Nonplasmacytoid, High IFN-α–Producing, Bone Marrow Dendritic Cells

Meredith O’Keeffe, Ben Fancke, Mark Suter, Georg Ramm, Joan Clark, Li Wu and Hubertus Hochrein
J Immunol April 15, 2012, 188 (8) 3774-3783; DOI: https://doi.org/10.4049/jimmunol.1101365

Abstract

Plasmacytoid dendritic cells (pDC) are the producers of type I IFNs in response to TLR9 ligands. However, we have found that when bone marrow is depleted of pDC, the IFN-α produced in response to TLR9 ligands is not fully removed. We assign the source of this non-pDC IFN-α as a newly described DC type. It displays the high IFN-α producing activity of pDC but to a more limited range of viruses. Unlike pDC, the novel DC display high T cell stimulation capacity. Moreover, unlike mouse pDC, they are matured with GM-CSF and are less prone to apoptosis upon activation stimuli, including viruses. We propose that these DC constitute a novel bone marrow inflammatory DC type, ideally geared to linking innate and adaptive immune responses in bone marrow via their potent IFN-α production and high T cell stimulatory capacity.

Dendritic cells (DC) can be divided into two major categories: conventional DC (cDC) and plasmacytoid DC (pDC). The cDC can be further divided into several subsets based on tissue location and surface phenotype. In mice, the surface markers CD4 and CD8α are useful markers to distinguish functionally different cDC subsets. The unifying function of all cDC is the ability to induce naive T cells into cell cycle. The exceptional ability of cDC to process and present Ag in the context of MHC class I and MHC class II endows them with the title “professional APCs.” The CD8α+ cDC have the additional feature of cross-presentation, being the ability to present exogenous Ag in the context of MHC class I (1). The differential functions of cDC also extend to cytokine and chemokine production. High IL-12p70 production is a hallmark of CD8α+ cDC (2–4), and high levels of chemokines including RANTES, MIP-1α, and MIP-1β are produced by CD8α− cDC (5).

Alternatively, the pDC, generally considered as part of the DC “family,” lack typical cDC characteristics, including surface phenotype and morphology, and they also normally lack the ability to stimulate naive T cells (6). When given specific PRR stimuli, they can induce some T cell division, more than B cells or macrophages but typically on the order of 10-fold or less that of the cDC (7). Unlike cDC, the pDC continually present Ags on MHC class II molecules once they are activated, and as a result they can continue to present new viral Ags during the course of infection (8). The importance of this function of pDC during an ongoing infection has not yet been elucidated. Instead the pDC, also referred to as natural IFN-producing cells, are renowned for their production of type I IFNs (IFN-I) in response to viral or bacterial stimuli and mimics thereof (9, 10). The categorization of the pDC as a member of the DC family rests on morphological and phenotypical features that they display upon activation. Namely, the pDC upregulate costimulation markers and MHC molecules to levels resembling the cDC and they rapidly acquire the typical stellate morphology of cDC. Based on these features, it was initially proposed that this IFN-I–producing DC subset would be the ultimate antiviral cell, combining within the same cell the innate IFN response and potent CTL stimulation. To date, these high hopes have not been realized. The IFN-I response of pDC is remarkable but their concomitant ability to induce CTL is in most cases relatively poor.

The pDC of both mice and humans recognize DNA via TLR9. As a consequence of endoplasmic reticulum to lysosome internal trafficking of TLR9, and differential expression of molecules that are involved in the TLR9 signaling complex, such as high constitutive expression of IRF7, the pDC, unlike any other cell previously described, have the ability to produce extremely high levels of IFN-I upon TLR9 ligation (9). Synthetic CpG-containing oligonucleotides (ODN), without addition of transfection reagent, are sufficient for the triggering of IFN-I from pDC and, in fact, the pDC are the only cell type known to produce IFN-I to CpG-ODN alone. This statement holds true in mice in most lymphoid organs. However, as we have previously shown (11), when bone marrow (BM) cells are depleted of pDC there remains IFN-α production in response to CpG-ODN. These data suggested that cells other than CD45R+CD45RA+ pDC were capable of TLR9-mediated IFN-α production. This finding contradicts the current dogma that pDC are the only cell type that produces IFN-I in response to TLR9-mediated ligation. The biological relevance of this observation extends beyond responses to CpG-ODN; many viruses have now been shown to activate cells via TLR9-mediated recognition (12). The BM is a site frequently infected by viruses, and yet our knowledge on the cellular responses to viruses or other pathogens in the BM is extremely limited. With the advent of stem cell transplantation and a desire to understand the cellular entities potentially involved in transplantation rejection, it is of the utmost importance to clarify the cell types within BM.

We have identified a novel BM cell that produces high levels of IFN-α in response to TLR9 ligands and certain viruses and that is highly efficient at stimulating naive T cells, particularly when presenting viral-encoded Ag. In Flt3 ligand (FL)-generated DC cultures, they undergo one or two divisions and resemble the surface phenotype of mouse CD8− cDC. However, together with their maturation with GM-CSF, upregulation of CD8α, IFN-α production to TLR7 and TLR9 ligands, TLR4 responsiveness, presentation capacity, and phenotype in viral responses, miDC stand apart as a unique cell type.

Materials and Methods

Mice

Female or male C57BL/6J mice were purchased from Harlan Winkelmann (Borchen, Germany) and used at 6–10 wk age or obtained from the Walter and Eliza Hall Institute breeding facility. C57BL/6 Pep3b mice (Ly5.1) were also obtained from the Walter and Eliza Hall Institute. T cells used in allostimulatory MLRs were from BALB/c mice and also purchased from Harlan Winkelmann. C57BL/6-Tg(TcraTcrb)1100Mjb/j (OTI) mice were originally purchased from The Jackson Laboratory and bred in-house. FLKO mice were described by McKenna et al. (13) and bred at the Institut für Labortierkunde (University of Zurich). Mice expressing a human Bcl-2 transgene expressed under the vav promoter have been described previously (14) and were obtained with permission (S. Cory, Walter and Eliza Hall Institute) from G. Hacker (Institute for Medical Microbiology, Immunology, and Hygiene, Technical University of Munich).

Animal experimentation was approved by the government of Upper Bavaria (Regierung von Oberbayern) or by the Walter and Eliza Hall Institute or Alfred Medical Research and Education Precinct Animal Ethics Committees.

Abs and reagents

Recombinant flag-tagged murine FL was expressed in CHO cells and purified in-house as previously described (15). Recombinant murine GM-CSF and M-CSF were from Tebu-Bio (Frankfurt, Germany). ODN containing CpG motifs (CpG-2216 and CpG-1668) were synthesized by TIB Molbiol (Berlin, Germany) according to published sequences (16). Imiquimod (R837) and palmitoyl-3-cysteine-serine-lysine-4 (Pam3Cys)were purchased from InvivoGen (San Diego, CA). Polyinosinic:polycytidylic acid and LPS were purchased from Sigma-Aldrich (Taufkirchen, Germany).

