science pharmaceutical expo biotech jobs
Originally published online as doi:10.1189/jlb.1006606 on January 16, 2007

Published online before print January 16, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1006606v1
81/4/1075    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gray, R. C.
Right arrow Articles by Harding, C. V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gray, R. C.
Right arrow Articles by Harding, C. V.
(Journal of Leukocyte Biology. 2007;81:1075-1085.)
© 2007 by Society for Leukocyte Biology

CpG-B ODNs potently induce low levels of IFN-{alpha}ß and induce IFN-{alpha}ß-dependent MHC-I cross-presentation in DCs as effectively as CpG-A and CpG-C ODNs

Reginald C. Gray*,{dagger}, John Kuchtey*,1 and Clifford V. Harding*,{dagger},{ddagger},2

* Departments of Pathology and
{dagger} Pharmacology and
{ddagger} Center for AIDS Research, Case Western Reserve University, Cleveland, Ohio, USA

2 Correspondence: Department of Pathology, WRB 5534, Case Western Reserve University, 10900 Euclid Ave., Cleveland, OH 44106-7288, USA. E-mail: cvh3{at}cwru.edu


arrow
ABSTRACT
 
Deoxycytidyl-deoxyguanosine [(CpG)3] oligodeoxynucleotides (ODNs) signal through TLR9 to induce type-I IFN (IFN-{alpha}ß) and IFN-{alpha}ß-dependent MHC-I cross-presentation of exogenous antigens by dendritic cells (DCs). A puzzle was presented by our observation that three ODN classes, CpG-A, CpG-B, and CpG-C, had similar efficacy for induction of IFN-{alpha}ß-dependent MHC-I antigen cross-presentation by myeloid DCs despite greatly differing for induction of IFN-{alpha}ß (CpG-A>CpG-C>>CpG-B). All ODN classes similarly enhanced plasmacytoid DC (pDC) presentation of exogenous MHC-I-restricted peptide, although pDCs did not cross-process protein antigen. MHC-I and the transporter for antigen presentation were induced by all ODN classes or IFN-{alpha}. CpG-B ODNs were slightly more potent than CpG-A or CpG-C ODNs for induction of low levels of IFN-{alpha}ß but less efficacious at high concentrations than CpG-A or CpG-C ODNs. Low levels of IFN-{alpha}ß induced by CpG-B ODNs sufficed for full induction of MHC-I cross-presentation. Thus, CpG-B ODNs are slightly more potent but less efficacious than CpG-A and CpG-C ODNs for induction of IFN-{alpha}ß. High sensitivity to IFN-{alpha}ß allows CpG-B ODNs to be equally efficacious for induction of MHC-I cross-presentation. CpG-B ODNs may be effective for inducing therapeutic responses that require low levels of IFN-{alpha}ß and may avoid unnecessarily high induction of IFN-{alpha}ß.

Key Words: CpG oligodeoxynucleotides • antigen presentation • Toll-like receptor 9 • dendritic cells • type I interferon


arrow
INTRODUCTION
 
Deoxycytidyl-deoxyguanosine [(CpG)3] oligodeoxynucleotides (ODNs) signal through TLR9 to induce innate immune responses and adjuvant effects that promote T cell responses [1 2 3 4 5 6 7 ]. CpG ODNs can be divided into Class A (also known as D-type), Class B (also known as K-type), and Class C ODNs [2 3 4 5 , 8 ]. CpG-B and CpG-C ODNs stimulate B cells efficiently to produce IgM and IL-6, whereas CpG-A ODNs are weak inducers of B cell responses. CpG-A and CpG-C ODNs are strong inducers of type-I IFN (IFN-{alpha}ß) [4 5 , 8 ], particularly in plasmacytoid dendritic cells (pDCs), whereas CpG-B ODNs are weak inducers of IFN-{alpha}ß [2 3 4 5 , 9 ], but they induce DCs to mature and produce TNF-{alpha} [2 , 5 , 9 ]. Thus, CpG-A and CpG-B ODNs have distinct effects, and CpG-C ODNs combine properties of CpG-A and CpG-B ODNs.

TLR9 signaling, particularly by CpG-A ODNs, induces IFN-{alpha}ß, which includes 12 functional isoforms of IFN-{alpha} and one of IFN-ß, which all signal through a single transmembrane receptor (IFN-{alpha}ßR) to induce key immunoregulatory functions that affect innate and adaptive immunity. Signaling by TLRs and other receptors initially leads to induction of IFN-ß and IFN-{alpha}4, and autocrine/paracrine signaling triggers a positive-feedback system that induces other forms of IFN-{alpha} and amplifies the IFN-{alpha}ß response.

Exogenous antigens are processed in endocytic vacuolar compartments for presentation by MHC-II molecules, whereas endogenous antigens are processed in the cytosol for presentation by MHC-I molecules. Some exogenous antigens, however, are presented by MHC-I molecules after processing in vacuolar compartments and/or penetration from vacuolar compartments into the cytosol for cytosolic processing [10 11 12 ]. Such cross-processing (or alternate MHC-I antigen processing) leads to MHC-I cross-presentation, which can produce cross-priming [13 ], particularly efficient for antigens of a particulate nature, e.g., antigens associated with beads, liposomes, microorganisms, apoptotic cells, or other cellular fragments.

TLR9 signaling regulates multiple aspects of antigen processing and APC function. In mice, TLR9 is widely expressed by APCs, including myeloid DCs (mDCs), pDCs, macrophages, and B cells. In humans, TLR9 expression is restricted to pDCs and B cells [1 , 3 , 14 ]. CpG ODNs promote maturation of mDCs, which results in enhanced expression of MHC-I, MHC-II, and costimulatory molecules (i.e., CD40, CD80, and CD86) and increased MHC-I and MHC-II antigen-presentation function [15 16 17 18 ]. Cross-processing of exogenous antigens for presentation by MHC-I is induced by CpG ODNs via a mechanism that is blocked by genetic deficiency of IFN-{alpha}ßR or the presence of IFN-{alpha}ß-neutralizing antibodies [19 20 21 ]. Thus, the ability of CpG ODNs to induce MHC-I cross-presentation is IFN-{alpha}ß-dependent.

As induction of MHC-I cross-processing is IFN-{alpha}ß-dependent, and CpG-A and CpG-C ODNs are more efficacious for induction of IFN-{alpha}ß, we tested whether CpG-A and CpG-C ODNs are more efficacious than CpG-B ODNs in the induction of IFN-{alpha}ß-dependent MHC-I cross-presentation. Our studies assessed the efficacy and potency of different ODN classes for induction of IFN-{alpha}ß and MHC-I antigen cross-processing. Potency, as defined, is the concentration of a drug or reagent needed to produce an effect. Efficacy is distinct from potency and relates to the maximum magnitude of the effect that can be produced by a drug or reagent, regardless of the concentration. Despite differences in efficacy for induction of IFN-{alpha}ß, the three ODN classes had similar efficacy for induction of IFN-{alpha}ß-dependent MHC-I antigen cross-processing of protein antigen by mDCs and MHC-I presentation of peptide by mDCs and pDCs. The focus of our studies was to explain the discrepancy among the similar efficacy of CpG-A, CpG-B, and CpG-C ODNs for induction of IFN-{alpha}ß-dependent MHC-I cross-presentation and their differing efficacies for induction of IFN-{alpha}ß. For induction of IFN-{alpha}ß, CpG-B ODNs were as potent or slightly more potent than CpG-A and CpG-C ODNs, although CpG-B ODNs had lower efficacy. Moreover, low levels of IFN-{alpha}ß induced by CpG-B ODNs sufficed for full induction of MHC-I cross-presentation. Thus, CpG-B ODNs are slightly more potent but less efficacious than CpG-A and CpG-C ODNs for induction of IFN-{alpha}ß. High sensitivity to IFN-{alpha}ß allows CpG-B ODNs to be similarly efficacious for induction of IFN-{alpha}ß-dependent MHC-I cross-presentation.


