Comparative analysis of IRF and IFN-alpha expression in human plasmacytoid and monocyte-derived dendritic cells

Transcription

1 Comparative analysis of IRF and IFN-alpha expression in human plasmacytoid and monocyte-derived dendritic cells Alexander Izaguirre,*,,1 Betsy J. Barnes,,1 Sheela Amrute,* Wen-Shuz Yeow, Nicholas Megjugorac,*, Jihong Dai, Di Feng,*, Eugene Chung,*,, Paula M. Pitha, 2,3 and Patricia Fitzgerald-Bocarsly*,,,2,3 *UMDNJ-New Jersey Medical School, UMDNJ Graduate School for Biomedical Sciences, Newark, NJ 07103; The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins School of Medicine, Baltimore, MD 21231; and the Cancer Institute of New Jersey Abstract: Plasmacytoid dendritic cells (PDC) produce high levels of type I IFN upon stimulation with viruses, while monocytes and monocyte-derived dendritic cells (MDDC) produce significantly lower levels. To find what determines the high production of type I IFN in PDC, we examined the relative levels of IRF transcription factors, some of which play critical roles in the induction of IFN. Furthermore, to determine whether the differences could result from expression of distinct IFNA subtypes, the profile of IFNA genes expressed was examined. PDC responded equally well to stimulation with HSV-1 and Sendai virus (SV) by producing high levels of type I IFN, whereas the MDDC and monocyte response to SV were lower, and neither responded well to HSV-1. All three populations constitutively expressed most of the IRF genes. However, real-time RT-PCR demonstrated increased levels of IRF-7 transcripts in PDC compared with monocytes. As determined by intracellular flow cytometry, the PDC constitutively expressed significantly higher levels of IRF-7 protein than the other populations while IRF-3 levels were similar among populations. Analysis of the profile of IFNA genes expressed in virus-stimulated PDC, monocytes and MDDC demonstrated that each population expressed IFNA1 as the major subtype but that the range of the subtypes expressed in PDC was broader, with some donor and stimulusdependent variability. We conclude that PDC but not MDDC are uniquely preprogrammed to respond rapidly and effectively to a range of viral pathogens with high levels of IFN- production due to the high levels of constitutively expressed IRF-7. J. Leukoc. Biol. 74: ; Among human peripheral blood mononuclear cells, there are two populations of cells that respond to viral challenge with the production of type I interferon (IFN): monocytes and the natural IFN-producing cells (NIPC) [1 3], the latter being recently identified as the precursors to the type 2 dendritic cell (pdc2, also known as the plasmacytoid dendritic cells, or PDC) [4, 5]. PDC can be identified in peripheral blood as cells that are negative for lineage-specific markers (CD3, CD19, CD14, and CD16), negative for CD11c, but positive for HLA-DR and are CD123 bright, the IL-3 receptor [6]. A portion of monocytes, when infected with certain viruses such as the Sendai virus (SV), respond to the infection by producing type I IFN, whereas the PDC produce type I IFN in response to a wide range of enveloped viruses as well as bacteria, and DNA containing unmethylated CpG sequences [3 5, 7 9]. The PDC do not have to be productively infected by the virus in order to produce IFN, but rather, produce IFN- vigorously in response to UV-inactivated virus or formalin-fixed, virus-infected cells [1, 2, 4]. The second population of peripheral blood DC, the DC1, are lineage-negative but positive for CD11c and are CD123 dim, and do not produce IFN- in response to stimulation with HSV-1 [4, 6, 10]. The peripheral blood DC1 are phenotypically and functionally similar to monocyte-derived dendritic cells (MDDC), which can be derived in vitro by culturing of peripheral blood monocytes with GM-CSF and IL-4 [11]. A population of DC that produces high levels of IFN- and which resemble the human PDC has been also identified in the mouse [12, 13]. The high IFN- -producing, HSV-responsive PDC are present in peripheral blood at fold lower frequency than Sendai virus-responsive monocytes [14, 15], but the PDC are 6 12 fold more efficient IFN producers than the monocytes [15]. However, in contrast to monocytes, PDC respond to a wider range of inducers [8, 14, 16], and the virus mediated induction of IFN- in these cells can be blocked by inhibitors of lysosomal function [14, 17] and by neutralizing antibody to the mannose receptor [18]. In addition to type I IFN, the PDC Key words: IRF-7 interferon producing cells PDC INTRODUCTION 1 Contributed equally to this study. 