The modified vaccinia Ankara virus (MVA) used for this study was MVA-BN, developed by Bavarian Nordic and deposited at the European Collection of Cell Cultures (V00083008), and MVA expressing OVA (MVA-OVA). MVA was propagated and titered on primary chicken embryo fibroblasts that were prepared from 11-d-old embryonated pathogen-free hen eggs (Charles River, Wilmington, MA) and cultured in RPMI 1640 medium supplemented with 10% FCS. Ectromelia virus (ECTV) strain Moscow was obtained from the American Type Culture Collection as VR-1372 and propagated and titered on Vero C1008 cells (European Collection of Cell Cultures 85020206). Shope fibroma virus (SFV) was obtained from the American Type Culture Collection (VR-364) and propagated and titered on the rabbit cornea cell line SIRC obtained from the American Type Culture Collection (CCL-60). All cell lines were maintained in DMEM (Invitrogen, Karlsruhe, Germany) supplemented with 10% FCS without antibiotics.

Abs were obtained from Becton Dickinson (Heidelberg, Germany), with the following exceptions: purified and FITC-conjugated anti-CD11c (rat clone 223H7) and anti-Ly49Q (Biozol Diagnostica Vertrieb, Eching, Germany), anti–mPDCA-1 (Miltenyi Biotec, Bergisch Gladbach, Germany), and anti-F4/80 (NatuTec, Frankfurt, Germany).

Isolation of cells from bone marrow and DC purification

Femur and tibiae from mice were removed and cells were flushed from the bones with a 20-ml syringe filled with RPMI 1640 (Life Technologies/BRL; adjusted to mouse osmolarity) containing 2% FCS and fitted with a 21-gauge needle. Cells were resuspended by repeatedly flushing through the syringe. FLDC were prepared from the BM cells as described (11).

For preparation of light density cells, the cells were pelleted by centrifugation and resuspended in Nycodenz 1.077A (Progen Biotechnik, Heidelberg, Germany; adjusted to mouse osmolarity [308 mOsm]). Density separation, incubation with Ab depletion mixture, and magnetic bead depletion were essentially as previously described (6) except that anti-CD49b mAb was also added to the depletion mixture and anti-Thy1.1 was omitted. Hybridomas, the supernatants of which were used in the depletion mixture, were provided by Prof. Ken Shortman (Walter and Eliza Hall Institute). For sorting of BM miDC and pDC, after depletion steps, the cells were routinely stained with CD11c-PE-Cy7, CD45RA-PE, CD11b (allophycocyanin or Pacific Blue), and CD24-FITC. Cells were sorted as pDC (CD11cintCD45RAhiCD11b−) or miDC (CD11cintCD45RA−CD11b−HSAint).

Splenic DC were isolated as previously described (6) using the Nycodenz and depletion mixture described above.

DC stimulation and cytokine quantitation

Stimulation of DC was carried out at 0.25–0.5 × 106 cells/ml in round-bottom plates or V-bottom plates for volumes of 50 μl or less. Cells were stimulated for 18–48 h in complete medium (RPMI 1640 medium supplemented with 10% FCS, 50 μM 2-ME, 100 IU/ml penicillin/streptomycin) with or without added stimulus. The stimuli used were as follows: 10 ng/ml GM-CSF, 1 μg/ml Pam3Cys, 100 μg/ml polyinosinic:polycytidylic acid, 1 μg/ml LPS, 1 μg/ml R837, and 0.5 μM CpG-2216 or 0.5 μM CpG-1668.

Culture supernatants were assayed for IFN-α by two-site ELISA as previously described (6). Other cytokines (IL-12p70, IL-6, TNF-α, MCP-1, and IFN-γ) were measured using a Cytometric Bead Array mouse inflammation kit (Becton Dickinson).

CFSE labeling

T cells or miDC were labeled with a final concentration of 2 μM CFSE using the CellTrace CFSE cell proliferation kit (Molecular Probes/Invitrogen, Karlsruhe, Germany) according to the manufacturer’s instructions.

Allogeneic MLR

DC subsets were purified from BM and spleen and sorted to >95% purity using a FACSAria. T cells from BALB/c mice were purified from s.c. and mesenteric lymph nodes. Triplicates of 50,000 CFSE-labeled T cells were incubated with DC subsets in media or with the addition of stimuli including GM-CSF, LPS, or CpG-2216. “No DC” controls in triplicate with each stimulus were also included. After 4 d at 37°C, the replicate samples were analyzed by FACS to ascertain the degree of T cell division (loss of CFSE fluorescence). An aliquot of beads used for quantitation was also added to each well so that absolute numbers of divided T cells could be enumerated.

OTI T cell stimulation

A cross-presentation assay was similarly carried out to the MLR above, whereby T cells from OTI transgenic mice were used instead of BALB/c T cells, and DC were first incubated with OVA for 1 h at 37°C, washed, and then added to the T cells.

Presentation of OVA expressed by MVA virus was analyzed by first incubating the DC with MVA-OVA for 1 h at 37°C, followed by washing in complete medium. CFSE-labeled T cell replicates were then incubated with 103, 3 × 103, or 104 infected DC. As controls, DC were also included that were incubated with MVA (not expressing OVA) or untreated. T cell controls included were T cells alone and T cells that were incubated with MVA-OVA as for the DC subpopulations, without any addition of DC. After 4 d incubation, cells from pooled replicate samples were analyzed for CFSE expression and quantitation beads were included in each sample to enumerate cell numbers.

Results

BM non-pDC IFN-α–producing cells are CD11cloCD11b−/lo cells that do not express B, T, or NK cell-specific markers

We have previously shown that conditions resulting in pDC removal in BM do not remove all IFN-α production in response to CpG-ODN stimulation (11). Liu and colleagues (17) have reported that human early pre-pDC that do not yet have the complete pDC phenotype also have the ability to produce high levels of IFN-I. In our hands, no other organ apart from BM retained any IFN-α–producing capacity to CpG-ODN when pDC were first removed, and we considered it a likely possibility that pDC precursors were the cells that remained after CD45R/CD45RA depletion of mouse BM cells and we set out to characterize them.

Light density separation of BM cells revealed that all of the IFN-α activity in response to CpG-ODN segregated with the light density fraction of BM cells (Fig. 1A).

FIGURE 1.
FIGURE 1.

