|
|
||||||||

,

* Mucosal Immunobiology Section, Laboratory of Molecular Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892;
Department of Molecular Microbiology and Immunology, Brown University, Providence, RI 02912;
Wyeth Nutrition, Collegeville, PA 19426; and
Schering-Plough Biopharma, Palo Alto, CA 94304
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Both human and murine pDCs express TLRs that recognize viral RNA (via TLR7), and CpG-rich unmethylated DNA from bacteria and DNA viruses (via TLR9). Signaling by TLRs expressed on pDCs drives the production of type I IFN that can directly activate antiviral responses and augment both innate and adaptive immunity to viral as well as nonviral infections (2).
We characterized the phenotype of pDCs in PPs and found that following stimulation with CpG-enriched oligodeoxynucleotides (CpG ODN) or influenza virus, PP pDCs were incapable of producing significant amounts of type I IFN while they were fully capable of producing IL-12. Furthermore, three factors present in the mucosal microenvironment, IL-10, PGE2, and TGFβ, inhibited type I IFN production by spleen pDCs in response to CpG ODN, suggesting that the mucosal microenvironment may condition pDCs for poor type I IFN production. Finally, in studies using spleen pDCs from IFN
R1–/– mice, we determined that TGFβ acts to inhibit initial TLR9-induced type I IFN mRNA expression while IL-10 acts at a later stage, most likely by blocking autocrine signaling via the type I IFN receptor. Together, these studies suggest a role for the mucosal microenvironment in preventing type I IFN production from pDCs.
| Materials and Methods |
|---|
|
|
|---|
Female BALB/c and C57BL/6 mice obtained from the National Cancer Institute (Frederick, MD) and in-house bred IFN
R1–/– mice on a C57BL/6 background were housed at the National Institutes of Health (Bethesda, MD), maintained in accordance with institutional guidelines for animal welfare, and used at 6- to 12-wk of age. In some experiments, 4 x 105 Fms-like tyrosine kinase 3 ligand (FLT3L)-expressing B16 cells (3) were injected s.c. 11 days before dendritic cell isolation.
Antibodies
Abs used include: 120G8 (4), anti-murine pDC Ag (mPDCA) (clone JF05-1C2.4.1; Miltenyi Biotec), anti-CD11c (clone HL3 or N418), anti-B220 (clone RA3-6B2), anti-Gr-1 (clone RB6-8C5), anti-CD19 (clone 1D3), anti-CD40 (clone 3/23), anti-CD80 (clone 1G10), anti-CD86 (clone GL1), anti-MHC class II (clone AMS-32.1), anti-CD4 (clone GK1.5), and anti-CD8 (clone 53-6.7) (BD Biosciences).
Confocal microscopy
Primary Abs were detected in frozen sections using Alexa 488 and Alexa 594 tyramide signal amplification (TSA) kits (Molecular Probes). Nuclei were counterstained with Hoechst (Sigma-Aldrich). Images were collected on a Leica TCS-SP2 confocal microscope (Leica Microsystems) at x40 magnification.
Cell isolation
Spleen and PP pDCs were obtained from single cell suspensions as previously described (5) except that collagenase was replaced with 420 µg/ml Liberase CI (Roche Applied Science). pDCs were enriched by negative selection (first step of pDC isolation kit; Miltenyi Biotec) and analyzed using a FACSCalibur cytometer and FlowJo software (Tree Star) or sorted using a FACS Vantage SE with DIVA option or FACS Aria cytometer (BD Biosciences).
Cell culture
Enriched pDCs or FACS-purified pDCs were cultured at 106 cells/ml in complete RPMI 1640 containing 10% FCS (BioSource International) and antibiotics at 37°C and 5% CO2. Cells were stimulated with UV-irradiated influenza virus (strain PR8; 40 hemagglutinating units/ml; gift of J. Yewdell, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD) and 1 µM CpG ODN 1826 (TCCATGACGTTCCTGACGTT) or 2216 (GGGGGACGATCGTCGGGGGG), both synthesized on a phosphorothioate backbone, and both of which have been demonstrated to induce type I IFN production in murine pDC (6, 7). Where indicated, cells were preincubated for 1 h with TGFβ 1, IL-10 (Peprotech), PGE2 (Sigma-Aldrich), anti-IL10R or control Ab (clones 1B1.3a and GL113, kindly provided by DNAX Research). Culture supernatants were collected 20 h after the addition of CpG ODN and assayed by ELISA for IFN
or β (PBL Biomedical Laboratories), IL-12 p40 or p70 (R&D Systems), or bioactive type I IFNs (8).
