By Specific Targeting of B Cells Systemic Immunity to Incorporated Antigens Vector Promotes Priming of Mucosal and the Combined

The Combined CTA1-DD/ISCOM Adjuvant Vector Promotes Priming of Mucosal and Systemic Immunity to Incorporated Antigens by Specific Targeting of B Cells This information is current as of September 28, 2021. Anja Helgeby, Neil C. Robson, Anne M. Donachie, Helen Beackock-Sharp, Karin Lövgren, Karin Schön, Allan Mowat and Nils Y. Lycke J Immunol 2006; 176:3697-3706; ; doi: 10.4049/jimmunol.176.6.3697 Downloaded from http://www.jimmunol.org/content/176/6/3697

References This article cites 57 articles, 16 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/176/6/3697.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 28, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

The Combined CTA1-DD/ISCOM Adjuvant Vector Promotes Priming of Mucosal and Systemic Immunity to Incorporated Antigens by Speciﬁc Targeting of B Cells1

Anja Helgeby,* Neil C. Robson,† Anne M. Donachie,† Helen Beackock-Sharp,† Karin Lo¨vgren,‡ Karin Scho¨n,* Allan Mowat,† and Nils Y. Lycke2*

The cholera toxin A1 (CTA1)-DD/QuilA-containing, immune-stimulating complex (ISCOM) vector is a rationally designed mucosal adjuvant that greatly potentiates humoral and cellular immune responses. It was developed to incorporate the distinctive properties of either adjuvant alone in a combination that exerted additive enhancing effects on mucosal immune responses. In this study we demonstrate that CTA1-DD and an unrelated Ag can be incorporated together into the ISCOM, resulting in greatly augmented immunogenicity of the Ag. To demonstrate its relevance for protection against infectious diseases, we tested the vector Downloaded from incorporating PR8 Ag from the influenza virus. After intranasal immunization we found that the immunogenicity of the PR8 proteins were significantly augmented by a mechanism that was enzyme dependent, because the presence of the enzymatically inactive CTA1R7K-DD mutant largely failed to enhance the response over that seen with ISCOMs alone. The combined vector was a highly effective enhancer of a broad range of immune responses, including specific serum Abs and balanced Th1 and Th2 CD4؉ T cell priming as well as a strong mucosal IgA response. Unlike unmodified ISCOMs, Ag incorporated into the combined vector could be presented by B cells in vitro and in vivo as well as by dendritic cells; it also accumulated in B cell follicles of draining http://www.jimmunol.org/ lymph nodes when given s.c. and stimulated much enhanced germinal center reactions. Strikingly, the enhanced adjuvant activity of the combined vector was absent in B cell-deficient mice, supporting the idea that B cells are important for the adjuvant effects of the combined CTA1-DD/ISCOM vector. The Journal of Immunology, 2006, 176: 3697–3706.

t has proved difficult to stimulate immune responses with Ags been limited, and vaccine-induced diarrhea is an unwanted side given at mucosal sites (1, 2). The natural response to mucosal effect when these holotoxins have been used as adjuvants in oral I Ag exposure is tolerance, and effective adjuvants are required vaccines (16). Studies have also shown that the toxins may accu- to facilitate effective priming of mucosal as well as systemic im- mulate and affect the CNS after i.n. administration (17–19). An by guest on September 28, 2021 munity after oral or intranasal (i.n)3 immunization (3–5). Few ad- increased incidence of Bell’s palsy recently caused withdrawal juvants have been found to work when given mucosally, but the from the market of an influenza vaccine containing the LT adju- effects of QuilA-containing, immune-stimulating complexes vant, which was administered i.n. (20, 21). (ISCOMs) have been well documented (6–11). Perhaps the most To circumvent the toxicity problem, we have developed an al- potent mucosal adjuvants, though, are the closely related bacterial ternative CT-based adjuvant, the CTA1-DD, a gene fusion protein enteroxins, cholera toxin (CT) and Escherichia coli heat-labile that does not bind to ganglioside receptors (22). The CTA1-DD toxin (LT) (4). These holotoxins are structurally AB5 complexes adjuvant is nontoxic in mice, but retained adjuvant function, com- that bind to most mammalian cells through ganglioside receptors parable to that of CT, when given i.n. (22, 23). It consists of the via their B subunits (12). The A1 subunit is an ADP-ribosylating enzymatically active CTA1 subunit fused in-frame with a gene enzyme that has been found to host strong adjuvant function, but encoding a dimer of the D domain from the Staphylococcus aureus is also responsible for the toxicity of the molecules (13–15). Be- protein A (24, 25), allowing it to specifically target B cells via cause of their relative toxicity, the clinical use of CT or LT has binding to their Ig receptors. In contrast, ISCOMs are taken up by dendritic cells (DCs) preferentially (22, 23, 26–30) and are more *Department of Clinical Immunology, University of Goteborg, Goteborg, Sweden; potent than CTA1-DD when given orally. Therefore, we combined †Division of Immunology, Infection, and Inflammation, University of Glasgow, Glas- CTA1-DD and ISCOMs to create one of the first rationally de- ‡ gow, Scotland; and Isconova, Uppsala, Sweden signed adjuvant vectors (31). We found that the CTA1-DD/ Received for publication August 2, 2005. Accepted for publication December ISCOM vector was highly immunogenic by the i.n. as well as the 29, 2005. oral route even with nanogram doses of Ag, inducing Ag-specific 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 serum Abs, CD4 T cell priming, and IFN-␥ production (31). with 18 U.S.C. Section 1734 solely to indicate this fact. Combinations of adjuvants have several potential advantages. In 1 This work was supported by the Swedish Research Council; the Swedish Cancer addition to targeting different APCs, they offer the possibility of Foundation, the Sahlgrenska University Hospital Foundation, European Union Grants QLK2-CT-2001-01702, QLK2-CT-1999-00228, and LSHP-CT-2003-503240, and improving the stability of pharmacologically active enzymes by the Welcome Trust. incorporating in a stable vector that can deliver them linked to Ag. 2 Address correspondence and reprint requests to Dr. Nils Lycke, Department of Micro- or nanoparticles especially have been found to be the most Clinical Immunology, University of Goteborg, 413 46 Goteborg, Sweden. E-mail effective in these adjuvant combinations and, apart from ISCOMs, address: [email protected] chitosan or virus-like particles have also been successfully tested 3 Abbreviations used in this paper: i.n, intranasal; BM, bone marrow; CLN, cervical lymph node; CT, cholera toxin; DC, dendritic cell; GC, germinal center; LT, E. coli in formulations together with mutant holotoxins or muramyl heat-labile toxin; ISCOM, QuilA-containing immune-stimulating complex. dipeptide (32, 33). Although our previous work with CTA1-DD

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00 3698 COMBINED CTA1-DD/ISCOM ADJUVANT ACTS THROUGH B CELLS and p323 peptides from OVA expressed as a gene fusion protein, by electron microscopy, and the various components were analyzed for CTA1-OVA-DD, supports the idea that greatly enhanced re- comigration of protein and Quillaja saponins into fractions isolated from an sponses to CTA1-DD/ISCOM-linked Ag can be achieved, it will analytical 10–50% (w/w) sucrose gradient after centrifugation (18 h at 200,000 ϫ g, 10°C). The contents of protein and saponins in the different not be possible to create fusion proteins between CTA1 and any fractions were determined as described, using ELISA and spectrophoto- given Ag or peptide (31). Therefore, the aim of the present study metric analysis (at A214 nm), respectively. The amino acid content in each was to extend the potential of CTA1-DD/ISCOMs as an effective preparation was assessed. In some experiments we labeled the ISCOMs mucosal vaccine delivery vehicle by incorporating CTA1-DD and with CFSE. Briefly, 2 ␮l of solution, freshly prepared from a stock solution (100 mM in DMSO) of CFSE (Molecular Probes) was added per milligram the influenza virus PR8 Ag into the same ISCOM particles. In of cholesterol as the ISCOMs were prepared by mixing the lipids. Endo- addition to the ability of the novel vector to stimulate local and toxin contaminations were determined in the ISCOM preparations using systemic Ag-specific immunity, we determined to what extent the the Limulus amebocyte lysate test (LAL Endochrome; Charles River En- augmenting effects were CTA1 enzyme dependent and dissected dostestafe). The endotoxin levels were Ͻ110 endotoxin units/mg protein in the mechanisms involved in the presentation and immunogenicity all ISCOM preparations, whereas CTA1-OVA-DD, CTA1R7K-OVA-DD, and CT had endotoxin levels Ͻ 50 endotoxin units/mg protein. of the vector. Quality assessment of ISCOMs Materials and Methods We routinely analyzed the protein content in all ISCOMs using specific Animals ELISA. When sucrose-derived fractions were used, we incubated 50 ␮lof BALB/c mice (H-2d) and, when indicated, C57BL/6 (H-2b) mice were each fraction in the first row of a 96-well microtiter Maxisorp plate (Nunc) obtained from B&K Universal or Harlan Olac. DO11.10 (H-2d) or OT-II at a 1/3 dilution of 50 mM carbonate buffer (pH 9.6). Three-fold dilutions b in subsequent subwells were performed, and the plates were incubated at (H-2 ) mice, hosting TCR specific for the OVA323–339 peptide, were bred Downloaded from and maintained in the central research facility at University of Glasgow. 4°C overnight. After washing, each well plate was blocked for 1 h at room The ␮MT mice on a BALB/c background were obtained from the animal temperature with PBS-Tween 20 containing 2% fat-free dried milk powder. facility at University of Goteborg. Mice were maintained under specific For detection of PR8, a biotinylated chicken anti-PR8 polyclonal serum pathogen-free conditions, and sex- and age-matched animals were used in (importantly, chicken Abs do not bind protein A and, thus, do not bind the experiments. DD), followed by HRP-conjugated avidin (DakoCytomation). For detection of the native or mutant CTA1-OVAp-DD, we used an HRP-conjugated Immunizations rabbit anti-mouse antiserum (DakoCytomation). All incubations were per-

formed with gentle agitation for1hatroom temperature. Plates were http://www.jimmunol.org/ BALB/c mice were immunized i.n. with 20 ␮l containing 2 ␮g of CTA1- washed three times in PBS-Tween 20 between all steps and before the OVA-DD/ISCOMs or mutant CTA1R7K-OVA-DD/ISCOMs or with incubation in tetramethylbenzidine substrate (Svanova), and the enzymatic ISCOMs containing the PR8 Ag. When indicated, the immunizations were reaction was read at A450 nm using a spectrophotometer. given as single s.c injections in the foot pad or complementing the i.n. immunizations of 10 ␮g of CTA1-DD or 2 ␮g of CT together with 2 ␮g Determination of immunogenicity of PR8 Ag. Alternatively, but only when indicated, C57BL/6 or ␮MT mice (on a BALB/c background) were used. Immunizations were given 10 days T cell-dependent immunity was assessed in cervical lymph nodes (CLNs) apart and were repeated once or twice. Animals were killed 7–9 days after or spleens from immunized or control mice as previously described (38). the final immunization, and tissues, cells, serum, genital tract secretions, Briefly, single-cell suspensions were prepared by passing the tissue through and bronchial lavage were recovered as described previously (34, 35). a nylon mesh. RBCs were lysed with a hypotonic ammonium chloride-Tris Specimens were freshly used (cells) or were stored at Ϫ80°C (tissues) or solution and washed in HBSS (Invitrogen Life Technologies). The cell by guest on September 28, 2021 ϫ 6 Ϫ20°C until analyzed. suspensions were resuspended at a final concentration of 2 10 cells/ml and cultured in 200-␮l aliquots in 96-well microtiter plates (Nunc) in Preparation of adjuvants and Ags Iscove’s medium (Biochrom) supplemented with 10% heat-inactivated FCS (Biochrom), 50 ␮M 2-ME (Sigma-Aldrich), 1 mM L-glutamine (Bio- CTA1-OVA-DD and CTA1R7K-OVA-DD fusion proteins, containing a chrom), and 50 ␮g/ml gentamicin (Sigma-Aldrich; Iscove’s complete me- single copy of OVA323–339 peptide, were produced in Escherichia coli as ␮ dium) and cultured for 72 h at 37°C in 5% CO2 either alone or with 1 M previously described (26, 36). Protein analysis was performed with SDS- p323 (KJ Ross-Petersen) or 1 ␮g/ml PR8 micelles (Isconova). Proliferation PAGE, and protein concentrations were determined using the Bio-Rad DC was assessed after addition of 1 ␮Ci/well [3H]thymidine (Amersham Bio- protein assay according to the manufacturer’s instructions. ADP-ribosyl- sciences) for the last6hofculturing. [3H]Thymidine uptake was deter- transferase enzymatic activity was tested using the NAD:agmatine assay as mined using a beta scintillation counter (Beckman Coulter). After 96 h of described previously (12, 37). PR8 Ags were produced from the influenza culture, in vitro unstimulated and PR8 micell-restimulated cell superna- virus envelope glycoproteins hemagglutinin and neuraminidase of human tants were stored at Ϫ70°C until assayed. Cytokine responses to recall Ag influenza A virus strain PR/8/34 (H1N1) as described. Briefly, the virus in vitro were analyzed by multicytokine analysis (Luminex), which in- was propagated in 11-day-old embryonic hen eggs; after allantoic fluid was volved incubation with Ab-conjugated beads directed against mouse IL-5, clarified by centrifugation, the virus was purified by sucrose density gra- IL-10, and IFN-␥ (Bio-Rad) according to the manufacturer’s instructions. dient centrifugation. PR8 Ag micelles, prepared from the envelope pro- The assay was read on a Luminex 100 and analyzed using Bio-Plex Man- teins, hemagglutinin and neuraminidase, were isolated by ultracentrifuga- ager software; the concentrations of cytokines were determined against a tion (30 min, 40,000 rpm, 20°C) from a sucrose layer (20% (w/v) sucrose panel of cytokine preparations of known concentrations. containing 0.5% MEGA-10) and formed after removal of detergent by Specific Ab responses in serum, genital tract secretions, or bronchial dialysis against PBS for 48 h at room temperature. Protein content in the lavage were determined using ELISA as described previously (26). PR8- preparations was assessed. specific total IgG, IgG1, IgG2a, and IgA concentrations were determined Preparation of ISCOMs using polystyrene, 96-well microtiter plates (Nunc) coated with PR8 micelles (1 ␮g/ml). Total IgE in serum was assessed using rat anti-mouse IgE Briefly, ISCOMs were prepared by mixing 1.0 mg of Ag (PR8 Ags alone (Serotec)-coated, soft, 96-well ELISA plates (Dynatech Laboratories). Af- or mixed with CTA1-OVA-DD or CTA1R7K-OVA-DD) with 1.0 mg of ter blocking with 0.1% BSA/PBS, serum samples were diluted 1/500 for cholesterol (C-8503; Sigma-Aldrich), 1.0 mg of phosphatidylcholine of specific responses or 1/20 for total IgE levels, followed by serial dilutions egg origin (Lipoid), and 5.0 mg of Quillaja saponins (Spikoside; Isconova) in 0.1% BSA/PBS in subsequent subwells. Bronchial lavage or vaginal in a total volume of 1.0 ml and a final concentration of 2% Mega-10 secretions were diluted 1/10 and serially diluted. Alkaline phosphatase- (Bachem). To determine the optimum conditions for efficient incorporation conjugated, isotype-specific, goat anti-mouse Abs at 1/500 (Southern Bio- of the different proteins with retained biological activities, several buffers technology Associates) or 0.125 ␮g/ml biotin-conjugated anti-mouse IgE were used to remove the detergent during the first overnight dialysis at (Serotec) were then added, followed by 2.1 ␮g/ml Extravidin peroxidase room temperature (0.1 M acetate (pH 4.5), 0.2 M acetate (pH 6.0), PBS, (Sigma-Aldrich). Nitrophenyl phosphatase (1 mg/ml; Sigma-Aldrich) in and 0.1 M phosphate (pH 8.0)). Thereafter, dialysis buffer was changed to ethanolamine buffer (pH 9.8) or o-phenylenediamine substrates (1 mg/ml;

PBS, and dialysis was continued for another 24 h at room temperature. All Sigma-Aldrich) in citrate buffer (pH 4.5) containing 0.04% H2O2 were references in the text to Ag or adjuvant concentrations refer to the protein used, and enzymatic reactions were read in a Titer-Tek Multiscan spectro- content of incorporated protein with a ratio of protein content to Quillaja photometer (Labsystems). IgE concentrations were calculated in micro- saponins of 1:2 in the ISCOMs. The formation of ISCOMs was conﬁrmed grams per milliliter from a standard curve generated by serial dilutions of The Journal of Immunology 3699

purified IgE of known concentration (BD Pharmingen). PR8-specific log10 37°C in 5% CO2. On days 0, 3, 6, and 8 of culture, the medium was Ab titers (means Ϯ SD) were defined as the interpolated reading giving rise supplemented with 10% supernatant from the X-63 fibroblast cell line to an absorbance of 0.4 above background, which consistently gave read- transfected with the murine GM-CSF gene. After 10 days, nonadherent ings on the linear part of the curve. DCs were harvested by gentle washing and were typically Ͼ85% CD11cϩ, class II MHCint, CD40low, B7.1low, and B7.2low. In vivo distribution of ISCOMs Assessment of Ag-presenting ability CFSE-labeled ISCOMs containing 5 ␮g of either CTA1-OVA-DD or OVA were injected s.c. into the footpad of BALB/c mice. Two, 4, 24, and 48 h Aliquots of 3 ϫ 106 cells to be used as APCs were plated in 12-well plates later, the draining popliteal lymph nodes were harvested; they were either or 24-well ultra low adherence plates in 1 ml (Costar) and pulsed with Ag frozen for immunohistochemistry, or single-cell suspensions were prepared for2hat37°C. For the last 45 min of culture, 50 ␮g/ml mitomycin C by forcing through a fine nylon mesh. The cells were washed in Iscove’s (Sigma-Aldrich) was added before the APCs were recovered and washed complete medium, followed by incubation for 5–15 min at 4°C with 1 ␮l four times in RPMI 1640. After washing, APCs were plated in triplicate at of the 2.4G2-FcR-blocking Ab (BD Pharmingen) in cold PBS containing 1 ϫ 105 cells/well in 96-well, flat-bottom microtiter plates (Costar) to- 0.1% BSA (BSA/PBS), and were added to cells aliquoted in 100 ␮l. The gether with 2 ϫ 105 lymph node cells from DO11.10 or OT-II mice in a cells were labeled with 1/100 PE-conjugated anti-mouse CD19 or CD11c total volume of 200 ␮l. To assess T cell proliferation, 1 ␮Ci/well [3H]TdR (BD Pharmingen) by incubation for 30 min at 4°C. After washing twice, (West of Scotland Radionucleotide Dispensary) was added for the last 16 h the cells were analyzed on a FACScan flow cytometer (BD Biosciences) of culture, and cell-bound DNA was harvested onto glass-fiber filter mats using CellQuest software. Cells were gated on the lymphocyte population (Wallac). [3H]TdR uptake was counted on a Betaplate counter (Wallac). with the gate set on either the CD19 or CD11c population, and 10,000 cells were collected, which were positive for CD19 or CD11c positive for CFSE Statistical analysis (detected in the FL-1 channel). Data were compared using Student’s t test. Uptake of ISCOMs in vitro Downloaded from Bone marrow (BM)-derived DCs or purified spleen B cells were incubated Results with CFSE-labeled CTA1-OVA-DD/ISCOMs or OVA/ISCOMs in RPMI Optimal conditions for incorporation of Ag and adjuvant into 1640/10% FCS medium for1hat37°C in 24-well, low adhesion plates, ISCOMs then washed three times, and the amount of uptake was assessed by FACS. In a previous study we showed that the CTA1-DD adjuvant could Immunohistochemistry be effectively incorporated into ISCOMs and that the combined http://www.jimmunol.org/ Frozen sections (6 ␮m) were prepared on microslides using a cryostat, vector acquired adjuvant potency greatly surpassing either system fixed in 100% acetone for 10 min at room temperature, and dried before used alone (31). In this study we explored the possibility of incor- washing in PBS. The slides were then treated with 5% horse serum in PBS porating unrelated proteins in the same ISCOM particle using the for 15 min in a humidified chamber. To identify germinal centers (GCs), proteins from influenza virus PR8 as a representative infectious sections were double labeled with FITC-conjugated GL-7 (BD Pharmin- gen) mAb and a Texas Red-conjugated anti-IgM Ab (Southern Biotech- agent together with CTA1-DD containing the MHC class II-re- nology Associates) at a 1/100 dilution. GCs were counted in each lymph stricted p323 peptide from OVA (CTA1-OVA-DD). In prelimi- node, and the total GL-7-positive area in the B cell follicles was measured nary studies we established that both CTA1-OVA-DD and PR8 in four CLNs per mouse and five mice per group. To assess the distribution Ags were incorporated optimally at pH 6.0. Electron microscopic of CFSE-labeled ISCOMs in vivo, the sections were stained with biotinylated anti-CD45R/B220 or CD4 (or 7-amino-4-methylcoumarin-3-acetic inspection and sucrose gradient analysis confirmed the successful by guest on September 28, 2021 acid-labeled anti-CD4; BD Pharmingen), and Texas Red-labeled strepta- construction of ISCOMs that carried both proteins in a balanced vidin (Vector Laboratories), diluted 1/100, was used as a secondary Ab. ratio of 1:1, with the peak distribution of each component being in The slides were mounted with a fluorescence mounting medium (Dako- fractions 5–7 of the sucrose gradient (Fig. 1). These fractions were Cytomation) and evaluated using a Leica LSC microscope, with digital subsequently used for the immunization studies. storing of photographs. FACS analysis Immunogenicity of combined CTA1-DD/ISCOM vector containing PR8 Ags The frequency of GL-7 B lymphocytes in the CLNs was analyzed by FACS. Briefly, single-cell suspensions of CLN cells were labeled with The immunogenicity of the combined vaccine vector was deter- anti-GL7-FITC and anti-B220-PE (BD Pharmingen) and analyzed by mined after i.n. immunization. Mice were immunized three times FACS. The frequency of double-positive CLN cells was calculated for each group of five mice, and the mean percentage was calculated. The background labeling of CLN cells was 1% from naive mice lacking GCs in frozen sections. Enrichment of B cells and BM-derived DCs Naive B cells were purified from the spleens or lymph nodes of BALB/c mice by negative selection using MACS. Briefly, after preparation of single-cell suspensions, RBCs were lysed by addition of 5 ml of NH4Cl (0.14 M) for 5 min at room temperature, and the remaining cells were washed in MACS medium (PBS and 2% FCS), before being resuspended in MACS- medium and counted. Following centrifugation, the cells were resuspended at a concentration of 1 ϫ 107 cells/100 ␮l MACS medium, and 10 ␮l/107 cells anti-CD43-coated microbeads (Miltenyi Biotec) were added for 15 min at 4°C. The labeled cells were then passed over a CS MACS column according to the manufacturer’s instructions (Miltenyi Biotec). B cells from Ag-immunized mice were positively selected by MACS using anti- FIGURE 1. Analysis of CTA1-OVA-DD and PR8 incorporation into CD19-coated microbeads (Miltenyi Biotec) according to the manufactur- ISCOMs. ISCOMs containing CTA1-OVA-DD and PR8 Ags were sub- Ͼ ϩ er’s instructions. Eluted cells were 96% B220 as assessed by flow jected to sucrose density gradient analysis, and the fractions were analyzed cytometry. for CTA1-DD/PR8 and saponin content as described in the text. The peaks To obtain DCs, BM cells were washed out of the femurs of adult mice in RPMI 1640 using a syringe and a 21-gauge needle. Aliquots of 3 ϫ 106 of incorporation of CTA1-DD, PR8 proteins, and saponin were superim- BM cells were seeded in 90-cm petri dishes (Bibby Sterilin) and cultured posed, illustrating the successful formation of ISCOMs. The y-axis values in RPMI 1640 containing 2 mM L-glutamine, 100 U/ml penicillin, 100 represent arbitrary OD units at A450 nm for immunodetection of HRP- ␮g/ml streptomycin, 1.25 ␮g/ml fungizone (all from Invitrogen Life Tech- labeled CTA1-DD or PR8 (left) and saponin at A214 nm (right) using a nologies), and 10% FCS (Harlan Sera Laboratories; complete medium) at spectrophotometer. 3700 COMBINED CTA1-DD/ISCOM ADJUVANT ACTS THROUGH B CELLS

FIGURE 2. Intranasal immunization with PR8/CTA1-OVA-DD/ISCOMs primes T cells at both systemic and mucosal sites. Priming of OVA- and PR8-specific proliferative responses in spleen (A and C) or cervical lymph nodes (B and D) by i.n. immunization was performed. Mice were immunized three times i.n. with PR8/ ISCOMs, PR8/CTA1-OVA-DD/ISCOMs, the enzymatically inactive PR8/ CTA1R7K-OVA-DD/ISCOMs, or PR8 alone (/) or were given PR8 together with the CTA1-DD adjuvant. Isolated lymphocytes were stimulated 7 days after the final immunization with recall Ag, OVA323– 339 or PR8 Ag in vitro. The results shown are from five individual mice per group and are representative of three identical experiments with similar results. Data are Downloaded from ,ءء) expressed as the mean cpm Ϯ SEM .(p Ͻ 0.05 ,ء ;p Ͻ 0.01 http://www.jimmunol.org/ with PR8/CTA1-OVA-DD/ISCOMs, and the specific immune re- with the combined vector was regularly 3- to 8-fold higher than sponses to recall Ag were assessed in spleen or CLN T cells 8 days that observed with the ISCOMs or the mutant inactive PR8/ after the final immunization. As we mentioned above, mice im- CTA1R7K-OVA-DD/ISCOMs (Fig. 2). For comparison, mice immunized with CTA1-OVA-DD/ISCOMs showed excellent prim- munized with PR8 and the CTA1-DD protein alone responded to ing to the OVA peptide, and these responses were much lower in the same magnitude as PR8/ISCOMs or the mutant combined PR8/ mice receiving the enzymatically inactive form of CTA1R7K- CTA1R7K-OVA-DD/ISCOM vector (Fig. 2, C and D). OVA-DD/ISCOMs (Fig. 2). A similar pattern of responses was Intranasal immunization with the combined vector also stimulated seen when CLNs or spleen cells were restimulated with PR8 Ags strong Ag-specific Ab responses in serum, with the enzymatically in vitro, confirming that the additional Ags that had been incor- active PR8/CTA1-OVA-DD/ISCOMs stimulating ϳ10-fold higher by guest on September 28, 2021 porated into the CTA1-DD/ISCOMs were immunogenic. In addi- serum anti-PR8 IgG Ab titers compared with PR8/ISCOMs alone tion, this process had not altered the enzyme activity of CTA1, (Fig. 3). For comparison, mice immunized with PR8 together with because the R7K form was still much less active than the ISCOMs CTA1-DD alone gave significantly better specific IgG responses than containing the intact CTA1 and induced responses similar to those mice immunized with only PR8 Ag, but had titers comparable to those using PR8/ISCOMs alone. The enhancing effect on T cell priming of mice immunized with PR8/ISCOMs (Fig. 3A). Moreover, marked

FIGURE 3. Induction of influenza-specific systemic and mucosal Ab production by immunization with PR8/CTA1-OVA-DD/ISCOMs. Serum (A), bronchial lavage (B), or vaginal secretions (C) were taken 7 days after the last of three i.n. immunizations with PR8/CTA1-OVA-DD/ISCOMs, the enzymatically inactive PR8/CTA1R7K-OVA-DD/ISCOMs, or PR8/ISCOMs alone. PR8-specific Abs were mea-

sured by ELISA and are expressed as log10 titers for ,ءء) individual mice; the mean is shown by the bar p Ͻ 0.05). The content of total IgA in ,ء ;p Ͻ 0.01 the bronchial lavage and genital tract secretions was similar in the respective groups, indicating that the total IgA production was not differently affected by the immunizations. Thus, there was 86 Ϯ 24, 43 Ϯ 20, and 40 Ϯ 16 ␮g/ml total IgA (in the order given above) in lavage and 1.9 Ϯ 0.7, 1.8 Ϯ 1.0, and 1.5 Ϯ 0.8 ␮g/ml total IgA in genital tract secretions in the respective groups. This is one representative experiment of three with similar results. The Journal of Immunology 3701

FIGURE 4. The combined PR8/ CTA1-OVA-DD/ISCOM vector induces a balanced Th1 and Th2 response. Mice were immunized three times i.n. with PR8/CTA1-OVA-DD/ ISCOMs, the enzymatically inactive PR8/CTA1R7K-OVA-DD/ISCOMs, PR8/ISCOMs alone, or PR8 Ag plus CT, and anti-PR8 IgG1 (f) and IgG2a (Ⅺ) serum titers as well as total serum IgE (nanograms per milliliter) responses were measured by ELISA 7–9 days after the ﬁnal immunization (A and B). Spleen cells were restimulated for 96 h with PR8 Ag in vitro and IL-10 (C), IL-5 (D), and IFN-␥ (E) concentrations in culture supernatants were determined by ELISA. The results (pico- grams per milliliter) are expressed as the mean Ϯ SEM of individual samples Downloaded from .(p Ͻ 0.05 ,ء) from ﬁve mice per group This is one of three experiments with similar results.

bronchioalveolar and genital tract IgA responses were observed in the mutant CTA1R7K-OVA-DD/ISCOM vector enhanced anti-PR8 http://www.jimmunol.org/ mice immunized with the combined vector (Fig. 3), which, again, titers 3-fold ( p Ͻ 0.05) above those of mice given PR8/ISCOMs were enhanced 10-fold compared with ISCOMs alone. Interestingly, alone in both serum and secretions, suggesting that the adjuvant by guest on September 28, 2021

FIGURE 5. The combined CTA1- DD/ISCOM vector augments GC formations. Mice were given a single i.n. immunization with PR8/CTA1-OVA- DD/ISCOMs, enzymatically inactive PR8/CTA1R7K-OVA-DD/ISCOMs, PR8/ISCOMs alone, or CT plus PR8 micelles (2 ␮g of each). A, Fourteen days later, frozen sections of the CLNs were prepared and double labeled with Texas Red-conjugated anti- IgM (red) and FITC-labeled GL-7 (green) to detect GC reactions (green/ yellow). B, the mean area of GL-7ϩ cells in the B cell follicles of four CLNs per mouse from ﬁve mice in each group was calculated. C, These calculations were complemented with FACS analysis of the frequency of GL-7ϩ cells per total B220ϩ CLN cells from ﬁve mice for the respective groups, using the background labeling of GC-negative naive CLN cells as a -p Ͻ 0.05. This experi ,ء .(/) control ment represents one of three identical experiments with similar results. 3702 COMBINED CTA1-DD/ISCOM ADJUVANT ACTS THROUGH B CELLS

function of the combined vector was not exclusively dependent on the enzymatic activity (Fig. 3). CTA1-DD-ISCOMs combined vector induced balanced Th1- and Th2-type response The CT adjuvant is known to skew CD4 T cell priming toward a Th2-type response, whereas ISCOMs stimulate strong Th1 immunity, with IFN-␥ and CTL activity (39–41). Therefore, we pre- dicted that the combined vector might promote a balanced Th1 and Th2 type of response to influenza Ags. In support of this, ISCOMs promoted higher relative PR8-specific IgG2a Ab responses than those stimulated by the combined vector, whereas specific IgG2a titers were lower compared with IgG1 titers in CT-immunized mice (Fig. 4A). Also, total serum IgE levels were higher in mice immunized i.n. with CT than in those given the combined vector or ISCOMs alone (Fig. 4). In addition, spleen and CLN (data not shown) cells restimulated in vitro with PR8 Ag produced high levels of both Th1- and Th2-dependent cytokines, which were