2 Contributed equally to this study. 3 Correspondence: UMDNJ-New Jersey Medical School, Dept. Pathology and Lab. Medicine, 185 So. Orange Ave., Newark, NJ bocarsly@umdnj.edu; also, Dr. Paula Pitha, The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins and The Cancer Research Building, Baltimore, MD parowe@jhmi.edu; Received June 3, 2003; revised July 11, 2003; accepted July 28, 2003; doi: /jlb Journal of Leukocyte Biology Volume 74, December

2 have been reported to produce IL-12 [19, 20], IL-6, and TNF- [21], as well as a number of inflammatory chemokines and chemokine receptors that stimulate migration of DC to lymph nodes ([5, 22 26] and Megjugorac et al. (unpublished data)). The molecular mechanisms that regulate the production of type I IFN and other inflammatory cytokines have yet to be determined; however, recent studies indicate the production of IFN, as well as other inflammatory cytokines in DC and monocytes/ macrophages, depends both on the distinct molecular patterns of the stimuli [27,28]. Type I IFNs are pleiotropic modulators of host resistance and play critical roles in the induction of an antiviral state, as well as have multiple immunomodulatory effects. Moreover, IFN- is an important regulator of DC subsets: It enhances the survival of PDC, augments the DC1-induced Th1 responses, up-regulates expression of costimulatory molecules on DC [29, 30], and enhances primary antibody responses and classswitching in a DC-dependent manner [31]. Induction of type I IFN gene expression is regulated at the transcriptional level. The family of interferon regulatory factors (IRF) plays important roles in the regulation of both IFNA and IFNB gene transcription in infected cells, as well as other cytokines. To date, nine cellular IRF have been identified [32, 33]. Three of the IRF (IRF-3, IRF-5, and IRF-7) function as direct transducers of virus-mediated signaling and play critical roles in the expression of type I IFN genes [34 39]. While IRF-3 is constitutively expressed in all cell types, expression of IRF-5 and IRF-7 has been primarily detected in lymphoid cells and can be further stimulated by type I IFN. In infected cells, these proteins are phosphorylated and transported to the nucleus where they interact with the transcription coactivator, transacetylases CBP/p300 [40 46]. It was shown that expression of IRF-3 is sufficient to support induction of IFNB, while IRF-5 or IRF-7 are needed for stimulation of IFNA gene expression in infected cells [36, 39]. Recent data suggest that the profile of IFN- subtypes produced in infected cells is determined by the levels of IRF-3, IRF-5, and/or IRF-7 in producing cells [36, 37, 39, 44, 46]. The aim of the current study was to characterize, for the first time, expression of IRF genes, as well as individual IFN- subtypes expressed in populations of PDC, MDDC, and monocytes and to determine whether the ability of the PDC to produce high levels of IFN- upon virus infection is associated with the high level of expression of distinct IRF or unique IFNA subtypes. We have shown that IRFs are constitutively expressed in monocytes, MDDC, and PDC isolated from human PBMC, but the PDC constitutively express higher levels of IRF-7 mrna and polypeptide than other cell types analyzed. Moreover, we demonstrate that the PDC respond more effectively to HSV stimulation than the other cell types and generally express a larger range of IFN- subtypes. MATERIALS AND METHODS Viruses Herpes simplex virus type I (HSV-1) strain 2931 and vesicular stomatitis virus (VSV) (the latter obtained from Dr. Nicholas Ponzio of New Jersey Medical School) were grown in VERO cells (originally obtained from the American Type Culture Collection) and titered by plaque forming assay. Sendai virus, strain Sendai/Cantell, was obtained from the American Type Culture Collection (Manassas, VA), grown in 10-day-old embryonic chicken eggs, and titered by hemagglutination assay using chicken RBC. Allantoic fluid was used in