Quantitative RT-PCR
Spleen mPDCA+ (Miltenyi Biotec) pDCs isolated from B16-FLT3L-treated mice were pretreated as described above and stimulated with CpG ODN for a subsequent 4 h. RNA was isolated (Qiagen), and cDNA was synthesized (SuperScript III; Invitrogen). Real-time RT-PCR was performed on an Applied Biosystems 7900 HT instrument. FAM-labeled IL-12 p40 probe/primers were obtained from Applied Biosystems (Assays-on-Demand). SYBR Green (Applied Biosystems) was used to detect IFN
(all genes) and IFNβ (9). GAPDH was the endogenous control. Fold difference in gene expression was determined with the formula 2
Ct, where Ct is threshold cycle.
Statistical analysis
Graphing and statistical analysis was conducted using Prism 4 software (GraphPad Software). Data were analyzed with a two-tailed, unpaired Students t test.
| Results and Discussion |
|---|
|
|
|---|
Luminal Ags and commensal bacteria can enter the organized mucosal lymphoid tissues of the Peyers patches and isolated colonic follicles via transport by M cells in the follicle-associated epithelium or via direct sampling by cDCs with intraepithelial extensions (1). Therefore, immune responses must be controlled at these inductive sites (10). We explored the role of pDCs in the immune regulation in PPs. We first identified these cells in tissue sections by immunofluorescence staining using the pDC-reactive Ab mPDCA (Fig. 1A) and the 120G8 Ab (Ref. 4 and data not shown). pDCs were found in the subepithelial dome and interfollicular regions of PPs, as well as in the lamina propria (LP) of PP-adjoining villi, but were not found in B cell follicles, consistent with prior studies (4, 11). Therefore, PP pDCs are present at sites of initial Ag entry and within regions where they could influence T cell responses.
|
pDCs from PPs were further characterized by the flow cytometry of pDC-enriched cell populations (see Materials and Methods; Fig. 1, B and C). As with pDCs from other tissues, PP pDCs were identified as CD19–, CD11cint, B220+, and Gr1+ cells that were uniformly positive for the 120G8 Ab (Fig. 1B) and mPDCA (data not shown). Prior studies identified CD8+CD11b–, CD8– CD11b+, and CD8–CD11b– populations of cDCs in PPs, all of which express high levels of CD11c and do not express B220 (1). Therefore, CD11cint and B220+ PP pDCs are not included in these previously identified populations. Costimulatory molecule expression was low, whereas the expression of MHC class II was intermediate on pDCs from both spleen and PP (Fig. 1C). PP pDCs contained a greater percentage of CD8+ cells than their splenic counterparts (40 vs 10%). However, both CD8– and CD8+ pDC populations from the spleen and PPs expressed similar levels of CD40, CD80, CD86, and MHC class II Ags, suggesting that CD8+ cells are not a more mature population of CD8– cells (7). Finally, purified pDCs from the spleen and PP had similar morphology on Wright-Giemsa staining (Fig. 1D).
PP pDCs are deficient in their ability to produce type I IFNs
PP pDCs purified to
98% by flow cytometry produced IL-12 p40 and IL-12 p70 at similar levels to those of spleen pDCs when stimulated with influenza virus strain PR8 or CpG ODN (Fig. 2, A and B). In contrast, the production of IFN
and IFNβ by PP pDCs was only a small fraction of that produced by spleen pDCs (Fig. 2, C and D). These results were confirmed with a bioassay for type I IFN (Fig. 2E). Consistent with these findings, IFN
production was undetectable from whole PP cells depleted of B cells, T cells, NK cells, and macrophages stimulated with CpG ODN whereas it was readily detectable from the depleted spleen preparations (Fig. 2F), indicating it was unlikely that a population of IFN-producing cells that did not express classic pDC markers was present in the PP.
|
than cells from untreated mice. However, the levels from PP pDCs were still at least 5- to 10-fold less than those from spleen pDCs. Therefore, despite the higher levels of type I IFN produced, the differential production between PPs and spleen pDCs was consistent with results from unmanipulated mice.