arrow
MATERIALS AND METHODS
 
Antibodies and reagents
Anti-mouse CD11c (HL3, FITC- or APC-conjugated), anti-B220-PE (RA3-6B2), and anti-CD11b (M1/70, PE-Cy7- or Alexa-Flour488-conjugated) were from BD Biosciences PharMingen (San Diego, CA, USA). Anti-myeloid pyrrole-2,3-dicarboxylic acid (anti-mPDCA)-1-APC was from Miltenyi Biotec (Auburn, CA, USA). Isotype controls (BD Biosciences PharMingen) were APC-hamster IgG1,{lambda}; FITC-hamster IgG2,{kappa}; PE-rat IgG2a,{kappa}; PE-Cy7-rat IgG2b,{kappa}; Alexa-Flour488-rat IgG2b; and APC-rat IgG2b,{kappa}, respectively. CpG ODNs included CpG-A-2216 (5'-ggG GGA CGA TCG TCg ggg gG-3'), CpG-A-2336 (5'-ggG GAC GAC GTC GTG ggg ggG-3'), CpG-B-2006 (5'-tcg tcg ttt tgt cgt ttt gtc gtt-3'), CpG-B-10105 (sequence proprietary, Coley Pharmaceutical Group, Wellesley, MA, USA), CpG-B-1668 (5'-tcc atg acg ttc ctg atg ct-3'), CpG-B-1826 (5'-tcc atg acg ttc ctg acg tt-3'), CpG-C-2395 (5'-tcg tcg ttt tcg gcg cgc gcc g-3'), CpG-C-2429 (5'-tcg tcg ttt tcg gcg gcc gcc g-3'), and CpG-C-M362 (5'-tcg tcg tcg ttc gaa cga cgt tga t-3'). Non-CpG (GpC) ODNs were non-CpG-A-2243 (5'-ggG GGA GCA TGC TGg ggg gG-3'), non-CpG-B-2138 (5'-tcc atg agc ttc ctg agc tt-3'), non-CpG-C-2137 (5'-tgc tgc ttt tgt gct ttt gtg ctt-3'), and M362 control (5'-tgc tgc tgc ttg caa gca gct tga t-3'). Lowercase letters in ODN sequences refer to nucleotides for which the 3' internucleotide linkage is phosphorothioate-modified, and uppercase letters refer to standard phosphodiester-linked nucleotides. ODNs contained phosphorothioate-modified linkages at the 5' and 3' ends (A class) or throughout (B and C classes) to resist nuclease degradation. ODNs were provided by Coley Pharmaceutical Group, purchased from InvivoGen (San Diego, CA, USA), or were described elsewhere [22 , 23 ] and synthesized by MWG-Biotech (High Point, NC, USA). TLR7 agonists Imiquimod (R-837) and ssRNA40 were purchased from InvivoGen. ODNs and TLR7 agonists were dissolved in endotoxin-tested (≤0.05 units) PBS (Cambrex, East Rutherford, NJ, USA) and sterile cell-culture water (Sigma-Aldrich, St. Louis, MO, USA). Mouse recombinant IFN-{alpha}A (rIFN-{alpha}) and rIFN-ß were purchased from PBL Laboratories (Piscataway, NJ, USA). Bead-conjugated OVA (latex-OVA) was prepared as described [20 , 24 ] by noncovalent, passive adsorption of chicken OVA (Sigma-Aldrich) to 2 µm polystyrene beads (Polysciences, Warrington, PA, USA).

Cells and media
Standard medium was RPMI (Invitrogen, Carlsbad, CA, USA, and Hyclone Laboratories, Logan, UT, USA) with L-glutamine and glucose and supplemented with 10% heat-inactivated FCS (HyClone Laboratories), 50 µM 2-ME, 1% penicillin-streptomycin (Invitrogen and HyClone Laboratories), and 1 mM sodium pyruvate (Invitrogen and HyClone Laboratories). Mice were housed under specific pathogen-free conditions. DCs were prepared from femur and tibia bone marrow cells of C57BL/6 or CBA/J mice (Jackson Laboratories, Bar Harbor, ME, USA), IFN-{alpha}ßR–/– A129 mice on a 129/SvEv background (B&K Universal Ltd., Grimston, Aldbrough, UK), 129S6/SvEv wild-type mice (Taconic Laboratories, Germantown, NY, USA), or TLR9–/– mice (generous gift from S. Akira Osaka University, Osaka, Japan [6 ]). Bone marrow cells were cultured at 106 cells/ml in six-well bacterial-grade dishes for 8–10 days in mouse rFMS-like tyrosine kinase 3 ligand (Flt3L;100 ng/ml, R&D Systems, Minneapolis, MN, USA) or Flt3L-Ig fusion protein (1 µg/ml, BioExpress, West Lebanon, NH, USA). Medium was replenished on Days 3 and 6. Alternatively, bone marrow cells were cultured at 2 x 105 cells/ml in 10 cm bacterial-grade dishes for 8 days in 10 ng/ml GM-CSF (R&D Systems). Medium was replenished on Days 3 and 6. On Day 8 or 9, nonadherent cells were removed from plates, pelleted, resuspended, and counted. For some experiments, Flt3L or GM-CSF DCs were used without further purification. Flt3L DC cultures contain a mixture of mDCs and pDCs (mixed DCs), whereas GM-CSF DC cultures contain mDCs (<1% pDCs) [25 ]. Alternatively, Flt3L DC cultures were used to purify pDCs (CD11c+, B220+, mPDCA-1+, CD11b–) and mDCs (CD11c+, CD11b+, mPDCA-1–, and B220–) by FACS or by use of magnetic beads and the MACS system (Miltenyi Biotec). pDCs were purified by positive selection using a mixture of anti-B220 beads and anti-mPDCA-1 beads (Miltenyi), and mDCs were purified by negative selection with the same bead mixture followed by positive selection with CD11b beads. DCs were plated at 4–6 x 106 cells/well in six-well plates in standard medium for 24 h, with or without CpG ODNs. Cells were incubated with 5–10 µg/ml Fc block (anti-CD16/CD32, Fc{gamma}III/II, BD Biosciences PharMingen) and stained with CD11c-APC or CD11c-FITC, CD11b-PE-Cy7, B220-PE, mPDCA-1-APC, and isotype controls (BD Biosciences PharMingen or Miltenyi Biotec). For FACS, 1–14 x 107 cells were sorted for pDCs (CD11c+, B220+, CD11b–) and mDCs (CD11b+, CD11c+, B220–) using a FACS Aria cell sorter and FACSDiva software (BD Biosciences PharMingen). Magnetic bead-separated pDC and mDC purities were >95%, as determined by postsort flow cytometry using the same markers and an LSRII flow cytometer (BD Biosciences PharMingen).

Antigen-presentation assay
Antigen-presentation assays were performed as described previously [20 , 24 ]. Murine DCs were plated at 7.5–10 x 104 cells/well in flat- or U-bottom, 96-well plates in standard medium, with or without CpG ODN, non-CpG ODN, TLR7 agonist (R-837 or ssRNA40), and/or rIFN{alpha} and rIFNß for 18–24 h. Latex-OVA or OVA257–264 peptide (SIINFEKL) was added for 2 h. Cells were fixed in 0.5% paraformaldehyde, washed, and incubated for 24 h with CD8OVA1.3 T hybridoma cells, which detect OVA257–264:Kb complexes [26 ]. IL-2 was measured using a colorimetric cytotoxic T cell 2 (CTLL-2) bioassay. CTLL-2 cells (5x103 cells/well) were incubated with supernatants for 24 h. Alamar blue (TREK Diagnostic Systems, Cleveland, OH, USA) was added (15 µl/well) for 24 h, and Alamar blue reduction was determined by the difference in OD at 550 and 595 nm. All experimental conditions were tested in triplicate.

ELISA and quantitative real-time PCR (qRT-PCR)
DCs were plated at 7.5–10 x 104 cells/well in flat-bottomed, 96-well plates or at 2 x 106 cells/well in 48-well, non-tissue, culture-treated plates in standard medium and incubated for 24 h, with or without CpG ODN. Plates were spun at ≥15,000 g for 5 min, and supernatants were removed and assessed immediately or stored at –80°C. Supernatants were diluted and tested for IFN-{alpha} and IFN-ß using mouse IFN-{alpha} and IFN-ß sandwich ELISAs (PBL Biomedical Laboratories, Piscataway, NJ, USA). Absorbance was determined at 450 nm. For RT-PCR, RNA was isolated from DCs using an RNeasy mini kit (Qiagen, Valencia, CA, USA). Cells were pelleted at 4°C and resuspended in RLT lysis buffer (Qiagen). Total RNA was extracted following on-column DNase digestion using RNeasy mini columns (Qiagen) and collected in RNase-free water. Yield was determined by OD. Oligo(dT)-primed RT of RNA into cDNA was performed with Superscript II first-strand synthesis kit (Invitrogen), and 5% of the product was used for each qRT-PCR sample using PCR buffer with hot-start Invitrogen Taq polymerase, Bio-Rad SYBR Green detection, and the Bio-Rad iCycler fluorescence detection system (Bio-Rad, Hercules, CA, USA). All conditions were tested in triplicate. Primer pairs for H-2Db (sense 5'-TCC GAG ATT GTA AAG CGT GAA-3' and antisense 5'-TGT GGT TGC TGG GAT TTG A-3'), H-2Kb (sense 5'-GCC CTC AGT TCT CTT TAG TCA-3' and antisense 5'-GCC CTA GGT CAA GAT GAT AAC-3'), transporter associated with antigen processing-1A (TAP-1A; sense 5'-CAG CGG CTC CTG TAT GAG A-3' and antisense 5'-CAG TCC AGA GGC CTT GTC AGT-3'), and GAPDH (sense 5'-AAC GAC CCC TTC ATT GAC-3' and antisense 5'-TCC ACG ACA TAC TCA GCA C-3') were as described previously [20 , 27 ] or designed using Clone Manager Suite v7.11 and Primers Designers v5.11 (Scientific and Educational Software, Cary, NC, USA). A BLAST search was then performed to verify specificity.

Statistical analysis
Data points are means of triplicate samples, except for ELISA data points, which are means of duplicate samples. Error bars represent SD and where not shown, are smaller than the symbol size. Statistical analysis was performed with SigmaStat v3.1. Data sets were analyzed with a two- or three-way ANOVA, and t and pvalues were obtained from a pairwise multiple comparison Bonferroni t-test comparing treated and untreated (or non-CpG ODN-treated) values as shown (#, P<0.05; +, P<0.01; *, P<0.001). qRT-PCR data were analyzed based on normalized copy number (NC): NC = (copy number of mRNA of interest)/(copy number of GAPDH) x (1000). Fold induction = (NC of treated cells)/(NC of untreated cells).


arrow
RESULTS
 
All three classes of CpG ODNs enhance MHC-I cross-processing and cross-presentation of particulate antigen by mDCs but not pDCs
To evaluate the efficacy of CpG ODNs to induce MHC-I cross-processing and presentation by mDCs and pDCs, we compared processing of latex-OVA at various concentrations by mixed DCs (cells from Flt3L-stimulated bone marrow cultures containing mDCs and pDCs), purified mDCs, and purified pDCs. Cells were incubated for 18–24 h, with or without CpG-A, CpG-B, CpG-C, or non-CpG ODN at 3 µg/ml (many published in vitro studies use CpG ODNs at 1–3 µg/ml; as demonstrated below, this condition is fully efficacious, allowing us to do antigen dose-response studies at a single, effective ODN dose). The cells were then incubated for 2 h with various doses of latex-OVA or OVA257–264 peptide and fixed. OVA257–264:Kb complexes were assessed with CD8OVA1.3 T hybridoma cells.

Cross-processing and presentation of latex-OVA were induced in Flt3L-induced, mixed DCs (Fig. 1A ) and purified mDCs (Fig. 1B) by CpG-A, CpG-B, and CpG-C ODNs (Fig. 1) . Non-CpG-A, non-CpG-B, and non-CpG-C ODNs failed to induce cross-processing (data not shown, and data reported elsewhere [20 ]), confirming CpG specificity of this effect. Similar results were obtained with GM-CSF-induced mDCs (data not shown), which are largely pDC-deficient [25 ]. In contrast, pDCs did not cross-process latex-OVA (Fig. 1C) . After stimulation with CpG ODN, mixed DCs, mDCs, and pDCs were able to present exogenous OVA257–264peptide, but without CpG ODN, exogenous peptide presentation was absent or inefficient (Fig. 1D 1E 1F) . Thus, the CpG-dependent cross-processing function was exhibited by mDCs but not pDCs, although both types of DCs presented exogenous peptide after stimulation with CpG ODN.


Figure 1
View larger version (33K):
[in this window]
[in a new window]

  		Figure 1. CpG-A, CpG-B, and CpG-C ODNs induce cross-processing of particulate antigen (Ag) by mDCs but not pDCs. C57BL/6 Flt3L-cultured, mixed DCs (A and D), immunoaffinity magnetic bead-purified mDCs (B and E), or pDCs (C and F) were incubated for 24 h, with or without 3 µg/ml CpG-A-2336 (A–F), CpG-B-1668 (A, D, and E), CpG-B-1826 (B, C, and F), CpG-C-2395 (A–F), or non-CpG ODN 2243 (D). The cells were then incubated for 2 h with various concentrations of latex-OVA (A–C) or OVA257–264peptide (D–F), fixed with paraformaldehyde, and incubated with CD8OVA1.3 T-hybridoma cells to detect presentation of OVA257–264:Kb complexes. IL-2 production was assessed with a CTLL colorimetric IL-2 bioassay. Error bars represent SD of triplicate wells and where not visible, are smaller than symbol size. Pvalues from two-way ANOVA and Bonferroni t-test comparing treated and untreated (or non-CpG-treated) values are shown (#, P<0.05; +, P<0.01; *, P<0.001). Data are representative of at least three independent experiments, all of which showed statistically significant (P<0.05), CpG-induced increases in T cell responses to 100, 300, 1000, and 3000 ng/ml latex-OVA presented by mixed DCs and mDCs, as well as statistically significant CpG-induced increases in pDC presentation of OVA257–264 (10–300 pM).

Moreover, the antigen cross-processing, dose-response curves and plateau responses were similar whether mDCs were stimulated with CpG-A, CpG-B, or CpG-C ODNs. These results indicate that the three ODN classes had similar efficacy (maximum response) for the induction of MHC-I cross-processing by mDCs.

CpG-A, CpG-B, and CpG-C ODNs have similar potency for induction of MHC-I antigen cross-processing and presentation
The preceding experiments demonstrated that all three classes of CpG ODNs had similar efficacy for induction of cross-processing, as assessed by detailed antigen dose-response studies, but they did not assess the relative potency (the minimum concentration needed to induce an effect) of the ODNs for induction of cross-processing. To evaluate the relative potency of CpG-A, CpG-B, and CpG-C ODNs for induction of cross-processing, C57BL/6 mDCs and pDCs were incubated with various concentrations of ODNs for 18–24 h, incubated with 1 µg/ml latex-OVA, and fixed prior to the T hybridoma assay. Figure 2 demonstrates data obtained with two CpG-A ODNs, four CpG-B ODNs, three CpG-C ODNs, and two non-CpG control ODNs, and ODNs of all three classes enhanced cross-processing at minimum ODN concentrations of 0.1–0.3 µg/ml or below (as low as 0.03 µg/ml for CpG-B ODNs). CpG-B ODNs were in general slightly more potent than CpG-A or CpG-C ODNs, but there was sufficient variation within ODN classes that we conclude conservatively only that CpG-B ODNs are as potent or slightly more potent than CpG-A or CpG-C ODNs for induction of cross-processing. Similar results were observed with mDCs from GM-CSF-stimulated bone marrow cultures (data not shown). None of the ODNs induced cross-processing by pDCs, consistent with Figure 1 . At higher concentrations (e.g., 1–3 µg/ml), ODNs of all classes produced similar, maximum plateau mDC-driven responses (i.e., all ODN classes had similar efficacy for induction of cross-processing, consistent with Fig. 1 ). Thus, although CpG-B ODNs may be slightly more potent, CpG-A, CpG-B, and CpG-C ODNs have similar potency and efficacy for induction of MHC-I antigen cross-processing and presentation by mDCs.


