T Lymphocytes + Bacteria to CD4 the Surface of Recombinant Gram ...

Human Dendritic Cells Very Efficiently Present a Heterologous Antigen Expressed on the Surface of Recombinant Gram-Positive Bacteria to CD41 T Lymphocytes1 Silvia Corinti,2* Donata Medaglini,† Andrea Cavani,* Maria Rescigno,‡ Gianni Pozzi,† Paola Ricciardi-Castagnoli,‡ and Giampiero Girolomoni* Recombinant Streptococcus gordonii expressing on the surface the C-fragment of tetanus toxin was tested as an Ag delivery system for human monocyte-derived dendritic cells (DCs). DCs incubated with recombinant S. gordonii were much more efficient than DCs pulsed with soluble C-fragment of tetanus toxin at stimulating specific CD41 T cells as determined by cell proliferation and IFN-g release. Compared with DCs treated with soluble Ag, DCs fed with recombinant bacteria required 102- to 103-fold less Ag and were at least 102 times more effective on a per-cell basis for activating specific T cells. S. gordonii was internalized in DCs by conventional phagocytosis, and cytochalasin D inhibited presentation of bacteria-associated Ag, but not of soluble Ag, suggesting that phagocytosis was required for proper delivery of recombinant Ag. Bacteria were also very potent inducers of DC maturation, although they enhanced the capacity of DCs to activate specific CD41 T cells at concentrations that did not stimulate DC maturation. In particular, S. gordonii dose-dependently up-regulated expression of membrane molecules (MHC I and II, CD80, CD86, CD54, CD40, CD83) and reduced both phagocytic and endocytic activities. Furthermore, bacteria promoted in a dosedependent manner DC release of cytokines (IL-6, TNF-a, IL-1b, IL-12, TGF-b, and IL-10) and of the chemokines IL-8, RANTES, IFN-g-inducible protein-10, and monokine induced by IFN-g. Thus, recombinant Gram-positive bacteria appear a powerful tool for vaccine design due to their extremely high capacity to deliver Ags into DCs, as well as induce DC maturation and secretion of T cell chemoattractans. The Journal of Immunology, 1999, 163: 3029 –3036.

D

endritic cells (DCs)3 are the critical APCs involved in the induction of primary T cell responses, and they are very efficient in the amplification of memory T cell responses (1). DCs are strategically distributed in tissues, they are constitutively rich in MHC class II molecules and can be readily induced to express the costimulatory molecules necessary for activation of naive or resting T cells. To elicit an immune response, DCs must undergo a maturation process, which is initiated by inflammatory signals and is completed after contact with T cells. Maturation enables DCs to migrate from peripheral tissues to lymphoid organs and to acquire a very potent Ag-presenting capacity. In addition, mature DCs are the most relevant and initial source of cytokines (e.g., IL-12) that govern the development of Th1 responses.

*Laboratory of Immunology, Istituto Dermopatico dell’Immacolata, Instituto di Ricovero e Cura a Carattere Scientifico, Rome, Italy; †Section of Microbiology, Department of Molecular Biology, University of Siena, Siena, Italy; and ‡Department of Biotechnology and Bioscience, University of Milan-Bicocca, Milan, Italy Received for publication January 26, 1999. Accepted for publication June 26, 1999. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by the Associazione Italiana per la Ricerca sul Cancro, the European Community (BIOMED 2 Contract No. BMH4 CT98-3713; BIOTECH Contract Nos. BIO2 CT94-3055 and BIO4 CT96-0542), the Istituto Superiore di Sanita` (Progetto AIDS, Contract No. 40A.0.83), and by the Consiglio Nazionale delle Ricerche (P. F. Biotecnologie, Contract No. 97.01185.PF49). 2 Address correspondence and reprint request to Dr. Silvia Corinti, Laboratory of Immunology, Istituto Dermopatico dell’Immacolata, Instituto di Ricovero e Cura a Carattere Scientifico, Via dei Monti di Creta 104, 00167 Rome, Italy. E-mail address: [email protected] 3 Abbreviations used in this paper: DC, dendritic cell; CCD, cytochalasin D; LTA, lipotheicoic acid; TTFC, C-fragment of tetanus toxin; IP-10, IFN-g-inducible protein; MIG, monokine induced by IFN-g.

Copyright © 1999 by The American Association of Immunologists

Given their central role in the immune system, DCs can represent an important target for vaccine development, and the possibility of generating large numbers of autologous DCs from precursors has made feasible and very attractive the use of DCs as natural adjuvants for inducing or boosting anti-tumor immune responses (2–5). A major problem in using DCs in immunotherapy is to optimize Ag administration so that MHC molecules are properly loaded. In addition, it is highly desirable to stimulate DC maturation to maximize their Ag-presenting capacity and render them resistant to the inhibitory effects of products (e.g., IL-10) released by tumor cells (6, 7). Bacteria or their components, including cell wall constituents and unmethylated CpG-containing DNA, are among the strongest inducers of DC maturation (8 –12). Because bacteria can directly stimulate DC maturation and can be easily modified, they appear very good candidates as delivery systems for vaccine Ags. Bacteria in fact can be engineered to express the gene product at the cell surface, in the cytosol or as secreted protein (13–15), and can also serve as carriers for introducing Agencoding DNA into APCs (16 –18). These different approaches have been employed for inducing protective immune responses against viral and tumor Ags in mouse models (15, 19). Recombinant strains of Streptococcus gordonii have been successfully used as vectors for parenteral and mucosal delivery of vaccine Ags and have been shown to induce both local and systemic Ab responses in mice (14, 20 –22). More recently, the efficacy of such vaccination approach has been confirmed in primates against viral Ags (23). However, the cell populations and mechanisms involved in these bacteria-based immunostimulating systems have been only marginally investigated. We have previously shown that mouse DCs very efficiently present a MHC class I-restricted heterologous Ag expressed on the surface of recombinant S. gordonii to T lymphocytes and that S. gordonii induces 0022-1767/99/$02.00

3030

PRESENTATION OF HETEROLOGOUS BACTERIA Ag BY HUMAN DC

neo-biosynthesis and membrane stabilization of MHC class I and class II molecules (10). In the present study, human DCs generated from peripheral blood monocytes were investigated for their capacity to present a MHC class II-restricted model Ag, the C-fragment of tetanus toxin (TTFC), expressed on the surface of recombinant S. gordonii to specific CD41 T cells. Results demonstrate that DCs pulsed with recombinant bacteria stimulate specific CD41 T cell response with high efficiency and that S. gordonii is a potent stimulus for human DC maturation and induces DC release of chemokines active on T cells. These findings can be of value for the design of new DC-based vaccines.

anti-MHC class I mAb (not shown). Autologous DCs were pulsed at 37°C with S. gordonii GP1253, control bacteria GP1221, TTFC-conjugated beads, or soluble TTFC in complete medium at the indicated concentrations. After 18 h, APCs were washed, examined for cell viability by the trypan blue exclusion test, and then cocultured with T cells (2–3 3 104 cells/well) in triplicate wells. Cocultures were pulsed with 1 mCi/well [3H]thymidine at day 2 or 3. In some experiments, DCs were pretreated for 30 min at 37°C with 10 mg/ml cytochalasin D (CCD; Sigma), and then washed twice with PBS before Ag pulsing. Radioactivity was measured in a beta counter (Topcount, Packard Instruments, Groningen, The Netherlands). Results are given as mean cpm 6 SD of triplicate cultures. Where indicated, supernatants from cocultures were analyzed for IFN-g contents by ELISA (R&D Systems, Minneapolis, MN).

Materials and Methods

Transmission electron microscopy

Recombinant bacteria and Dot blot analysis The recombinant strain of S. gordonii expressing TTFC (named GP1253) was constructed using the host-vector system GP1221-pSMB55, as already described in detail (24). Surface expression of TTFC on GP1253 was achieved using the M6 protein as fusion partner (14, 25). Recombinant S. gordonii GP1253 and control strain GP1221 were grown at 37°C in tryptic soy broth without dextrose (Difco, Detroit, MI) and harvested by centrifugation at the end of the exponential phase of growth. Bacterial cells were then washed and resuspended in fresh medium containing 10% glycerol at 1:500 of the original culture volume. Aliquots were stored frozen at 280°C until use. Surface expression of TTFC was determined by flow cytometry analysis on whole cells and Western blotting of cell fractions (not shown). The concentration of TTFC expressed on bacteria was determined by Dot blot analysis of whole cells using a standard curve of purified TTFC. Twofold dilutions of recombinant S. gordonii GP1253, control strain GP1221 (from 4 3 109 to 6.2 3 107 CFU), and 200 –3.12 ng soluble TTFC (Boerhinger Mannheim, Mannheim, Germany) in a volume of 100 ml were transfered onto the nitrocellulose membrane using a HYBRI.DOT instrument (Life Technologies, Gaithersburg, MD). The membrane was than reacted with a TTFC-specific rabbit serum (Calbiochem, San Diego, CA; 1:1,000 dilution) followed by anti-rabbit serum conjugated to alkaline phosphatase (Sigma, St. Louis, MO; 1:5,000 dilution). The concentration of TTFC expressed on the bacterial surface was calculated by densitometric analysis using the Fluor-s Multimager (Bio-Rad, Hercules, CA)

Bead preparations TTFC was bound to 1-mm latex beads (Polysciences, Warrington, PA) by passive absorption, according to the manufacturer’s instructions. The amount of TTFC coupled to beads was measured by difference in optical densities at 595 nm of the protein solution before and after the linkage using Bradford assay (Bio-Rad). Bead preparations were washed with culture medium before use.

DCs DCs were prepared from peripheral blood monocytes of healthy individuals, as described by Sallusto and Lanzavecchia (26). Briefly, PBMC isolated by standard density gradient centrifugation were further separated on multistep Percoll gradients (Pharmacia, Uppsala, Sweden). Cells from the light density fraction (42.5–50%; .90% CD141) were recovered and cultured at 1 3 106 cells/ml in RPMI 1640 (Life Technologies) complemented with 10% FBS (HyClone, Logan, UT), 1 mM sodium pyruvate, 0.1 mM nonessential amino acids, 2 mM L-glutamine, 25 mM HEPES, 100 U/ml penicillin, 100 mg/ml streptomycin (all from Life Technologies), and 0.05 mM 2-ME (Merck, Darmstadt, Germany) (complete medium) at 37°C with 5% CO2 in the presence of 200 ng/ml recombinant human GM-CSF (Mielogen; Schering-Plough, Milan, Italy) and 200 U/ml recombinant human IL-4 (Genzyme, Cambridge, MA). Medium was changed after 3 days, and at day 6 of culture, cells were recovered and depleted of CD21 and CD191 cells by means of immunomagnetic beads coated with specific mAbs (Dynal, Oslo, Norway). This procedure resulted in .97% pure CD1a1 and CD142 DC preparation.

Ag presentation assay CD41 T cell clones specific for TTFC (CP7, ALS4) were prepared by limiting dilution of TTFC-specific CD41 T cell lines generated from PBMC of two healthy individuals. These T cell clones were CD41, CD82, TCR ab1, TCR gd2, and CD281 and secreted high IFN-g, but no IL-4, upon activation with anti-CD3 and anti-CD28 mAbs or in Ag-specific stimulation assay. TTFC-specific CD41 T cell clones were strictly MHC class II-dependent, as determined by inhibition studies with anti-HLA-DR or

DCs were incubated with live S. gordonii GP1253 (cells-to-bacteria ratio, 1:50) in complete medium at 37°C with 5% CO2. After 2–18 h, cells were pelleted, washed, fixed in 1% glutaraldehyde in a Tyrode buffer, postfixed in 1% osmium tetroxide, dehydrated in graded alcohols, and finally embedded in Durcopan ACM (Fluka, Buchs, Switzerland). Thin sections were stained with uranyl acetate and lead citrate and then examined under a Philips CM100 electron microscope.