TGFβ, PGE2, and IL-10 can inhibit IFN
production by spleen pDC
To address the mechanism by which PP pDCs cannot produce type-I IFN in response to CpG ODN, we initially determined whether differences in the expression of TLR9 or IRF7, a major transcriptional regulator of type I IFN, could account for differences in type I IFN production by PP and spleen pDCs. We found that mRNA for TLR9 or IRF7 were expressed and <2-fold different between purified spleen and PP pDCs (data not shown). In addition, surface level expression of TLR9, as indicated by flow cytometry, was similar between the two cell types (data not shown).
We next evaluated the ability of the regulatory factors most commonly associated with mucosal tissues to affect IFN
production from purified spleen pDCs (Fig. 3). Pretreatment of spleen pDCs with TGFβ, PGE2, or IL-10 resulted in a dose-dependent reduction in IFN
production (Fig. 3A). In addition, IL-10, PGE2, and, to a lesser extent, TGFβ inhibited the production of IL-12 p40 (Fig. 3B). Finally, an additive effect on type I IFN production was seen with PGE2 and IL-10 or TGFβ. Baseline survival of pDCs was low in overnight cultures with only 25% of cells surviving after 24 h. IL-10 (10 ng/ml) and TGFβ (10 ng/ml) resulted in an enhancement of cell death by 14–27% at 8 h after culture, a time when the majority of type I IFN is present in the culture supernatants; however, at this time point type I IFN production is blocked by >80% in these cultures (data not shown). Therefore, the increase in cell death could not account for the observed suppression of type I IFN.
|
production by IL-10 is consistent with studies of human pDCs (13). Furthermore, suppression of IL-12 production by IL-10 and PGE2 is consistent with effects on cDCs (14). These data suggest that factors present in the mucosal microenvironment have direct effects on the ability of pDC to produce type I IFN as well as IL-12. Of the factors tested, treatment with TGFβ most closely recapitulated the phenotype observed in PP pDCs in that type I IFN was suppressed while IL-12 production was maintained, at least to some degree. Furthermore, TGFβ is constitutively expressed in the intestine (15) and plays a key role in the PP for inducing B cell switching to IgA (16), supporting the possibility that TGFβ is involved in vivo in conditioning PP pDCs for low type I IFN production.
Additional roles for IL-10 and PGE2 in conditioning mucosal pDCs are suggested because cDCs (1) and T cells (17) from mucosal tissues produce IL-10, and IL-10–/– mice develop colitis (18). Furthermore, PGE2 is expressed by stromal cells in the intestinal LP and at low levels in PPs (19); and PGE2-deficient Cox-2 –/– mice are less susceptible to the induction of oral tolerance (20). Whether pDCs in the LP of the intestine are similar to PP pDCs in their inability to produce significant amounts of type I IFN is not yet known, but TGFβ expressed by intestinal epithelial cells, PGE2 by stromal cells, and IL-10 by activated LP dendritic cells and macrophages suggests that this may be the case. It has been reported that mesenteric lymph node pDCs can produce type I IFN after exposure in vivo to R-848 given orally and drive cDC activation (21). In addition, pDCs do not appear to readily migrate from the intestine to the mesenteric lymph node (22), indicating that direct exposure to the intestinal microenvironment may be important for the phenotype of PP pDCs.
We are currently exploring the direct role of IL-10, TGFβ, and PGE2 in conditioning pDCs in vivo. We found that pDCs from IL-10–/– mice still had a defect in type I IFN production so that IL-10 does not appear to be required for the suppression of PP pDC type I IFN production (data not shown). TGFβ-1–/– as well as wild-type and IL-10–/– mice treated with the cyclooxygenase inhibitor piroxicam develop intestinal inflammation. Therefore, to date, we have been unable to directly determine the in vivo role of TGFβ and/or PGE2, with or without IL-10, on PP pDC type I IFN production. However, given their additive effects on pDC function in vitro (Fig. 3C), it is likely that a combined effect of these factors may be important to control pDC responses in PPs.