much enhanced compared with those of cells obtained from mice Downloaded from immunized with enzymatically inactive PR8/CTA1R7K-OVA-DD or PR8/ISCOMs (Fig. 4). CTA1-DD/ISCOM vector stimulates germinal center reactions In search of an early in vivo marker that reﬂected the adjuvant

effect of the combined vector, we investigated the ability of CTA1- http://www.jimmunol.org/ OVA-DD/ISCOMs to stimulate GC formations in the regional lymph nodes. We found strong GC reactions 14 days after a single i.n. immunization with PR8/CTA1-OVA-DD/ISCOMs, whereas the enzymatically inactive PR8/CTA1R7K-OVA-DD/ISCOMs induced smaller and fewer GC reactions in the CLNs, similar to those seen after treatment with ISCOMs alone (Fig. 5). The size and frequency of GCs stimulated by the CTA1-OVA-DD/ISCOM vector were comparable with those seen after i.n. immunization with intact CT holotoxin, which is known to induce prominent GC by guest on September 28, 2021 reactions (Fig. 5) (42–44). A ﬂow cytometric analysis of GL7ϩ B FIGURE 6. Uptake and presentation of ISCOM-associated Ag by DCs cells from CLNs also revealed the difference between immuniza- and B cells in vitro. Puriﬁed spleen B cells (A) or BM-derived DCs (C) tions with the combined vector and CT, on the one hand, and from BALB/c mice were pulsed with either CFSE-labeled CTA1-OVA- DD/ISCOMs or CFSE-labeled OVA/ISCOMs for 1 h, and uptake was an- ISCOMs alone or the mutant vector, on the other (Fig. 5C). Thus, alyzed by FACS analysis. Although DCs took up both forms of ISCOMs the combined vector was as effective as CT in stimulating GC equally above background levels (open curve), B cells only acquired the formations in CLNs in i.n. immunized mice. combined CTA1-OVA-DD/ISCOM vector, not the OVA/ISCOMs. To assess the Ag-presenting ability, spleen B cells (B) or BM-derived DCs (D) Contrary to ISCOMs, the combined vector acts on B cells were pulsed for 2 h with 10 ␮g/ml OVA/ISCOMs or a molar equivalent,

We next examined how the enhanced immune responses found based on the OVA323–339 peptide content, of CTA1-OVA-DD/ISCOMs or ␮ after immunization with the CTA1-OVA-DD/ISCOM vector cor- CTAR7K-OVA-DD/ISCOMs or of OVA323–339 peptide (0.32 g/ml) related with the APC population involved in their uptake. Purified alone. The APC function of DCs and B cells were determined after 72 h in BM, DCs, or derived spleen B cells were pulsed for different time culture with DO11.10 T cells. Results are from one representative exper- 3 Ϯ periods with CFSE-labeled CTA1-OVA-DD/ISCOMs or CFSE- iment of three and are expressed as the mean [ H]TdR incorporation SD p Ͻ 0.01 vs B cells pulsed with OVA/ISCOMs ,ءء) from triplicate cultures labeled ISCOMs containing equimolar amounts of the OVA323–339 or OVA peptide alone). peptide, and the uptake was analyzed by FACS. These studies showed that although DCs took up the OVA/ISCOMs and CTA1- OVA-DD/ISCOMs with equal efficiency, B cells only took up the combined vector and not the modified ISCOMs (Fig. 6A). In par- Differential distribution of CTA1-DD/ISCOMs and ISCOMs in allel, BM-derived DCs presented the OVA peptide to DO11.10 draining lymph nodes OVA-specific CD4ϩ T cells in vitro with identical efficiency when To confirm these properties of the combined vector in vivo, mice pulsed with all the different ISCOM constructs, whereas purified B were injected s.c. into the footpad with CFSE-labeled ISCOMs or cells presented OVA peptide only when pulsed with the combined CFSE-labeled CTA1-OVA-DD/ISCOMs, and their distribution in vectors containing CTA1-OVA-DD and could not present OVA/ the draining popliteal lymph nodes was examined at various time ISCOMs themselves (Fig. 6, B and D). The ability of B cells to points. Two hours after injection, both OVA/ISCOMs and CTA1- present Ag in the combined vector in vitro appeared not to depend OVA-DD/ISCOMs could be found in the deep cortical regions of on the CTA1 enzyme, because the mutant CTA1R7K-OVA-DD/ the lymph node, consistent with access via the afferent lymphatics ISCOMs also triggered peptide-specific T cell proliferation. Thus, and subcapsular sinuses (Fig. 7). Thereafter, the CTA1-OVA-DD/ the combined CTA1-OVA-DD/ISCOM vector gains access to and ISCOMs began to concentrate in the B cell follicles, whereas the can be presented by a broader repertoire of APCs than conven- ISCOMs never appeared in follicles and were progressively lost tional ISCOMs. from the lymph node, with only small amounts remaining in the The Journal of Immunology 3703

FIGURE 7. CTA1-DD/ISCOMs and ISCOMs localize in different anatomical compartments of the lymph node. Mice were injected with 5 ␮g of CFSE-labeled CTA1- OVA-DD/ISCOMs or OVA/ISCOMs into the footpad. Two and 24 h after injection, the popliteal lymph nodes were excised, and frozen sections were stained with Texas Red- Downloaded from labeled anti-B220 (red; A–D) to highlight B cell follicles or with Texas Red-labeled anti- CD4 (red; E and F)/7-amino-4-methylcoumarin-3-acetic acid-labeled anti-CD4 (blue; H) to identify the T cell-dependent areas (TDA). CFSE-labeled ISCOMs (green) of

both types localized initially in the deep cor- http://www.jimmunol.org/ tical areas between follicles (A and B). Thereafter, CTA1-OVA-DD/ISCOMs accumulated in the B cell follicles (C and E), whereas only small amounts of OVA/ ISCOMs remained in the TDA (blue) by 24 h (G and H). This is one representative experiment of three with similar results. by guest on September 28, 2021

deep cortex after 24 h. These results were also confirmed by FACS and cultured with PR8 Ag in vitro, spleen cells from wild-type and analysis of isolated cells from the draining lymph node, which ␮MT mice showed equivalent responses after immunization with showed that CFSE-labeled CTA1-OVA-DD/ISCOMs could be de- PR8/ISCOMs. However, in striking contrast to wild-type mice, tected in CD19ϩ B cells, whereas both the combined vector and ␮MT mice demonstrated no enhancement of T cell priming in- the normal ISCOMs were found in DCs (data not shown). duced by the combined CTA1-OVA-DD/ISCOM vector (Fig. 8B). These findings also support the idea that B cells play a central role CTA1-DD/ISCOMs target B cells as APCs in vivo in vivo in the immunoenhancing effect of the combined CTA1- To address the role of B cells in the immunogenicity of the com- OVA-DD/ISCOMS vector. bined vector, we purified B cells from the popliteal lymph nodes of mice immunized with OVA/ISCOMs or CTA1-OVA-DD/ Discussion ISCOMs and examined their ability to present OVA to TCR-trans- In this study we have confirmed and extended our previous find- genic OT-II CD4ϩ T cells in vitro. This showed that B cells from ings that a potent mucosal vaccine vector can be constructed by mice injected with the combined CTA1-OVA-DD/ISCOM vector, combining the distinctive adjuvants CTA1-DD and ISCOMs. We but not those injected with OVA/ISCOMs alone, could present Ag show that the combined vector can be exploited to host additional to the specific T cells (Fig. 8A). Ags, such as influenza surface proteins, and the resulting formu- Finally, we determined whether B cells were needed for the lation not only induces strong mucosal and systemic immune re- immunogenicity of the combined CTA1-OVA-DD/ISCOMs in sponses to all incorporated Ags, but also retains the enzymatic vivo by immunizing ␮MT, B cell-deficient mice i.n. with PR8/ function of CTA1, which is required for its biological function. CTA1-OVA-DD/ISCOMs or PR8/ISCOMs. When restimulated Our results also indicate that the explanation for the enhanced 3704 COMBINED CTA1-DD/ISCOM ADJUVANT ACTS THROUGH B CELLS

FIGURE 8. The combined CTA1-DD/ISCOM vector is presented by B cells ex vivo and requires B cells for its enhanced adjuvant activity in vivo. A, Ex vivo presentation of ISCOM-associated Ag by B cells. C57BL/6 mice were immunized with ISCOMs containing 2.5 ␮g of either OVA or a molar equivalent of CTA1-OVA-DD in each rear footpad. After 4 h, B cells were purified by MACS and cultured together with OT-II T lymphocytes for 72 h. 3 p Ͻ 0.01 vs Downloaded from ,ءء) The results from one representative experiment of three are shown as the mean [ H]TdR incorporation Ϯ 1 SD from triplicate cultures OVA ISCOMs). BALB/c (B)or␮MT (B cell-deficient; C) mice were immunized twice i.n. with PR8/CTA1-OVA-DD/ISCOMs or PR8/ISCOMs; 7 days after the last immunization, spleen cells were restimulated for 72 h with PR8 Ag in vitro. The results from one representative experiment of three are shown .(p Ͻ 0.01 vs PR8/ISCOMs ,ءء) as the mean incorporation of [3H]TdR Ϯ 1 SD for triplicate cultures with five individual mice/group

adjuvant function of ISCOMs containing CTA1-DD compared ISCOMs containing the R7K mutant of CTA1-DD were barely http://www.jimmunol.org/ with conventional ISCOMs is that the CTA1-DD component al- more immunogenic than ISCOMs with PR8 alone. This underlines lows additional targeting to B cells as APCs. the critical role of enzyme function in the activity of CT-related Previous studies have shown that many protein Ags, including adjuvants (47). Nevertheless, ISCOMs containing enzymatically those prepared from influenza PR8, are highly immunogenic when inactive CTA1R7K-DD retained some enhanced adjuvant func- incorporated into ISCOMs (39–44) as is the CTA1-DD construct tion, such as specific Ab production in vivo and presentation of Ag containing the OVA323–339 peptide (31). We demonstrate that by B cells in vitro. Together, these results indicate the potentially CTA1-DD/ISCOMs can be modified to incorporate PR8 Ags, and beneficial effects of targeting B cells as additional APCs. In vitro the resulting vector induces mucosal and systemic immune re- studies have shown that ISCOM particles themselves are taken up sponses when given i.n. Not only was the combined vector highly and presented preferentially by DCs (27, 28, 48), a finding we have by guest on September 28, 2021 immunogenic, but it was also, because, as batch-to-batch quality extended in this study by showing that ISCOMs accumulate in control responses to the integrated OVA peptide reflected, a stable, DCs in vitro and in the DC-rich, T cell-dependent areas of lymph reproducible, and effective vaccine vector for strongly enhanced nodes in vivo. Interestingly, the pattern of uptake in vivo was immune responses (31). This occurs using very low doses of Ag, consistent with localization in the fibroreticular conduits of T cell with as little as 2 ␮g of PR8 protein and the equivalent of 150 ng areas, which others have shown to be important sites of accumu- of OVA peptide per dose being immunogenic by the nasal or par- lation of Ag-loaded DCs and initial interactions between DCs and enteral routes (31). In addition, the combined CTA1-DD/ISCOM T cells (49–52). In contrast, ISCOMs containing CTA1-DD were vector induces a balanced Th1 and Th2 response, which comprises taken up very efficiently by B cells as well as by DCs in vitro and IFN-␥ and IL-5 production as well as T cell-mediated immune were presented by both APCs to CD4ϩ T cells, presumably re- responses, such as delayed-type hypersensitivity and CTL activity flecting the ability of the DD portion to bind to B cells selectively and serum IgG and local IgA Abs (31), with no evidence of prim- via their surface Ig. Furthermore, s.c. injected CTA1-DD/ISCOMs ing of IgE production. These properties distinguish the combined had a unique ability to accumulate in B cell follicles of lymph vector from its individual components, first by enabling much nodes, where they were retained for at least 24 h, by which time smaller doses of Ag to be used. In addition, the use of CT is conventional ISCOMs were virtually undetectable in the lymph frequently associated with Th2 polarization and potentially harm- node. Most importantly, the augmented adjuvant properties of ful IgE production, whereas ISCOMs stimulate marked Th1 and CTA1-DD/ISCOMs were largely absent in B cell-deficient, ␮MT CD8ϩ T cell responses when used alone (39–41, 45, 46). These mice. For these reasons, we propose that this novel combined vec- features were also confirmed in the present study and demonstrated tor is so effective because it targets both DCs and B cells in vivo. in the combined vector’s balanced IgG2a- and IgG1-specific serum Unfortunately, attempts to document the same effects of the responses in contrast to the relative skewing of these responses combined vector on B cells in the nasal-associated lymphoid tis- toward Th1 and Th2 by ISCOMs and CT, respectively. In fact, in sues or CLNs after i.n. immunizations failed because no CFSE- this regard the combined vector mimicked the balanced Th1 and labeled DCs and B cells were detected. We have previously ex- Th2 responses seen with CTA1-DD adjuvant alone (22). Together, perienced this problem when using labeled CTA1-DD or CT given these findings highlight the potential usefulness of CTA1-DD/ i.n. (22, 53). Despite this, we have no reason to believe that DCs ISCOMs as practical mucosal vaccine vectors that will provide a or B cells were not targeted in the nasal-associated lymphoid tis- flexible and stable means of inducing protective immunity against sues and CLNs by the combined vector, especially because distinct a variety of pathogens. GCs developed in the CLNs, as shown in the present study. Also, As we have found previously (31), the improved adjuvant prop- ␮MT mice failed to exhibit augmented responses after i.n. immu- erties of CTA1-DD incorporated into ISCOMs were largely de- nization with the combined vector, suggesting a relative depen- pendent on the presence of enzymatically active toxin, because dence on B cells for the effect. The Journal of Immunology 3705

The ISCOM particle provides a stable formulation that interacts 12. Spangler, B. D. 1992. Structure and function of cholera toxin and the related efficiently with and activates DCs (27, 28, 48), whereas CTA1-DD Escherichia coli heat-labile enterotoxin. Microbiol. Rev. 56: 622. 13. Kawamura, Y. I., R. Kawashima, Y. Shirai, R. Kato, T. Hamabata, allows preferential delivery into B cells. These B cells are likely to M. Yamamoto, K. Furukawa, K. Fujihashi, J. R. McGhee, H. Hayashi, et al. be activated as a result of both the binding of sIg and the presence 2003. Cholera toxin activates dendritic cells through dependence on GM1-ganglioside which is mediated by NF-␬B translocation. Eur. J. Immunol. 33: of pharmacologically active CTA1 enzyme. Activated B cells have 3205–3212. been shown to be highly efficient APCs in other systems (52, 54) 14. Rappuoli, R., M. Pizza, G. Douce, and G. Dougan. 1999. Structure and mucosal and together with the efficient localization of CTA1-DD/ISCOMs adjuvanticity of cholera and Escherichia coli heat-labile enterotoxins. Immunol. Today 20: 493–500. in B cell follicles, these properties presumably encourage cognate 15. Soriani, M., L. Bailey, and T. R. Hirst. 2002. Contribution of the ADP-ribosy- interactions between Ag-specific T and B cells in evolving germi- lating and receptor-binding properties of cholera-like enterotoxins in modulating nal centers. This idea is supported by the fact that GC formations cytokine secretion by human intestinal epithelial cells. Microbiology 148: 667–676. were greatly enhanced after administration of CTA1-DD/ISCOMs 16. Wu, A. L., and W. A. Walker. 1976. Immunological control mechanism against compared with ISCOMs alone. As well as enhancing primary im- cholera toxin: interference with toxin binding to intestinal receptors. Infect. Im- mune responses, as we show in this study, recent work demon- mun. 14: 1034–1042. 17. van Ginkel, F. W., J. R. McGhee, J. M. Watt, A. Campos-Torres, L. A. Parish, strates that such interactions are of crucial importance in sustaining and D. E. Briles. 2003. Pneumococcal carriage results in ganglioside-mediated memory T and B cell responses (55–57). This would be a major olfactory tissue infection. Proc. Natl. Acad. Sci. USA 100: 14363–14367. factor in the success of any vaccine vector, and we are currently 18. van Ginkel, F. W., R. J. Jackson, Y. Yuki, and J. R. McGhee. 2000. Cutting edge: the mucosal adjuvant cholera toxin redirects vaccine proteins into olfactory tis- studying the effects of B cell-targeted ISCOMs on the induction of sues. J. Immunol. 165: 4778–4782. immunological memory. 19. Fujihashi, K., T. Koga, F. W. van Ginkel, Y. Hagiwara, and J. R. McGhee. 2002.

A particular asset to the combined vector was its strong aug- A dilemma for mucosal vaccination: efficacy versus toxicity using enterotoxin- Downloaded from based adjuvants. Vaccine 20: 2431–2438. menting effect on mucosal IgA responses. Contrary to ISCOMs 20. Mutsch, M., W. Zhou, P. Rhodes, M. Bopp, R. T. Chen, T. Linder, C. Spyr, and themselves, which are fairly poor inducers of mucosal IgA (58), R. Steffen. 2004. Use of the inactivated intranasal influenza vaccine and the risk combination with the CTA1-DD adjuvant rendered ISCOMs much of Bell’s palsy in Switzerland. N. Engl. J. Med. 350: 896–903. Ͼ 21. Gluck, R., R. Mischler, P. Durrer, E. Furer, A. B. Lang, C. Herzog, and S. J. Cryz, better IgA-stimulating properties. We consistently observed 10- Jr. 2000. Safety and immunogenicity of intranasally administered inactivated fold stronger mucosal IgA responses in CTA1-OVA-DD/ISCOMs trivalent virosome-formulated influenza vaccine containing Escherichia coli heat- labile toxin as a mucosal adjuvant. J. Infect. Dis. 181: 1129–1132. compared with i.n. ISCOM-immunized mice. Importantly, regard- http://www.jimmunol.org/ 22. Eriksson, A. M., K. M. Scho¨n, and N. Y. Lycke. 2004. The cholera toxin-derived less of the total IgA production at the mucosal sites, the specific CTA1-DD vaccine adjuvant administered intranasally does not cause inflamma- responses were truly augmented and were not a reflection of poly- tion or accumulate in the nervous tissues. J. Immunol. 173: 3310–3319. clonal stimulation of IgA production. In vaccine formulations, 23. Ågren, L., B. Lo¨wenadler, and N. Lycke. 1998. A novel concept in mucosal adjuvanticity: the CTA1-DD adjuvant is a B cell-targeted fusion protein that such strong mucosal IgA immunity may be critical for protection incorporates the enzymatically active cholera toxin A1 subunit. Immunol. Cell against many infectious diseases. Biol. 76: 280–287. In conclusion, we have shown that a combined vector compris- 24. Uhlen, M., B. Guss, B. Nilsson, S. Gatenbeck, L. Philipson, and M. Lindberg. 1984. Complete sequence of the staphylococcal gene encoding protein A. A gene ing CTA1-DD incorporated into ISCOMs has considerable poten- evolved through multiple duplications. J. Biol. Chem. 259: 1695–1702. tial as a vaccine for mucosal immunization with a variety of pro- 25. Ljungberg, U. K., B. Jansson, U. Niss, R. Nilsson, B. E. Sandberg, and

B. Nilsson. 1993. The interaction between different domains of staphylococcal by guest on September 28, 2021 tein Ags. By targeting the CTA1 adjuvant to both DCs and B cells Ј protein A and human polyclonal IgG, IgA, IgM and F(ab )2: separation of affinity as APCs, it allows powerful immune responses to be induced using from specificity. Mol. Immunol. 30: 1279–1285. low doses of Ags and points the way to a new generation of ra- 26. Ågren, L. C., L. Ekman, B. Lo¨wenadler, and N. Y. Lycke. 1997. Genetically engineered nontoxic vaccine adjuvant that combines B cell targeting with immu- tionally designed vaccines. nomodulation by cholera toxin A1 subunit. J. Immunol. 158: 3936–3946. 27. Robson, N. C., H. Beacock-Sharp, A. M. Donachie, and A. M. Mowat. 2003. The Disclosures role of antigen-presenting cells and interleukin-12 in the priming of antigen- specific CD4ϩ T cells by immune stimulating complexes. Immunology 110: 95– The authors have no financial conflict of interest. 104. 28. Robson, N. C., H. Beacock-Sharp, A. M. Donachie, and A. M. Mowat. 2003. References Dendritic cell maturation enhances CD8ϩ T-cell responses to exogenous antigen via a proteasome-independent mechanism of major histocompatibility complex 1. Medina, E., and C. A. Guzman. 2000. Modulation of immune responses follow- class I loading. Immunology 109: 374–383. ing antigen administration by mucosal route. FEMS Immunol. Med. Microbiol. 29. Ågren, L., E. Sverremark, L. Ekman, K. Scho¨n,B.Lo¨wenadler, C. Fernandez, 27: 305–311. and N. Lycke. 2000. The ADP-ribosylating CTA1-DD adjuvant enhances T cell- 2. Del Giudice, G., M. Pizza, and R. Rappuoli. 1999. Mucosal delivery of vaccines. dependent and independent responses by direct action on B cells involving anti- Methods 19: 148–155. apoptotic Bcl-2- and germinal center-promoting effects. J. Immunol. 164: 3. Garside, P., and A. M. Mowat. 2001. Oral tolerance. Semin. Immunol. 13: 6276–6286. 177–185. 30. Eriksson, A., and N. Lycke. 2003. The CTA1-DD vaccine adjuvant binds to 4. Salmond, R. J., J. A. Luross, and N. A. Williams. 2002. Immune modulation by human B cells and potentiates their T cell stimulating ability. Vaccine 22: the cholera-like enterotoxins. Expert Rev. Mol. Med. 2002: 1–16. 185–193. 5. Lavelle, E. C., A. Jarnicki, E. McNeela, M. E. Armstrong, S. C. Higgins, O. Leavy, and K. H. Mills. 2004. Effects of cholera toxin on innate and adaptive 31. Mowat, A. M., A. M. Donachie, S. Ja¨gewall, K. Scho¨n,B.Lo¨wenadler, immunity and its application as an immunomodulatory agent. J. Leukocyte Biol. K. Dalsgaard, P. Kaastrup, and N. Lycke. 2001. CTA1-DD-immune stimulating 75: 756–763. complexes: a novel, rationally designed combined mucosal vaccine adjuvant ef- 6. Harandi, A. M. 2004. The potential of immunostimulatory CpG DNA for induc- fective with nanogram doses of antigen. J. Immunol. 167: 3398–3405. ing immunity against genital herpes: opportunities and challenges. J. Clin. Virol. 32. Moschos, S. A., V. W. Bramwell, S. Somavarapu, and H. O. Alpar. 2005. Com- 30: 207–210. parative immunomodulatory properties of a chitosan-MDP adjuvant combination 7. Sajic, D., A. J. Patrick, and K. L. Rosenthal. 2005. Mucosal delivery of CpG following intranasal or intramuscular immunisation. Vaccine 23: 1923–1930. oligodeoxynucleotides expands functional dendritic cells and macrophages in the 33. Baudner, B. C., M. M. Giuliani, J. C. Verhoef, R. Rappuoli, H. E. Junginger, and vagina. Immunology 114: 213–224. G. D. Giudice. 2003. The concomitant use of the LTK63 mucosal adjuvant and 8. McCluskie, M. J., and H. L. Davis. 1999. CpG DNA as mucosal adjuvant. Vac- of chitosan-based delivery system enhances the immunogenicity and efficacy of cine 18: 231–237. intranasally administered vaccines. Vaccine 21: 3837–3844. 9. Bacon, A., J. Makin, P. J. Sizer, I. Jabbal-Gill, M. Hinchcliffe, L. Illum, 34. Haneberg, B., D. Kendall, H. M. Amerongen, F. M. Apter, J. P. Kraehenbuhl, and S. Chatfield, and M. Roberts. 2000. Carbohydrate biopolymers enhance antibody M. R. Neutra. 1994. Induction of specific immunoglobulin A in the small intes- responses to mucosally delivered vaccine antigens. Infect. Immun. 68: tine, colon-rectum, and vagina measured by a new method for collection of se- 5764–5770. cretions from local mucosal surfaces. Infect. Immun. 62: 15–23. 10. Baldridge, J. R., Y. Yorgensen, J. R. Ward, and J. T. Ulrich. 2000. Monophos- 35. Bromander, A. K., L. Ekman, M. Kopf, J. G. Nedrud, and N. Y. Lycke. 1996. phoryl lipid A enhances mucosal and systemic immunity to vaccine antigens IL-6-deficient mice exhibit normal mucosal IgA responses to local immunizations following intranasal administration. Vaccine 18: 2416–2425. and Helicobacter felis infection. J. Immunol. 156: 4290–4297. 11. Marciani, D. J. 2003. Vaccine adjuvants: role and mechanisms of action in vac- 36. Ågren, L. C., L. Ekman, B. Lo¨wenadler, J. G. Nedrud, and N. Y. Lycke. 1999. cine immunogenicity. Drug Discov. Today 8: 934–943. Adjuvanticity of the cholera toxin A1-based gene fusion protein, CTA1-DD, is 3706 COMBINED CTA1-DD/ISCOM ADJUVANT ACTS THROUGH B CELLS

critically dependent on the ADP-ribosyltransferase and Ig-binding activity. J. Im- 48. Beacock-Sharp, H., A. M. Donachie, N. C. Robson, and A. M. Mowat. 2003. A munol. 162: 2432–2440. role for dendritic cells in the priming of antigen-specific CD4ϩ and CD8ϩ T 37. Tsuji, T., Inoue, T., Miyama, A., Noda, M. 1991. Glutamatic acid-112 of the lymphocytes by immune-stimulating complexes in vivo. Int. Immunol. 15: A-subunit of heat-labile enterotoxin from enterotoxigenic Escherichia coli is im- 711–720. portant for ADP-ribosyltransferase activity. FEBS Lett. 291: 319–321. 49. Lindquist, R. L., G. Shakhar, D. Dudziak, H. Wardemann, T. Eisenreich, 38. Ga¨rdby, E., P. Lane, and N. Y. Lycke. 1998. Requirements for B7-CD28 co- M. L. Dustin, and M. C. Nussenzweig. 2004. Visualizing dendritic cell networks stimulation in mucosal IgA responses: paradoxes observed in CTLA4-H␥1 trans- in vivo. Nat. Immunol. 5: 1243–1250. genic mice. J. Immunol. 161: 49–59. 50. Katakai, T., T. Hara, J. H. Lee, H. Gonda, M. Sugai, and A. Shimizu. 2004. A 39. Smith, R. E., A. M. Donachie, and A. M. Mowat. 1998. Immune stimulating novel reticular stromal structure in lymph node cortex: an immuno-platform for complexes as mucosal adjuvants. Immunol. Cell Biol. 76: 263–269. interactions among dendritic cells, T cells and B cells. Int. Immunol. 16: 40. Sjo¨lander, A., D. Drane, E. Maraskovsky, J. P. Scheerlinck, A. Suhrbier, 1133–1142. J. Tennent, and M. Pearse. 2001. Immune responses to ISCOM formulations in 51. Itano, A. A., S. J. McSorley, R. L. Reinhardt, B. D. Ehst, E. Ingulli, animal and primate models. Vaccine 19: 2661–2665. A. Y. Rudensky, and M. K. Jenkins. 2003. Distinct dendritic cell populations sequentially present antigen to CD4 T cells and stimulate different aspects of 41. Mowat, A. M., K. J. Maloy, and A. M. Donachie. 1993. Immune stimulating cell-mediated immunity. Immunity 19: 47–57. complexes as adjuvants for inducing local and systemic immunity after oral im- 52. Yan, J., M. J. Wolff, J. Unternaehrer, I. Mellman, and M. J. Mamula. 2005. munization with protein antigens. Immunology 80: 527–534. Targeting antigen to CD19 on B cells efficiently activates T cells. Int. Immunol. 42. Mowat, A. M., and G. Reid. 2001. The preparation of immune stimulating com- 17: 869–877. plexes (ISCOMS) as adjuvants for local and systemic immunisation with protein 53. Grdic, D., L. Ekman, K. Scho¨n, K. Lindgren, J. Mattsson, K. E. Magnusson, antigens. In Current Protocols in Immunology. J. E. Coligan, A. M. Kruisbeek, P. Ricciardi-Castagnoli, and N. Lycke. 2005. Splenic marginal zone dendritic D. Margulies, E. Shevach, and W. Strober, eds. Wiley & Sons, New York, pp. cells mediate the cholera toxin adjuvant effect: dependence on the ADP-ribosyl- 2.11.11–2.11.12. transferase activity of the holotoxin. J. Immunol. 175: 5192–5202. 43. Maloy, K. J., A. M. Donachie, D. T. O’Hagan, and A. M. Mowat. 1994. Induction 54. Lanzavecchia, A. 1990. Receptor-mediated antigen uptake and its effect on an- of mucosal and systemic immune-responses by immunization with ovalbumin tigen presentation to class II-restricted T lymphocytes. Annu. Rev. Immunol. 8: entrapped in poly(lactide-co-glycolide) microparticles. Immunology 81: 773–793. 661–667. 55. Gourley, T. S., E. J. Wherry, D. Masopust, and R. Ahmed. 2004. Generation and Downloaded from 44. Maloy, K. J., A. M. Donachie, and A. M. Mowat. 1995. Induction of Th1 and Th2 ϩ maintenance of immunological memory. Semin. Immunol. 16: 323–333. CD4 T cell responses by oral or parenteral immunization with ISCOMS. Eur. 56. Gaspal, F. M., M. Y. Kim, F. M. McConnell, C. Raykundalia, V. Bekiaris, and J. Immunol. 25: 2835–2841. P. J. Lane. 2005. Mice deficient in OX40 and CD30 signals lack memory anti- 45. Mowat, A. M., A. M. Donachie, G. Reid, and O. Jarrett. 1991. Immune stimu- body responses because of deficient CD4 T cell memory. J. Immunol. 174: lating complexes containing Quil A and protein antigen prime class I MHC- 3891–3896. restricted T lymphocytes in vivo and are active by the oral route. Immunology 72: 57. Smith, K. M., J. M. Brewer, A. M. Mowat, Y. Ron, and P. Garside. 2004. The 317–322. influence of follicular migration on T-cell differentiation. Immunology 111:

46. Takahashi, H., T. Takeshita, B. Morein, S. Putney, R. N. Germain, and 248–251. http://www.jimmunol.org/ ϩ J. Berzofsky. 1990. Induction of CD8 cytotoxic T cells by immunisation with 58. Grdic, D., R. Smith, A. Donachie, M. Kjerrulf, E. Ho¨rnquist, A. Mowat, and puriﬁed HIV-1 envelope protein in ISCOMS. Nature 344: 873–875. N. Lycke. 1999. The mucosal adjuvant effects of cholera toxin and immune- 47. Lycke, N. 2005. From toxin to adjuvant: basic mechanisms for the control of stimulating complexes differ in their requirement for IL-12, indicating different mucosal IgA immunity and tolerance. Immunol. Lett. 97: 193–198. pathways of action. Eur. J. Immunol. 29: 1774–1784. by guest on September 28, 2021