Figure 2
View larger version (40K):
[in this window]
[in a new window]

  		Figure 2. CpG-A, CpG-B, and CpG-C ODNs have similar potency and efficacy for induction of MHC-I cross-processing and presentation of particulate antigen. From Flt3L-cultured C57BL/6 DCs, mDCs (A–C) were isolated with anti-CD11b magnetic beads, and pDCs (D) were isolated with magnetic beads conjugated with anti-CD45R (B220) and anti-mPDCA-1. Cells were stimulated with or without CpG-A, CpG-B, CpG-C, or non-CpG ODN for 18–24 h. The cells were then incubated for 2 h with 1 µg/ml latex-OVA. Presentation of OVA257–264:Kb complexes was assessed with CD8OVA1.3 T hybridoma cells as in Figure 1 . Error bars represent SD of triplicate wells and where not shown, are smaller than symbol size.

Induction of MHC-I antigen cross-processing and cross-presentation by CpG-A, CpG-B, and CpG-C ODNs are IFN-{alpha}ßR-dependent
Previous studies showed that CpG DNA enhances MHC-I cross-presentation by a mechanism that is IFN-{alpha}ßR-dependent [19 , 20 ], but the relative activity of CpG-A, CpG-B, and CpG-C ODNs was not assessed. To evaluate the role of IFN-{alpha}ß, we studied DCs from wild-type 129S6/SvEv and IFN-{alpha}ßR–/– mice. Mixed Flt3L-induced DCs were incubated for 24 h, with or without CpG or non-CpG ODN (1 µg/ml), incubated for 2 h with 1 µg/ml latex-OVA, and fixed with paraformaldehyde. Presentation of OVA257–264:Kb complexes was assessed using CD8OVA1.3 T hybridoma cells and an IL-2 bioassay. CpG-A, CpG-B, and CpG-C ODNs induced cross-processing and presentation by wild-type 129S6/SvEv but not IFN-{alpha}ßR–/– DCs (Fig. 3 ). Non-CpG ODN failed to induce cross-processing (data not shown). Thus, all three classes of CpG ODNs induce MHC-I cross-processing and cross-presentation by an IFN-{alpha}ßR-dependent mechanism, indicating a general requirement for IFN-{alpha}ß autocrine/paracrine signaling for CpG ODN-induced MHC-I cross-presentation.


Figure 3
View larger version (21K):
[in this window]
[in a new window]

  		Figure 3. CpG-A-, CpG-B-, and CpG-C-induced cross-processing is IFN-{alpha}ß-dependent. DCs were prepared from wild-type 129S6/SvEv or IFN-{alpha}ßR–/–-deficient mice and incubated with or without 1 µg/ml CpG-A-2336, CpG-B-1668, or CpG-C-2395 ODN for 24 h. The cells were then incubated for 2 h with 1 µg/ml latex-OVA and fixed. Presentation of OVA257–264:Kb complexes was assessed with CD8OVA1.3 T hybridoma cells as in Figure 1 . Error bars represent SD of triplicate wells and where not shown, are too small for visualization. Data are representative of at least three independent experiments. Two-way ANOVA and Bonferroni t-test showed statistical significance (P<0.05) for the difference in responses by wild-type and IFN-{alpha}ßR–/– DCs for all ODNs. In addition, significantly (P<0.05) greater responses were seen with wild-type DCs after incubation with all of these CpG ODNs at 0.01, 0.03, 0.1, 0.3, 1, and 3 µg/ml (data not shown).

CpG-A, CpG-B, and CpG-C ODNs differentially induce IFN-{alpha}ß protein expression in mDCs and pDCs
In contrast to their similar potency and efficacy for induction of IFN-{alpha}ß-dependent cross-processing (Figs. 1 and 2) , previous studies demonstrated that CpG-A, CpG-B, and CpG-C ODNs have different efficacy for the induction of IFN-{alpha}ß [2 , 4 , 5 , 8 , 23 ]. To investigate this issue, we first tested the relative ability of CpG-A, CpG-B, and CpG-C ODNs to induce IFN-{alpha}ß protein expression in different types of DCs from C57BL/6 mouse bone marrow cultures (mixed Flt3L-cultured DCs, mDCs, or pDCs). Magnetic bead isolation and FACS were used to purify mDCs (CD11c+, CD11b+, and B220–) and pDCs (CD11c+, CD11b–, B220+, and mPDCA-1+). The mDC and pDC populations were >95% pure, as determined by subsequent flow cytometry for these markers (data not shown). DCs were incubated with or without CpG or non-CpG ODNs (3 µg/ml, as used in Fig. 1 ), and supernatants were assessed by ELISA for IFN-{alpha} and IFN-ß. As mixed DCs are easier to prepare in large quantities, they were used to test a large battery of two CpG-A ODNs, three CpG-B ODNs, two CpG-C ODNs, and two non-CpG control ODNs. Under these conditions, IFN-{alpha} (Fig. 4A ) and IFN-ß (Fig. 4B) were induced with greatest efficacy by CpG-A ODNs, followed by CpG-C ODNs, and CpG-B ODNs had low efficacy for induction of IFN-{alpha} and IFN-ß, although some CpG-B ODNs matched or exceeded the efficacy of CpG-C-2429 for induction of IFN-ß. To compare responses of mDCs and pDCs, these cell types were purified and incubated with a representative ODN from each class. pDCs and mDCs expressed IFN-{alpha} (Fig. 4C) and IFN-ß (Fig. 4D) in response to CpG ODNs, but pDCs produced much more IFN-{alpha}ß than mDCs with all CpG ODN classes. As seen with mixed DCs and in other studies [4 , 8 ], the rank order of ODN classes for efficacy of IFN-{alpha} production was CpG-A > CpG-C >> CpG-B (a ratio of 18:14:1 for IFN-{alpha} production in response to CpG-A, CpG-C, and CpG-B, respectively).


Figure 4
View larger version (35K):
[in this window]
[in a new window]

  		Figure 4. CpG-A and CpG-C ODNs are more efficacious than CpG-B ODNs for induction of IFN-{alpha} and IFN-ß in pDCs and mDCs. C57BL/6 Flt3L-cultured, mixed DCs (A and B) or purified mDCs or pDCs (C and D) were incubated for 24 h, with or without ODNs at 3 µg/ml. (C and D) The ODNs were CpG-A-2336, CpG-B-1826, CpG-C-2395, and non-CpG-2243 ODN; (A and B) data with a larger set of ODNs, as indicated in the figure. Expression of IFN-{alpha} and IFN-ß was measured by ELISA. Data are representative of three independent experiments, all of which showed statistically significant, CpG-induced production of IFN-{alpha} and IFN-ß by ANOVA and Bonferroni t test (comparing CpG-treated to untreated or non-CpG-treated, mixed DCs, mDCs, and pDCs; +, P<0.01).

CpG ODN-induced production of IFN-{alpha} protein is dependent on IFN-{alpha}ß autocrine/paracrine signaling
To determine the role of IFN-{alpha}ß autocrine/paracrine signaling in the induction of IFN-{alpha} by CpG ODNs, mixed DCs derived from IFN-{alpha}ßR–/– and 129S6/SvEv wild-type mice were incubated for 18–24 h, with or without CpG ODN at 0.3, 1, or 3 µg/ml. Supernatants were harvested and evaluated for IFN-{alpha} protein by ELISA. Induction of IFN-{alpha} by CpG ODNs was IFN-{alpha}ßR-dependent and therefore, dependent on IFN-{alpha}ß autocrine/paracrine signaling (Fig. 5 ), consistent with previous findings [9 , 20 , 28 ]. In addition, the dose-response curve for CpG-B appeared biphasic, and induction of IFN-{alpha} was higher at 1 µg/ml than 0.3 or 3 µg/ml. This experiment explored only three ODN doses, however, necessitating more detailed dose-response studies.