Non-pDC, CD11c+CD45RA− BM cells also produce high levels of IFN-α in response to CpG-2216. (A) BM cells were separated over a 10-ml 1.077 g/cm3 Nycodenz gradient and the “light” cells (within the top 6 ml) and “heavy” cells (bottom 4 ml) were collected. The cells were stimulated with the A-type ODN CpG-2216 (filled bars) or the B-type ODN CpG-1668 (open bars) and the supernatants were tested for IFN-α production. Data shown are one experiment representative of two separate experiments; error bars represent the SD between duplicate samples. (B) Light density cells were stained with CD45RA and CD11c. (C) The surface phenotype of pDC (filled light gray histograms) is compared with the boxed CD11c+CD45RA− cells in (A) (black histograms). A junk channel containing Abs to CD3, CD19, and CD49b was used. Unstained cells are depicted by dotted light gray line. (D) Cells in CD11c+CD45RA− box in (B) were stained with CD24 and CD11b. All of the IFN-α production was located in the CD24intCD11b− cells (shown boxed). Their IFN-α production in response to CpG-2216 is shown (filled bar) and compared with BM pDC (open bar). (E) The IFN-α–producing cells were sorted to high purity after first depleting light density cells with a mixture of lymphocyte and granulocyte Abs and staining the cells with CD45RA, CD11c, CD24, and CD11b. (F) The expression of typical pDC markers and other cell surface markers (G) on CD11c+CD45RA−CD24intCD11b− cells boxed in (E) is shown (filled black histograms). Light gray histogram indicates unstained background control. Filled light gray histogram indicates staining on BM pDC. (H) The morphology of the BM pDC and the IFN-α–producing CD11c+CD45RA−CD24intCD11b− cells boxed in (E) are compared after overnight incubation in media alone, GM-CSF, or CpG-2216 (original magnification, ×40). (A) is representative of 2 experiments; (B) and (C) are representative of 3 experiments; (D) and (E) are representative of >10 experiments; and (F) and (G) are representative of at least 2 and 3 experiments, respectively.

Light density BM cells were then purified and stained with CD45RA and CD11c (Fig. 1B). Three crude cell populations were sorted: CD45RAhiCD11c+ pDC, CD11c+CD45RA− putative BM cDC (boxed in Fig. 1B), and “the rest.” Each of these three populations was stimulated with CpG-2216 overnight and the supernatants were tested for IFN-α activity. Apart from pDC, the CD11c+CD45RA− cell population also produced IFN-α. The CD11c+ cell population was heterogeneous, containing small numbers of cells expressing CD4, CD8, and Sca-1 (Ly6A/E), and they were heterogeneous for CD11b and CD24 expression, with 40–50% of the cells expressing medium to high levels of CD11b and the remaining were CD11b−/lo (Fig. 1C). The cells could be sorted to four distinct populations by staining with CD11c, CD45RA, CD24, and CD11b (Fig. 1D). In response to CpG-2216, the IFN-α–producing capacity segregated precisely with the CD24intCD11b− cells, and the amount produced was similar to that produced by BM pDC (Fig. 1D).

The CD11b−CD24int cells were sorted to high purity by first depleting the light-density cells with a mixture of Abs directed to the likely contaminating cells, lymphocytes, NK cells, and granulocytes (Fig. 1E). Freshly isolated they existed at an ∼1:5 ratio with the BM pDC and shared with pDC high expression of 120G8 (BST-2) and expression of Siglec-H (lower than BM pDC; Fig. 1F). They expressed low levels of costimulation markers CD40, CD80, and CD86 and very low levels of MHC class II (lower than BM pDC; Fig. 1F). They expressed intermediate levels of CD172a and Ly6C, in both cases much lower than the levels on BM pDC (Fig. 1F). They also lacked expression of CCR9, unlike most BM pDC (Fig. 1F). Thus, by phenotypic comparison, the non-pDC IFN-α–producing cells of BM differed from lineage-negative macrophage DC progenitor or common DC progenitor precursors (18) and from the spleen CD3−CD19−NK1.1−Ter119−CD45RA−/loCD11c+CD43+SIRP-αlo pre-cDC (19) or the CD11b+ BM pre-DC (20) or spleen or BM CD11c− pro-DC (20). Electron microscopy did not reveal any major morphological differences between the CD11b−CD24int cells and the BM pDC (Supplemental Fig. 1).

The CD11c+CD45RA−CD24intCD11b− cells or BM pDC were stimulated overnight with CpG-2216, GM-CSF, or media and their morphologies were compared (Fig. 1H). In media, the CD11c+CD45RA−CD24intCD11b− cells and pDC behaved similarly, with little sign of typical DC morphology in either population, although the former were recovered in higher numbers. However, the CD11c+CD45RA−CD24intCD11b− cells, unlike the pDC, were clearly matured in GM-CSF (Fig. 1H). Moreover, they responded incredibly well to CpG-2216, forming large clusters of cells with typical cytoplasmic processes, unlike the small clusters of cells formed in the pDC cultures.

Activation of BM non-pDC IFN-α–producing cells induces a DC-like transformation, but miDC respond to TLR ligands and growth factors differently to pDC

The CD11c+CD3−CD19−CD49b−Ly6G−NK1.1−CD24intCD11b− cells were potent producers of IFN-α to CpG-2216 and they appeared activated in GM-CSF. We tested their cytokine response to a range of TLR ligands and GM-CSF. As shown in Fig. 2, the cells produced high levels of IFN-α to CpG-2216 and reproducibly low levels to a TLR7 ligand, R837. Τhe IFN-α production was only detected among the CD11c+CD3−CD19−CD49b−Ly6G−NK1.1−CD24intCD11b− cells, with no production detectable by the CD11b+ cells.

FIGURE 2.
FIGURE 2.

CD11c+CD45RA−CD24intCD11b− cells respond with cytokine production to multiple TLR ligands. (A) The CD11c+CD24intCD45RA−CD11b− and CD11c+CD24intCD45RA−CD11b+ BM cells were sorted and stimulated with TLR ligands, MVA, or GM-CSF as shown. Supernatants were tested by ELISA for IFN-α production. (B) The CD11c+CD24intCD45RA−CD11b− cells were further analyzed for IL-6 and TNF-α production and (C) their production of IL-6 was compared with BM pDC. Error bars represent SD of duplicate samples. Data for CD11b− cells and BM pDC are representative of more than five experiments. Data for CD11b+ cells are representative of two experiments.

The CD11c+CD3−CD19−CD49b−Ly6G−NK1.1−CD24intCD11b− cells produced high levels of IL-6 and TNF-α in response to TLR9 and 7 ligands and lower, although reproducible, levels of IL-6 in response to TLR2 (Pam3Cys) and TLR4 ligands (LPS) (Fig. 2). The IL-6 response to CpG-2216 was particularly high. In the same conditions using A-type ODN, IL-6 is barely detectable from BM pDC (Fig. 2). We also tested IL-10 and MIP-3α and did not find any produced in any conditions. Very low levels of IL-12p70 (<100 pg/ml) were produced in some experiments in response to CpG-2216 but only when GM-CSF was additionally included in the medium. We could not detect any cytokines produced in response to GM-CSF alone. We also tested the response of the cells to M-CSF and FL and none of the tested cytokines was produced, nor were any morphological changes noted.

The activated surface phenotype of CD11c+CD3−CD19−CD49b−Ly6G−NK1.1−CD24intCD11b− cells resembled in some ways the phenotype of activated pDC, showing upregulation of CD8α, CD40, and CD86 in response to CpG-2216 (Fig. 3A). However, the expression of MHC class II and costimulation markers including CD40, CD80, and CD86 (shown in Fig. 3C) was at least 5- to 10-fold higher on the activated CD49b−Ly6G−NK1.1−CD24intCD11b−/locells in all conditions tested.

FIGURE 3.
FIGURE 3.

CD11c+CD24intCD11b− cells respond with surface activation to multiple TLR ligands and upregulate CD45R on their surface. BM pDC or CD11c+CD24intCD11b− cells were purified by FACS sorting from pooled BM of at least 20 mice. (A) Cells were stimulated with CpG-2216 for 18 h and then stained with the surface markers shown. Dotted gray line depicts pDC in GM-CSF (identical to media only control that is not shown); solid gray line shows staining of CD11c+CD24intCD11b− cells in GM-CSF; filled histogram depicts staining of pDC in CpG-2216; black line shows staining of CD11c+CD24intCD11b− cells in CpG-2216. Unstained control for CD11c+CD24intCD11b− cells was in first log for all markers but is not shown for the purpose of clarity. Data are from one experiment representative of three experiments for CD40 staining and more than five experiments for CD8α and CD86 staining. (B) The expression of CD45R (B220) is shown on pDC (filled gray line), CD11c+CD24intCD11b− cells in GM-CSF (light gray line), or in CpG-2216 (black line). Data are representative of more than five experiments. (C) CD86 expression (black line) is shown on CD11c+CD24intCD11b− cells after 18 h incubation with the stimuli shown. Gray line is unstained background control. Data are from 1 experiment representative of >10 experiments (CpG-2216 and media) or at least 3 experiments for other stimuli.

The expression of CD45R was upregulated on the CD49b−Ly6G−NK1.1−CD24intCD11b− cells upon activation (Fig. 3B), but never as high as on pDC; nevertheless, the expression of CD45R did suggest a relationship to the pDC. However, TLR2 and TLR4 ligands also induced upregulation of costimulation markers (Fig. 3C) but had no effect on pDC. Moreover, unlike pDC, low levels of costimulation markers were observed after overnight incubation in media alone, which was further enhanced by GM-CSF (Fig. 3A).

These observations, together with the differential functions described below, led us to name the CD11c+CD3−CD19−CD49b−Ly6G−NK1.1−CD24intCD11b− cells as a novel BM DC population, myelos (bone marrow in Greek) IFN DC (or miDC).

miDC are infected by virus but survive well and respond differently to various viruses than do pDC

The activation of miDC with various viruses shed further light on the differences between these cells and pDC. HSV, known to activate BM pDC via TLR9 and TLR9-independent pathways (11), was also a strong inducer of miDC IFN-α production (Fig. 4A). miDC produced substantial levels of IFN-α in response to HSV, although this was still ∼50% that produced by pDC. MVA induced CD86 upregulation on miDC (Fig. 4B) and low levels of IFN-α production (Fig. 4A), being about a third that produced by pDC. Note that no IFN-α was detected in supernatants from spleen cDC (Fig. 4A).

FIGURE 4.
FIGURE 4.

Viral activation of miDC. (A) BM pDC, miDC, or spleen cDC were purified by FACS sorting and stimulated for 20 h with the viruses as shown, and supernatants were tested for IFN-α production by ELISA. Data are from one experiment representative of at least two experiments for each virus. Error bars represent SD of duplicate samples. (B) Staining of CD86 and CD45R on miDC (black filled) or pDC (dark gray line) stimulated as shown. Light gray line represents staining of miDC after 20 h in media only. Data are from one experiment representative of three experiments. (C) The survival of pDC or miDC after 20 h culture is shown. Data are from one experiment representative of at least three experiments.

ECTV and SFV are both members of the pox virus family. ECTV has previously been shown by us to activate pDC via TLR9 but to potently inhibit the largely TLR9- independent activation pathways of cDC (21). Both SFV and ECTV did not induce either surface activation of miDC or measurable IFN-α production (Fig. 4). Thus, miDC responded to these pox viruses similarly as did cDC, exhibiting no activation in response to these pathogenic pox viruses.

A general feature of the response of miDC to all of the viruses (and other stimuli) tested was that they displayed high survival (>65%, Fig. 4C) whether activated or not, differentiating these cells from the pDC that exhibited poor survival upon stimulation with viruses or in the presence of GM-CSF.

We have previously found that pDC from Bcl2-transgenic mice survive extremely well upon culture and activation (22). We found that the miDC purified from Bcl2 transgenic mice did not have a survival advantage over that of C57BL/6 mice; in fact, the survival of C57BL/6 miDC paralleled that of the pDC from the Bcl2 transgenics. We found that miDC from Bcl2 transgenic mice produced similar levels of IFN-α as did wild-type mice in response to MVA and HSV and also did not produce IFN-α in response to SFV (Supplemental Fig. 2). In contrast, the pDC from Bcl2 transgenic mice displayed increased IFN-α production to these viruses.

Qualitative differences between miDC and pDC

Pelayo et al. (23) have shown that the BM pDC of mice expressing Rag-GFP are composed of two groups, that is, GFP+ (pDC1) and GFP− (pDC2) cells that display differences in cytokine production. They demonstrated that pDC2 produced substantially more IFN-α than did the pDC1 cells and, additionally, the pDC2 produced quite high levels of IL-6 and IFN-γ in response to CpG-ODN. We examined the GFP expression in miDC purified from these Rag-GFP mice. We found that miDC did not express GFP and thus lacked a Rag “signature,” clearly differentiating them from the pDC1 (Fig. 5A). Moreover, although miDC produced substantial amounts of IL-6, resembling pDC2, they lacked IFN-γ production to any TLR or viral stimulus tested and their surface activation upon stimulation (Fig. 3) was substantially stronger than that shown by Pelayo et al. (23) for pDC1 or pDC2 or to total BM pDC (Fig. 3A).

FIGURE 5.
FIGURE 5.

Expression of Rag and PU.1 in miDC. DC were enriched from light density BM cells of Rag-GFP mice (A) or PU.1-GFP mice (B) and BM pDC (black open histograms) or miDC (black filled histograms) were gated and analyzed for expression of GFP. In (B), the lower panel compares levels of PU.1-GFP expressed by spleen pDC (light gray unfilled histogram) and spleen cDC (filled gray histogram) to the BM pDC and miDC. Results are from one experiment of three pooled RAG-GFP mice or representative of three separately analyzed PU.1-GFP mice.

PU.1 is a transcription factor that is absolutely required by DC precursors for DC development (24). Studies using PU.1-GFP mice have elegantly shown that pDC and cDC do not equally express this transcription factor (24). Indeed, pDC express substantially lower levels of PU.1 (GFP) than do cDC in these mice. Given this clear difference of PU.1 expression between pDC and cDC, we examined the PU.1-GFP levels in miDC and BM pDC of the PU.1-GFP mice. The miDC expressed higher levels of GFP than did the BM pDC (Fig. 5B). The level of GFP expressed by BM pDC was lower than that of spleen pDC, and likewise the miDC GFP expression was lower than that of the spleen cDC, although still clearly distinct from the spleen pDC (Fig. 5B).