all IFN inductions using Sendai virus. All virus stocks were stored at 70 C until use. Cell lines GM-0459A cells (National Institute of General Medicine Sciences Human Genetic Mutant Cell Line Repository, Camden, NJ), a primary fibroblast cell line trisomic for chromosome 21, were grown in DMEM (JHR Biosciences, Lenexa, KS) supplemented with 15% FCS (HyClone, Logan, UT), 2 mm L-glutamine, 100 U/ml penicillin, 100 g/ml streptomycin (DMEM, 15%). VERO cells were grown in DMEM-10% FCS. Cytokines and Abs Recombinant IL-4 was purchased from R&D Systems (Minneapolis, MN), recombinant Sargramostim-Leukine (GM-CSF) from Immunex Corp. (Seattle, WA), and recombinant IFN- was from Schering-Plough (Kenilworth, NJ). APC-conjugated anti-hla-dr, PE-conjugated anti-cd123, PE-conjugated anti-cd3, PE-conjugated anti-cd19, FITC-conjugated lineage cocktail, APCconjugated CD11c, and PE-conjugated anti-cd14 were purchased from Becton-Dickinson (Sunnyvale, CA), and BDCA-4 conjugated immunobeads and anti-bdca-2 were purchased from Miltenyi Biotec (Auburn, CA). Unconjugated polyclonal rabbit, anti-irf-3, and anti-irf-7 and rabbit IgG (as a control for the IRF-3 and IRF-7 antisera) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). FITC Conjugated goat anti-rabbit IgG was purchased from PharMingen (Sunnyvale, CA). Preparation of peripheral blood mononuclear cells (PBMC) PBMC were isolated by Ficoll-Hypaque density centrifugation (Lymphoprep: Accurate Chemical and Scientific Co., Westbury, NY) from fresh heparinized peripheral blood obtained with informed consent from healthy volunteers. The human studies were approved by the IRB of the New Jersey Medical School. The PBMC were resuspended in MACS Buffer (PBS with 0.5% BSA and 2 mm EDTA (Sigma)) or in RPMI 1640 media containing 10% heat-inactivated fetal bovine serum (Hyclone Labs, Logan, UT), 2 mm L-glutamine, 100 U/ml penicillin, 100 g/ml streptomycin and 25 mm HEPES and enumerated electronically with a Coulter Counter Series Z1 (Coulter Electronics, Inc., Hialeah, FL). Preparation of peripheral blood monocytes (Mo) Monocytes were positively selected from PBMC using anti-cd14 labeled microbeads (Miltenyi), as described by the manufacturer. The purity of the CD14 populations was typically 98% or greater. Preparation of monocyte-derived dendritic cells (MDDC) MDDC were prepared by culture of plastic-adherent mononuclear cells in RPMI containing 1% autologous plasma, 500 IU/ml of IL-4 (R&D Systems, Minneapolis, MN) and 1000 IU/ml of GM-CSF. The cells were fed with IL-4 and GM-CSF-containing media on days 2, 4, and 6, and DC were collected on day 7 and evaluated by flow cytometry. Typically, 85% of cells were MDDC, as defined by the phenotype HLA-DR, CD3 -, CD14, CD16, CD19, CD20, CD56, and CD83. Enrichment of PDC PBMC were depleted of T, B, NK, red blood, and monocytic cells by using a modification of the Miltenyi Blood Dendritic Cell isolation kit. Briefly, PBMC were washed and resuspended in MACS buffer, treated with FcR-blocking reagent, labeled with haptenized murine antibodies against CD3, CD11b, and CD16 and placed at 4 C for 10 min. Cells were then washed twice, treated with antihapten microbeads, CD19 Microbeads, and Glycophorin A microbeads (to deplete B cells and red blood cells, respectively). After washing, the labeled cells were applied to an LD-negative selection column pretreated with 2 ml of 1126 Journal of Leukocyte Biology Volume 74, December

3 IRF-1S IRF-1AS IRF-2S IRF-2AS IRF-3S IRF-3AS IRF-4S IRF-4AS IRF-5S IRF-5AS IRF-7S IRF-7AS IRF-8S IRF-8AS IFNA-S IFNA-AS IFNA2-S IFNA2-AS IFNB-S IFNB-AS -actin-s -actin-as TABLE 1. MACS buffer and washed with 2 rounds of 1 ml of MACS buffer. Negatively selected DC were counted and evaluated for purity using flow cytometry. Typically, 60% of cells were PDC as determined by the phenotype HLA-DR, Lin, CD11c, and CD123. For preparation of the PDC that were purified to 99%, PDC were negatively enriched as described above; the populations were stimulated with HSV or mock stimulated for 5 h, and then the cells were finally purified using a further round of negative selection, followed by a final positive selection using the Miltenyi DC isolation kit. Following isolation, the purified PDC were analyzed for purity by flow cytometry, and mrna was extracted