Figure 5
View larger version (23K):
[in this window]
[in a new window]

  		Figure 5. CpG-A-, CpG-B-, and CpG-C-induced IFN-{alpha} protein expression is IFN-{alpha}ßR-dependent. Flt3L-cultured, mixed DCs were prepared from wild-type (WT) 129S6/SvEv or IFN-{alpha}ßR–/– mice and incubated for 24 h, with or without CpG-A-2336, CpG-B-1668, or CpG-C-2395 ODNs (0.3, 1, or 3 µg/ml). IFN-{alpha} protein expression was measured by ELISA. Data are representative of three independent experiments. Two-way ANOVA and Bonferroni t-test showed significantly higher IFN-{alpha} expression by wild-type DCs relative to IFN-{alpha}ßR–/– DCs under the indicated conditions (#, P<0.05).

CpG-B ODNs are slightly more potent but less efficacious than CpG-A or CpG-C ODNs for induction of IFN-{alpha} protein expression by DCs
We evaluated the relative potency and efficacy of CpG-A, CpG-B, and CpG-C ODNs for induction of IFN-{alpha} expression by Flt3L-cultured, mixed DCs, derived from C57BL/6 mice (note that the ODN dose/IFN-{alpha}ß response curve for these mice is slightly different from that observed with 129S6/SvEv mice, which were used as wild-type controls in experiments with IFN-{alpha}ßR–/– mice). Mixed DCs were stimulated with ODNs for 24 h, and supernatants were assessed by ELISA. Representative ODNs from all three ODN classes induced IFN-{alpha} expression in a dose-dependent manner (Fig. 6 ). CpG-A-2336 and CpG-C-2395 were more efficacious and at relatively high concentrations (at or above ~2 µg/ml, corresponding to ~0.3 µM), produced far more IFN-{alpha} than CpG-B-1826. Conversely, CpG-B-1826 was slightly more potent, as it induced IFN-{alpha} expression at lower ODN concentrations (0.06–0.6 µg/ml and 0.01–0.1 µM). Dose response curves for all three ODNs were biphasic. IFN-{alpha} responses declined at high ODN concentrations, and this was more pronounced and occurred at lower concentrations with CpG-B-1826 (IFN-{alpha} responses virtually ceased with CpG-B ODNs at ≥2.0 µg/ml). Figure 6 shows representative data with one ODN of each class. Other experiments confirmed these results with other ODNs (CpG-A-2216, CpG-B-1668, and CpG-C-2429), and these data are consistent with the findings of Hemmi et al. [9 ]. These data indicate that CpG-B ODNs are slightly more potent but much less efficacious than CpG-A and CpG-C ODNs for induction of IFN-{alpha}.


Figure 6
View larger version (21K):
[in this window]
[in a new window]

  		Figure 6. CpG-B ODN is more potent but less efficacious than CpG-A or CpG-C ODN for induction of IFN-{alpha} protein expression by DCs. C57BL/6 Flt3L-cultured, mixed DCs (106/well) were incubated for 24 h with CpG-A-2336, CpG-B-1826, CpG-C-2395, or non-CpG-2243 ODN. IFN-{alpha} protein expression was measured by ELISA. (A) Results for all ODNs; (B) results for CpG-B-1826 and non-CpG-2243 with a reduced y-axis scale. Data are representative of three independent experiments. In addition, experiments with other ODNs (CpG-A-2216, CpG-B-1668, and CpG-C-2429) confirmed this pattern of results.

Synergy involving IFN-{alpha}ß-independent and IFN-{alpha}ß-dependent signaling does not explain the similar efficacy of ODNs for induction of IFN-{alpha}ß-dependent cross-processing
The preceding studies demonstrated that CpG-A, CpG-B, and CpG-C ODNs have similar efficacy for induction of IFN-{alpha}ß-dependent cross-processing, despite differing efficacy for induction of IFN-{alpha}ß. One possible explanation for this is a rate-limiting role for an IFN-{alpha}ß-independent signaling mechanism that synergizes with IFN-{alpha}ß signaling (hence, dependence on IFN-{alpha}ß with efficacy limited by an IFN-{alpha}ß-independent mechanism). To determine whether CpG ODNs induce IFN-{alpha}ß-independent (TLR9-dependent) signaling, which synergizes with signaling by CpG-induced IFN-{alpha}ß, we examined cross-presentation by Flt3L-cultured, mixed DCs stimulated for 18–24 h with low concentrations of CpG-B-1826 (0.03, 0.1, and 0.3 µg/ml) plus various concentrations of rIFN-{alpha} (Fig. 7 ) or rIFN-ß (data not shown). Cross-processing was induced by all three concentrations of CpG-B-1826 or by rIFN-{alpha} or IFN-ß alone, at or above 0.02–0.2 pg/ml. The addition of rIFN-{alpha} or IFN-ß to CpG-B-1826 produced little or no increase in cross-processing beyond that achieved by the ODN alone (Fig. 7) . Significant cross-processing was not detected with lower ODN concentrations (e.g., 0.01 µg/ml CpG-B-1826), even in the presence of exogenous IFN-{alpha}ß, unless the exogenous IFN-{alpha}ß concentration alone was sufficient to induce cross-processing (data not shown). Other experiments that assessed cross-processing at different kinetic points or with GM-CSF-cultured mDCs also failed to detect synergy. Thus, no synergy was detected between IFN-{alpha}ß and CpG-B-1826 under the conditions that we tested, although we cannot completely exclude the existence of a synergy mechanism by these experiments.


Figure 7
View larger version (27K):
[in this window]
[in a new window]

  		Figure 7. The discrepancy between CpG-induced IFN-{alpha}ß protein expression and MHC-I antigen cross-presentation is not a result of synergy of IFN-{alpha}ß signaling with IFN-{alpha}ßR-independent, TLR9-dependent signaling. C57BL/6 Flt3L-cultured, mixed DCs were incubated for 24 h with CpG-B-1826 (0, 0.03, 0.1, or 0.3 µg/ml) and varying concentrations of rIFN-{alpha}. Processing of latex-OVA (1 µg/ml) was assessed as in Figure 1 . Error bars represent SD of triplicate wells and where not visible, are smaller than the symbol size. Two-way ANOVA and Bonferroni t-test comparing CpG- or rIFN-{alpha}-treated DCs to untreated DCs showed that statistically significant (P<0.01) increases in cross-processing were induced by rIFN-{alpha} at 0.2, 2, 20, and 200 pg/ml and by CpG-B-1826 at 0.03, 0.1, and 0.3 µg/ml.

We also tested the potential contribution of synergistic signaling to induction of mRNA for TAP-1 and MHC-I (H-2Db/Kb), which are regulated by IFN-{alpha}ß. Flt3L-cultured, mixed DCs were incubated with CpG ODN and/or rIFN-{alpha} for 3 or 6 h (when CpG induces maximum TAP-1 mRNA expression [20 ]), and mRNA levels were determined by qRT-PCR. CpG-A ODN, CpG-B ODN, CpG-C ODN, and rIFN-{alpha} individually induced TAP1 mRNA expression at 3 h and 6 h and H-2Db at 6 h, but synergy between CpG ODNs and IFN-{alpha}ß was not detected (data not shown). Thus, CpG ODNs can induce IFN-{alpha}ß-dependent regulation of TAP and MHC-I, as well as DC cross-presentation, but we were unable to detect an important, synergistic role for IFN-{alpha}ß-independent signaling by CpG ODNs.

These studies also revealed the extreme sensitivity with which DC cross-processing is regulated by IFN-{alpha}ß. Cross-processing was induced by rIFN-{alpha} at concentrations as low as 0.02–0.2 pg/ml (0.1–1 U/ml), a level well below the lower limit of detection by the IFN-{alpha} ELISA (1–10 pg/ml). The amount of IFN-{alpha}ß induced by 0.03 µg/ml CpG-B-1826 was below the limit of detection by ELISA, although the induction of cross-presentation by CpG-B-1826 was dependent on IFN-{alpha}ßR. In summary, CpG induction of MHC-I cross-processing is IFN-{alpha}ß-dependent, but the requisite amount of IFN-{alpha}ß is low and undetectable by ELISA. The extreme sensitivity to IFN-{alpha}ß may explain the high efficacy of CpG-B ODNs in the induction of cross-processing, despite their relatively low efficacy for induction of IFN-{alpha}ß.

IFN-{alpha}ß autocrine/paracrine signaling is necessary and sufficient for induction of MHC-I cross-processing by CpG ODN
As the similar efficacy of different CpG ODN classes for induction of cross-processing is not explained by synergy of IFN-{alpha}ß-dependent and IFN-{alpha}ß-independent signaling, we further tested whether CpG-induced IFN-{alpha}ß is sufficient for induction of cross-processing. We used a combination of wild-type and receptor-knockout cells of different MHC haplotypes to test whether paracrine signaling by CpG-induced IFN-{alpha}ß modulates cross-processing, independent of other CpG-induced signaling. Taking advantage of the H-2b restriction of the CD8OVA1.3 T cell read-out, we tested whether IFN-{alpha}ß produced by CpG-stimulated wild-type H-2k cells was sufficient to induce cross-processing by TLR9–/– H-2b DCs present in the same culture well. Flt3L-cultured, mixed DCs were prepared from wild-type C57BL/6 (H-2b), TLR9–/– (H-2b), IFN-{alpha}ßR–/– (H-2b), and wild-type CBA/J (H-2k) mice. Various combinations of these DCs were cocultured in the presence of CpG ODN or R837 (TLR7 agonist), exposed to latex-OVA, fixed, and incubated overnight with CD8OVA1.3 T hybridoma cells. As CD8OVA1.3 T hybridoma cells recognize OVA257–264:Kb complexes, they will respond to H-2b DCs but not H-2k DCs. As expected, CpG-A, CpG-B, and CpG-C ODNs induced cross-presentation of OVA257–264:Kb complexes by wild-type H-2b DCs (Fig. 8A ) but not wild-type H-2k DCs (Fig. 8B) or TLR9–/– H-2b DCs alone. R837, a TLR7 agonist, induced cross-presentation by wild-type and TLR9–/– DCs (Fig. 8C) . The observation that CpG ODNs induced cross-presentation of OVA257–264:Kb complexes by a mixture of TLR9–/– H-2b DCs and wild-type H-2k DCs was most important (Fig. 8D) , and the sensitivity of this system to CpG ODNs was similar to that of wild-type H-2b DCs alone (Fig. 8A) or a mixture of wild-type H-2b and wild-type H-2k DCs (Fig. 8E) . These data indicate that induction of cross-presentation by TLR9–/– H-2b DCs could be explained fully by the activity of CpG-induced factors expressed by the wild-type H-2k DCs. Moreover, CpG ODNs did not induce cross-presentation by a mixture of IFN-{alpha}ßR–/– H-2b DCs and wild-type H-2k DCs, confirming that IFN-{alpha}ß was an essential, CpG-induced factor. These data suggest that autocrine/paracrine effects of CpG-induced IFN-{alpha}ß are necessary and sufficient to explain the induction of cross-processing by CpG ODNs, and they further exclude the necessity of synergistic IFN-{alpha}ß-independent signaling.


Figure 8
View larger version (33K):
[in this window]
[in a new window]

  		Figure 8. CpG-induced IFN-{alpha}ß autocrine/paracrine signaling alone is sufficient to induce cross-processing. Flt3L-cultured, mixed DCs were prepared from wild-type C57BL/6 (H-2b), wild-type CBA/J (H-2k), TLR9–/– (H-2b), and IFN-{alpha}ßR–/– (H-2b) mice. These DC populations were incubated alone or as 1:1 mixtures (105 total cells/well) with varying concentrations of CpG-A-2336, CpG-B-1668, and CpG-C-2395 ODN or R837 (TLR7 agonist) for 24 h and assessed for cross-processing of latex-OVA as in Figure 1 . Error bars represent SD of triplicate wells and where not visible, are smaller than symbol size. Two-way ANOVA and Bonferroni t-test showed that ODNs induced statistically significant increases in responses where indicated (*, P<0.001). Data are representative of at least three independent experiments.


arrow
DISCUSSION
 
To address a discrepancy between the efficacy of different ODNs to induce IFN-{alpha}ß and IFN-{alpha}ß-dependent cross-presentation, we compared the ability of multiple CpG-A, CpG-B, and CpG-C ODNs to induce IFN-{alpha}ß-dependent MHC-I antigen cross-presentation by DCs. Experimental conditions included a wide range of ODN concentrations (and antigen concentrations) to fully assess the efficacy and potency of different ODN classes for induction of MHC-I antigen cross-processing and IFN-{alpha}ß (including ODN concentrations, e.g., 3 µg/ml, which produced maximal differential of IFN-{alpha}ß expression between CpG-B ODNs and CpG-A or CpG-C ODNs). Despite their lower efficacy for induction of IFN-{alpha}ß, CpG-B ODNs were slightly more potent than CpG-A and CpG-C ODNs for induction of IFN-{alpha}ß. As low levels of IFN-{alpha}ß induced by CpG-B ODNs sufficed for full induction of MHC-I cross-presentation, CpG-B ODNs were as potent and as efficacious for induction of cross-processing.

Induction of cross-processing by CpG ODNs is dependent on IFN-{alpha}ß [19 , 20 ]. CpG ODNs induce production of IFN-{alpha}ß by DCs, particularly pDCs, but different classes of CpG ODNs vary in their efficacy for induction of IFN-{alpha}ß [3 , 4 , 8 ]. CpG-A and CpG-C ODNs are much more efficacious than CpG-B ODNs for induction of IFN-{alpha}ß. Our prior experiments indicated that CpG-B ODN 1826 effectively induced IFN-{alpha}ß-dependent MHC-I antigen cross-processing. The higher efficacy of CpG-A and CpG-C ODNs for induction of IFN-{alpha}ß suggested the hypothesis that these ODN classes would be more efficacious than CpG-B ODNs for induction of MHC-I cross-presentation, but direct comparison of CpG-A, CpG-B, and CpG-C ODNs in the current studies revealed that the three ODN classes had similar efficacy and potency for induction of cross-presentation (CpG-B ODNs may even be slightly more potent than CpG-A or CpG-C in this regard, consistent with their slightly higher potency for induction of IFN-{alpha}ß). The key focus of these studies was to resolve the apparent discrepancy between efficacy for induction of IFN-{alpha}ß (CpG-A>CpG-C>>CpG-B) and efficacy for induction of IFN-{alpha}ß-dependent cross-processing (CpG-A{approx}CpG-C{approx}CpG-B).

Our data are consistent with prior observations [3 , 4 , 8 ] that CpG-A and CpG-C ODNs are more efficacious than CpG-B ODNs for induction of IFN-{alpha}ß expression. It should be noted that there are minor differences in potency and efficacy within each ODN class, which have been described elsewhere [4 , 5 ]; these differences are dependent on cell type and species as well as ODN sequence [2 3 4 5 ]. Nevertheless, the rank order for efficacy was CpG-A > CpG-C >> CpG-B for induction of IFN-{alpha} (Figs 4 5 6) and CpG-A >> CpG-C ≥ CpG-B for induction of IFN-ß (Fig. 4) . Conversely, we observed that CpG-B ODNs were slightly more potent than CpG-A or CpG-C ODNs, as they elicited low levels of IFN-{alpha} expression at lower ODN concentrations (Fig. 6) , consistent with prior observations [9 ]. Based on dose response studies and the concentrations required to achieve significant IFN-{alpha} secretion, CpG-B ODNs were approximately threefold more potent than CpG-A ODNs and three- to tenfold more potent than CpG-C ODNs. The differences in potency were relatively minor, however, and may have little impact on the relative efficacy of CpG ODNs for induction of MHC-I cross-processing. Despite the low efficacy of CpG-B ODNs, their high potency for induction of low levels of IFN-{alpha}ß indicated a potential mechanism to explain their efficacy for induction of MHC-I antigen cross-processing and presentation.