Thus, in addition to the stimulation and survival data, the lack of expression of Rag and high PU.1 expression by miDC clearly differentiates them from the pDC of BM.

miDC do not divide upon activation

Upon activation, the miDC displayed features similar to pDC, including the expression of CD45R, CD8α, and the production of high levels of IFN-α upon CpG-2216-ODN stimulation. Concomitantly, the miDC displayed features of cDC: maturation in response to GM-CSF, responsiveness to TLR4 ligands, and high levels of PU.1 expression. Moreover, the survival of the miDC both in media and upon stimulation was extremely high (Fig. 4C). Given these factors, we considered the possibility that the miDC were in fact a dividing precursor population that gave rise to both pDC and cDC-like progeny. To test this hypothesis the miDC were labeled with CFSE and cultured for 40 h in media alone or in the presence of ligands to TLR4, TLR7, or TLR9, or GM-CSF. FACS analysis of the miDC after the culture period revealed that there was no loss of CFSE label in any of the culture conditions, indicating that the miDC did not divide upon stimulation and differentiation from an immature DC to an activated, mature DC (Fig. 6A). CFSE-labeled and unlabeled miDC were compared for their ability to produce IFN-α, IL-6, and TNF-α after stimulation. The CFSE-labeled cells behaved exactly as did their nonlabeled counterparts.

FIGURE 6.
FIGURE 6.

Proliferative capacity of miDC. (A) miDC were labeled with CFSE, incubated with CpG-2216, R837, GM-CSF, MVA-BN, or media, and analyzed by FACS 24 h later for loss of CFSE fluorescence. All traces overlayed each other. Shown are the results for GM-CSF, R837, and media. Data are of one experiment of DC purified from BM of 20 pooled mice and are representative of two experiments. (B) CFSE-labeled miDC (Ly5.2+) were spiked into FLDC cultures (Ly5.1+ cells) and analyzed 4 or 5 d later. Surface expression of CD45R is shown within the undivided peak 1 or divided peaks 2 or 3. Data are of one experiment of miDC purified from BM of 12 pooled C57BL/6 mice and are representative of two separate experiments. (C) A third experiment analyzing the Ly5. 2+ miDC gave similar results with respect to surface phenotype, with CD45R expression essentially absent by day 6, and (D) Ly5.2+ cells (thick black histogram) showing high levels of CD172 expression compared with Ly5.1+ cells (thin histogram).

miDC divide one to two times within FL BM cultures

FL BM cultures drive cDC and pDC development from DC precursors with FL. To further investigate the proliferative capacity of the miDC, they were purified from C57BL/6J Ly5.2+ mice, labeled with CFSE, and “spiked” into Ly5.1+ FL BM cultures After day 4, most of the miDC had undergone one division, and by day 5 the cells were mainly cells that had divided either once or twice (Fig. 6B). In culture, the cells upregulated CD45R, but upon division, CD45R was lost from the cell surface. By day 6 all of the CD45R expression was essentially gone (Fig. 6C) and the cells expressed extremely high levels of Sirpα (CD172a; Fig. 6D). This level was higher than that expressed by the Ly5.1+ FLDC day 6 culture, which would include CD172+ cDC and pDC and also the CD11clo cells that represented the dividing pre-DC in the FL cultures.

miDC display potent stimulation of allogeneic and Ag-specifc T cells

To determine the stimulatory capacity of miDC, they were first tested in an allogeneic MLR. Similar to pDC, they were relatively poor stimulators of allogeneic T cells without prior activation. With LPS activation the miDC stimulated naive T cells (Fig. 7A) more efficiently than did pDC, albeit to a lesser extent than CD8− spleen DC. CpG activation of the miDC led to a poor recovery of the T cells.

FIGURE 7.
FIGURE 7.

miDC are potent stimulators of Ag-specific T cells. (A) Allogeneic MLR miDC, BM pDC, or spleen CD8− cDC from C57BL/6 mice were purified and incubated with BALB/c lymph node T cells labeled with CFSE. Stimulants were added to the cultures as shown. CFSE expression by the T cells was analyzed 3 d later. FACS data are representative of pooled triplicate wells. Data are representative of two separate experiments. (B) miDC, BM pDC, or spleen CD8+ or CD8− cDC were incubated with MVA-OVA for 1 h at 37°C and then washed. Titrated DC were then added to 50,000 CFSE-labeled OTI T cells. Data shown are for 104 DC of each subtype. CFSE fluorescence of OTI T cells was measured by FACS 3 d later and the number of proliferating cells was quantitated by reference to a known number of quantitation beads spiked into each well before harvesting the cells. Shown are values of pooled triplicate wells. Data are representative of three separate experiments. (C) miDC were incubated with soluble OVA and incubated at 37°C for 1 h and then washed. DC (104) with added stimulus as shown were added to 50,000 OTI T cells and CFSE proliferation was analyzed 3 d later. Data are results from pooled triplicate wells of one experiment and are representative of three experiments.

To determine whether the miDC had the capacity to stimulate Ag-specific T cells, we coincubated them with MVA-expressing OVA (MVA-OVA) under an early viral promoter. The cells were washed after 1 h and then coincubated with OTI T cells. The MVA-OVA–stimulated miDC clearly presented the virally encoded Ag and were far more effective at stimulating the OTI T cells than were the pDC (Fig. 7B). In fact, they were at least as efficient as CD8− spleen cDC and more efficient than CD8+ spleen cDC in stimulating the OTI T cells.

We also tested the capacity of the miDC to cross-present Ag. Similar to pDC and CD8− cDC, the miDC, whether activated or not, were inefficient at cross-presenting peptides from soluble OVA (Fig. 7C).

Discussion

The BM is well established as a primary lymphoid organ, but it also provides niches for B cells (25–27) and naive T cells (28) that, in situ, respond to blood-borne Ags. The BM is thus established as an important and unique secondary lymphoid organ. It is also a preferential site for memory T cell homing and homeostatic proliferation (29) and for long-lived plasma cells (30). Given these factors, it is important to understand the cell types that may interact with lymphocytes in situ in the BM. Within this work, we have identified a novel DC type in BM.

We have characterized that, apart from the pDC, there is another DC in BM that produces high amounts of IFN-α, the miDC. Similar to the pDC, the miDC express Bst-2 and Siglec-H (albeit at lower levels than do pDC) and have a potent ability to produce IFN-α in response to CpG-2216 and upon activation acquire the expression of CD8α and CD45R. However, this is where the similarity between the two cell types ends.

The miDC begin to mature in media alone, which has no effect on pDC, and they further mature in response to GM-CSF in vitro, a factor that has no effect on the surface phenotype of the mouse BM pDC. Human pDC have been shown to be activated in response to both IL-3 and GM-CSF, but this has not been reported in the mouse system. Human pDC, however, are not normally activated in media alone (31). The miDC clearly respond differently to A-type ODN than do the pDC. They are strongly matured at the cell surface and produce extremely high levels of IL-6 to this stimulus, and yet pDC hardly produce detectable IL-6 to CpG-A-ODN. Similar to the pDC, the miDC respond to the TLR7 ligand R837, but reproducibly they also produce low levels of IFN-α to this stimulus. The miDC, but not pDC, also respond to TLR2 and TLR4 ligands with cell surface activation and low IL-6 production.