for RT-PCR. In some later experiments, PDC were positively selected using BDCA-4 microbeads and used for real-time PCR and immunocytochemistry. The purity of these populations was typically 98% or greater. Stimulation of cell with virus PBMC or enriched cell populations at /ml were stimulated with HSV (a multiplicity of infection of 1), SV (16 HAU/ml), or IFN- (10,000 IU/ml) and used for RNA purification, flow cytometry, or harvesting of supernatants. Interferon bioassay Primers for Amplification of Human IRFs and IFN Genes a 5 -GCCAGTCGACGAGGATGAGGAAGGGAA-3 5 -CCAGCGGCCGCCTGCTACGGTGCACAGG-3 5 -CCAGTCGACTACCTCAGCAACATGGGG-3 5 -CCAGCGGCCGCGGCTTAACAGCTTGAC-3 5 -CGGAAGCTTCTGAAGCGGCTGTTGGTG-3 5 -GTGCTCGAGACCATGAGGAGCGAGGGC-3 5 -CCAGTCGACGCAAGCTCTTTGACACAC-3 5 -CCAGCGGCCGCCTTTTCATTCTTGAATAG-3 5 -GCCTTGTTATTGCATGCCAGC-3 5 -AGACCAAGCTTTTCAGCCTGG-3 5 -TGCAAGGTGTACTGGGAG-3 5 -TCAAGCTTCTGCTCCAGCTCCATAAG-3 5 -GCCGAATTCTCCGAGAGCTGCAGCA-3 5 -CGGCTCGAGGCTTAGACGGTGATC-3 5 -GTACTGCAGAATCTCTCCTTTCTCCTG-3 5 -GTGTCTAGATCTGACAACCTCCCAGGCACA-3 5 -GTACTGCAGAATCTCTCTTTTCTCCTG-3 5 -GTGTCTAGATCTGACAACCTCCCAGGCACA-3 5 -TTGTGCTTCTCCACTACAGC-3 5 -CTGTAAGTCTGTTAATGAAG-3 5 -ACAATGAGCTGCTGGTGGCT-3 5 -GATGGGCACAGTGTGGGTGA-3 a Primers were synthesized at Life Technologies Custom Primers. Restriction sites were underlined. S sense strand; AS antisense strand. IFN bioassays were performed on supernatants harvested from microtiter plates using a cytopathic effect reduction assay with GM-0459A (GM) cells infected by vesicular stomatitis virus (VSV) as the challenging virus. GM cells were plated in 96-well flat bottom tissue culture-treated plates. An IFN- standard starting at 100 IU/ml (NIAID Standard G ) was included in each assay, and the IFN in the test supernatants was quantified relative to the NIAID reference standard. Reverse transcription PCR analysis RNA was isolated by using an RNeasy Mini Kit (Qiagen, Santa Clarita, CA). One microgram of DNase-treated total RNA was reverse-transcribed to cdna with oligo(dt) primers in a 30 l reaction. From this mixture of cdnas, human IFNA (consensus primers designed to recognize all human IFNA subtypes), IFNA2 (primers designed to specifically recognize the IFNA2 subtype), IFNB, IRF-1, IRF-2, IRF-3, IRF-4, IRF-5, IRF-7, IRF-8, and -actin cdna were amplified by polymerase chain reaction (PCR) using the primer sets indicated in Table 1. The IFNA and IFNA2 specific primers contain a PstI (sense primer) and an XbaI (antisense primer) restriction enzyme recognition sequence to facilitate the cloning of the PCR fragments. The conditions of the PCR assay, described previously [36, 39], gave a linear amplification within the amplification conditions. The amplified IFNA and IFNA2 fragments were cloned into pbluescript KS II (Stratagene) at the PstI/XbaI site. The clones were randomly chosen and individual IFNA subtypes identified by sequencing of the PCR-amplified fragments, as described [36, 47]. Real time reverse transcription PCR analysis RNA was isolated using RNeasy kit (Qiagen) and each RNA sample was treated with RQ1 RNase-free DNase (Promega) according to the manufacturer s instructions. Reverse transcription was carried out using Oligo (dt) at 42 C for 30 min and 95 C for 5 min. Real-time PCR assays were carried out with the ABI PRISM 7700 Sequence Detector using SYBR Green I fluorogenic dye. The specificity of the reaction was examined by dissociation curve analysis, and data normalization was carried out as described [48] using the geometric means of three housekeeping genes. The following primers were used to measure the IRF and housekeeping gene expression in RNA obtained from highly purified monocytes and PDC: IRF3, 5 - ACCAGCCGTGGACCAA- GAG-3 and 5 -TACCAAGGCCCTGAGGCAC-3 ; IRF4, 5 -CGGGCAAG- CAGGACTACAAC-3 and 5 -CCTTTAAACAGTGCCCAAGCC-3 ; IRF5, 5 -AAGCCGATCCGGCCAA-3 and 5 -GGAAGTCCCGGCTCTTGTTAA-3 ; IRF7, 5 -TGGTCCTGGTGAAGCTGGAA-3 and 5 -GATGTCGTCATAGAG- GCTGTTGG-3 ; Beta-2-microglobulin, 5 -TGCTGTCTCCATGTTTGATG- TATCT-3 and 5 -TCTCTGCTCCCCACCTCTAAGT-3 ; Ubiquitin C AT- TTGGGTCGCGGTTCTTG-3 and 5 - TGCCTTGACATTCTCGATGGT-3 ; and Beta actin 2,., 5 -ATTGCCGACA GGATGCAGAA-3 and 5 -CAAGATCATT GCTCCTCCTG AGCGCA-3. Intracellular flow cytometry with surface phenotyping Intracellular flow for IRF-3 and IRF-7 was performed using indirect staining combined with surface staining. Briefly, PBMC or MDDC at a concentration of cells/ml were stimulated with either HSV-1 strain 2931 (MOI of 1) or 10,000 IU/ml recombinant IFN- 2b or mock stimulated for 6hat37 Cin5% CO 2. The cells were washed with cold 0.1% BSA (Sigma, St. Louis, Mo.) in PBS, blocked with 5% heat-inactivated human serum, and stained with the appropriate fluorochrome conjugated surface cellular marker antibodies for 20 min at 4 C. The PDC population was defined as cells that are HLA-DR, CD123 bright, the T cells as CD3, the B cells as CD19, the monocytes as CD14, the DC1 as lineage-negative, HLA-DR, CD-11C, and the MDDC as HLA-DR, lineage. The cells were washed and fixed overnight with 1% paraformaldehyde in PBS at 4 C. The cells were washed twice with 2% FCS in PBS and permeabilized with 0.5% saponin (Sigma) in 2% FCS-PBS for 30 min at RT. Cells were pelleted and incubated with rabbit anti-hu IRF-7 (400 ng) or IRF-3 (2 g) in a final volume of 50 l for 30 min at RT. Normal rabbit IgG was used for isotype controls at the same concentration as that for anti- IRF-3 or IRF-7, respectively. The cells were washed twice with permeabilization buffer, resuspended, and incubated with goat anti-rabbit IgG-FITC for 30 min. The cells were washed, resuspended in 1% paraformaldehyde in PBS, and analyzed using the FACSCalibur flow cytometer with CellQuest Analysis software (BD Biosciences). Immunofluorescence PDC and monocytes were purified from PBMC by positive selection with BDCA-4 and CD14 microbeads, respectively, as described above. Cells were washed with 0.1% BSA in PBS, stained with anti-bdca2 PE, and fixed with 1% paraformaldehyde overnight. Cells were adhered to slides via cytospin and permeabilized with wash buffer (0.2% Triton X-100 in PBS). Cells were blocked with wash buffer containing 3% bovine serum albumin and 10% normal goat serum for 30 min and incubated with the primary antibody; rabbit anti-human IRF-7 at a 1:20 dilution, using normal rabbit IgG as an isotype control (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). After washing, cells were incubated with secondary antibody (FITC-conjugated goat anti-rabbit IgG), and DAPI (4,6-diamidino-2-phenylindole) at 50 ng/ml. Following extensive washing, slides were viewed under fluorescence with an Olympus fluorescence microscope. Izaguirre et al. IRF and IFN-alpha expression in human dendritic cells 1127

4 RESULTS Expression of type I IFN mrna and protein in virus stimulated populations of myeloid and dendritic cells HSV-1 and the Sendai virus are prototypical inducers of type I IFN in PDC and monocytes, respectively [3]. In the current study, we have compared the ability of PBMC, a population of enriched PDC, monocytes, and MDDC to produce biologically active IFN in response to these two viruses. The PDC used in these studies were enriched by negative selection rather than by positive selection since positive selection (with the final purification based on their expression of CD4) yielded cells that had decreased IFN-producing capacity that could be attributed to cross-linking of the CD4 (Olshalsky et al., in preparation). As previously demonstrated, the population of enriched PDC was the most potent producer of type I IFN in response to both HSV-1 and the Sendai virus, with each of these stimuli inducing IU/10 6 cells/ml of biologically active type I IFN (Fig. 1). The PDC purity, as measured by flow cytometry (Lineage-negative, HLA-DR, CD123 bright ) was 60%, compared with 0.2% in PBMC. Highly purified monocytes (98%) stimulated with Sendai virus produced IU/ml of IFN, while HSV-1 challenge of these cells induced only a minimal amount of IFN. Similarly, MDDC derived from monocytes cultured for 7 days in the presence of IL-4 and GM-CSF consistently produced relatively small amounts of IFN ( 3,000 IU/ml) in response to the Sendai virus, but failed to produce IFN in response to HSV-1. Although the results shown are for a single donor, they are typical of results obtained from many different donors. Our previous results indicated that the majority of the biologically active IFN produced by PBMC in response to HSV-1 and Sendai virus infection was IFN-, as the activity could be neutralized by the addition of anti-ifn- antiserum [14]. However, we and others have found a small but significant production of IFN- by PDC, as well as relatively high levels of IFNB mrna [4, 49, 50]. To determine whether both IFNA and IFNB mrna are expressed in the enriched cell populations following either mock stimulation or stimulation with HSV-1 or the Sendai virus, mrna was isolated at 5 h postexposure to virus and analyzed by semiquantitative RT-PCR using universal primers that amplify all of the IFNA subtypes or primers detecting IFNB (Table 1, Fig. 2A and 2B). Figure 2A depicts relative levels of type I IFN obtained from a different donor for each of the 3 cell types: monocytes, MDDC, and PDC. While high levels of IFNA mrna were detected in the Sendai virus but not in HSV-1-challenged monocytes, there was a low constitutive expression of IFNB mrna in these cells that was enhanced upon stimulation with Sendai virus but not HSV-1. In a similar manner, only Sendai virus-induced IFNA mrna from MDDC, yet IFNB mrna was also induced by HSV-1 but at a much lower level than in Sendai virus-stimulated cells. In the enriched PDC population of cells, both HSV-1 and the Sendai virus induced high levels of IFNA and IFNB mrna. Taken together, these data correlate with the production of biologically active type I IFN shown in Fig. 1. Because each of the cell populations analyzed in these experiments (Fig. 2A) were prepared from a different PBMC donor, we had to eliminate the possibility that the differences in type I IFN gene expression were a result of genetic variability. We therefore prepared the cell populations from PBMC of a single donor and repeated the analysis. Figure 2B reveals that HSV-1 and the Sendai virus both induce high levels of IFNA mrna in PBMC, but only the Sendai virus induced IFNB mrna. The pattern of IFNA and IFNB expression from the other cell populations from a single donor was the same as seen in cell populations from multiple donors. These data indicate that the observed difference in the expression of IFNA and IFNB mrna are representative of a cell type variation and not a genetic variation. Differential expression of IRFs in subpopulations of dendritic cells and monocytes Because the PDC are the most potent type I IFN-producing cell population, we asked whether the high production of type I IFN in these cells could be attributed to the high levels of IRFs in these cells. We have shown previously that both IRF-5 and IRF-7 play critical roles in the induction of IFNA genes while expression of IRF-3 is sufficient only for induction of IFNB Fig. 1. Analysis of antiviral activity in virally stimulated populations. Levels of IFN- and IFN- synthesized in response to HSV-1 or Sendai virus (SV) stimulation of PBMC, PDC, monocytes, and MDDC are presented. PBMC or enriched populations of PDC (obtained by negative selection), monocytes, and MDDC were stimulated overnight with either HSV or SV, and supernatants were tested for antiviral activity in a cytopathic effect reduction assay. Data are expressed as IU/ml of antiviral activity. Data are representative of 3 similar experiments Journal of Leukocyte Biology Volume 74, December

5 Fig. 2. Differential expression of IRFs and type I IFN in subpopulations of monocytic and dendritic cells. (A) The expression of IRFs, IFNA, and IFNB genes in mock-stimulated and virus-infected monocytes, MDDC, and PDC isolated from PBMC were examined; a separate donor was used for each cell type. PDC were enriched by negative selection as described in Materials and Methods. Cell populations were either mock-stimulated (lane 1), infected with HSV-1 (lane 2), or infected with the Sendai virus (lane 3) for 5 h. Total RNA was isolated, and RT-PCR analysis was carried out using primers described in Table 1. Amplification of -actin mrna was used as an internal loading control. (B) The expression of IRFs, IFNA, and IFNB gene transcripts in PBMC, monocytes, MDDC, and PDC from a single donor. Lane 1 corresponds to freshly isolated cell subpopulations, lane 2, mock-stimulated for 5 h, lane 3, HSV-1- stimulated for 5 h, and lane 4, Sendai virusstimulated for 5 h. [36, 39, 51]. Because the expression of IRF family members has not been previously characterized in connection with type I IFN production in myeloid and dendritic cell populations, we analyzed expression of IRF genes by semiquantitative RT-PCR in mock-stimulated and virus-stimulated cells. Primers used for these studies are shown in Table 1. Data from two experiments are shown in Fig. 2A and 2B. As indicated above, the data shown in Fig. 2A were derived from a separate donor for Izaguirre et al. IRF and IFN-alpha expression in human dendritic cells 1129

6 each of the cell types, whereas the data from Fig. 2B was derived from a single donor. As shown in Fig. 2A, IRF-1, IRF-2, IRF-3, IRF-5, and IRF-7 mrna were detected in all the cell populations analyzed, and the virus stimulation was not required for the expression of the IRF genes. Interestingly, for this donor, IRF4 was present only in the PDC. To eliminate the possibility that the expression of IRF genes in cell populations studied was modulated by in vitro culture, we analyzed the levels of IRF transcripts both in freshly isolated and cells cultured in vitro for 5 h. When the cell populations derived from PBMC of a single donor were analyzed (Fig. 2B), some differences were observed. MDDC freshly harvested from the flasks and freshly isolated PDC did not show expression of IRF-2, IRF-4 or IRF-5, and IRF-1 or IRF-2, respectively. Compared with the results from multiple donors, the constitutive expression of IRF-4 in this single donor was not restricted to PDC, but was also weakly detected in PBMC and virus-stimulated MDDC. Furthermore, expression of IRF-5 was not detected in mock-stimulated MDDC. These data indicate that the expression of several IRFs is enhanced/induced by in vitro culture. Although the levels of -actin indicate that similar amounts of RNA were analyzed in the different samples from Fig. 2, this analysis is only semiquantitative. Therefore, in order to carry out more quantitative analysis of the IRF expression, we designed real-time PCR primers for the IRF-3, IRF-4, IRF-5, and IRF-7 and compared the expression of each of these in highly purified monocytes and PDC freshly isolated from five separate donors. The data were normalized vs. 3 housekeeping genes as described [48] (Fig. 3). Taken together, there were no significant differences in expression of IRF-3, IRF-4 or IRF-5 expression in donors as determined by Student s t analysis. However, for IRF-5, all but one of the doors expressed higher levels in the PDC rather than the monocytes. In contrast, the PDC expressed significantly more IRF-7 mrna than the monocytes, with all five donors showing the same pattern. Detection of IRF-3 and IRF-7 in subpopulations by intracellular flow cytometry The RT-PCR and real-time PCR results strongly indicate that PDC express high levels of the broad range of IRFs. Of particular interest was the expression of IRF-3 and IRF-7 since these are specifically associated with the expression of IFNA genes in both human and mouse [34, 35, 38, 39, 51 53]. To study the expression of these IRFs at the single-cell level in DCs, as well as in other PBMC subsets, we used intracellular staining for IRF-3 or IRF-7 protein combined with cell surface staining. This technique has frequently been used in the study of intracellular cytokine expression in specific cell subsets but has not been applied to studies of IRFs. For these experiments, we used monoclonal antibodies to define cell populations within freshly isolated PBMC. The cells were either mockstimulated, stimulated with HSV-1, or treated with 10,000 IU of exogenous IFN- for 5 hours. Because IFNA was shown previously to up-regulate IRF-7 levels, we used treatment with high levels of exogenous IFN- as a positive control for the up-regulation of IRF-7 levels. The stimulated cells were surface stained for identification of cell populations, followed by fixation and intracellular staining for IRF-7. Fig. 3. Real-time RT-PCR analysis of IRF expression in PDC and monocytes. Monocytes and PDC were prepared from 5 individual donors using positive selection with anti-cd14 and anti-bdca-4 microbeads, respectively, and RNA was extracted. The real-time RT-PCR was carried out as described in Materials and Methods and normalized to the mean of 3 housekeeping genes. (A) IRF-3 in monocytes (solid bars) vs. PDC (hatched bars); (B) IRF-4; (C) IRF-5; and (D) IRF Journal of Leukocyte Biology Volume 74, December