Flow cytometry analysis After 18 h incubation with S. gordonii, lipoteichoic acid (LTA; Sigma), or medium alone, cells were washed and then incubated in PBS added of 2% FBS and 0.01% NaN3 with the following mAbs: FITC-conjugated antiHLA-DR (L243, IgG2a) and FITC-conjugated anti-CD14 (MFP9, IgG2b) from Becton Dickinson (San Jose, CA); FITC-conjugated anti-CD1a (HI149, IgG1) and FITC-conjugated anti-CD86 (FUN-1, IgG1) from PharMingen (San Diego, CA); FITC-conjugated anti-CD40 (BB20, IgG1) from Ylem (Avezzano, Italy); FITC-conjugated anti-CD54 (84H10, IgG1), FITC-conjugated anti-CD80 (MAB104, IgG1), and anti-CD83 (HB15, IgG2b) from Immunotech (Marseille, France); anti-CSF-1R (CD115) (204A5-4, rat IgG1) from Calbiochem, and anti-MHC class I (W6/32, IgG1) from Dako (Glostrup, Denmark). FITC-conjugated anti-mouse Ig F(ab9)2, FITC-conjugated anti-rat IgG, and control rat IgG came from Southern Biotechnology Associates (Birmingham, AL). In control samples, the mAb was substituted with matched isotype control mouse or rat Ig (Becton Dickinson). Cells were analyzed in a FACScan equipped with Cell Quest software (Becton Dickinson, Mountain View, CA).

Phagocytosis and endocytosis assays DCs incubated with S. gordonii or 10 mg/ml LTA for 18 h were washed, resuspended in complete medium, and finally pulsed with 1 mg/ml Texas red-conjugated BSA or FITC-labeled 2-mm latex beads (Molecular Probes, Eugene, OR). DCs were then incubated at 37°C or 4°C, and, at selected time points, uptake was stopped by adding cold PBS containing 2% FBS and 0.01% NaN3. Cells were then washed four times and analyzed in a FACScan. Surface binding values obtained by incubating cells at 4°C were subtracted from values measured at 37°C.

Cytokine and chemokine secretion DCs treated as described above were cultured in 6-well plates (1 3 106 cells/well) for 18 –24 h at 37°C, and then supernatants were collected and stored at 280°C. IL-1b, TNF-a, IL-6, IL-12 (p70), TGF-b, IL-10, IL-8, and RANTES were measured by commercially available ELISA kits (R&D Systems), according to the manufacturer’s instructions. IP-10 (IFN-g-inducible protein of 10 kDa) and MIG (monokine induced by IFN-g) were detected with sandwich ELISAs performed with Abs pairs from R&D (antiIP-10: mouse mAb 33036.211 and goat polyclonal Ab) and PharMingen (anti-MIG: mouse mAbs B8-11 and B8-6) following the manufacturers’ protocol. Adsorbance (450 –540 nm) was read in an ELISA reader model 3550 UV (Bio-Rad).

Results

DCs very efficiently present TTFC to specific CD41 T cells clones when expressed on the surface of recombinant S. gordonii S. gordonii, a Gram-positive coccus and a normal commensal present in the human oral cavity, has been used in previous studies as vector for mucosal delivery of heterologous vaccine Ags (20 – 22). The strain “Challis” allows easy genetic manipulation because

The Journal of Immunology

3031

FIGURE 1. Dot blot analysis of recombinant S. gordonii expressing TTFC. Two-fold dilutions of recombinant S. gordonii GP1253 expressing TTFC and control strain GP1221 were analyzed by Dot blotting against a standard curve of soluble TTFC. The nitrocellulose membrane was reacted with a TTFC-specific polyclonal Ab. The concentration of TTFC expressed on the bacterial surface was calculated by densitometric analysis: 5 3 108 CFU of bacteria corresponded to 50 ng of soluble TTFC.

it is naturally competent for genetic transformation and is capable of colonizing the mouse oral and vaginal mucosa (14). In the present study, we used a recombinant strain of S. gordonii (GP1253) expressing TTFC on the cell surface as fusion with the M6 protein (25). The S. gordonii parental strain GP1221 served as control. Western blot analysis of bacterial cell fractions showed the presence of a 84-kDa protein band reactive with the TTFC-specific Ab only in the envelope fraction (data not shown). The amount of TTFC expressed on bacteria was determined by Dot blot analysis and resulted in 50 ng/5 3 108 CFU as calculated by densitometric analysis with a standard curve of purified TTFC (Fig. 1). Thus, incubation of DCs (106 cells/ml) with GP1253 at a bacteria-to-DC ratio of 50:1 corresponded to 5 ng/ml of TTFC. In the first set of experiments, we investigated the efficiency of DCs to present TTFC expressed on the surface of S. gordonii compared with soluble TTFC. DCs were incubated for 18 h with the recombinant strain GP1253, the control strain GP1221, or with soluble TTFC and then were used as APCs for stimulating autologous TTFC-specific CD41 Th1 cells. DC culture in the presence of bacteria (at a bacteria-to-DC ratio of 1:1 to 100:1) or LTA for 18 h did not decrease cell viability, as measured by the trypan blue exclusion test (not shown). Fig. 2 shows that specific CD41 T cells were much more readily activated when the same amount of Ag was administered to DCs as a surface bacterial protein than as soluble Ag, with the T cell response determined as cell proliferation (Fig. 2A) and IFN-g release (Fig. 2B). When DCs were pulsed with recombinant bacteria in respect to soluble TTFC, 102-103 times less Ag was required to obtain a comparable T cell response, and incubation of DCs with as few as one bacterium per DC (corresponding to 0.1 ng/ml TTFC) was sufficient for inducing a significant T cell proliferation. On a per-cell basis, DCs fed with recombinant bacteria were at least 102 times more effective than DCs pulsed with an equal dose of soluble Ag (Fig. 2C). Control

FIGURE 2. DCs present TTFC expressed on the surface of recombinant Gram-positive bacteria with higher efficiency than soluble TTFC to specific CD41 T cells. In A and B, a fixed number of monocyte-derived DCs (2000 cells/well) were incubated for 18 h with increasing doses of recombinant S. gordonii GP1253 (F) or soluble TTFC (M) and then used to stimulate a TTFC-specific Th1 clone (CP7; 30,000 cells/well). At each Ag dose, the amount of soluble TTFC and bacteria-associated TTFC were equivalent. DCs incubated with wild-type S. gordonii GP1221 (E) served as control. Cocultures were carryed out for 3 days, and the T cell response measured as [3H]thymidine uptake and IFN-g release, respectively. In C, graded numbers of DCs were pulsed with soluble TTFC (5 ng/ml), S. gordonii GP1253, or GP1221. Both bacterial strains were used at a DC-to-bacteria ratio of 50, corresponding in the case of strain GP1253 to about 5 ng/ml of TTFC. Thymidine uptake is expressed as mean cpm 6 SD, whereas IFN-g levels as pg/106 cells/ml. Similar results were obtained in seven (A and C) or two (B) independent experiments.

strain GP1221 did not induce a significant T cell response in any case. Immature DCs uptake S. gordonii via conventional phagocytosis Next, we determined whether and how DCs internalize S. gordonii. Immature DCs were incubated with live bacteria at a bacteriato-DC ratio of 50:1, and then analyzed by transmission electron microscopy. As soon as 2 h after incubation at 37°C (the shortest time tested), bacteria were found in a small proportion (,20%) of DCs within membrane-bound phagosomal organelles (not shown). After 18 h incubation, bacteria were visualized in most (.90%) DCs and appeared at various stages of degradation always included in vacuoles and phagosomes and not free in the cytosol (Fig. 3). The number of bacteria present in each DC section varied, with the majority of DCs having phagocytosed 5–20 bacteria. In

3032


FIGURE 3. DCs internalize S. gordonii via conventional phagocytosis. DCs were incubated for 18 h with S. gordonii GP1253 at a bacteria-to-DC ratio of 50 and then processed for transmission electron microscopy. Bacteria are visible within membrane-bound phagosomes and at various stages of degradation (arrows). Bar, 1 mm.

phagosomes, cell membranes were tightly or loosely opposed to bacteria, and in many instances a single organelle contained multiple bacteria. S. gordonii GP1253 and GP221 were internalized similarly, and no bacteria were found when DCs where incubated at 4°C. Finally, no bacteria enclosed by cell membrane protrusions or pseudopod coils were observed, suggesting that S. gordonii was taken up exclusively by conventional phagocytosis, as previously observed with mouse DCs (10). Because S. gordonii was internalized by DCs via phagocytosis, we then sought to determine whether phagocytosis was necessary

FIGURE 5. Control bacteria increase the capacity of DCs to present very limited amount of particulate Ag or a suboptimal dose of soluble Ag. A, DCs were incubated with recombinant S. gordonii GP1253, a corresponding dose of soluble TTFC (5 ng/ml), or soluble TTFC in the presence of control strain GP1221 or LTA (10 mg/ml). B, DCs were treated with 100 ng/ml soluble TTFC or soluble TTFC in the presence of control strain GP1221 or LTA. C, DCs were incubated with equal amount (5 ng/ml) of TTFC in the form of recombinant bacteria, bead-bound Ag, or soluble Ag. DCs were also treated with bead-conjugated Ag in the presence of control bacteria. Both bacterial strains were used at a bacteria-to-DC ratio of 50. After 18 h, DCs were cocultured for 3 days with 30,000 specific CD41 T cells (clone ALS4), and then thymidine uptake was measured. Data are expressed as mean cpm 6 SD and are representative of two or three experiments.

for the presentation of TTFC expressed on bacteria. To this end, we examined the effects of CCD, a drug which disrupts actin filament rearrangements and inhibits phagocytic events. Pretreatment of DCs with CCD inhibited the presentation of bacteria-bound TTFC to specific CD41 T cells (Fig. 4A), but did not affect presentation of soluble TTFC (Fig. 4B). Thus, presentation of bacteria-associated Ag by DCs required phagocytosis and was not the result of membrane processing. Recombinant Ag expressed on S. gordonii is presented more effectively to CD41 T cells through different mechanisms FIGURE 4. Phagocytosis is required for effective presentation of TTFC expressed on recombinant S. gordonii. Untreated (F, f) or CCD (10 mg/ ml) pretreated (E, M) DCs were pulsed with S. gordonii GP1253 at a bacteria-to-DC ratio of 50 (corresponding to 5 ng/ml of TTFC) (A) or 1 mg/ml of soluble TTFC (B) and then were cocultured for 3 days with 30,000 specific CD41 T cells (clone CP7). Data are expressed as mean cpm 6 SD and are representative of two experiments.

To dissect the mechanisms by which TTFC on S. gordonii is presented with higher efficiency by DCs, T cell responses to DCs fed with S. gordonii GP1253 and to DCs pulsed with equal doses of soluble TTFC or TTFC-conjugated to latex particles and concomitantly treated with control strain GP1221 were compared. Fig. 5A shows that pulsing DCs with 5 ng/ml soluble TTFC in the presence

The Journal of Immunology

FIGURE 6. DCs incubated with S. gordonii show enhanced expression of membrane molecules. DCs were cultured for 18 h in medium alone (thin lines) or in the presence of either strain S. gordonii GP1253 or GP1221 at a bacteria-to-DC ratio of 50:1 (bold lines) and then were analyzed by flow cytometry for the indicated surface markers. Dotted histograms represent staining with matched-isotype Abs.

of strain GP1221 or LTA did not change the magnitude of the T cell response. In contrast, incubation of DCs with GP1221 or LTA increased their capacity to present a higher dose (100 ng/ml) of soluble TTFC (Fig. 5B). Strain GP1253 could not be included in this last experiment because incubation with 1000 bacteria/DC drastically reduced cell viability. In contrast, DCs treated with 5 ng/ml of TTFC bound to 1-mm particles showed a better T cell stimulating activity compared with DCs pulsed with an equal dose of soluble Ag, and incubation of DCs with TTFC-conjugated beads together with control bacteria further augmented the T cell proliferation, although not to the levels measured with strain GP1253 (Fig. 5C). The results indicated that particulate TTFC was more efficiently presented than soluble TTFC, and that S. gordonii could effectively exert an adjuvant activity when suboptimal doses of soluble Ag or minute amounts of Ag in particulate form were provided. Therefore, in the following experiments, we tested whether S. gordonii was capable of stimulating DC maturation in terms of cells-surface marker expression, phagocytic and endocytic functions, and cytokine release. Interaction of DCs with S. gordonii induces cell-surface maturation and inhibits DC phagocytic and endocytic activities Cells treated with S. gordonii (either GP1253 or GP1221) for 18 h showed a dramatic increase in the expression of molecules involved in Ag presentation such as MHC class I and class II molecules, CD80 and CD86 costimulatory molecules, and CD54. In addition, CD40 and CD83 were both up-regulated (Figs. 6 and 7). In contrast, expression of CD1a was slightly but consistently di-

3033

FIGURE 7. S. gordonii induces membrane maturation of DCs in a dosedependent manner. DCs were cultured in medium alone or in the presence of increasing numbers of bacteria (either strain GP1253 or GP1221) and, after 18 h, were analyzed for surface expression of the indicated molecules by flow cytometry. Values represent the mean fluorescence intensity subtracted of the fluorescence of matched-isotype control Ab. The experiment was conducted four times with similar results.

minished in DCs treated with S. gordonii, whereas expression of the M-CSF receptor (CD115) was abolished (not shown). These effects of S. gordonii on DCs were strictly dose-dependent (Fig. 7) and could be detected also using heat-killed bacteria or LTA, but not following phagocytosis of 2-mm latex beads (not shown). Interestingly enough, doses of bacteria (bacteria-to-DC ratio of 1:1) that allowed an efficient presentation of the recombinant Ag did not induce significant changes in the surface phenotype of DCs. Maturation of DCs is associated with reduced phagocytic and endocytic activities. This phenomenon has been previously demonstrated in response to inflammatory cytokines or bacterial products (27, 28). Incubation of DCs with S. gordonii or LTA for 24 h at 37°C induced a marked down-regulation of DC capacity to internalize Texas red-BSA (Fig. 8A) and 2-mm FITC-conjugated latex beads (Fig. 8B). To control for potential nonspecific association of BSA or beads with the exterior of DCs, experiments were performed in parallel at 4°C. S. gordonii induces release of regulatory cytokines and T cell chemoattractants from DCs To analyze whether the observed phenotypic maturation was associated with cytokine production, supernatants from 24-h DC cultures performed in the presence of different doses of bacteria were tested by ELISA. Both strains of S. gordonii induced a dose-dependent release of large amounts of TNF-a and IL-6 and less IL-1b and TGF-b. As previously observed in mouse DCs (10), and in human DCs exposed to other stimuli that promote DC maturation (29, 30), S. gordonii also caused the release of IL-10, an important DC regulatory factor that maintains DC in an immature state and inhibits their Ag-presenting activity. In addition, S. gordonii promoted substantial production of IL-12 heterodimer, with

3034


FIGURE 10. S. gordonii induces chemokine release by DCs. DCs were cultured (106 cells/ml) in medium alone or in the presence of increasing numbers of bacteria. After 24 h, culture supernatants were collected, and chemokine were levels determined by ELISA. Results are representative of three different experiments.

FIGURE 8. DCs incubated with S. gordonii or LTA show decreased phagocytic and endocytic activities. A, DCs were cultured in medium alone (M) or in the presence of S. gordonii (50 bacteria/DC) (E) or 10 mg/ml LTA (‚) for 18 h at 37°C. Thereafter, DCs were collected and then incubated with 1 mg/ml Texas red-BSA at 37°C. At various time points, cells were washed with cold PBS, and the BSA accumulation was measured by flow cytometry. Dashed line with open squares represents the fluorescence of untreated DCs pulsed with Texas red-BSA at 4°C. Fluorescence of cells incubated at 4°C was subtracted from the fluorescence of cells incubated at 37°C. B, DCs treated as described above were incubated with 2-mm FITCconjugated latex beads for 24 h at 37°C or 4°C and then were analyzed by flow cytometry.

levels up to 4 ng/ml at the maximum dose of bacteria (Fig. 9). Finally, monocyte-derived DCs were found to release constitutively significant amounts of IL-8 and low levels of IP-10. Upon incubation with S. gordonii, DCs increased secretion of IL-8 and IP-10 as well as started to dose-dependently release RANTES and

FIGURE 9. DCs incubated with S. gordonii release inflammatory and immunoregulatory cytokines. DCs were cultured (106 cells/ml) in medium alone or in the presence of increasing numbers of bacteria for 24 h. Thereafter, culture supernatants were collected, and the concentration of the indicated cytokines was measured by ELISA. The experiment was repeated three times with similar results.

MIG (Fig. 10). Similar amounts of cytokines or chemokines were detected in supernatants from DC cultures stimulated with LTA (not shown).

Discussion In this study, we demonstrate that human monocyte-derived DCs can phagocytose Gram-positive bacteria and that feeding DCs with a recombinant strain of S. gordonii engineered to express on the cell surface a heterologous Ag (TTFC) is a very efficient system for activating specific Th1 cells in vitro, both in terms of cell proliferation and IFN-g release. Compared with soluble TTFC, a markedly reduced amount of Ag or much less DCs were necessary to obtain similar T cell responses. Two principal explanations can account for these findings: first, recombinant bacteria can allow a more efficient Ag delivery into the class II pathway of Ag processing, and, second, bacteria can stimulate DC maturation and thus potentiate their Ag-presenting function. Results indicate that both mechanisms cooperate in rendering heterologous Ag on recombinant bacteria very effectively presented by DCs to CD41 T cells. S. gordonii was internalized by DCs via conventional phagocytosis and delivered in membrane-bound phagosomes, and phagocytosis was required for effective Ag presentation to specific T cells. Although DCs have been considered poorly phagocytic compared with macrophages, the capacity of mouse or human DCs and epidermal Langerhans cells to phagocytose bacteria has been documented for a variety of microorganisms including Corynebacterium parvum, Staphylococcus aureus, Borrelia burgdorferi, Chlamydia species, Mycobacterium tubercolosis, and Leishmania and has been confirmed both in vitro and in vivo (3, 31–33). The administration of exogenous class I-restricted Ags in a form that requires phagocytosis is essential for an efficient presentation to T cells (10, 34, 35). Our results show that phagocytosis of particulate Ag may represent a very effective means to supply Ags to the class II processing pathway, and this can partly explain why TTFC expressed on the bacterial surface was presented better than soluble TTFC. In keeping with this hypothesis is the observation that DCs presented bacteria-associated TTFC with a greater efficiency when low numbers of recombinant bacteria not sufficient for inducing DC maturation were used. In addition, Ag adsorbed to latex particles was presented better than soluble Ag. The fact that adsorptive phagocytosis is very effective for the generation of MHC class II-peptide complexes from Ag-carrying bacteria has been previously reported for mouse DCs (35) and recently confirmed in DCs phagocytosing apoptotic or necrotic allogeneic B cell blasts (36).

The Journal of Immunology In addition, bacteria effectively induced DC maturation and enhanced the APC function of DCs. In fact, treatment of DCs with control bacteria markedly increased T cell proliferation to either 100 ng/ml of soluble Ag or 5 ng/ml of Ag bound to latex particles. However, cotreatment of DCs with particulate Ag and control bacteria partially reconstituted (30 – 60%) the efficiency of recombinant bacteria, most likely because DCs possess specific receptors for recognizing bacteria and their delivery in the Ag-processing pathway may be favored (35). DCs are critical for the induction of primary immune responses against pathogens, but to perform this function they need to undergo a series of changes, collectively known as “maturation,” which enables DCs to migrate to lymphoid organs and to acquire high T cell-stimulating capacity (1, 2). In this study, we show that S. gordonii is very effective in inducing DC maturation, although this bacterium is a normal commensal of the oral cavity and considered nonpathogenic. In particular, S. gordonii dose-dependently induced increased membrane expression of MHC, CD80, CD86, ICAM-1, CD40, and CD83 and, in parallel, considerably increased the capacity of DCs to activate allogeneic T cells in the primary MLR assay (data not shown). In addition, DCs exposed to S. gordonii released higher amounts of proinflammatory cytokines such as IL-6, IL-1b, and TNF-a, which can mediate DC maturation in an autocrine fashion. However, incubation of DCs with bacteria in the presence of neutralizing antiTNF-a Ab had only marginal effects on DC membrane molecule expression (data not shown), indicating that autocrine release of TNF-a is not relevant to DC maturation in this system, as previously reported in mouse DCs (10) or in human DCs incubated with bacillus Calmette-Gue´rin (37). S. gordonii was effective in inducing the release of IL-12, a cytokine pivotal for the induction of Th1 cell differentiation, as observed for human blood-derived DCs stimulated with M. tubercolosis or B. burgdorferi (8, 31). Stimulation with bacteria also induced secretion of IL-10, which may serve as a counter-regulatory signal for blocking DC activation and thus exaggerated T cell responses. All these effects could be reproduced by using LTA, suggesting that this cell wall component play a crucial role in Gram-positive bacteria-induced DC maturation. Following incubation with bacteria or LTA, DCs decreased their capacity to endocytose a soluble protein and phagocytose latex beads, confirming that mature DCs down-regulate mechanisms of Ag capture and switch from an “Ag capturing” to an “Ag presenting” mode (2). To maximize their Ag-presenting potential, mature DCs transiently increase the biosynthesis of MHC class II molecules, and, most strikingly, MHC molecules are massively exported to the cell membrane where their half-life is prolonged because the rate of endocytosis is lowered (10, 38, 39). The accumulation of high numbers MHC class II molecules on the cell membrane together with increased expression of costimulatory molecules allow a highly efficient Ag presentation to T lymphocytes. An interesting finding of this study was that DCs stimulated with bacteria released chemokines very active in attracting T cells. Indeed, recent results in the mouse system demonstrated that DCs produce an array of chemokines in response to bacterial stimulation, and that the pattern of chemokines expressed changes during the maturation stages of DCs (40). Here, we show that blood-born human DCs constitutively secrete IL-8 and IP-10 and, upon maturation, high levels of IL-8, IP-10, RANTES, and MIG. RANTES can bind to different chemokine receptors and is active on a variety of leukocytes, including memory lymphocytes, NK cells, monocytes, and immature DCs (40, 41). IP-10 and MIG are selective and efficient chemotactic stimuli for memory and activated T lymphocytes, and their main receptor, CXCR3, is preferentially expressed by Th1 cells (42, 43). Other than functioning as che-

3035 moattractants, these chemokines can directly mediate an Ag-independent activation signal on T cells and potentiate Th1 as well as CTL responses (44, 45). Thus, DC release of these chemokines, together with the secretion of IL-12, can be very important during the development of an immune response for recruiting and favor the functions of Th1 cells. In conclusion, recombinant bacteria appear very good candidates for the design of new DC-based vaccine strategies. In fact, compared with soluble Ag or Ag conjugated to inert particles, recombinant bacteria combine a high efficiency of delivering Ag into DCs with the capacity to induce a robust DC maturation. The use of mature DCs in vaccination procedures can offer several advantages: in vivo experiment using mouse bone marrow-derived DCs have clearly shown that the ability of DCs to induce protective antitumor immune responses correlates with their degree of maturation (46). Additionaly, mature DCs may be resistant to the suppressive effects of IL-10 or other factors released by tumor cells (7, 47, 48) and may have a higher propensity to home to lymphoid organs (49). Finally, mature DCs have an impaired capability to acquire and process exogenous Ags and thus may carry a reduced potential of inducing undesirable autoimmune responses (5, 50).

Acknowledgments We thank Dr. G. Zambruno and C. Albanesi for help with transmission electron microscopy and chemokine ELISA, respectively, and for contributive discussion.

References 1. Banchereau, J., and R. M. Steinman. 1998. Dendritic cells and the control of immunity. Nature 392:245. 2. Young, J. W., and K. Inaba. 1996. Dendritic cells as adjuvants for class I major histocompatibility complex-restricted antitumor immunity. J. Exp. Med. 183:7. 3. Girolomoni, G., and P. Ricciardi-Castagnoli. 1997. Dendritic cells hold promise for immunotherapy. Immunol. Today 18:102. 4. Hsu, F. J., C. Benike, F. Fagnoni, T. M. Liles, D. Czerwinski, B. Taidi, E. G. Engleman, and R. Levy. 1996. Vaccination of patients with B-cell lymphoma using autologous antigen-pulsed dendritic cells. Nat. Med. 2:52. 5. Nestle, F. O., S. Alijagic, M. Gilliet, Y. Sun, S. Grabbe, R. Dummer, G. Burg, and D. Schadendorf. 1998. Vaccination of melanoma patients with peptide- or tumor lysate-pulsed dendritic cells. Nat. Med. 4:328. 6. Schuler, G., and R. M. Steinman. 1997. Dendritic cells as adjuvants for immunemediated resistance to tumors. J. Exp. Med. 186:1183. 7. Enk, A. H., H. Jonuleit, J. Saloga, and J. Knop. 1997. Dendritic cells as mediators of tumor-induced tolerance in metastatic melanoma. Int. J. Cancer 73:309. 8. Winzler, C., P. Rovere, M. Rescigno, F. Granucci, G. Penna, L. Adorini, V. S. Zimmermann, J. Davoust, and P. Ricciardi-Castagnoli. 1997. Maturation stages of mouse dendritic cells in growth factor-dependent long-term cultures. J. Exp. Med. 185:317. 9. Henderson, R. A., S. C. Watkins, and J. L. Flynn. 1997. Activation of human dendritic cells following infection with Mycobacterium tuberculosis. J. Immunol. 159:535. 10. Rescigno, M., S. Citterio, C. The`ry, M. Rittig, D. Medaglini, G. Pozzi, S. Amigorena, and P. Ricciardi-Castagnoli. 1998. Bacteria-induced neo-biosynthesis, stabilization, and surface expression of functional class I molecules in mouse dendritic cells. Proc. Natl. Acad. Sci. USA 95:5229. 11. Jakob, T., P. S. Walker, A. M. Krieg, M. C. Udey, and J. C. Vogel. 1998. Activation of cutaneous dendritic cells by CpG-containing oligodeoxynucleotides: a role for dendritic cells in the augmentation of Th1 responses by immunostimulatory DNA. J. Immunol. 161:3042. 12. Ha¨cker, H., H. Mischak, T. Miethke, S. Liptay, R. Schmid, T. Sparwasser, K. Heeg, G. B. Lipford, and H. Wagner. 1998. CpG-DNA specific activation of antigen-presenting cells requires stress kinase activity and is preceded by nonspecific endocytosis and endosomal maturation. EMBO J. 17:6230. 13. Verjans, G. M. G. M., R. Janssen, F. G. C. M. UytdeHaag, C. E. M. van Doornik, and J. Tommassen. 1995. Intracellular processing and presentation of T cell epitopes, expressed by recombinant Escherichia coli and Salmonella typhimurium, to human T cells. Eur. J. Immunol. 25:405. 14. Pozzi, G., M. R. Oggioni, and D. Medaglini. 1997. Recombinant Streptococcus gordonii as a live vehicle for vaccine antigens. In Gram-Positive Bacteria as Vaccine Vehicles for Mucosal Immunization. G. Pozzi and J. M. Wells, eds. Landes Biosciences, Austin, TX, p. 35. 15. Ru¨ssmann, H., H. Shams, F. Poblete, Y. Fu, J. E. Galań, and R. O. Donis. 1998. Delivery of epitopes by the Salmonella type III secretion system for vaccine development. Science 281:565. 16. Sizemore, D., A. Branstrom, and J. Sadoff. 1997. Attenuated bacteria as DNA delivery vehicle for DNA-mediated immunization. Vaccine 15:804.

3036


17. Darij, A., C. A. Guzmań, B. Gerstel, P. Wachholz, K. N. Timmis, J. Wehland, T. Chakraborty, and S. Weiss. 1997. Oral somatic transgene vaccination using attenuated S. typhimurium. Cell 91:765. 18. Dietrich, G., A. Bubert, I. Gentschev, Z. Sokolovic, A. Simm, A. Catic, S. H. E. Kaufmann, J. Hess, A. A. Szalay, and W. Goebel. 1998. Delivery of antigen-encoding plasmid DNA into the cytosol of macrophages by attenuated suicide Listeria monocytogenes. Nat. Biotech. 16:181. 19. Paglia, P., E. Medina, I. Arioli, C. A. Guzmań, and M. P. Colombo. 1998. Gene transfer in dendritic cells, induced by oral DNA vaccination with Salmonella typhimurium, results in protective immunity against a murine fibrosarcoma. Blood 92:3172. 20. Oggioni, M. R., R. Manganelli, M. Contorni, M. Tommasino, and G. Pozzi. 1995. Immunisation of mice by oral colonisation with live recombinant commensal streptococci. Vaccine 13:775. 21. Medaglini, D., G. Pozzi, T. P. King, and V. A. Fischetti. 1995. Mucosal and systemic immune responses to a recombinant protein expressed on the surface of the oral commensal bacterium Streptococcus gordonii after oral colonization. Proc. Natl. Acad. Sci. USA 92:6868. 22. Oggioni, M. R., D. Medaglini, L. Romano, F. Peruzzi, T. Maggi, L. Lozzi, L. Bracci, M. Zazzi, F. Manca, P. E. Valensin, and G. Pozzi. 1999. Antigenicity and immunogenicity of the V3 domain of HIV type 1 glycoprotein 120 expressed on the surface of Streptococcus gordonii. AIDS Res. Hum. Retroviruses 15:451. 23. Di Fabio, S., D. Medaglini, C. M. Rush, F. Corrias, G. L. Panzini, M. Pace, P. Verani, G. Pozzi, and F. Titti. 1998. Vaginal immunization of Cynomolgus monkeys with Streptococcus gordonii expressing HIV-1 and HPV-6 antigens. Vaccine 16:485. 24. Oggioni, M. R., and G. Pozzi. 1996. A host-vector system for heterologous gene expression in Streptococcus gordonii. Gene 169:85. 25. Pozzi, G., M. Contorni, M. R. Oggioni, R. Manganelli, M. Tommasino, F. Cavalieri, and V. A. Fischetti. 1992. Delivery and expression of a heterelogous antigen on the surface of streptococci. Infect. Immun. 60:1902. 26. Sallusto, F., and A. Lanzavecchia. 1994. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor a. J. Exp. Med. 179:1109. 27. Reis e Sousa, C., P. D. Stahl, and J. M. Austin. 1993. Phagocytosis of antigens by Langerhans cells in vitro. J. Exp. Med. 178:509. 28. Sallusto, F., M. Cella, C. Danieli, and A. Lanzavecchia. 1995. Dendritic cells use macropinocytosis and the mannose receptor to concentrate macromolecules in the major histocompatibility complex class II compartment: downregulation by cytokines and bacterial products. J. Exp. Med. 182:389. 29. Corinti, S., E. Fanales-Belasio, C. Albanesi, A. Cavani, P. Angelisova, and G. Girolomoni. 1999. Crosslinking of membrane CD43 mediates dendritic cell maturation. J. Immunol. 162:6331. 30. De Saint-Vis, B., I. Figier-Vivier, C. Massacrier, C. Gaillard, B. Vanbervliet, S. Aı¨t-Yahia, J. Banchereau, Y.-J. Liu, S. Lebecque, and C. Caux. 1998. The cytokine profile expressed by human dendritic cells is dependent on cell subtype and mode of activation. J. Immunol. 160:1666. 31. Filigueira, L., F. O. Nestle, M. Rittig, H. I. Joller, and P. Groscurth. 1996. Human dendritic cells phagocytose and process Borrelia burgdorferi. J. Immunol. 157: 2998. 32. Ojcius, D. M., Y. Bravo de Alba, J. M. Kanellopoulos, R. A. Hawkins, K. A. Kelly, R. G. Rank, and A. Dautry-Varsat. 1998. Internalization of Chlamydia by dendritic cells and stimulation of Chlamidia-specific T cells. J. Immunol. 160:1297. 33. Blank, C., H. Fuchs, K. Rappersberger, M. Ro¨llinghoff, and H. Moll. 1993. Parasitism of epidermal Langerhans cells in experimental cutaneous leishmaniasis with Leishmania major. J. Infect. Dis. 167:418.

34. Albert, M. L., B. Sauter, and N. Bhardwaj. 1998. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs. Nature 392:86. 35. Svensson, M., B. Stockinger, and M. J. Wick. 1997. Bone marrow-derived dendritic cells can process bacteria for MHC-I and MHC-II presentation to T cells. J. Immunol. 158:1997. 36. Inaba, K., S. Turley, F. Yamaide, T. Yyoda, K. Mahnke, M. Inaba, M. Pack, M. Subklewe, B. Sauter, D. Sheff, M. Albert, N. Bhardwaj, I. Mellman, and R. M. Steinman. 1998. Efficient presentation of phagocytosed cellular fragments on the major histocompatibility complex class II products of dendritic cells. J. Exp. Med. 188:2163. 37. Thurnher, M., R. Ramoner, G. Gastl, C. Radmayr, G. Bo¨ck, M. Herold, H. Klocker, and G. Bartsch. 1997. Bacillus Calmette-Gue´rin mycobacteria stimulate human blood dendritic cells. Int. J. Cancer 70:128. 38. Pierre, P., S. J. Turley, E. Gatti, M. Hull, J. Meltzer, A. Mirza, K. Inaba, R. M. Steinman, and I. Mellman. 1997. Developmental regulation of MHC class II transport in mouse dendritic cells. Nature 388:787. 39. Cella, M., A. Engering, V. Pinet, J. Pieters, and A. Lanzavecchia. 1997. Inflammatory stimuli induce accumulation of MHC class II complexes on dendritic cells. Nature 388:782. 40. Foti, M., F. Granucci, D. Aggujaro, W. Luini, S. Minardi, A. Mantovani, S. Sozzani, and P. Ricciardi-Castagnoli. 1999. Upon dendritic cell activation chemokines and chemokine receptors expression are rapidly regulated for recruitment and maintenance of dendritic cells at the inflammatory site. Int. Immunol. 6:979. 41. Sallusto, F., P. Schaerli, P. Loetscher, C. Schaniel, D. Lenig, C. R. Mackay, S. Qin, and A. Lanzavecchia. 1998. Rapid and coordinated switch in chemokine receptor expression during dendritic cell maturation. Eur. J. Immunol. 28:2760. 42. Bonecchi, R., G. Bianchi, P. Panina-Bordignon, D. D’Ambrosio, R. Lang, A. Borsatti, S. Sozzani, P. Allavena, P. A. Gray, A. Mantovani, and F. Sinigaglia. 1998. Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J. Exp. Med. 187:129. 43. Sallusto, F., D. Lenig, C. R. Mackay, and A. Lanzavecchia. 1998. Flexible programs of chemokine receptor expression on human polarized T helper 1 and 2 lymphocytes. J. Exp. Med. 187:875. 44. Gangur, V., F. E. R. Simons, and K. T. Hayglass. 1998. Human IP-10 selectively promotes dominance of polyclonally activated and environmental antigen-driven IFN-g over IL-4 responses. FASEB J. 12:705. 45. Hadida, F., V. Vieillard, B. Autran, I. Clark-Lewis, M. Baggiolini, and P. Debre´. 1998. HIV-specific T cell cytotoxicity mediated by RANTES via the chemokine receptor CCR3. J. Exp. Med. 188:609. 46. Lauber, M. S., B. Roberts, B. Pers, A. Mehling, T. A. Luger, T. Schwarz, and S. Grabbe. 1999. Generation of tumor immunity by bone marrow-derived dendritic cells correlates with dendritic cell maturation stage. J. Immunol. 162:168. 47. Ishida, T., T. Oyama, D. P. Carbone, and D. I. Gabrilovich. 1998. Defective function of Langerhans cells in tumor-bearing animals is the result of defective maturation from hemopoietic progenitors. J. Immunol. 161:4842. 48. Menetrier-Caux, C., G. Montmain, M. C. Dieu, C. Bain, M. C. Favrot, C. Caux, and J. Y. Blay. 1998. Inhibition of the differentiation of dendritic cells from CD341 progenitors by tumor cells: role of interleukin-6 and macrophage colonystimulating factor. Blood 92:4778. 49. Sozzani, S., P. Allavena, A. Vecchi, and A. Mantovani. 1999. The role of chemokines in the regulation of dendritic cell trafficking. J. Leuokocyte Biol. 66:1. 50. Ludewig, B., B. Odermatt, S. Landmann, H. Hengartner, and R. M. Zinkernagel. 1998. Dendritic cells induce autoimmune diabetes and maintain disease via de novo formation of local lymphoid tissue. J. Exp. Med. 188:1493.