In response to viruses, the miDC and pDC also differ somewhat in their reactivities. Both cell types respond vigorously with maturation and IFN-α production to HSV-1. However, the miDC response to MVA-BN is weaker than that of pDC, and they show no sign of activation in response to the pox viruses ECTV and SFV.

In stark contrast to the pDC, the survival of miDC in all conditions tested was high, at least 65%. In contrast to pDCs, the transgenic Bcl2 expression did not further enhance survival in miDC. The high survival was not due to detectable proliferation of the cells. The miDC appear to have a capacity to undergo one to two divisions, but this was only observed when they were “spiked” into FLDC cultures. Whether this would occur in vivo during, for example, FL spikes occurring during viral infection was not further investigated in this study. It was clear, however, that the miDC did not undergo detectable division when activated by CpG or other stimuli in vitro.

Schlitzer et al. (32) have recently shown that BM pDC (selected as 120G8+ cells) could be separated by CCR9 into cells that responded to GM-CSF (CCR9−) and cells that did not (CCR9+). The 120G8+CCR9−/lo miDC show similarities to the CCR9− pDC, including the production of high IL-6 production to CpG2216; however, the CCR9− pDC described by Schlitzer et al. (32) express quite high levels of B220 ex vivo. We have shown that miDC do not express B220 ex vivo but do upregulate B220 in culture. B220 appears to be transiently expressed, at least in vitro, as miDC mature to conventional-type DC (Fig. 6). Our data do not suggest that miDC are the direct precursors of CCR9+pDC. In our hands, gating on 120G8+CCR9− cells (Fig. 1F) would result in a population of cells containing both miDC and pDC. Note that although pDC have a relatively low number of cells that are CCR9−, there are in total ∼5-fold more pDC than the miDC. Thus, the data of Schlitzer et al. (32) probably correspond to a mixed population of CCR9− miDC (high in vitro survival) and pDC (short in vitro survival), although the expression of B220 on all of the cells does not fit with our data.

The striking feature of the miDC is their ability to present virally encoded Ag and to induce proliferation of specific T cells to this Ag. Given the high survival of virally infected miDC, they are a potential long-lived source of viral Ag and an inducer of CTL in the bone marrow. Alternatively, they could potentially act as a sink for latent viruses that “hide out” in the BM.

Several viruses are known to be found in BM, including lymphocytic choriomeningitis virus. Acute lymphocytic choriomeningitis virus infections in mice lead to BM failure during the first week of infection as a direct result of IFN-I–mediated ablation of hematopoietic precursor cells (33). HSV-1 (34) and dengue virus (35–37) infection of BM cells have been implicated in hematopoietic defects. Additionally, hematopoietic abnormalities in HIV patients are attributable, at least in part, to the replication of HIV in hematopoietic cells in the BM (38).

CMV specifically targets hematopoietic precursor cells in the BM and is of particular concern in immunosuppressed individuals. In mouse pox infections, ECTV is found in BM of both resistant and susceptible mouse strains (39). Replicating poxviruses, including ALVAC and vaccinia, used as vectors in human vaccine trials, infect as yet uncharacterized CD33+ cells within BM (40).

The significance of miDC within the BM, cells that produce large amounts of IFN-I and have the ability to strongly stimulate naive T cells, relates to both the primary and secondary lymphoid organ roles of this organ. IFN-I produced by miDC and pDC in response to viruses potentially affects hematopoiesis since it is established that high levels of IFN-I can ablate hematopoietic precursors leading to acquired anemia (33) or skewing of precursor development (41). IFN-I act as adjuvants and activate and enhance DC, CTL, and B cell responses (42) and thus potentially enhance innate and adaptive immune responses to BM-encountered Ag. In the case of miDC, the production of IFN-I as well as their ability to stimulate T cells suggests their direct involvement in T cell activation. This also raises the possibility that miDC may be deleterious in BM transplant situations, having the ability to strongly stimulate graft or host-derived T cells. Their response to TLR9 and TLR7 stimulation also marks them, along with pDC, as potential contributors to systemic lupus erythematosus pathology. The high IL-6 production of miDC also suggests they may play a role in survival of BM plasma cells (43).

In this study, we have identified a novel cell type, the miDC. The miDC appear to be a cell type related to but distinct from pDC. Others have identified pDC subsets in vivo, including the CD9+ pDC described by Björck et al. (44). The miDC do not appear to be related to these pDC subsets, they are CD9− (data not shown), and they do not express B220 ex vivo. We propose that the miDC are a unique IFN-producing cell type that, upon proinflammatory signals, can mature into a conventional type DC. These cells have the potential to aid viral clearance from the BM but also have attributes that could be deleterious in transplant situations or in autoimmune conditions such as systemic lupus erythematosus.

Disclosures

H.H. and M.S. are current employees of Bavarian Nordic GmbH. M.O.K. and B.F. are prior employees of Bavarian Nordic GmbH and currently receive research funding from Bavarian Nordic GmbH. The remaining authors have no financial conflicts of interest.

Acknowledgments

We thank animal technicians at Bavarian Nordic, the Walter and Eliza Hall Institute, and the Alfred Medical and Research and Education Precinct for their services. The members of the FACS facilities at the Walter and Eliza Hall Institute and the Alfred Medical and Research and Education Precinct are gratefully acknowledged for their expertise.

Footnotes

  • This work was supported by Bavarian Nordic GmbH.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    BM
    bone marrow
    cDC
    conventional dendritic cell
    DC
    dendritic cell
    ECTV
    ectromelia virus
    FL
    Flt3 ligand
    IFN-I
    type I IFN
    miDC
    myelos IFN-producing dendritic cell
    MVA
    modified vaccinia Ankara virus
    ODN
    oligonucleotide
    Pam3Cys
    palmitoyl-3-cysteine-serine-lysine-4
    pDC
    plasmacytoid dendritic cell
    SFV
    Shope fibroma virus.

  • Received May 10, 2011.
  • Accepted February 6, 2012.

References

  1. ↵
    1. Villadangos J. A.,
    2. P. Schnorrer
    . 2007. Intrinsic and cooperative antigen-presenting functions of dendritic-cell subsets in vivo. Nat. Rev. Immunol. 7: 543–555.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Reis e Sousa C.,
    2. S. Hieny,
    3. T. Scharton-Kersten,
    4. D. Jankovic,
    5. H. Charest,
    6. R. N. Germain,
    7. A. Sher
    . 1997. In vivo microbial stimulation induces rapid CD40 ligand-independent production of interleukin 12 by dendritic cells and their redistribution to T cell areas. J. Exp. Med. 186: 1819–1829.
    OpenUrlAbstract/FREE Full Text
    1. Hochrein H.,
    2. K. Shortman,
    3. D. Vremec,
    4. B. Scott,
    5. P. Hertzog,
    6. M. O’Keeffe
    . 2001. Differential production of IL-12, IFN-α, and IFN-γ by mouse dendritic cell subsets. J. Immunol. 166: 5448–5455.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Maldonado-López R.,
    2. T. De Smedt,
    3. P. Michel,
    4. J. Godfroid,
    5. B. Pajak,
    6. C. Heirman,
    7. K. Thielemans,
    8. O. Leo,
    9. J. Urbain,
    10. M. Moser
    . 1999. CD8α+ and CD8α− subclasses of dendritic cells direct the development of distinct T helper cells in vivo. J. Exp. Med. 189: 587–592.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Proietto A. I.,
    2. M. O’Keeffe,
    3. K. Gartlan,
    4. M. D. Wright,
    5. K. Shortman,
    6. L. Wu,
    7. M. H. Lahoud
    . 2004. Differential production of inflammatory chemokines by murine dendritic cell subsets. Immunobiology 209: 163–172.
    OpenUrlCrossRefPubMed
  6. ↵
    1. O’Keeffe M.,
    2. H. Hochrein,
    3. D. Vremec,
    4. I. Caminschi,
    5. J. L. Miller,
    6. E. M. Anders,
    7. L. Wu,
    8. M. H. Lahoud,
    9. S. Henri,
    10. B. Scott,
    11. et al
    . 2002. Mouse plasmacytoid cells: long-lived cells, heterogeneous in surface phenotype and function, that differentiate into CD8+ dendritic cells only after microbial stimulus. J. Exp. Med. 196: 1307–1319.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Villadangos J. A.,
    2. L. Young
    . 2008. Antigen-presentation properties of plasmacytoid dendritic cells. Immunity 29: 352–361.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Young L. J.,
    2. N. S. Wilson,
    3. P. Schnorrer,
    4. A. Proietto,
    5. T. ten Broeke,
    6. Y. Matsuki,
    7. A. M. Mount,
    8. G. T. Belz,
    9. M. O’Keeffe,
    10. M. Ohmura-Hoshino,
    11. et al
    . 2008. Differential MHC class II synthesis and ubiquitination confers distinct antigen-presenting properties on conventional and plasmacytoid dendritic cells. Nat. Immunol. 9: 1244–1252.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Gilliet M.,
    2. W. Cao,
    3. Y. J. Liu
    . 2008. Plasmacytoid dendritic cells: sensing nucleic acids in viral infection and autoimmune diseases. Nat. Rev. Immunol. 8: 594–606.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Kadowaki N.
    2009. The divergence and interplay between pDC and mDC in humans. Front. Biosci. 14: 808–817.
    OpenUrlPubMed
  11. ↵
    1. Hochrein H.,
    2. B. Schlatter,
    3. M. O’Keeffe,
    4. C. Wagner,
    5. F. Schmitz,
    6. M. Schiemann,
    7. S. Bauer,
    8. M. Suter,
    9. H. Wagner
    . 2004. Herpes simplex virus type-1 induces IFN-α production via Toll-like receptor 9-dependent and -independent pathways. Proc. Natl. Acad. Sci. USA 101: 11416–11421.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Hochrein H.,
    2. M. O’Keeffe
    . 2008. Dendritic cell subsets and Toll-like receptors. Handb. Exp. Pharmacol. 2008: 153–179.
    OpenUrl
  13. ↵
    1. McKenna H. J.,
    2. K. L. Stocking,
    3. R. E. Miller,
    4. K. Brasel,
    5. T. De Smedt,
    6. E. Maraskovsky,
    7. C. R. Maliszewski,
    8. D. H. Lynch,
    9. J. Smith,
    10. B. Pulendran,
    11. et al
    . 2000. Mice lacking flt3 ligand have deficient hematopoiesis affecting hematopoietic progenitor cells, dendritic cells, and natural killer cells. Blood 95: 3489–3497.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Ogilvy S.,
    2. D. Metcalf,
    3. C. G. Print,
    4. M. L. Bath,
    5. A. W. Harris,
    6. J. M. Adams
    . 1999. Constitutive Bcl-2 expression throughout the hematopoietic compartment affects multiple lineages and enhances progenitor cell survival. Proc. Natl. Acad. Sci. USA 96: 14943–14948.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. O’Keeffe M.,
    2. H. Hochrein,
    3. D. Vremec,
    4. J. Pooley,
    5. R. Evans,
    6. S. Woulfe,
    7. K. Shortman
    . 2002. Effects of administration of progenipoietin 1, Flt-3 ligand, granulocyte colony-stimulating factor, and pegylated granulocyte-macrophage colony-stimulating factor on dendritic cell subsets in mice. Blood 99: 2122–2130.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Spies B.,
    2. H. Hochrein,
    3. M. Vabulas,
    4. K. Huster,
    5. D. H. Busch,
    6. F. Schmitz,
    7. A. Heit,
    8. H. Wagner
    . 2003. Vaccination with plasmid DNA activates dendritic cells via Toll-like receptor 9 (TLR9) but functions in TLR9-deficient mice. J. Immunol. 171: 5908–5912.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Blom B.,
    2. S. Ho,
    3. S. Antonenko,
    4. Y. J. Liu
    . 2000. Generation of interferon α-producing predendritic cell (Pre-DC)2 from human CD34+ hematopoietic stem cells. J. Exp. Med. 192: 1785–1796.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Liu K.,
    2. C. Waskow,
    3. X. Liu,
    4. K. Yao,
    5. J. Hoh,
    6. M. Nussenzweig
    . 2007. Origin of dendritic cells in peripheral lymphoid organs of mice. Nat. Immunol. 8: 578–583.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Naik S. H.,
    2. D. Metcalf,
    3. A. van Nieuwenhuijze,
    4. I. Wicks,
    5. L. Wu,
    6. M. O’Keeffe,
    7. K. Shortman
    . 2006. Intrasplenic steady-state dendritic cell precursors that are distinct from monocytes. Nat. Immunol. 7: 663–671.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Naik S. H.,
    2. P. Sathe,
    3. H. Y. Park,
    4. D. Metcalf,
    5. A. I. Proietto,
    6. A. Dakic,
    7. S. Carotta,
    8. M. O’Keeffe,
    9. M. Bahlo,
    10. A. Papenfuss,
    11. et al
    . 2007. Development of plasmacytoid and conventional dendritic cell subtypes from single precursor cells derived in vitro and in vivo. Nat. Immunol. 8: 1217–1226.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Samuelsson C.,
    2. J. Hausmann,
    3. H. Lauterbach,
    4. M. Schmidt,
    5. S. Akira,
    6. H. Wagner,
    7. P. Chaplin,
    8. M. Suter,
    9. M. O’Keeffe,
    10. H. Hochrein
    . 2008. Survival of lethal poxvirus infection in mice depends on TLR9, and therapeutic vaccination provides protection. J. Clin. Invest. 118: 1776–1784.