Murine pDCs and mDCs produced IFN-{alpha}ß in response to all three ODN classes, although pDCs produced much more IFN-{alpha}ß than mDCs (Fig. 4) . At 3 µg/ml CpG ODN, mDCs produced 1.6 ng/ml IFN-{alpha} with CpG-A-2336, 0.47 ng/ml IFN-{alpha} with CpG-C-2395, and 0.21 ng/ml IFN-{alpha} with CpG-B-1826, whereas pDCs produced 14 ng/ml IFN-{alpha} with CpG-A-2336, 11 ng/ml IFN-{alpha} with CpG-C-2395, and 1.2 ng/ml IFN-{alpha} with CpG-B-1826. As murine mDCs express TLR9 [6 ], they may respond directly to CpG ODNs to produce IFN-{alpha}ß (in humans, TLR9 is expressed by pDCs but not mDCs). Although pDCs contaminating the mDC preparations could produce IFN-{alpha}ß, similar findings with mDCs purified from Flt3L-cultured DCs by immunoaffinity magnetic bead separation and FACS, as well as GM-CSF-cultured mDCs preparations (which are largely pDC-deficient) [25 ], indicate that murine mDCs do express IFN-{alpha}ß after stimulation with CpG ODNs. This conclusion is consistent with other observations that stimulation of GM-CSF-cultured mDCs with CpG DNA induces IFN-{alpha}ß-dependent responses, including increased MHC-I cross-processing function and associated increases in expression of TAP, MHC-I, and costimulatory molecules [19 ]. As cross-processing is induced by extremely low concentrations of IFN-{alpha}ß (0.02–0.2 pg/ml, Fig. 7 ), the low level of IFN-{alpha}ß produced by mDCs in response to CpG ODNs is consistent with induction of cross-processing function.

Although mDCs and pDCs responded to CpG-induced IFN-{alpha}ß with enhanced MHC-I presentation of exogenous peptide, antigen cross-processing function was only induced in mDCs (Figs. 1 and 2) . Thus, pDCs are deficient in cross-processing, although they present an exogenous peptide in a CpG-dependent manner. CpG ODNs do not affect microsphere uptake, and phagocytic uptake of latex-OVA was similar in mDCs and pDCs (data not shown and ref. [20 ]), indicating that the pDC defect in cross-processing is not a result of defective antigen uptake. It is possible that this defect is explained by reduced transfer of exogenous antigen from vacuolar compartments to the cytosol or other limitations in MHC-I antigen-processing mechanisms, but further studies will be necessary to address this question.

We considered a signaling synergy hypothesis to resolve the apparent discrepancy between efficacy for induction of IFN-{alpha}ß (CpG-A>CpG-C>>CpG-B) and efficacy for induction of IFN-{alpha}ß-dependent cross-processing (CpG-A{approx}CpG-C{approx}CpG-B). This hypothesis proposed that induction of cross-processing requires two synergistic signals, one being signaling by IFN-{alpha}ß induced by CpG/TLR9/MyD88-dependent signaling and the other being IFN-{alpha}ß-independent, CpG/TLR9/MyD88-dependent signaling mechanisms. Our experiments, however, did not support this hypothesis, and synergy between CpG ODNs and rIFN-{alpha} or rIFN-ß was not detected (Fig. 7 and data not shown). Cross-presentation was induced by low concentrations of IFN-{alpha} or IFN-ß alone (Fig. 7 and data not shown). The lowest stimulatory concentration of CpG-B-1826 (0.03 µg/mL) induced a level of IFN-{alpha}ß below the limit of detection by ELISA (1–10 pg/ml) but apparently above the amount needed to induce cross-presentation (0.02–0.2 pg/ml). Together, these data indicate that IFN-{alpha}ß was sufficient as well as necessary for induction of cross-processing.

To confirm that autocrine or paracrine signaling by CpG-induced IFN-{alpha}ß is sufficient to explain induction of MHC-I cross-processing and presentation, we performed experiments with various combinations of TLR9–/– and IFN-{alpha}ßR–/– DCs. TLR9–/– Flt3L-cultured DCs did not respond directly to CpG ODNs, but when these cells were cocultured with wild-type Flt3L-cultured DCs, they responded indirectly to the addition of CpG ODNs with enhanced cross-processing function (Fig. 8C and 8D) . Moreover, the dose-response efficiency of CpG induction was similar in cocultured TLR9–/– DCs (Fig. 8D) and wild-type DCs (Fig. 8A) , consistent with a central role for an indirect mechanism. As predicted, cross-processing was not induced in IFN-{alpha}ßR–/– DCs, even when these cells were cocultured with wild-type DCs (Fig. 8F) . Similar indirect induction of cross-processing was observed with TLR9–/– GM-CSF-cultured mDCs and when supernatants from CpG-stimulated wild-type DCs were transferred to TLR9–/– DCs (data not shown). These data suggest that IFN-{alpha}ß autocrine/paracrine signaling is sufficient and necessary to induce cross-processing, independent of other CpG-induced signaling mechanisms.

An IFN-{alpha}ß-negative regulatory mechanism was suggested by biphasic dose-response curves (decreased responses above or below an optimal dose) for induction of IFN-{alpha} by CpG-B ODNs (Fig. 6) . A similar effect was observed with CpG-A and CpG-C ODNs but at much higher ODN concentrations. These data are consistent with other reports showing loss of IFN-{alpha} (and perhaps IFN-ß) production at high concentrations of CpG-B ODN, which still elicit TNF-{alpha}, IL-12p40, and IL-6 [4 , 9 ]. Furthermore, our data reveal a decline in the induction of cross-processing above (and sometimes at) 3 µg/ml CpG ODN (e.g., Figs. 2A and 8 ), although some experiments did not include ODN concentrations high enough to reveal this effect.

In summary, CpG-B ODNs are slightly more potent but less efficacious than CpG-A and CpG-C ODNs for induction of IFN-{alpha}ß. High sensitivity to IFN-{alpha}ß allows CpG-B ODNs to be equally (or slightly more) efficacious for induction of MHC-I cross-presentation. In vivo correlates of these experiments remain to be tested, and human systems must be assessed as a result of differential expression of TLR9 and species-to-species variability [1 , 3 , 14 ], but these findings suggest that CpG-B ODNs may be effective for inducing therapeutic responses, which require low levels of IFN-{alpha}ß and may avoid unnecessarily high induction of IFN-{alpha}ß.


arrow
ACKNOWLEDGEMENTS
 
This work was supported by National Institutes of Health (NIH) Grants AI35726, AI47255, and AI055793 to C. V. H. Technical support was provided by NIH-supported core facilities of the Case Comprehensive Cancer Center (CA43703) and Case Center for AIDS Research (AI36219). Coley Pharmaceutical generously provided CpG ODNs. The authors have no financial or commercial conflict of interest. Institutional Animal Care and Use Committee approval was obtained for all animal procedures. We appreciate the helpful advice from Arthur Krieg, Steven Emancipator, Yu-Chung Yang, and Michael Maguire.


arrow
FOOTNOTES
 
1 Current address: Vanderbilt Eye Institute, Vanderbilt University, Nashville, TN 37232, USA. Back

Received October 4, 2006; revised December 7, 2006; accepted December 8, 2006.


