced into the C region of h␥3, whereas in the vaccibody molecules the ␭2315 CD4⫹ T cell epitope is located in its original position in the CDR3 loop of the VL region in the scFv315 fragment. Both vaccine proteins were tested in in vitro T cell proliferation assays. Irradiated BALB/c splenocytes were used as APC, and either polarized ␭2315-specific Th2 cells from TCR transgenic mice (Fig. 3A, upper panel) or cloned ␭2315-specific Th1 cells (Fig. 3A, lower panel) were used as responders. The dose-response curves demonstrate that the CD40-specific vaccine proteins were 100 –1,000 times more efficient at activating Ag-specific CD4⫹ T cells when compared with the corresponding nontargeted (NIPspecific) controls or synthetic peptide, respectively (Fig. 3A). The results above could be explained by activation of DC by agonistic CD40 stimulation, increased loading of MHC class II molecules by targeting Ag to CD40, or both. To distinguish between these possibilities, we did an experiment in which cultures received a mixture of anti-CD40 Ab lacking the T cell epitope in the C region and NIP-specific control Ab with the ␭2315 epitope. In this experimental situation, the CD40 specificity and the T cell epitope resided on separate molecules and were thus unlinked. Such physical uncoupling of CD40 specificity and T cell epitope abolished T cell responses. These results demonstrate a requirement for a physical link between CD40-specific V regions and the T cell epitope for enhancement of T cell responses (Fig. 3B). The results of Fig. 3A could be extended to human cells because a mouse mAb to human CD40 (clone 5C3) was ⬃100 –1,000 times more efficient at stimulating a mouse C␬40⫺8-specific/DR4-restricted CD4⫹ T cell clone (25) than was isotype-matched IgG1,␬ mAb (data not shown). Targeting CD40 in vivo induces priming of APC and activation and proliferation of Ag-specific CD4⫹ T cells To investigate whether CD40-specific vaccine proteins are in fact delivered to APC in vivo, DNA encoding CD40-specific vaccibodies with scFv315 were injected i.m. in the quadriceps of BALB/c mice, immediately followed by electroporation to increase uptake and expression of plasmids (16, 26). Eight days after injection, the draining (sacral and lumbar) and nondraining lymph nodes were removed and used as APC in an in vitro proliferation assay using polarized ␭2315-specific Th2 cells as responders. The results show that CD40-specific vaccibodies primed APC in the

4172

TARGETING CD40 INDUCES PROTECTIVE IMMUNE RESPONSES Table I. Vaccine constructs used in this studya Nameb

Vaccibodies (FvCD40Fv315)2 (FvCD40FvA20)2 (FvNIPFv315)2 (FvNIPFvA20)2 Troybodies ␣CD40.L6-␭2315 ␣NIP.L6-␭2315 Control Ab ␣CD40.wt FGK-45

Specificityc

C Regiond

CD40 CD40 NIP NIP

Hinge ⫹ CH3 h␥3 Hinge ⫹ CH3 h␥3 Hinge ⫹ CH3 h␥3 Hinge ⫹ CH3 h␥3

CD40 NIP CD40 CD40

Age

MHC Restrictionf

Reference

M315 A20 M315 A20

I-Ed Kd I-Ed Kd

Present study Present study 16 16

h␥3 h␥3

␭2315(91–101) ␭2315(91–101)

I-Ed I-Ed

Present study 11

h␥3 Rat ␥1

None None

Present study 13

a

See Fig. 1A for schematic overview. The names indicate the specificity and the Ag introduced. L6 denotes that the T cell epitope replaced loop 6 (loop F-G) connecting ␤-strands in the CH1 domain of h␥3. ()2 indicates homodimers. c CD40-specific V regions were cloned from hybridoma FGK-45 (13). NIP-specific V regions corresponds to that of the hybridoma B1-8 (34). d The homodimerization motif is composed of hinge exon 1 and 4 and the CH3 domain from h␥3 C region (16). Troybodies have complete h␥3 C region. e The Ag in vaccibodies were cloned as scFv corresponding to monoclonal Ig produced by MOPC315 myeloma (19) or the A20 B cell lymphoma (20). The T cell epitope in troybodies were from ␭2315 L chain (14) and replaced the 4-aa sequence naturally occurring in L6 of h␥3 (15, 35). Amino acids are indicated. f MHC restriction of defined idiotypic T cell epitopes (aa 91–101 of the V␭2315 fragment of M315 (17), aa 106 –114 of VH of A20 (36)) is indicated. b