    OpenUrlCrossRefPubMed
  22. ↵
    1. O’Keeffe M.,
    2. R. J. Grumont,
    3. H. Hochrein,
    4. M. Fuchsberger,
    5. R. Gugasyan,
    6. D. Vremec,
    7. K. Shortman,
    8. S. Gerondakis
    . 2005. Distinct roles for the NF-κB1 and c-Rel transcription factors in the differentiation and survival of plasmacytoid and conventional dendritic cells activated by TLR-9 signals. Blood 106: 3457–3464.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Pelayo R.,
    2. J. Hirose,
    3. J. Huang,
    4. K. P. Garrett,
    5. A. Delogu,
    6. M. Busslinger,
    7. P. W. Kincade
    . 2005. Derivation of 2 categories of plasmacytoid dendritic cells in murine bone marrow. Blood 105: 4407–4415.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Carotta S.,
    2. A. Dakic,
    3. A. D’Amico,
    4. S. H. Pang,
    5. K. T. Greig,
    6. S. L. Nutt,
    7. L. Wu
    . 2010. The transcription factor PU.1 controls dendritic cell development and Flt3 cytokine receptor expression in a dose-dependent manner. Immunity 32: 628–641.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Sapoznikov A.,
    2. Y. Pewzner-Jung,
    3. V. Kalchenko,
    4. R. Krauthgamer,
    5. I. Shachar,
    6. S. Jung
    . 2008. Perivascular clusters of dendritic cells provide critical survival signals to B cells in bone marrow niches. Nat. Immunol. 9: 388–395.
    OpenUrlCrossRefPubMed
    1. Pillai S.,
    2. A. Cariappa
    . 2009. The bone marrow perisinusoidal niche for recirculating B cells and the positive selection of bone marrow-derived B lymphocytes. Immunol. Cell Biol. 87: 16–19.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Cariappa A.,
    2. I. B. Mazo,
    3. C. Chase,
    4. H. N. Shi,
    5. H. Liu,
    6. Q. Li,
    7. H. Rose,
    8. H. Leung,
    9. B. J. Cherayil,
    10. P. Russell,
    11. et al
    . 2005. Perisinusoidal B cells in the bone marrow participate in T-independent responses to blood-borne microbes. Immunity 23: 397–407.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Feuerer M.,
    2. P. Beckhove,
    3. N. Garbi,
    4. Y. Mahnke,
    5. A. Limmer,
    6. M. Hommel,
    7. G. J. Hämmerling,
    8. B. Kyewski,
    9. A. Hamann,
    10. V. Umansky,
    11. V. Schirrmacher
    . 2003. Bone marrow as a priming site for T-cell responses to blood-borne antigen. Nat. Med. 9: 1151–1157.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Becker T. C.,
    2. S. M. Coley,
    3. E. J. Wherry,
    4. R. Ahmed
    . 2005. Bone marrow is a preferred site for homeostatic proliferation of memory CD8 T cells. J. Immunol. 174: 1269–1273.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Slifka M. K.,
    2. R. Antia,
    3. J. K. Whitmire,
    4. R. Ahmed
    . 1998. Humoral immunity due to long-lived plasma cells. Immunity 8: 363–372.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Ghirelli C.,
    2. R. Zollinger,
    3. V. Soumelis
    . 2010. Systematic cytokine receptor profiling reveals GM-CSF as a novel TLR-independent activator of human plasmacytoid predendritic cells. Blood 115: 5037–5040.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Schlitzer A.,
    2. J. Loschko,
    3. K. Mair,
    4. R. Vogelmann,
    5. L. Henkel,
    6. H. Einwächter,
    7. M. Schiemann,
    8. J. H. Niess,
    9. W. Reindl,
    10. A. Krug
    . 2011. Identification of CCR9− murine plasmacytoid DC precursors with plasticity to differentiate into conventional DCs. Blood 117: 6562–6570.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Binder D.,
    2. J. Fehr,
    3. H. Hengartner,
    4. R. M. Zinkernagel
    . 1997. Virus-induced transient bone marrow aplasia: major role of interferon-α/β during acute infection with the noncytopathic lymphocytic choriomeningitis virus. J. Exp. Med. 185: 517–530.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Lamontagne L.,
    2. P. Jolicoeur
    . 1994. Herpes simplex virus type 1 pathogenicity in mice expressed by lytic infection in pre-B and B cells. Clin. Immunol. Immunopathol. 73: 38–44.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Balsitis S. J.,
    2. J. Coloma,
    3. G. Castro,
    4. A. Alava,
    5. D. Flores,
    6. J. H. McKerrow,
    7. P. R. Beatty,
    8. E. Harris
    . 2009. Tropism of dengue virus in mice and humans defined by viral nonstructural protein 3-specific immunostaining. Am. J. Trop. Med. Hyg. 80: 416–424.
    OpenUrlAbstract/FREE Full Text
    1. Kyle J. L.,
    2. P. R. Beatty,
    3. E. Harris
    . 2007. Dengue virus infects macrophages and dendritic cells in a mouse model of infection. J. Infect. Dis. 195: 1808–1817.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Albuquerque P. L.,
    2. G. B. Silva Júnior,
    3. S. S. Diógenes,
    4. H. F. Silva
    . 2009. Dengue and aplastic anemia: a rare association. Travel Med. Infect. Dis. 7: 118–120.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Alexaki A.,
    2. B. Wigdahl
    . 2008. HIV-1 infection of bone marrow hematopoietic progenitor cells and their role in trafficking and viral dissemination. PLoS Pathog. 4: e1000215.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Jacoby R. O.,
    2. P. N. Bhatt
    . 1987. Mousepox in inbred mice innately resistant or susceptible to lethal infection with ectromelia virus, II: Pathogenesis. Lab. Anim. Sci. 37: 16–22.
    OpenUrlPubMed
  40. ↵
    1. Yu Q.,
    2. N. Hu,
    3. M. Ostrowski
    . 2009. Poxvirus tropism for primary human leukocytes and hematopoietic cells. Methods Mol. Biol. 515: 309–328.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Sevilla N.,
    2. D. B. McGavern,
    3. C. Teng,
    4. S. Kunz,
    5. M. B. Oldstone
    . 2004. Viral targeting of hematopoietic progenitors and inhibition of DC maturation as a dual strategy for immune subversion. J. Clin. Invest. 113: 737–745.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Bracci L.,
    2. V. La Sorsa,
    3. F. Belardelli,
    4. E. Proietti
    . 2008. Type I interferons as vaccine adjuvants against infectious diseases and cancer. Expert Rev. Vaccines 7: 373–381.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Kawano M. M.,
    2. K. Mihara,
    3. N. Huang,
    4. T. Tsujimoto,
    5. A. Kuramoto
    . 1995. Differentiation of early plasma cells on bone marrow stromal cells requires interleukin-6 for escaping from apoptosis. Blood 85: 487–494.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Björck P.,
    2. H. X. Leong,
    3. E. G. Engleman
    . 2011. Plasmacytoid dendritic cell dichotomy: identification of IFN-α producing cells as a phenotypically and functionally distinct subset. J. Immunol. 186: 1477–1485.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section