arrow
REFERENCES
 
    1
  1. Wagner, H. (2004) The immunobiology of the TLR9 subfamily Trends Immunol. 25,381-386[CrossRef][Medline]
    2. 2
  2. Klinman, D. M. (2004) Immunotherapeutic uses of CpG oligodeoxynucleotides Nat. Rev. Immunol. 4,249-258[CrossRef][Medline]
    3. 3
  3. Krieg, A. M. (2002) CpG motifs in bacterial DNA and their immune effects Annu. Rev. Immunol. 20,709-760[CrossRef][Medline]
    4. 4
  4. Vollmer, J., Weeratna, R., Payette, P., Jurk, M., Schetter, C., Laucht, M., Wader, T., Tluk, S., Liu, M., Davis, H. L., Krieg, A. M. (2004) Characterization of three CpG oligodeoxynucleotide classes with distinct immunostimulatory activities Eur. J. Immunol. 34,251-262[CrossRef][Medline]
    5. 5
  5. Hartmann, G., Battiany, J., Poeck, H., Wagner, M., Kerkmann, M., Lubenow, N., Rothenfusser, S., Endres, S. (2003) Rational design of new CpG oligonucleotides that combine B cell activation with high IFN-{alpha} induction in plasmacytoid dendritic cells Eur. J. Immunol. 33,1633-1641[CrossRef][Medline]
    6. 6
  6. Hemmi, H., Takeuchi, O., Kawai, T., Kaisho, T., Sato, S., Sanjo, H., Matsumoto, M., Hoshino, K., Wagner, H., Takeda, K., Akira, S. (2000) A Toll-like receptor recognizes bacterial DNA Nature 408,740-745[CrossRef][Medline]
    7. 7
  7. Takeda, K., Kaisho, T., Akira, S. (2003) Toll-like receptors Annu. Rev. Immunol. 21,335-376[CrossRef][Medline]
    8. 8
  8. Krug, A., Rothenfusser, S., Hornung, V., Jahrsdorfer, B., Blackwell, S., Ballas, Z. K., Endres, S., Krieg, A. M., Hartmann, G. (2001) Identification of CpG oligonucleotide sequences with high induction of IFN-{alpha}/ß in plasmacytoid dendritic cells Eur. J. Immunol. 31,2154-2163[CrossRef][Medline]
    9. 9
  9. Hemmi, H., Kaisho, T., Takeda, K., Akira, S. (2003) The roles of Toll-like receptor 9, MyD88, and DNA-dependent protein kinase catalytic subunit in the effects of two distinct CpG DNAs on dendritic cell subsets J. Immunol. 170,3059-3064[Abstract/Free Full Text]
    10. 10
  10. Harding, C. V. (1995) Phagocytic processing of antigens for presentation by MHC molecules Trends Cell Biol. 5,105-109[CrossRef][Medline]
    11. 11
  11. Jondal, M., Schirmbeck, R., Reimann, J. (1996) MHC class I-restricted CTL responses to exogenous antigens Immunity 5,295-302[CrossRef][Medline]
    12. 12
  12. Ackerman, A. L., Cresswell, P. (2004) Cellular mechanisms governing cross-presentation of exogenous antigens Nat. Immunol. 5,678-684[CrossRef][Medline]
    13. 13
  13. Bevan, M. J. (1976) Cross-priming for a secondary response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay J. Exp. Med. 143,1283-1288[Abstract/Free Full Text]
    14. 14
  14. Hornung, V., Rothenfusser, S., Britsch, S., Krug, A., Jahrsdorfer, B., Giese, T., Endres, S., Hartmann, G. (2002) Quantitative expression of Toll-like receptor 1–10 mRNA in cellular subsets of human peripheral blood mononuclear cells and sensitivity to CpG oligodeoxynucleotides J. Immunol. 168,4531-4537[Abstract/Free Full Text]
    15. 15
  15. Askew, D., Chu, R. S., Krieg, A. M., Harding, C. V. (2000) CpG DNA induces maturation of dendritic cells with distinct effects on nascent and recycling MHC-II antigen processing mechanisms J. Immunol. 165,6889-6895[Abstract/Free Full Text]
    16. 16
  16. Sparwasser, T., Koch, E. S., Vabulas, R. M., Heeg, K., Lipford, G. B., Ellwart, J. W., Wagner, H. (1998) Bacterial DNA and immunostimulatory CpG oligonucleotides trigger maturation and activation of murine dendritic cells Eur. J. Immunol. 28,2045-2054[CrossRef][Medline]
    17. 17
  17. Hartmann, G., Weiner, G. J., Krieg, A. M. (1999) CpG DNA: a potent signal for growth, activation, and maturation of human dendritic cells Proc. Natl. Acad. Sci. USA 96,9305-9310[Abstract/Free Full Text]
    18. 18
  18. Jakob, T., Walker, P. S., Krieg, A. M., Udey, M. C., Vogel, J. C. (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. 161,3042-3049[Abstract/Free Full Text]
    19. 19
  19. Cho, H. J., Hayashi, T., Datta, S. K., Takabayashi, K., Van Uden, J. H., Horner, A., Corr, M., Raz, E. (2002) IFN-{alpha} ß promote priming of antigen-specific CD8+ and CD4+ T lymphocytes by immunostimulatory DNA-based vaccines J. Immunol. 168,4907-4913[Abstract/Free Full Text]
    20. 20
  20. Kuchtey, J., Chefalo, P. J., Gray, R. C., Ramachandra, L., Harding, C. V. (2005) Enhancement of dendritic cell antigen cross-presentation by CpG DNA involves type I IFN and stabilization of class I MHC mRNA J. Immunol. 175,2244-2251[Abstract/Free Full Text]
    21. 21
  21. Van Uden, J. H., Tran, C. H., Carson, D. A., Raz, E. (2001) Type I interferon is required to mount an adaptive response to immunostimulatory DNA Eur. J. Immunol. 31,3281-3290[CrossRef][Medline]
    22. 22
  22. Vicari, A. P., Chiodoni, C., Vaure, C., Ait-Yahia, S., Dercamp, C., Matsos, F., Reynard, O., Taverne, C., Merle, P., Colombo, M. P., O’Garra, A., Trinchieri, G., Caux, C. (2002) Reversal of tumor-induced dendritic cell paralysis by CpG immunostimulatory oligonucleotide and anti-interleukin 10 receptor antibody J. Exp. Med. 196,541-549[Abstract/Free Full Text]
    23. 23
  23. Jurk, M., Schulte, B., Kritzler, A., Noll, B., Uhlmann, E., Wader, T., Schetter, C., Krieg, A. M., Vollmer, J. (2004) C-Class CpG ODN: sequence requirements and characterization of immunostimulatory activities on mRNA level Immunobiology 209,141-154[CrossRef][Medline]
    24. 24
  24. Tobian, A. A., Potter, N. S., Ramachandra, L., Pai, R. K., Convery, M., Boom, W. H., Harding, C. V. (2003) Alternate class I MHC antigen processing is inhibited by Toll-like receptor signaling pathogen-associated molecular patterns: Mycobacterium tuberculosis 19-kDa lipoprotein, CpG DNA, and lipopolysaccharide J. Immunol. 171,1413-1422[Abstract/Free Full Text]
    25. 25
  25. Brawand, P., Fitzpatrick, D. R., Greenfield, B. W., Brasel, K., Maliszewski, C. R., De Smedt, T. (2002) Murine plasmacytoid pre-dendritic cells generated from Flt3 ligand-supplemented bone marrow cultures are immature APCs J. Immunol. 169,6711-6719[Abstract/Free Full Text]
    26. 26
  26. Pfeifer, J. D., Wick, M. J., Roberts, R. L., Findlay, K. F., Normark, S. J., Harding, C. V. (1993) Phagocytic processing of bacterial antigens for class I MHC presentation to T cells Nature 361,359-362[CrossRef][Medline]
    27. 27
  27. Pai, R. K., Pennini, M. E., Tobian, A. A., Canaday, D. H., Boom, W. H., Harding, C. V. (2004) Prolonged Toll-like receptor signaling by Mycobacterium tuberculosis and its 19-kilodalton lipoprotein inhibits {gamma} interferon-induced regulation of selected genes in macrophages Infect. Immun. 72,6603-6614[Abstract/Free Full Text]
    28. 28
  28. Muller, U., Steinhoff, U., Reis, L. F., Hemmi, S., Pavlovic, J., Zinkernagel, R. M., Aguet, M. (1994) Functional role of type I and type II interferons in antiviral defense Science 264,1918-1921[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 BloodHome page
J. Mouries, G. Moron, G. Schlecht, N. Escriou, G. Dadaglio, and C. Leclerc
Plasmacytoid dendritic cells efficiently cross-prime naive T cells in vivo after TLR activation
Blood, November 1, 2008; 112(9): 3713 - 3722.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
F. Meyer-Wentrup, D. Benitez-Ribas, P. J. Tacken, C. J. A. Punt, C. G. Figdor, I. J. M. de Vries, and G. J. Adema
Targeting DCIR on human plasmacytoid dendritic cells results in antigen presentation and inhibits IFN-{alpha} production
Blood, April 15, 2008; 111(8): 4245 - 4253.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1006606v1
81/4/1075    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gray, R. C.
Right arrow Articles by Harding, C. V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gray, R. C.
Right arrow Articles by Harding, C. V.