TGFβ but not IL-10 inhibits the induction of mRNA for type I IFN production by pDCs from IFN
R1–/– mice
We next addressed whether two factors likely affecting type I IFN in the PP, TGFβ and IL-10, act to inhibit type I IFN production by spleen pDCs by preventing primary signaling via TLR9 or by blocking autocrine stimulation of the type I IFN receptor. To do this, we tested the ability of TGFβ and IL-10 to suppress the early induction of type I IFN from the pDCs of wild-type mice without functional type I IFN receptors (IFN
R1–/– mice; Fig. 4). Although both TGFβ and IL-10 inhibited the transcription of mRNA for IFNβ by CpG ODN stimulation in pDCs from wild-type mice, only TGFβ was able to suppress the expression of IFNβ mRNA in pDCs from IFN
R1–/– mice. In contrast, TGFβ was not able to suppress the expression of mRNA for IL-12 p40 in pDCs from either mouse strain, whereas IL-10 suppressed IL-12 p40 from both. These studies imply that in the regulation of type I IFN production, TGFβ acts primarily on proximal signaling via TLR9 while IL-10 acts on type I IFN production downstream of initial TLR signaling and type I IFN production, most likely by blocking autocrine type I IFN signaling.
|
by NK cells and CD8+ T cells. Alternatively, type I IFN can act independently of IL-12 to activate cDC maturation and cross-presentation and to enhance CD8+ T cell responses. An enhanced Th1 response or possibly an aberrant CD8+ T cell response (23, 24) to commensal bacteria driven by type I IFN could contribute to intestinal inflammation. Therefore, one consequence of the lack of type I IFN production by pDC may be to shift the immunological balance away from potentially damaging immune responses against commensal organisms. A second possible consequence of our findings is that pathogens dependent on the production of type I IFN by pDCs for their clearance may gain a selective advantage by replicating in cells of the mucosa. The most intriguing possibility in this regard is HIV, which readily infects mucosal tissues (25), activates pDCs to produce type I IFN (26), and progresses coincidentally with a decline in pDC numbers (27). However, the actual role of type 1 IFN in HIV infection or progression is not yet clear. In contrast, during certain viral infections such as those caused by type I reovirus (28) or Herpes simplex virus-2 (29) or during intestinal inflammation (30), pDCs may be newly recruited from the blood, and thus may be fully capable of producing type I IFN. Therefore, local microenvironmental conditions may dictate whether type I IFN-competent pDCs are present within mucosal tissues.
Together, our studies support the hypothesis that dendritic cells are conditioned by their tissue microenvironment to drive tissue-specific immune responses. Further studies are needed to elucidate the in vivo mechanism by which PP pDC type I IFN production is inhibited and to determine the specific signaling pathways involved in the TGFβ-, PGE2-, and IL-10-mediated suppression of type I IFN production by pDCs.
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This research was supported by the Intramaral Research Program of the National Institute of Allergy and Infectious Diseases of the National Institutes of Health (NIH). In addition, C.B. and J.L. were supported by NIH Grant number RO1 AI 55677. ![]()
2 Address correspondence and reprint requests to Dr. Brian Kelsall, 10/11N111, 10 Center Drive, Bethesda, MD 20892. E-mail address: kelsall{at}nih.gov ![]()
3 Abbreviations used in this paper: PP, Peyers patch; cDC, conventional dendritic cell; FLT3L, FMS-like tyrosine kinase 3 ligand; int, intermediate; LP, lamina propria; mPDCA, murine pDC Ag; ODN, oligodeoxynucleotide; pDC, plasmacytoid dendritic cell. ![]()
Received for publication November 6, 2006. Accepted for publication July 5, 2007.
| References |
|---|
|
|
|---|
. Blood 103: 4201-4206.
β, and TNF effects on antiviral state and NK cell responses during murine cytomegalovirus infection. J. Immunol. 156: 4746-4756. [Abstract]
producing (plasmacytoid) dendritic cells. Hum. Immunol. 65: 104-113. [Medline]
+ DC, with a plasmacytoid phenotype, induce differentiation and support function of T cells with regulatory properties. Immunology 108: 481-492. [Medline]
and β in colonic mucosa in inflammatory bowel disease. Gastroenterology 110: 975-984. [Medline]
and type 1 IFNs after feeding a TLR7/8 ligand. J. Immunol. 176: 5205-5212.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |