cholera toxin and escherichia coli heat-labile enterotoxin, but … · edge of their modes of...

INFECTION AND IMMUNITY, May 2009, p. 1924–1935 Vol. 77, No. 50019-9567/09/$08.00�0 doi:10.1128/IAI.01559-08Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Cholera Toxin and Escherichia coli Heat-Labile Enterotoxin, but NotTheir Nontoxic Counterparts, Improve the Antigen-Presenting Cell

Function of Human B Lymphocytes�

Donatella R. M. Negri,1* Dora Pinto,1 Silvia Vendetti,1 Mario Patrizio,2 Massimo Sanchez,3Antonella Riccomi,1 Paolo Ruggiero,4 Giuseppe Del Giudice,4

and Maria Teresa De Magistris1

Department of Infectious, Parasitic and Immune-Mediated Diseases,1 Department of Therapeutic Research andMedicines Evaluation,2 and Department of Cell Biology and Neurosciences,3 Istituto Superiore di Sanità,

Rome, Italy, and Research Center, Novartis Vaccines, Siena, Italy4

Received 23 December 2008/Accepted 9 February 2009

B lymphocytes play an important role in the immune response induced by mucosal adjuvants. In this studywe investigated the in vitro antigen-presenting cell (APC) properties of human B cells upon treatment withcholera toxin (CT) and Escherichia coli heat-labile enterotoxin (LT) and nontoxic counterparts of these toxins,such as the B subunit of CT (CT-B) and the mutant of LT lacking ADP ribosyltransferase activity (LTK63).Furthermore, forskolin (FSK), a direct activator of adenylate cyclase, and cyclic AMP (cAMP) analogues wereused to investigate the role of the increase in intracellular cAMP caused by the A subunit of CT and LT. Blymphocytes were cultured with adjuvants and polyclonal stimuli necessary for activation of B cells in theabsence of CD4 T cells. Data indicated that treatment with CT, LT, FSK, or cAMP analogues, but not treatmentwith CT-B or LTK63, upregulated surface activation markers on B cells, such as CD86 and HLA-DR, andinduced inhibition of the proliferation of B cells at early time points, while it increased cell death in long-termcultures. Importantly, B cells treated with CT, LT, or FSK were able to induce pronounced proliferation of bothCD4� and CD8� allogeneic T cells compared with untreated B cells and B cells treated with CT-B and LTK63.Finally, only treatment with toxins or FSK induced antigen-specific T-cell proliferation in Mycobacteriumtuberculosis purified protein derivative or tetanus toxoid responder donors. Taken together, these resultsindicated that the in vitro effects of CT and LT on human B cells are mediated by cAMP.

The development of effective mucosal vaccines has beenhindered by the lack of useful adjuvants and our limited knowl-edge of their modes of action. Cholera toxin (CT) from Vibriocholerae and Escherichia coli heat-labile enterotoxin (LT) arepotent immunological adjuvants, as indicated by mouse vac-cine studies, although their mechanisms of action are not fullyunderstood. These toxins are holotoxins composed of an en-zymatically active A subunit that is noncovalently linked to apentamer of B subunits binding a variety of galactose-contain-ing molecules present in the plasma membranes of eukaryoticcells. CT binds mostly to the ganglioside GM1, which is be-lieved to be the major toxin receptor, whereas LT binds notonly to GM1 but also to other glycosphingolipids. Once inter-nalized, the A subunit ADP ribosylates the � subunit of theGTP-binding regulatory protein Gs, thereby inducing perma-nent adenylate cyclase activation, resulting in an increase in thelevel of intracellular cyclic AMP (cAMP) (reviewed in refer-ence 34).

The potentiation of antigen-presenting cell (APC) functionis a major aspect of adjuvant action, and it has been shown thatCT and LT induce maturation of both murine dendritic cells(DC) (26, 36) and human DC (5, 14, 15). Several studies

demonstrated the ability of these toxins to promote B-cellisotype switch differentiation in mice (19, 27) and upregulationof activation markers in both murine and human B cells (2–4).While these toxins are potent adjuvants, their toxicity makesthem unsuitable for human use. For this reason, a number ofinvestigators have tried to develop nontoxic derivatives of CTand LT that retain adjuvanticity either by removing the Adomain or by rendering it enzymatically inactive by site-di-rected mutagenesis (34). Although the current data suggestthat the enzymatic activity of CT and LT holotoxins is respon-sible for the most potent adjuvant activity, a number of reportsproposed that there are multiple immune modulating pathwaysthat are triggered by CT and LT, including mechanisms inde-pendent of ADP ribosyltransferase activity (11, 13, 30, 33, 42).Numerous studies have suggested that engagement of the gan-glioside GM1, the major receptor for CT and LT, is requiredfor the ability of these molecules to modulate immune re-sponses (22, 31). Recently, workers demonstrated that in theabsence of the toxic A subunit, the B subunit of CT (CT-B)induces intracellular signaling associated with the in vitro ac-tivation of murine B cells and macrophages (37).

The majority of these studies have been performed withmurine cells and have confirmed the in vivo adjuvanticity ofnontoxic compounds, such as CT-B and LTK63, a mutant ofLT lacking the ADP ribosyltransferase enzymatic activity,when they were mucosally delivered into animals, even if theimmune responses observed in the in vivo studies were usuallyweaker than those induced by the wild-type toxins (6, 11, 20,

* Corresponding author. Mailing address: Department of Infectious,Parasitic and Immune-Mediated Diseases, Istituto Superiore di Sanità,Viale Regina Elena 299, 00161 Rome, Italy. Phone: 39-6-49902734.Fax: 39-6-49902886. E-mail: [email protected].

� Published ahead of print on 17 February 2009.

1924

on May 30, 2021 by guest

http://iai.asm.org/

Dow

nloaded from

https://crossmark.crossref.org/dialog/?doi=10.1128/IAI.01559-08&domain=pdf&date_stamp=2009-05-01http://iai.asm.org/

36, 40, 41). In order to develop a mucosal adjuvant for humanvaccine, the mechanism(s) of action of potential nontoxic ad-juvants should be investigated in vitro by using human APC. Ithas been shown that the B-cell antigen-presenting functionsmay be important for the induction of optimal vaccine-inducedresponses (10, 35). Moreover, B cells are present in mucosa-associated lymphoid tissues (8), and their function in thesesites is related not only to immunoglobulin (Ig) production butalso to their antigen-presenting properties (24). To elucidatethe mechanisms by which enterotoxins modulate antigen-pre-senting properties, we decided to carry out a comprehensiveand comparative analysis of the effects of the toxins and theirnontoxic derivatives on the APC function of human B cells.Here we present evidence that CT and LT, as well as forskolin(FSK) and cAMP analogues, but not CT-B and LTK63, in-crease the activation of human B cells and induce improve-ment in their APC capability, indicating that the presence ofthe enzymatic subunit is critical for their adjuvanticity.

MATERIALS AND METHODS

Recombinant enterotoxins. CT and CT-B were purchased from List BiologicalLaboratories (Campbell, CA); E. coli LT was purchased from Swiss SerumVaccine Institute (Berne, Switzerland); LTK63 was provided by Novartis (Siena,Italy); and FSK, dibutyryl-cAMP (Db-cAMP), and 8-bromo-cAMP (8Br-cAMP)were purchased from Sigma Chemical Co. (St. Louis, MO). Endotoxin contam-ination in adjuvant preparations was evaluated by Limulus amoebocyte lysateanalysis (Pyrochrome; Associates of Cape Cod, Falmouth, MA). The concentra-tion of endotoxin was less than 0.09 endotoxin unit /�g in all of the preparationsutilized in this study.

Isolation and activation of B cells. Human B cells were isolated from periph-eral blood mononuclear cells (PBMC) from healthy donors by positive selectionusing anti-CD19 microbeads and the manufacturer’s suggested protocol (Milte-nyi Biotec S.r.l., Bologna, Italy). The cells obtained were �95% CD19 positive,as assessed by flow cytometry analysis. B cells were cultured in 24-well plates orin 96-well plates at a concentration of 1.5 � 106 to 2 � 106 cells/ml in RPMI 1640medium (GIBCO Invitrogen, Paisley, United Kingdom) supplemented with 100U/ml of penicillin-streptomycin-glutamine (GIBCO Invitrogen, Paisley, UnitedKingdom), 10% heat-inactivated fetal bovine serum (Euroclone, Life SciencesDivision, Pero, Italy), sodium pyruvate (Euroclone), and nonessential aminoacids (Euroclone). In order to obtain polyclonal stimulation, B cells were cul-tured in the presence of 2.5 �g/ml of CpG ODN 2006 (MWG Biotech, M-Medical, Milan, Italy), 50 U/ml of interleukin-2 (IL-2) (BD Biosciences, SanDiego, CA), and 2 �g/ml of anti-Ig monoclonal antibody (MAb) (Jackson Im-munoResearch Laboratories, Suffolk, United Kingdom). In addition, togetherwith the stimuli, on day zero B cells were either treated with 3 �g/ml of CT, 10�g/ml of CT-B, 0.1 �g/ml of LT, 10 �g/ml of LTK63, 50 �M of FSK, 0.5 mMDb-cAMP, or 8Br-cAMP or left untreated. In some experiments CD27� andCD27� B cells were isolated by sorting total B cells with FACSAria (BD Bio-sciences, San Diego, CA). Briefly, B cells were isolated by using anti-CD19microbeads, as described above. CD27� and CD27� B-cell subsets were purifiedbased on CD27 cell surface expression by FACSAria after staining with phyco-erythrin-conjugated anti-CD27 MAb (Immunological Sciences, Rome, Italy).Dead cells were excluded on the basis of propidium iodide (PI) (5 �g/ml; BDBiosciences) fluorescence intensity. The two subpopulations were stimulatedwith polyclonal stimuli and treated with adjuvants as described above for theunfractionated B cells.

Determination of intracellular cAMP content. B cells (2 � 105 cells in 200 �l,seeded in duplicate) were stimulated and treated for 24 h with adjuvants or FSKor left untreated in the presence of 100 �M 3-isobutyl-1-methylxanthine (SigmaChemical Co., St. Louis, MO), which inhibits cAMP-hydrolyzing phosphodies-terases, in order to avoid cAMP degradation. The culture medium was removedafter 10 min of centrifugation at 1,300 rpm, and cold 0.1 N HCl was used to lysethe cells. The intracellular cAMP content was measured by an enzyme-linkedimmunoassay by following the manufacturer’s instructions (Biotrak EIA, GEHealthcare).

Flow cytometric immunofluorescence analysis of surface markers and apop-totic cells. B cells were stained with the following mouse anti-human MAbsobtained from Becton Dickinson (BD Biosciences, San Diego, CA): phyco-

erythrin-labeled anti-CD86, anti-HLA class I, anti-CD80, and anti-CD40 andperidinin chlorophyll protein-labeled anti-CD20 and anti-HLA class II. Isotype-matched mouse IgG MAbs were used as controls. To evaluate B-cell death,stimulated B cells, treated as indicated above for 3 and 5 days, were stained withAnnexin V-fluorescein isothiocyanate (FITC) plus PI (Annexin V-FITC apop-tosis detection kit II; BD PharMingen, San Diego, CA) by following the manu-facturer’s instructions. Flow cytometric analysis of the cells was performed usinga FACSCalibur and CellQuest software (BD Biosciences).

B-cell proliferation assay. PBMC or purified B cells were labeled with 2.5 �Mcarboxyfluorescein succinimidyl ester (CFSE) (Molecular Probes, Eugene, OR)in phosphate-buffered saline containing 1% fetal bovine serum for 10 min at37°C, washed in complete RPMI 1640 medium, and then seeded in 24-well or96-well culture plates (1.5 � 106 cells/ml) containing polyclonal stimuli (seeabove) and subjected to different treatments. After 3 and 5 days of culture, thecells were washed and stained with MAbs against human CD4, CD8, or CD20from Becton Dickinson. The amounts of cell proliferation in the cell populationswere quantified by monitoring the sequential loss of fluorescence intensity of theCFSE-labeled cells using a FACSCalibur.

Mixed allogeneic cultures and antigen-specific presentation assay. Purified,stimulated B cells treated as indicated above or left untreated for 3 days, care-fully washed, and irradiated (3,000 rads) were cocultured with allogeneic PBMClabeled with CFSE (see above). A total of 1 � 105 PBMC per well were seededonto 96-well plates (Sigma-Aldrich S.r.l., Milan, Italy) with titrated numbers ofirradiated B cells (the B cell/PBMC ratio ranged from 1:16 to 1:1). After 3 daysof coculture, cells were collected and stained with anti-human CD4 or anti-human CD8 MAbs. The levels of PBMC, CD4�, and CD8� T-cell proliferationwere evaluated by fluorescence-activated cell sorting (FACS) analysis. To exam-ine the antigen-specific response, B cells isolated from Mycobacterium tubercu-losis purified protein derivative (PPD) responder donors or tetanus toxoid (TT)responder donors were stimulated with CpG, IL-2, and PPD (Statens SerumInstitute, Copenhagen, Denmark) or with TT (Novartis, Siena, Italy) at day zero.Toxins, CT-B, LTK63, and FSK were added on the same day. PPD and TT werealso added on day 2. On day 3, cells were extensively washed, irradiated, andcocultured with autologous PBMC previously labeled with CFSE. CD4� andCD8� T-cell proliferation was evaluated after 5 days of coculture using a FAC-SCalibur and CellQuest software.

In some experiments anti-human CD86 and/or anti-human HLA-DR MAbswere used in order to block the interaction between APC and T cells in a mixedleukocyte reaction (MLR) assay. Briefly, B cells were incubated for 2 h at 4°Cwith 20 �g/ml of anti-CD86 (BU63; mouse IgG1; Ancell Immunology ResearchProducts, Bayport, MN) and/or anti-HLA-DR (G46-6; mouse IgG2a; BD Bio-sciences, San Diego, CA) MAbs, extensively washed to remove free MAbs, andthen added to CFSE-labeled PBMC.

Cytokine production. Cytokine concentrations in supernatants collected fromstimulated B cells treated as indicated above or left untreated for 3 days weredetermined by enzyme-linked immunosorbent assays (ELISA), including assaysfor tumor necrosis factor alpha (TNF-�) and IL-1� (Pierce Endogen, Rockford,IL), IL-6 (BD Biosciences, San Diego, CA), and IL-12 (R&D Systems, Minne-apolis, MN).

Statistical analysis. Microsoft Excel (Microsoft Corporation, Redmond, WA)was used for statistical analysis. Data were expressed as means � standarddeviations, and statistical significance was determined by Student’s t test. A Pvalue of 0.05 was considered statistically significant.

RESULTS

CT and LT induce upregulation of surface activation mark-ers. The levels of expression of activation markers CD86,CD80, HLA class I and II molecules, and CD40 on the B-cellsurface were examined after treatment with CT, CT-B, LT,LTK63, FSK, or cAMP analogues, such as Db-cAMP and8Br-cAMP, for 3 and 5 days. As expected, stimulation of Bcells cultured in the presence of polyclonal stimuli (stimulatedB cells), such as CpG ODN 2006, anti-Ig MAb, and IL-2,induced upregulation of CD86 and HLA-DR markers that wasdetectable after 3 days of culture compared to the results forunstimulated B cells (Fig. 1). Treatment of both unstimulatedand stimulated B cells with CT, LT, FSK, or cAMP analoguesinduced upregulation of these markers. As shown in Fig. 1, for

VOL. 77, 2009 EFFECTS OF CT AND LT ON HUMAN B-CELL APC FUNCTION 1925


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

1926


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

CD86 expression both the percentage of positive cells and themean fluorescence intensity (MFI) were strongly increased,whereas for HLA-DR, already expressed on the majority of Bcells, there was a marked increase in MFI following treatmentwith CT, LT, FSK, or cAMP analogues. None of the treat-ments was able to induce an evident change in CD40 and HLAclass I expression, whereas slight downregulation of CD80 ex-pression was observed in stimulated B cells treated with toxin,FSK, or cAMP analogues (data not shown). The presence ofCT-B or LTK63 in the culture did not induce modulation ofthe expression of any of the markers analyzed (Fig. 1). Asshown in dot plots in Fig. 1, isolated B cells cultured withoutpolyclonal stimulation showed high mortality. For this reasonwe decided to perform the next experiments in the presence ofpolyclonal stimuli, which were also necessary for analysis ofB-cell proliferation and cytokine production.

CT and LT induce an increase in the intracellular cAMPlevels. In order to check the enzymatic activity of adjuvants,intracellular cAMP was evaluated in B cells treated with thedifferent compounds. As expected, CT, LT, and FSK, but notCT-B or LTK63, induced increases in intracellular cAMP lev-els in B cells (Fig. 2). The results showed that there was astatistically significant difference (P 0.05) between B cellstreated with CT, LT, or FSK and untreated control (5,436 �1,001, 5,054 � 1,142, 11,613 � 2,178, and 1,458 � 266 fmol/106

cells, respectively). Moreover, FSK-treated B cells containedstatistically significant larger amounts of intracellular cAMPthan toxin-treated cells (Fig. 2).

CT and LT inhibit proliferation of B cells and increase theirsusceptibility to death. In order to evaluate the effect of adju-vants on cell proliferation, B lymphocytes were stained withCSFE and cultured for 3 and 5 days with polyclonal stimuli andwith CT, CT-B, LT, LTK63, FSK, or cAMP analogues or leftuntreated. Figure 3 shows the results of a representative ex-periment. Polyclonal stimuli induced proliferation after 3 daysof culture (67.7% proliferating B cells), which increased after5 days (83.1%). Similar percentages of dividing B cells wereevident when the cells were treated with CT-B or LTK63(65.8% and 63.6%, respectively, at 3 days and 81.7% and 86.2%,respectively, at 5 days), suggesting that these compounds didnot act on the ability of stimulated B cells to proliferate. Con-versely, proliferation of CT-treated B cells, and to a lesserextent LT-treated B cells, was significantly less (P 0.05) bothat day 3 (35.3% and 45.5%, respectively) and at day 5 (62.9%and 71.7%, respectively). Similar results were obtained when Bcells were treated with Db-cAMP and 8Br-cAMP (37.1% and37.5%, respectively, at day 3 and 57.7% and 57.1%, respec-tively, at day 5). The inhibition of cell proliferation induced bytoxins was evident not only based on the percentage of prolif-erating cells but also based on MFI, which indirectly indicatedthe rounds of B-cell division under each condition analyzed(Fig. 3). When stimulated B cells were treated with FSK, theirability to proliferate was inhibited even more than it was whenthey were treated with toxins (18.7% at day 3 and 25.3% at day5), probably because of the high dose of FSK used (50 �M).The selection of this dose was based on the observation thatlower doses were not sufficient to induce activation of humanDC (5).

At the same time, in order to understand if the evidentinhibition of proliferation observed was related to an increasein B-cell death, the effects of toxins on B-cell viability wereevaluated. The percentage of live cells was initially assessed bygating the events (R1) on a dot plot with forward and sidescatter parameters (Fig. 4A). On day 3, even if treatment withtoxins, FSK, and cAMP analogues resulted in a slight reductionin viability, the viability of B cells was high under all conditionstested, suggesting that at this time cell death was not the causeof the marked inhibition of B-cell proliferation described above.After 5 days of culture the percentage of live B cells decreased tothe same level in the control (untreated) or CT-B- or LTK63-treated samples. Treatment with FSK resulted in a further de-crease in viability (P 0.05), whereas CT, LT, and cAMP ana-logues induced a remarkable reduction in viability (P 0.05). Toconfirm these results, we labeled the cells with Annexin V and PIafter 3 and 5 days of culture in the presence or absence of adju-vants or FSK. As shown by the results of a representative exper-iment (Fig. 4B), after 3 days of culture the percentages of labeledcells (PI-positive cells, Annexin V-positive cells, and PI- and An-

FIG. 1. Effect of adjuvants on the expression of cell surface activation markers. Enzymatic activity is required for CT and LT to upregulateactivation markers on the B-cell surface. Unstimulated B cells (left panel) or B cells stimulated with 2.5 �g/ml of CpG ODN 2006, 50 U/ml of IL-2,and 2 �g/ml of anti-Ig MAb (right panel) were simultaneously treated with the compounds indicated or left untreated (NT) for 3 days. Theexpression of cell surface markers was evaluated by FACS analysis of B cells stained with MAbs directed to CD86 and HLA-DR. Dot plots withforward scatter (FSC) and side scatter (SSC) parameters are shown to visualize the amounts of live cells (in the gate) analyzed for surface markers.The percentage of positive cells and the MFI are indicated in each graph. Representative data from five independent experiments are shown.PerCP, peridinin chlorophyll protein.

FIG. 2. Effect of adjuvants on intracellular cAMP. The level ofintracellular cAMP increases upon treatment with CT, LT, or FSK.Stimulated B cells were treated with the compounds indicated for 24 hin the presence of 3-isobutyl-1-methylxanthine and lysed with HCl.The amount of cAMP was evaluated by an enzyme immunoassay andwas expressed in fmol/106 B cells. The asterisks indicate statisticallysignificant differences (P 0.05) between CT-, LT-, or FSK-treated Bcells and untreated B cells (NT). The P values indicate the statisticallysignificant differences between FSK- and CT-treated cells and betweenFSK- and LT-treated cells. Representative data from three indepen-dent experiments are shown.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

1928


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

nexin V-positive cells) were similar under all conditions analyzedexcept for LT-treated B cells, for which there was a slight higherpercentage of total labeled cells. After 5 days, treatment with CT,LT, or FSK resulted in evident increases in the percentages of

both PI- and Annexin V-positive cells, suggesting that these treat-ments induced increased susceptibility to death compared to thatof B cells that were treated with CT-B or LTK63 or were leftuntreated.

FIG. 3. Effect of adjuvants on B-cell proliferation. The inhibitory effect of adjuvants on B-cell proliferation is related to the increase in theintracellular cAMP level. CFSE-labeled B cells were stimulated with polyclonal stimuli, including 2.5 �g/ml of CpG ODN 2006, 50 U/ml of IL-2,and 2 �g/ml of anti-Ig MAb, and simultaneously treated with the compounds indicated or left untreated (NT) for 3 and 5 days. The percentagesof dividing cells and the MFI of CFSE proliferating cells are indicated in the graphs. The data shown are data from one representative experimentof five experiments performed. DB, Db-cAMP; 8Br, 8Br-cAMP.

FIG. 4. Effect of adjuvants on B-cell viability. The enzymatic activity of CT and LT renders B cells more susceptible to death in long-termculture. B cells were stimulated with polyclonal stimuli, including 2.5 �g/ml of CpG ODN 2006, 50 U/ml of IL-2, and 2 �g/ml of anti-Ig MAb, andsimultaneously treated with the compounds indicated or left untreated (NT) for 3 and 5 days. (A) Percentage of live cells as evaluated by gatingthe events (R1) on a dot plot with forward scatter (FSC) and side scatter (SSC) parameters. The histogram shows the percentages of gated B cellsfrom five different donors analyzed at day 3 (open bars) and day 5 (filled bars). The error bars indicate standard deviations. Db, Db-cAMP; 8Br,8Br-cAMP. (B) Stimulated B cells, treated as indicated for 3 and 5 days, were stained with Annexin V-FITC plus PI. The percentages of AnnexinV-positive cells, PI-positive cells, and Annexin V-positive PI-positive cells are indicated in the plots. The data shown are data from onerepresentative experiment of five experiments performed.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

Toxin-treated B cells are efficient APC. In order to investi-gate the APC function of stimulated B cells treated with toxins,MLR assays were performed. In this setting, to maintain agood activation state and to avoid cell death, B cells were usedas APC after 3 days of stimulation and treatment. AllogeneicT-cell proliferation was analyzed by FACS to determine thecontent of CFSE after 3 days of coculture with irradiated Bcells as stimulators. Figure 5 shows the proliferation of PBMCand CD4� and CD8� T cells from three different donors. Bcells stimulated with polyclonal stimuli were able to induce lowlevels of PBMC proliferation. Treatment with toxins or withFSK clearly improved the APC function of stimulated B cells,as shown by the increased proliferation of allogeneic PBMC. Inparticular, CT-, LT-, or FSK-treated B cells induced bothCD4� and CD8� T cells to proliferate at any ratio of B cells toPBMC used (1:16 to 1:1) in a dose-dependent manner. Theincrease in T-cell proliferation ranged from two- to fivefoldcompared with the T-cell proliferation observed for the un-treated control. Similar results were obtained when B cellswere treated with Db-cAMP (data not shown). Conversely,treatment of B cells with CT-B and LTK63 did not inducevariation in T-cell proliferation compared to the results ob-tained with untreated B cells at any ratio. These data stronglysuggested that CT and LT improved the efficiency of B cellsacting as APC and that the mechanism of this effect was re-lated to the increase in the intracellular cAMP level, as indi-cated by the effect of FSK-treated B cells on allogeneic T-cellproliferation.

To analyze the antigen-specific T-cell activation induced byB-cell antigen presentation, toxin-treated B cells were used as

APC in the autologous system in order to expand the antigen-specific T cells of either PPD or TT responder donors. Theisolated B cells were stimulated with CpG, IL-2, and PPDprotein or TT, and CT, LT, CT-B, LTK63, or FSK was addedon the same day. After 3 days, cells were cocultured withautologous PBMC previously labeled with CFSE. The prolif-eration of PBMC, CD4�, and CD8� T cells was evaluatedafter 5 days of coculture by FACS analysis. As shown in Fig. 6,only treatment with CT, LT, or FSK increased proliferation ofautologous PBMC, CD4�, and CD8� T cells at any ratio of Bcells to PBMC used, whereas CT-B or LTK63 did not enhancethe antigen-presenting capacity of B lymphocytes.

In order to check the role of CD86 and HLA-DR in theimprovement of the APC function of toxin-treated B cells,MLR experiments were performed in the presence of blockingMAb anti-CD86 and/or MAb anti-HLA-DR. As expected, thepresence of both blocking reagents inhibited up to 80% ofPBMC proliferation (Table 1) at ratio of B cells to PBMC of1:4, indicating the evident contribution of these molecules inthis system.

CT and LT inhibit TNF-� production while increasing IL-12, IL-1�, and IL-6 production. To examine the effect of ad-juvants on the production of cytokines, supernatants from Bcells cultured for 3 days with polyclonal stimuli and treatedwith toxins, FSK, or nontoxic counterparts or left untreatedwere analyzed by ELISA. Figure 7 shows the results obtainedfor five donors. Treatment with CT, LT, or FSK induced asignificant decrease in TNF-� production (56.3, 120.4, and 64.9pg/106 cells, respectively) compared to the untreated control(270 pg/106 cells). In contrast, treatment with CT-B or LTK63

FIG. 5. Proliferation of allogeneic PBMC cocultured with B cells as APC. Activation of B cells by an increased level of intracellular cAMPenhances their ability to present alloantigen in the allogeneic T-cell response. B cells were stimulated with polyclonal stimuli, including 2.5 �g/mlof CpG ODN 2006, 50 U/ml of IL-2, and 2 �g/ml of anti-Ig MAb, and simultaneously treated with the compounds indicated or left untreated (NT)for 3 days and irradiated. B cells were cocultured with allogeneic PBMC labeled with CFSE at different B cell/PBMC ratios. After 3 days ofcoculture, cells were collected and stained with anti-human CD4 or anti-human CD8 MAb. The level of PBMC, CD4�, and CD8� T-cellproliferation was evaluated by FACS analysis. The error bars indicate standard deviations for duplicates. Data from three independent experimentsare shown. PBMC TOT, total PBMC.

1930 NEGRI ET AL. INFECT. IMMUN.


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

did not significantly influence the cytokine levels (228 and 237pg/106 cells). As shown in Fig. 7, toxin- and FSK-treated B cellsshowed significant production of IL-12 and IL-1� cytokines. Inparticular, high levels of IL-1� were detected in toxin- andFSK-treated samples (1,520, 1,211, and 1,781 pg/106 B cellstreated with CT, LT, and FSK, respectively), whereas low butdetectable concentrations of IL-12 were present in B-cell su-pernatants from CT-, LT-, and FSK-treated samples (260, 243,and 250 pg/106 B cells). Finally, treatment with toxins and FSKinduced production of amounts of IL-6 larger than that inuntreated B cells (1,947, 1,542, and 1,273 pg/106 cells for Bcells treated with CT, LT, and FSK, respectively, versus 970pg/106 cells for untreated B cells). However, the increase inIL-6 production was not statistically significant, probably dueto the high variability among donors. Treatment with CT-B orLTK63 did not result in any significant difference in IL-12,IL-1�, or IL-6 production compared to untreated samples.

Effect of adjuvants on sorted CD27� and CD27� B-cellsubpopulations. Finally, in order to determine if the differentsubsets of B cells were targeted differently by adjuvants,CD27� and CD27� B cells were isolated and treated as de-scribed above for unsorted B cells. In particular, CD86 expres-sion and susceptibility to death were analyzed for the twodifferent B-cell populations after 3 days of culture. As shown inFig. 8A, for both subsets a dramatic increase in the level ofCD86 was evident in toxin- or FSK-treated cells compared tothe corresponding untreated sample, while the presence ofCT-B or LTK63 did not affect CD86 expression. Likewise, forinduction of cell death, the results indicated that treatmentwith toxins had similar effects on the two subsets (Fig. 8B).Indeed, at 3 days there was not a statistically significant differ-ence in the percentages of Annexin V- and PI-labeled cellsbetween the two subpopulations.

DISCUSSION

The specificity, magnitude, and quality of T-cell-mediatedimmune responses become conditioned during the early phaseof antigen presentation. For this reason, analysis of the effectsof adjuvants on APC could help identify the mechanism ofaction. In the population of APC that includes DC, B cells, andmacrophages, DC are the most potent, and for this reason themajority of adjuvant studies are focused on these cells. We andother authors have previously shown that CT and LT were ableto mature human DC and inhibit IL-12 and TNF-� production,whereas CT-B and LTK63 (5, 14, 15) did not have these abil-ities. Several papers described the APC function of B cells,indicating that in vivo B cells provide extra and essential anti-gen presentation capacity above that provided by DC, optimiz-ing expansion and allowing generation of memory and effectorT cells (9, 10, 17, 18, 23). Therefore, we decided to investigate

FIG. 6. Antigen-specific T-cell proliferation. Activation of B cells by an increased level of intracellular cAMP enhances their ability to presentPPD or TT in the autologous T-cell response. B cells derived from a PPD or TT responder donor were stimulated with 2.5 �g/ml of CpG ODN2006 and 50 U/ml IL-2 in the presence of PPD or TT and treated as indicated or left untreated (NT) for 3 days. Cells were then irradiated andcocultured with autologous PBMC labeled with CFSE at different B cell/PBMC ratios. After 5 days of coculture cells were collected and stainedwith anti-human CD4 or anti-human CD8 MAb. The levels of PBMC, CD4�, and CD8� T cell proliferation were evaluated by FACS analysis. Theerror bars indicate standard deviations for duplicates. PBMC TOT, total PBMC.

TABLE 1. MLR analysis in the presence of blocking MAbsa

B-celltreatment

% of proliferating cells (mean � SD)b

No blocking

Blocking MAbs

Anti-CD86 Anti-HLA-DR

Anti-CD86� anti-

HLA-DR

None 10.17 � 0.81 5.36 � 0.37 3.54 � 0.08 2.07 � 0.31CT 18.00 � 1.02 9.67 � 0.35 4.58 � 0.29 3.61 � 0.42CT-B 11.89 � 1.17 5.90 � 0.21 4.35 � 0.61 2.76 � 0.40LT 20.30 � 0.28 11.42 � 0.43 5.99 � 0.45 3.03 � 0.74LTK63 9.34 � 0.62 5.52 � 0.31 4.43 � 0.04 3.20 � 0.35FSK 18.86 � 0.73 12.00 � 1.12 6.48 � 0.50 6.05 � 0.78

a B cells preincubated with anti-CD86, with anti-HLA-DR, or with both anti-bodies were not able to induce proliferation of CFSE-labeled PBMC.

b The percentage of proliferating cells was determined using a B cell/PBMCratio of 1:4.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

the effects of toxins and their nontoxic counterparts on theantigen-presenting capacity of human B cells. In the presentstudy we performed an in vitro comparative evaluation of theAPC function of human B cells after treatment with CT, CT-B,LT, LTK63, or FSK, a direct activator of adenylate cyclase, asa positive control for an increase in intracellular cAMP, ordirectly with cAMP analogues, such as Db-cAMP and 8Br-cAMP. Our results show that the enzymatic activity of toxins iscrucial for in vitro activation of human B cells and improve-ment of their APC capacity. Indeed, CT and LT, which in-crease intracellular cAMP levels, induced an evident activa-tion state of human B cells, as judged by changes in surfacephenotype, whereas none of the enzymatically inactive de-rivatives of CT or LT tested in this study were able to modifyactivation markers. In addition, the functional changes in Bcells, including inhibition of proliferation, susceptibility tocell death, cytokine production, and an increase in the an-tigen-presenting capability induced by CT or LT, can bemimicked consistently by using the pharmacological agonistFSK or cAMP analogues.

To avoid the high rate of mortality of unstimulated B cellsand to prolong the in vitro cultures, we decided to performexperiments in the presence of polyclonal stimuli. These stim-uli, including antibody to human Ig as an antigen surrogate,CpG as a Toll-like receptor agonist, and IL-2 as a growthfactor, are required for activation of both naïve and memory Bcells in the absence of CD4� T cells (7). Treatment with toxinswas able to increase the expression of CD86 and HLA class II,confirming results obtained by other workers (2). Conversely,CT-B and LTK63, even if they were used at concentrationshigher than those used for the toxins, were unable to upregu-late the activation markers on human B cells. Recently, Schnit-

FIG. 8. Effects of CT and LT on CD27� and CD27� B-cell popu-lations. Toxins act similarly in both subsets in terms of activation andinduction of cell death. CD27� and CD27� B cells were sorted byFACSAria, stimulated with polyclonal stimuli, including 2.5 �g/ml ofCpG ODN 2006, 50 U/ml of IL-2, and 2 �g/ml of anti-Ig MAb, andsimultaneously treated with the compounds indicated or left untreated(NT) for 3 days. (A) Expression of CD86 was evaluated by FACSanalysis of CD27� and CD27� B-cell subsets. The histograms show thepercentages of CD86� cells for CD27� (open bars) or CD27� (filledbars) B lymphocytes from two donors. The error bars indicate standarddeviations. (B) Stimulated B-cell subsets, treated as indicated for 3days, were stained with Annexin V-FITC plus PI and analyzed byFACS. The percentages of Annexin V-positive and PI-positive cells areindicated. The error bars indicate standard deviations.

FIG. 7. Cytokine production. CT and LT strongly inhibit the production of TNF-� but increase the production of IL-12, IL-1�, and IL-6. B cellswere stimulated with polyclonal stimuli, including 2.5 �g/ml of CpG ODN 2006, 50 U/ml of IL-2, and 2 �g/ml of anti-Ig MAb, and simultaneouslytreated with the compounds indicated or left untreated (NT) for 3 days. B-cell culture supernatants from five different donors were collected andanalyzed for the presence of the cytokines indicated by ELISA. The error bars indicate standard deviations. The asterisks indicate a statisticallysignificant difference (P 0.05) between the treatment and control (NT) samples.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

zler et al. showed that treatment of murine B cells with CT-Binduced a sequence of signaling events related to cellularactivation and surface molecule expression (37). In ourstudy, treatment of both unstimulated and polyclonal acti-vated human B cells with CT-B did not induce any variationsin surface activation markers and antigen presentation, sug-gesting that there is a difference in behavior between humanand murine B cells. The precise mechanism of action ofthese adjuvants has not been completely elucidated, andthere are controversies concerning the requirements for androles of the A and B subunits of these toxins both in vitroand in vivo (reviewed in reference 16). Factors involved inthe dissimilar findings include the route of administration,the characteristics of the vaccine antigen, contaminationof the adjuvant with endotoxin or with holotoxin, and thespecies of animal used. The difference between the humanand murine cell responses to nontoxic derivatives of toxinscould be another important factor that should be taken intoaccount in the design and development of mucosal adjuvantssuitable for human vaccination.

It is known that the second messenger cAMP can haveimmunosuppressive effects on T and B lymphocytes (28, 32,38). We confirmed these findings, showing that CT, LT, FSK,and cAMP analogues inhibited the proliferation of B cellsinduced by polyclonal stimuli. FSK-treated B cells producedlarger amounts of intracellular cAMP than toxin-treated cells,suggesting that the difference could represent a possible reasonfor the more pronounced inhibition of proliferation seen forFSK-treated cells. These results are in agreement with thoseobtained previously by our group (39) and by Johnson et al.(21), showing that increased levels of cAMP were able toinhibit in a dose-dependent manner anti-CD3- or IL-2-inducedT-cell proliferation. In addition, toxins and FSK made B cellsmore susceptible to death. Conversely, treatment with CT-B orLTK63, which lack enzymatic activity, did not alter either theability of stimulated B cells to proliferate or the induction ofcell death. These results were expected, since it has beenshown that cAMP is involved in the regulation of apoptosis inB progenitor and mature B cells by inducing activation ofprotein kinase A (25, 29), suggesting that physiological ligandsthat control cellular cAMP levels could play an important rolein the regulation of B-cell maturation in vivo. Indeed, com-pounds acting on the increase in intracellular cAMP contentcould have pleiotropic effects on the immune cells, inducingboth suppressive (inhibition of proliferation) and stimulatorysignals (activation) at the same time. The final effect observedin vitro and even more in vivo upon treatment with toxins isprobably due to a balance of these signals.

In order to determine if the effects of toxins were directedmostly toward a particular subset of B cells, such as CD27� orCD27� populations, we first evaluated the expression of GM1on gated CD27� and CD27� B cells by using FITC-labeledCT-B. The results indicated that there was a slightly higherlevel of binding to CD27� B cells (data not shown). However,when CD86 expression and induction of cell death after treat-ment with toxins were evaluated for the two different B-cellpopulations, the results indicated that CT and LT acted simi-larly in both subsets. Taken together, our data indicated thatthe toxins induced a real increase in CD86 expression in total

B cells and did not cause selective depletion of the populationwith a low level of CD86 expression.

Finally, to investigate the effects of the adjuvants on theantigen-presenting function of human B cells, both allo-MLRand antigen-specific T-cell proliferation tests were performed.The ability of B cells treated with CT, LT, or FSK to induceT-cell proliferation was evident in both the assays. The datacould be explained by the fact that toxins and FSK were able toinduce an activation state with upregulation of costimulatorymolecules and HLA class II, which was responsible for theincreased antigen-presenting function observed. Indeed, as ex-pected, the use of blocking MAbs against CD86 and/orHLA-DR in the MLR assay resulted in a high level of inhibi-tion of PBMC proliferation. Again, treatment with CT-B andLTK63 did not improve the APC capacity of stimulated B cells,further confirming the role of cAMP in the adjuvant activity oftoxins. We and other authors reported that CT and LT inhibitIL-12 and TNF-� production by human DC, partially explain-ing the polarization of CD4� T cells toward a Th2 phenotypeobserved when CT- or LT-treated DC were used as APC (5,14, 15). In this study, we detected strong inhibition of TNF-�production upon treatment of B cells with CT, LT, or FSK,further supporting the role of cAMP in the modulation ofB-cell functions. Surprisingly, in contrast to the results for DCtreated with CT or LT, we observed IL-12 production in toxin-treated B cells and strong production of IL-1�. As reported inother papers (reviewed in reference 16), we confirmed that CTand LT induced an increase in IL-6 production. This peculiarpattern of cytokines induced by treatment with toxins could beimportant for the generation of an environment able to drivethe Th1/Th2 polarization of T cells. Additional studies arerequired to investigate this issue.

In this work, several observations support the hypothesisthat CT and LT directly activate human B cells predominantlyby elevating the intracellular cAMP level. We cannot excludethe possibility that there is concomitant involvement of otherfactors, including the signaling induced by binding with theirreceptors. Although in our model system the presence of en-zymatic activity is required for the adjuvanticity of toxins, sug-gesting that LTK63 and CT-B do not act directly on theseAPC, we cannot rule out the possibility that there is an indirecteffect on APC that is induced by these compounds in vivo. Ourresults are in agreement with the results of other studies basedon a requirement for the A subunit of CT for the induction ofadjuvanticity of B cells. Indeed, it has been demonstrated thatCTA1-DD, an adjuvant based on the CT A subunit geneticallylinked to two Ig-binding domains (DD) of staphylococcal pro-tein A, but not enzymatically inactive mutants, was able totarget and activate B cells and act as a good mucosal adjuvantin vivo (1, 12). In this study, our objective was to determine ifthe enzymatic activity is mandatory for the APC function of Bcells, and therefore we focused on reagents with 0% or 100%activity. It would be interesting to check the minimum level ofenzymatic activity required for adjuvanticity with human Bcells by using recombinant enterotoxins with greatly reducedenzymatic activity.

Together with our previous data (14, 15), results obtained inthis study allow us to conclude that the adjuvanticity of toxins,measured as in vitro activation and the antigen-presentingability of human APC (both DC and B cells), is stringently



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

correlated to the presence of the enzymatic activity involved inthe increase in the intracellular cAMP content.

ACKNOWLEDGMENTS

We thank Andrea Cara for suggestions and critical reading of themanuscript, Laura Pancotto for purification of LT, and EmanueleFanales-Belasio and Maria Rosaria Pavone-Cossut for performing theendotoxin analysis.

This study was carried out with financial support from Commissionof the European Communities Sixth Framework Programme contractLSHP-CT-2003-503240 (Mucosal Vaccines for Poverty-Related Dis-eases) (M.T.D.M.) and from grants from the Italian AIDS NationalProgram (contract 45G/C) (M.T.D.M.).

P.R. and G.D.G. are Novartis employees. The other authors declareno conflict of interest.

REFERENCES

1. Agren, L. C., L. Ekman, B. Löwenadler, J. G. Nedrud, and N. Y. Lycke. 1999.Adjuvanticity of the cholera toxin A1-based gene fusion protein, CTA1-DD,is critically dependent on the ADP-ribosyltransferase and Ig-binding activity.J. Immunol. 162:2432–2440.

2. Anastassiou, E. D., H. Yamada, M. L. Francis, J. J. Mond, and G. C. Tsokos.1990. Effects of cholera toxin on human B cells. Cholera toxin induces B cellsurface DR expression while it inhibits anti-mu antibody-induced cell pro-liferation. J. Immunol. 145:2375–2380.

3. Arce, S., H. Nawar, M. W. Russell, and T. D. Connell. 2005. Differentialbinding of Escherichia coli enterotoxins LT-IIa and LT-IIb and of choleratoxin elicits differences in apoptosis, proliferation, and activation in lymphoidcells. Infect. Immun. 73:2718–2727.

4. Arce, S., H. F. Nawar, G. Muehlinghaus, M. W. Russell, and T. D. Connell.2007. In vitro induction of immunoglobulin A (IgA)- and IgM-secretingplasma blasts by cholera toxin depends on T-cell help and is mediated byCD154 up-regulation and inhibition of gamma interferon synthesis. Infect.Immun. 75:1413–1423.

5. Bagley, K. C., S. F. Abdelwahab, R. G. Tuskan, T. R. Fouts, and G. K. Lewis.2002. Cholera toxin and heat-labile enterotoxin activate human monocyte-derived dendritic cells and dominantly inhibit cytokine production through acyclic AMP-dependent pathway. Infect. Immun. 70:5533–5539.

6. Baudner, B. C., O. Balland, M. M. Giuliani, P. Von Hoegen, R. Rappuoli, D.Betbeder, and G. Del Giudice. 2002. Enhancement of protective efficacyfollowing intranasal immunization with vaccine plus a nontoxic LTK63 mu-tant delivered with nanoparticles. Infect. Immun. 70:4785–4790.

7. Bernasconi, N. L., E. Traggiai, and A. Lanzavecchia. 2002. Maintenance ofserological memory by polyclonal activation of human memory B Cells.Science 298:2199–2202.

8. Brandtzaeg, P., and F. E. Johansen. 2005. Mucosal B cells: phenotypiccharacteristics, transcriptional regulation, and homing properties. Immunol.Rev. 206:32–63.

9. Constant, S., N. Schweitzer, J. West, P. Ranney, and K. Bottomly. 1995. Blymphocytes can be competent antigen-presenting cells for priming CD4� Tcells to protein antigens in vivo. J. Immunol. 155:3734–3741.

10. Crawford, A., M. Macleod, T. Schumacher, L. Corlett, and D. Gray. 2006.Primary T cell expansion and differentiation in vivo requires antigen presen-tation by B cells. J. Immunol. 176:3498–3506.

11. Douce, G., M. Fontana, M. Pizza, R. Rappuoli, and G. Dougan. 1997. Intra-nasal immunogenicity and adjuvanticity of site-directed mutant derivatives ofcholera toxin. Infect. Immun. 65:2821–2828.

12. Eriksson, A., and N. Y. Lycke. 2003. The CTA1-DD vaccine adjuvant bindsto human B cells and potentiates their T cell stimulating ability. Vaccine22:185–193.

13. Fontana, M. R., R. Manetti, V. Giannelli, C. Magagnoli, A. Marchini, R.Olivieri, M. Domenighini, R. Rappuoli, and M. Pizza. 1995. Construction ofnontoxic derivatives of cholera toxin and characterization of the immuno-logical response against the A subunit. Infect. Immun. 63:2356–2360.

14. Gagliardi, M. C., F. Sallusto, M. Marinaro, A. Langenkamp, A. Lanzavec-chia, and M. T. De Magistris. 2000. Cholera toxin induces maturation ofhuman dendritic cells and licences them for Th2 priming. Eur. J. Immunol.30:2394–2403.

15. Gagliardi, M. C., and M. T. De Magistris. 2003. Maturation of humandendritic cells induced by the adjuvant cholera toxin: role of cAMP onchemokine receptor expression. Vaccine 21:856–861.

16. Hajishengallis, G., S. Arce, C. M. Gockel, T. D. Connell, and M. W. Russell.2005. Immunomodulation with enterotoxins for the generation of secretoryimmunity or tolerance: applications for oral infections. J. Dent. Res. 84:1104–1116.

17. Heit, A., K. M. Huster, F. Schmitz, M. Schiemann, D. H. Busch, and H.Wagner. 2004. CpG-DNA aided cross-priming by cross-presenting B cells.J. Immunol. 172:1501–1507.

18. Hoft, D. F., C. S. Eickhoff, O. K. Giddings, J. R. Vasconcelos, and M. M.Rodrigues. 2007. Trans-sialidase recombinant protein mixed with CpG mo-tif-containing oligodeoxynucleotide induces protective mucosal and systemicTrypanosoma cruzi immunity involving CD8� CTL and B cell-mediatedcross-priming. J. Immunol. 179:6889–6900.

19. Holmgren, J., N. Lycke, and C. Czerkinsky. 1993. Cholera toxin and choleraB subunit as oral-mucosal adjuvant and antigen vector systems. Vaccine11:1179–1184.

20. Holmgren, J., J. Adamsson, F. Anjuère, J. Clemens, C. Czerkinsky, K.Eriksson, C. F. Flach, A. George-Chandy, A. M. Harandi, M. Lebens, T.Lehner, M. Lindblad, E. Nygren, S. Raghavan, J. Sanchez, M. Stanford, J. B.Sun, A. M. Svennerholm, and S. Tengvall. 2005. Mucosal adjuvants andanti-infection and anti-immunopathology vaccines based on cholera toxin,cholera toxin B subunit and CpG DNA. Immunol. Lett. 97:181–188.

21. Johnson, K. W., B. H. Davis, and K. A. Smith. 1988. cAMP antagonizesinterleukin 2-promoted T-cell cycle progression at a discrete point in earlyG1. Proc. Natl. Acad. Sci. USA 85:6072–6076.

22. Kawamura, Y. I., R. Kawashima, Y. Shirai, R. Kato, T. Hamabata, M.Yamamoto, K. Furukawa, K. Fujihashi, J. R. McGhee, H. Hayashi, and T.Dohi. 2003. Cholera toxin activates dendritic cells through dependence onGM1-ganglioside which is mediated by NF-kappaB translocation. Eur. J. Im-munol. 33:3205–3212.

23. Lenschow, D. J., A. I. Sperling, M. P. Cooke, G. Freeman, L. Rhee, D. C.Decker, G. Gray, L. M. Nadler, C. C. Goodnow, and J. A. Bluestone. 1994.Differential up-regulation of the B7-1 and B7-2 costimulatory moleculesafter Ig receptor engagement by antigen. J. Immunol. 153:1990–1997.

24. Liu, Y.-J., C. Barthélémy, O. de Bouteiller, C. Arpin, I. Durand, and J.Banchereau. 1995. Memory B cells from human tonsils colonize mucosalepithelium and directly present antigen to T cells by rapid up-regulation ofB7-1 and B7-2. Immunity 2:239–248.

25. Lomo, J., H. K. Blomhoff, K. Beiske, T. Stokke, and E. B. Smeland. 1995.TGF�1 and cyclic AMP promote apoptosis in resting human B lymphocytes.J. Immunol. 154:1634–1643.

26. Lycke, N., A. K. Bromander, L. Ekman, U. Karlsson, and J. Holmgren. 1989.Cellular basis of immunomodulation by cholera toxin in vitro with possibleassociation to the adjuvant function in vivo. J. Immunol. 142:20–27.

27. Lycke, N., E. Severinson, and W. Strober. 1991. Molecular effects of choleratoxin on isotype differentiation. Immunol. Res. 10:407–412.

28. Majumdar, S., and B. B. Aggarwal. 2003. Adenosine suppresses activation ofnuclear factor-jB selectively induced by tumor necrosis factor in different celltypes. Oncogene 22:1206–1218.

29. Myklebust, J. H., D. Josefsen, H. K. Blomhoff, F. O. Levy, S. Naderi, J. C.Reed, and E. B. Smeland. 1999. Activation of the cAMP signaling pathwayincreases apoptosis in human B-precursor cells and is associated with down-regulation of Mcl-1 expression. J. Cell. Physiol. 180:71–80.

30. Nashar, T. O., H. M. Webb, S. Eaglestone, N. A. Williams, and T. R. Hirst.1996. Potent immunogenicity of the B subunits of Escherichia coli heat-labileenterotoxin: receptor binding is essential and induces differential modulationof lymphocyte subsets. Proc. Natl. Acad. Sci. USA 93:226–230.

31. Nashar, T. O., Z. E. Betteridge, and R. N. Mitchell. 2001. Evidence for a roleof ganglioside GM1 in antigen presentation: binding enhances presentationof Escherichia coli enterotoxin B subunit (EtxB) to CD4� T cells. Int.Immunol. 13:541–551.

32. Neumann, M., T. Grieshammer, S. Chuvpilo, B. Kneitz, M. Lohoff, A.Schimpl, B. R. Franza, Jr., and E. Serfling. 1995. RelA/p65 is a moleculartarget for the immunosuppressive action of protein kinase A. EMBO J.14:1991–2004.

33. Plant, A., R. Williams, M. E. Jackson, and N. A. Williams. 2003. The Bsubunit of Escherichia coli heat labile enterotoxin abrogates oral tolerance,promoting predominantly Th2-type immune responses. Eur. J. Immunol.33:3186–3195.

34. Rappuoli, R., M. Pizza, G. Douce, and G. Dougan. 1999. Structure andmucosal adjuvanticity of cholera and Escherichia coli heat-labile enterotox-ins. Immunol. Today 20:493–500.

35. Rodriguez-Pinto, D., and J. Moreno. 2005. B cells can prime naive CD4� Tcells in vivo in the absence of other professional antigen-presenting cells ina CD154-CD40-dependent manner. Eur. J. Immunol. 35:1097–1105.

36. Ryan, E. J., E. McNeela, M. Pizza, R. Rappuoli, L. O’Neill, and K. H. G.Mills. 2000. Modulation of innate and acquired immune responses by Esch-erichia coli heat-labile toxin: distinct pro- and anti-inflammatory effects ofthe nontoxic AB complex and the enzyme activity. J. Immunol. 165:5750–5759.

37. Schnitzler, A. C., J. M. Burke, and L. M. Wetzler. 2007. Induction of cellsignaling events by the cholera toxin B subunit in antigen-presenting cells.Infect. Immun. 75:3150–3159.

38. Vendetti, S., M. Patrizio, A. Riccomi, and M. T. De Magistris. 2006. HumanCD4� T lymphocytes with increased intracellular cAMP levels exert regu-latory functions by releasing extracellular cAMP. J. Leukoc. Biol. 80:880–888.

39. Vendetti, S., A. Riccomi, A. Sacchi, E. Sciaraffia, L. Gatta, C. Pioli, and M. T.De Magistris. 2008. Inhibition of T cell proliferation by cholera toxin in-



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

volves the modulation of costimulatory molecules CTLA-4 and CD28. Im-munol. Lett. 115:59–69.

40. Verweij, W. R., L. De Haan, M. Holtrop, E. Agsteribbe, R. Brands, G. J. M. VanScharrenburg, and J. Wilschut. 1998. Musosal immunoadjuvant activity of recom-binant Escherichia coli heat-labile enterotoxin and its B subunit: induction of sys-temic IgG and secretory IgA responses in mice by intranasal immunization withinfluenza virus surface antigen. Vaccine 16:2069–2076.

41. Wu, H.-Y., and M. W. Russell. 1998. Induction of mucosal and systemicimmune responses by intranasal immunization using recombinant choleratoxin B subunit as an adjuvant. Vaccine 16:286–292.

42. Yamamoto, S., Y. Takeda, M. Yamamoto, H. Kurazono, K. Imaoka, M.Yamamoto, K. Fujihashi, M. Noda, H. Kiyono, and J. R. McGhee. 1997.Mutants in the ADP-ribosyltransferase cleft of cholera toxin lack diarrhea-genicity but retain adjuvanticity. J. Exp. Med. 185:1203–1210.

Editor: S. R. Blanke



http://iai.asm.org/

Dow

nloaded from
http://iai.asm.org/

cholera toxin and escherichia coli heat-labile enterotoxin, but … · edge of their modes of...

Documents