draining, but not in the nondraining lymph node (Fig. 4A). By contrast, nontargeted (NIP-specific) vaccibodies failed to prime APC. A likely explanation for these data is that transfected muscle cells synthesize and secrete CD40-specific vaccibodies that target

FIGURE 2. Cloned CD40-specific V regions in troybodies and vaccibodies bind agonistically and induce maturation of DC. A, Splenocytes were cultured with CD40-specific vaccine constructs or controls in medium supplemented with IL-4. Proliferation (incorporation of [3H]Thd) was measured. B, The immature DC line D2SC/1 (left) or immature bone marrowderived DC (middle and right) were incubated with CD40-specific troybodies (thick line), NIP-specific troybodies (thin line), LPS (hatched line), or medium alone (filled) for 48 h before staining for MHC class II, CD54 and CD86 expression on CD11c-gated cells. C, Cultured immature bone morrow-derived DC were incubated for 24 h with the indicated CD40specifc troybodies and controls before IL-12p40 was measured.

APC. The primed APC evidently stimulated T cells because in TCR transgenic mice on a SCID background, where virtually all T cells express a ␭2315-specific TCR, a major fraction of the CD4⫹ T cells expressed the T cell activation marker CD69 10 days after DNA injection of the CD40-specific construct. By contrast, the nontargeted vaccine (NIP-specific) did not induce expression of CD69 (Fig. 4B). We proceeded to investigate whether CD40-specific vaccine constructs could induce proliferation of CD4⫹ T cells in vivo. First, CD40-specific vaccibody constructs with scFv315 were tested. BALB/c mice were immunized with naked DNA encoding CD40-specific vaccine construct, or nontargeted controls, and electroporated. On day 7, mice were adoptively transferred with ␭2315-specific TCR transgenic lymph node cells so that Ag-specific T cells constituted ⬃1–2% of total CD4⫹ T cells in the recipient. On day 10, BrdU was given and the mice were sacrificed and analyzed on day 13. We found that in mice immunized with CD40-specific vaccibodies, ⬎66% of the ␭2315-specific CD4⫹ T cells had incorporated BrdU in draining lymph nodes by day 14, whereas mice immunized with NIP-specific vaccibodies, or CD40specific vaccibodies with an irrelevant scFv in the antigenic unit, showed negligible incorporation of BrdU compared with reconstituted mice injected with NaCl alone (Fig. 4C). We also tested whether CD40-specific troybody injected as protein in PBS without adjuvant could induce T cell activation. Indeed, s.c. injection of 100 ␮g of CD40-specific troybody protein in BALB/c mice adoptively transferred with ␭2315-specific TCR transgenic lymph node cells induced a significant incorporation of BrdU into ␭2315-specific CD4⫹ T cells in draining lymph nodes 6 days after protein immunization. Mice that had been injected with the corresponding nontargeted NIP-specific troybodies or CD40specific Ab lacking the epitope showed a much lower incorporation of BrdU (Fig. 4D). When titrated amounts of CD40-specific troybodies were injected, the percentage of BrdU-positive ␭2315specific CD4⫹ T cells gradually decreased, as would be expected (data not shown). These findings demonstrate that CD40-specific vaccine constructs induce potent T cell activation in vivo, either when delivered as naked plasmid DNA or as purified protein. CD40-specific proteins induce memory T cell responses The ␭2315 epitope is known to be a very weak Ag (21), and DNA vaccination with CD40-specific vaccibodies induced only minor


FIGURE 3. CD40-specific vaccine proteins to CD40 enhance proliferation of Ag-specific CD4⫹ T cells in vitro: a physical linkage between the targeting V regions and the T cell epitope is required. A, Irradiated BALB/c splenocytes were cultured with titrated amounts of CD40-specific troybodies (left column), CD40-specific vaccibodies (right column), the corresponding nontargeted controls, or ␭2315 synthetic peptide. Polarized ␭2315specific Th2 cells (upper row, proliferation) or cloned ␭2315-specific CD4⫹ Th1 cells (lower row, IFN-␥ production) were used as responder T cells. B, Irradiated BALB/c splenocytes were cultured with titrated amounts of CD40-specific troybodies, the corresponding nontargeted troybodies, CD40-specific Ab with wt h␥3 C region, or a mixture of both the nontargeted troybodies and the CD40-specific Ab with wt h␥3 C region (unlinked condition). Cloned ␭2315-specific CD4⫹ Th1 cells were used as responder T cells.

responses in normal mice (Fig. 5A). Therefore, to study whether CD40-specific molecules could induce memory T cell responses, we resorted to mice adoptively transferred with lymph node cells from ␭2315-specific TCR transgenic mice so that Ag-specific T cells constituted ⬃1–2% of total CD4⫹ T cells in the recipient. Adoptively transferred BALB/c mice were either protein-injected s.c. with CD40-specific troybodies with ␭2315 epitope, the corresponding nontargeted NIP-specific troybodies, CD40-specific Ab without the epitope (␣CD40.wt), or PBS alone (Fig. 5B, left), or DNA immunized i.m. with CD40-specific vaccibodies or controls (Fig. 5B, right). Three weeks later, the spleens were removed. Upon restimulation in vitro with ␭2315 peptide, the number of IFN␥-producing splenocytes was significantly increased when mice

4173

FIGURE 4. CD40-specific vaccine constructs prime APC with Ag in vivo and induce activation and proliferation of Ag-specific CD4⫹ T cells. A, BALB/c mice were injected i.m. with plasmids encoding CD40-specific vaccibodies with scFv315, the corresponding nontargeted vaccibodies, CD40-specific vaccibodies with irrelevant scFvA20, or NaCl. Muscle cells were immediately electroporated after injection. After 10 days, draining and nondraining lymph nodes were removed, irradiated, and used as APC in an in vitro T cell proliferation assay using polarized ␭2315-specific Th2 cells as responders. B, ␭2315-specific TCR-transgenic SCID mice were immunized as described in A. On day 10, draining lymph nodes were stained for CD69 expression. C, BALB/c mice were immunized as described in A, adoptively transferred (day 7) with lymph node cells from ␭2315-specific TCR transgenic mice, and given BrdU (day 11). Three days later, draining lymph nodes were analyzed for incorporation of BrdU into gated ␭2315specific GB113⫹F23.1⫹CD4⫹ T cells. D, BALB/c mice adoptively transferred with lymph node cells from ␭2315-specific TCR transgenic mice were injected s.c. with 100 ␮g of CD40-specific troybody protein expressing the ␭2315 epitope, the corresponding nontargeted troybodies, CD40specific control Ab with wt h␥3 C region, or PBS alone. The mice were given BrdU and 6 days later, gated ␭2315-specific CD4⫹ T cells in draining lymph nodes were analyzed for incorporation of BrdU as described in C.

had been immunized with CD40-specific vaccine constructs compared with the controls (Fig. 5B). Thus, targeting Ag to CD40 in vivo induced memory T cell responses in contrast to delivery of a nontargeted vaccine. CD40-specific DNA vaccines induce Abs specific for tumor Ag The current vaccibodies contain tumor-specific scFv from monoclonal Ig derived either from a mouse multiple myeloma (MOPC315) or lymphoma (A20) as antigenic units. Such scFv express serologically defined Id determinants in addition to T cell epitopes. To investigate whether CD40-specific vaccibodies could induce production of tumor-specific anti-Id Ab, normal BALB/c

4174


FIGURE 6. CD40-specific vaccines induce protection against multiple myeloma and B cell lymphoma. BALB/c mice were immunized with naked DNA encoding CD40-specific vaccibodies with scFv antigenic units from MOPC315 or A20, the corresponding nontargeted controls, or NaCl alone, in combination with electroporation (10 mice in each group). Two weeks later, the mice were challenged with a lethal dose of MOPC315 myeloma cells (A) or A20 B cell lymphoma cells (B) s.c.

that because the troybody molecules only express short T cell epitopes, their ability to induce Ab was not investigated. CD40-specific DNA constructs induce protection against the MOPC315 myeloma and the A20 B cell lymphoma

FIGURE 5. CD40-specific vaccines induce Th1 memory T cells and Ab responses. A, BALB/c mice were immunized i.m. with DNA encoding CD40-specific vaccibodies. After 3 wk, spleens were removed and restimulated with ␭2315 peptide in an IFN-␥-specific ELISPOT assay. B, BALB/c mice were adoptively transferred with ␭2315-specific TCR transgenic CD4⫹ T cells and either injected s.c. with a single dose of CD40-specific troybody protein expressing the ␭2315 epitope, the corresponding nontargeted troybodies, CD40-specific troybodies with wt h␥3 C region, or PBS alone (left). Alternatively, BALB/c mice were DNA immunized i.m. with CD40-specific vaccibodies with scFv315, NIP-specific vaccibody with scFv315, CD40-specific vaccibody with irrelevant Ag (scFvA20), or NaCl (right). IFN-␥-producing spleen cells were detected in ELISPOT. ⴱ, Indicate statistically significant differences (p ⬍ 0.01). C, BALB/c mice were injected i.m. with naked DNA encoding CD40-specific vaccibodies with either scFv315 or scFvA20, nontargeted controls, or NaCl alone, followed by electroporation. Levels of serum Ab against the M315 Id (anti-Id315) were measured on the indicated time points (six mice in each group).

mice where injected i.m. with DNA-encoding CD40-specific vaccibodies with either scFv315 or scFvA20. Mice injected with CD40specific vaccibodies containing scFv315 produced high amounts of tumor-specific anti-Id Ab. Mice injected with the corresponding nontargeted vaccibody, or with anti-CD40 vaccibody with scFvA20, produced no anti-Id315 Ab (Fig. 5C). On day 28, Abs were predominately of IgG1 isotype, while at later time points IgG1 and IgG2a were equally expressed (data not shown). Note

Next, we wanted to test whether immunization with CD40-specific vaccine constructs could induce protection against challenges with cancer cells. Multiple myeloma and B lymphoma cells express monoclonal Ig with V regions that function as tumor-specific Ag called Id. Normal BALB/c mice were injected i.m. with naked DNA encoding CD40-specific vaccibodies with the scFv from the MOPC315 myeloma, the nontargeted control, CD40-specific vaccibody with scFvA20, or NaCl alone. All mice were electroporated. Two weeks after immunization, the mice were challenged with a lethal dose of MOPC315 cells s.c. Immunization with the CD40specific DNA vaccine expressing scFv315 induced protection in ⬎50% of the mice, whereas all mice in the other groups developed tumors within 25 days (Fig. 6A). In a second tumor model, similar groups of immunized mice were challenged with A20 lymphoma cells. The results (Fig. 6B) show that ⬎50% of the mice immunized with CD40-specific DNA vaccine with scFvA20 were protected, whereas mice injected with vaccine constructs not targeting CD40, or expressing the irrelevant scFv315 Ag, were not (Fig. 6B). Thus, CD40-specific vaccines with tumor-specific scFv induced Id-specific tumor resistance.

Discussion Ligation of CD40 is essential for activation of DC (3) and may release immature DC from suppression by CD4⫹CD25⫹ regulatory T cells (27). Under physiological conditions, CD40 is bound by CD40L expressed on activated CD4⫹ T cells. However, agonistic anti-CD40 mAb can substitute for CD40L. Thus, administration of agonistic anti-CD40 mAb in combination with Ag

The Journal of Immunology loaded via DEC-205 on DC resulted in memory and protective CD4⫹ and CD8⫹ T cells responses, whereas in the absence of anti-CD40 mAb, tolerance was induced (4, 9). In these previous studies, anti-CD40 mAb and delivery of Ag to DEC205 was unlinked. In this study, we show that a single molecule can deliver both an activation signal through CD40, as well as the Ag to APC. This strategy induced potent Ab and memory T cell responses, and protective immunity in two tumor models. The Ig-like vaccine constructs used in this study target CD40 by agonistic V regions cloned from the FGK-45 hybridoma (13). The current two types of recombinant Ab-based vaccines differ, however, in their ability to bind to FcRs. The troybody molecule expresses a complete C region of h␥3 and binds to FcRs (28). In contrast, the vaccibody molecule lacks the CH2 domain, and therefore the FcR binding site, and should only enter APC via their V regions. So, because vaccibodies were efficient in the present experiments, targeting of CD40 is efficient in itself and does not require simultaneous engagement of FcRs on APC. Our experiments demonstrated a need for physical linkage between the Ag and the CD40-specific V regions to activate Agspecific T cells in vitro. This finding is in agreement with previous findings demonstrating that efficiency of vaccine molecules consisting of scFv linked to chemokines (i.e., MCP, IFN-␥-inducible protein-10) is dependent on a physical link between the Ag (scFv) and targeting unit (chemokine) (29). However, the requirement for a physical linkage between the Ag and the maturation signal in the present experiments superficially appears to contradict recent in vivo studies demonstrating induction of long-lasting T cell responses even when anti-CD40 mAb was not linked to the Ab-Ag fusion protein (4, 9, 10). The difference might be explained by the fact that in the latter case, the recombinant Ab-Ag fusion was targeted to DEC205, an efficient endocytic receptor expressed on DC (30), but lacking the ability to induce a maturation signal. Thus, linkage between the CD40-targeting unit and the Ag might only be required if the Ag is otherwise inefficiently endocytosed by APC, as was the case in the design of the present study. CD40-specific recombinant Ab-based vaccines were effective not only as protein, but also as naked DNA plasmids delivered i.m. followed by electroporation. Electroporation enhances transfection of muscle cells (26, 31) and results in higher levels of vaccine constructs in serum, most likely due to increased secretion by muscle cells (16). Electroporation might also inflict some tissue damage to muscle cells and influx of MHC class II⫹ cells (32). Presumably, CD40-specific vaccibodies secreted by muscle cells target muscle-infiltrating APC, drain to lymph nodes, and stimulate T cells. This scenario is consistent with the present demonstration that DNA injection i.m. in combination with electroporation results in Ag-primed APC, as well as proliferating T cells, in draining lymph nodes. CD40-specific vaccibody DNA with scFv from B cell tumors induced potent Ab and CD4⫹ T cell responses against tumor-specific Ig V region Id, and protection against tumor challenges both in a myeloma and a lymphoma model, whereas the corresponding nontargeted vaccine construct failed to do so. Anti-Id Ab could be of significance for tumor protection in the A20 model where B lymphoma cells express monoclonal Ig in the cell membrane. However, in the MOPC315 myeloma model, where huge amounts of myeloma protein is secreted, and where tumor cells have little surface Ig in the membrane, anti-Id Ab is expected to play a minor role in protection. More likely, activation of Id-specific CD4⫹ T cells is of greater importance for rejection of MOPC315 myeloma cells (33). The present tumor vaccination experiments were performed in a prophylactic setting, because the rapid growth of the MOPC315

4175 plasmacytoma and A20 lymphoma tumors used are major obstacles to therapeutic vaccination of tumor-bearing mice. Nevertheless, the results suggest that targeting tumor Ag to CD40 on APC could be a valuable approach to eradicate minimal residual disease after conventional treatment, e.g., after highdose chemotherapy and reconstitution of the immune system by bone marrow transplantation. The Ab-based vaccine constructs are quite flexible molecules. troybodies have been successfully equipped with T cell epitopes in several different loops both in CH1, CH2, and CH3 domains (M. Flobakk, J. B. Rasmussen, E. Lunde, G. Berntzen, T. E. Michaelsen, B. Bogen, and I. Sandlie, manuscript in preparation). Moreover, epitopes up to 37 aa, proline-rich gluten epitopes, and tandemly linked T cell epitopes have been successfully incorporated (G. Tunheim, K. W. Schjetne, I. B. Rasmussen, L. M. Sollid, I. Sandlie, and B. Bogen, submitted for publication). As for vaccibodies, they can in their antigenic unit express large Ag-like scFv from myeloma patients (M. Frøyland, K. M. Thompson, T. GeddeDahl, A. B. Fredriksen, and B. Bogen, submitted for publication), fragment C of tetanus toxin (G. Tunheim, K. M. Thompson, A. B. Fredriksen, T. Espevik, K. W. Schjetne, and B. Bogen, submitted for publication), hemagglutinin from influenza virus (G. Tunheim, H. von Boehmer, and B. Bogen, unpublished data), and gp120 of HIV and streptavidin (A. B. Fredriksen, K. M. Thompson, D. Barouch, and B. Bogen, unpublished data). From our current work described above, it appears that there is a large potential for which Ag that the vaccine protein can accommodate.

Acknowledgments We thank Hilde Omholt, Peter Hofgaard, Mona Lindeberg, and Tom-Ole Løvås for excellent technical assistance.

Disclosures B. Bogen and A. B. Fredriksen are inventors on a pending patent on vaccibodies, and B. Bogen is inventor on a patent on troybodies.

References 1. Banchereau, J., and R. M. Steinman. 1998. Dendritic cells and the control of immunity. Nature 392: 245–252. 2. Steinman, R. M., and M. C. Nussenzweig. 2002. Avoiding horror autotoxicus: the importance of dendritic cells in peripheral T cell tolerance. Proc. Natl. Acad. Sci. USA 99: 351–358. 3. Cella, M., D. Scheidegger, K. Palmer-Lehmann, P. Lane, A. Lanzavecchia, and G. Alber. 1996. Ligation of CD40 on dendritic cells triggers production of high levels of interleukin-12 and enhances T cell stimulatory capacity: T-T help via APC activation. J. Exp. Med. 184: 747–752. 4. Hawiger, D., K. Inaba, Y. Dorsett, M. Guo, K. Mahnke, M. Rivera, J. V. Ravetch, R. M. Steinman, and M. C. Nussenzweig. 2001. Dendritic cells induce peripheral T cell unresponsiveness under steady state conditions in vivo. J. Exp. Med. 194: 769 –779. 5. Grewal, I. S., and R. A. Flavell. 1998. CD40 and CD154 in cell-mediated immunity. Annu. Rev. Immunol. 16: 111–135. 6. Ridge, J. P., R. F. Di, and P. Matzinger. 1998. A conditioned dendritic cell can be a temporal bridge between a CD4⫹ T-helper and a T-killer cell. Nature 393: 474 – 478. 7. Bennett, S. R., F. R. Carbone, F. Karamalis, R. A. Flavell, J. F. Miller, and W. R. Heath. 1998. Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature 393: 478 – 480. 8. 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-CD40L interactions. Nature 393: 480 – 483. 9. Bonifaz, L., D. Bonnyay, K. Mahnke, M. Rivera, M. C. Nussenzweig, and R. M. Steinman. 2002. Efficient targeting of protein antigen to the dendritic cell receptor DEC-205 in the steady state leads to antigen presentation on major histocompatibility complex class I products and peripheral CD8⫹ T cell tolerance. J. Exp. Med. 196: 1627–1638. 10. Bonifaz, L. C., D. P. Bonnyay, A. Charalambous, D. I. Darguste, S. Fujii, H. Soares, M. K. Brimnes, B. Moltedo, T. M. Moran, and R. M. Steinman. 2004. In vivo targeting of antigens to maturing dendritic cells via the DEC-205 receptor improves T cell vaccination. J. Exp. Med. 199: 815– 824. 11. Lunde, E., L. A. Munthe, A. Vabø, I. Sandlie, and B. Bogen. 1999. Antibodies engineered with IgD specificity efficiently deliver integrated T-cell epitopes for antigen presentation by B cells. Nat. Biotechnol. 17: 670 – 675.

4176


12. Lunde, E., K. H. Western, I. B. Rasmussen, I. Sandlie, and B. Bogen. 2002. Efficient delivery of T cell epitopes to APC by use of MHC class II- specific troybodies. J. Immunol. 168: 2154 –2162. 13. Rolink, A., F. Melchers, and J. Andersson. 1996. The SCID but not the RAG-2 gene product is required for S␮-S␧ heavy chain class switching. Immunity 5: 319 –330. 14. Bogen, B., and J. D. Lambris. 1989. Minimum length of an idiotypic peptide and a model for its binding to a major histocompatibility complex class II molecule. EMBO J. 8: 1947–1952. 15. Rasmussen, I. B., E. Lunde, T. E. Michaelsen, B. Bogen, and I. Sandlie. 2001. The principle of delivery of T cell epitopes to antigen-presenting cells applied to peptides from influenza virus, ovalbumin, and hen egg lysozyme: implications for peptide vaccination. Proc. Natl. Acad. Sci. USA 98: 10296 –10301. 16. Fredriksen, A. B., I. Sandlie, and B. Bogen. 2006. DNA vaccines increase immunogenicity of idiotypic tumor antigen by targeting novel fusion proteins to antigen-presenting cells. Mol. Ther. 13: 776 –785. 17. Bogen, B., L. Gleditsch, S. Weiss, and Z. Dembic. 1992. Weak positive selection of transgenic T cell receptor-bearing thymocytes: importance of major histocompatibility complex class II, T cell receptor and CD4 surface molecule densities. Eur. J. Immunol. 22: 703–709. 18. Bogen, B., L. Munthe, A. Sollien, P. Hofgaard, H. Omholt, F. Dagnaes, Z. Dembic, and G. F. Lauritzsen. 1995. Naive CD4⫹ T cells confer idiotypespecific tumor resistance in the absence of antibodies. Eur. J. Immunol. 25: 3079 –3086. 19. Lynch, R. G., R. J. Graff, S. Sirisinha, E. S. Simms, and H. N. Eisen. 1972. Myeloma proteins as tumor-specific transplantation antigens. Proc. Natl. Acad. Sci. USA 69: 1540 –1544. 20. Kim, K. J., C. Kanellopoulos-Langevin, R. M. Merwin, D. H. Sachs, and R. Asofsky. 1979. Establishment and characterization of BALB/c lymphoma lines with B cell properties. J. Immunol. 122: 549 –554. 21. Bogen, B., B. Malissen, and W. Haas. 1986. Idiotope-specific T cell clones that recognize syngeneic immunoglobulin fragments in the context of class II molecules. Eur. J. Immunol. 16: 1373–1378. 22. Granucci, F., G. Girolomoni, M. B. Lutz, M. Foti, G. Marconi, P. Gnocchi, L. Nolli, and P. Ricciardi-Castagnoli. 1994. Modulation of cytokine expression in mouse dendritic cell clones. Eur. J. Immunol. 24: 2522–2526. 23. Tollefsen, S., T. Tjelle, J. Schneider, M. Harboe, H. Wiker, G. Hewinson, K. Huygen, and I. Mathiesen. 2002. Improved cellular and humoral immune responses against Mycobacterium tuberculosis antigens after intramuscular DNA immunisation combined with muscle electroporation. Vaccine 20: 3370 –3378. 24. French, R. R., H. T. Chan, A. L. Tutt, and M. J. Glennie. 1999. CD40 antibody evokes a cytotoxic T-cell response that eradicates lymphoma and bypasses T-cell help. Nat. Med. 5: 548 –553.

25. Schjetne, K. W., K. M. Thompson, T. Aarvak, B. Fleckenstein, L. M. Sollid, and B. Bogen. 2002. A mouse C␬-specific T cell clone indicates that DC-SIGN is an efficient target for antibody-mediated delivery of T cell epitopes for MHC class II presentation. Int. Immunol. 14: 1423–1430. 26. Tjelle, T. E., A. Corthay, E. Lunde, I. Sandlie, T. E. Michaelsen, I. Mathiesen, and B. Bogen. 2004. Monoclonal antibodies produced by muscle after plasmid injection and electroporation. Mol. Ther. 9: 328 –336. 27. Serra, P., A. Amrani, J. Yamanouchi, B. Han, S. Thiessen, T. Utsugi, J. Verdaguer, and P. Santamaria. 2003. CD40 ligation releases immature dendritic cells from the control of regulatory CD4⫹CD25⫹ T cells. Immunity 19: 877– 889. 28. Schjetne, K. W., J. E. Thommesen, A. B. Fredriksen, E. Lunde, I. Sandlie, and B. Bogen. 2005. Induction of central T cell tolerance: recombinant antibodies deliver peptides for deletion of antigen-specific CD4⫹8⫹ thymocytes. Eur. J. Immunol. 35: 3142–3152. 29. Biragyn, A., K. Tani, M. C. Grimm, S. Weeks, and L. W. Kwak. 1999. Genetic fusion of chemokines to a self tumor antigen induces protective, T-cell dependent antitumor immunity. Nat. Biotechnol. 17: 253–258. 30. Jiang, W., W. J. Swiggard, C. Heufler, M. Peng, A. Mirza, R. M. Steinman, and M. C. Nussenzweig. 1995. The receptor DEC-205 expressed by dendritic cells and thymic epithelial cells is involved in antigen processing. Nature 375: 151–155. 31. Mathiesen, I., M. Rimer, O. Ashtari, I. Cohen, U. J. McMahan, and T. Lomo. 1999. Regulation of the size and distribution of agrin-induced postsynaptic-like apparatus in adult skeletal muscle by electrical muscle activity. Mol. Cell. Neurosci. 13: 207–217. 32. Gronevik, E., S. Tollefsen, L. I. Sikkeland, T. Haug, T. E. Tjelle, and I. Mathiesen. 2003. DNA transfection of mononuclear cells in muscle tissue. J. Gene Med. 5: 909 –917. 33. Corthay, A., D. K. Skovseth, K. U. Lundin, E. Rosjo, H. Omholt, P. O. Hofgaard, G. Haraldsen, and B. Bogen. 2005. Primary antitumor immune response mediated by CD4⫹ T cells. Immunity 22: 371–383. 34. Neuberger, M. S. 1983. Expression and regulation of immunoglobulin heavy chain gene transfected into lymphoid cells. EMBO J. 2: 1373–1378. 35. Lunde, E., B. Bogen, and I. Sandlie. 1997. Immunoglobulin as a vehicle for foreign antigenic peptides immunogenic to T cells. Mol. Immunol. 34: 1167–1176. 36. Armstrong, A. C., S. Dermime, C. G. Allinson, T. Bhattacharyya, K. Mulryan, K. R. Gonzalez, P. L. Stern, and R. E. Hawkins. 2002. Immunization with a recombinant adenovirus encoding a lymphoma idiotype: induction of tumor-protective immunity and identification of an idiotype-specific T cell epitope. J. Immunol. 168: 3983–3991.