7 IRF-7 was expressed at relatively low levels in monocytes, T cells, B cells, and MDDC, but was expressed constitutively at very high levels in PDC (Fig. 4). The addition of exogenous IFN- resulted in a slight up-regulation of IRF-7 in all cells, including the PDC (mean fluorescence intensities of 412, 437, 390, and 728 for the PDC at 0 h or after mock-stimulation, HSV-stimulation, or IFN-stimulation of the PDC, respectively). Although we did not observe up-regulation of IRF-7 in the PDC after incubation with HSV-1 for 5 hours, subsequent studies indicated that IRF-7 was reproducibly up-regulated after 9 h of incubation with HSV-1, most likely due to the virus-induced synthesis of IFN- (data not shown). Figure 5 shows expression of IRF-3 in the same subpopulations of cells described in Fig. 4. PDC expressed low but significant levels of IRF-3 that were not dramatically changed by infection with HSV-1 or treatment with IFN-. It has previously been shown that IRF-3 levels are not up-regulated by IFN- or viral infection [34]. Likewise, T cells and B cells expressed low levels of IRF-3. MDDC expressed slightly higher levels, and monocytes expressed little, if any, IRF-3. MDDC are reported to be similar to peripheral blood myeloid DC1, but dissimilar in other ways. To rule out the possibility that phenotypic differences between the MDDC and PDC were a function of in vitro derivation of the MDDC from monocytes, we analyzed expression of IRF-3 and IRF-7 in PDC vs. DC1 by intracellular flow. DC1 within PBMC was identified as cells that express CD11C and HLA-DR in the absence of lineage-specific markers. Peripheral blood DC1, like MDDC and unlike PDC, did not express high levels of IRF-7 (Fig. 6). To visually confirm the high levels of IRF-7 expression in PDC as compared with monocytes, enriched PDC and monocytes were stained with BDCA2 or CD14, then permeabilized and stained for IRF-7. In agreement with the real-time PCR data and intracellular flow cytometry, the freshly isolated PDC expressed high levels of IRF-7 compared with the monocytes (Fig. 7). In contrast, and also in agreement with our other results, PDC did not preferentially express IRF-3 as compared with monocytes (data not shown). IRF-4 is up-regulated in HSV-stimulated PDC Because IRF-4 expression has been reported in T and B cells and monocytes [54, 55] and shown here in unstimulated PDC, we wanted to confirm that the up-regulation in IRF-4 expression by virus stimulation seen in PDC (Fig. 2) was not due to contaminants in our enriched PDC population. Although we attempted to use a commercial IRF-4 antiserum as we had the antisera to IRF-3 and IRF-7, the available antibody did not work in intracellular flow. We, therefore, analyzed IRF-4 expression by RT-PCR in PDC that were enriched after viral stimulation. To accomplish this, we used a technique that purified PDC but did not compromise their ability to function, as is problematic with positively selected PDC. Partially purified PDC were stimulated or mock-stimulated with HSV for 6 h, followed by an additional round of negative selection plus Fig. 4. IRF-7 expression in subpopulations of PBMC and MDDC. PBMC were stained for lineage-specific cell surface markers along with intracellular anti-irf-7 as described in Materials and Methods. The data shown are for PBMC from a single donor (first four columns), as well as for MDDC derived from another donor as described. The top row demonstrates the gating criteria for determination of PDC (HLA- DR, CD123 ), monocytes (CD14 ), T cells (CD3 ), B cells (CD20 ), and MDDC, respectively, for the time 0 IRF-7 determination. Similar matched gates were used for each of the other time points and treatment conditions (not shown). For IRF-7 determination, cells were stained with anti-irf-7 (gray histograms) or matched rabbit IgG isotype control (black histograms) with FITC goat-anti-rabbit as the secondary stain. IRF-7 panels are shown for each cell population when freshly isolated, and after 6 hours mock-stimulation, stimulation with HSV or stimulation with rifn- at 10,000 IU/ml. Data are representative of 3 similar experiments. Izaguirre et al. IRF and IFN-alpha expression in human dendritic cells 1131

8 Fig. 5. IRF-3 expression in subpopulations of PBMC and MDDC. PBMC were stained for lineage-specific cell surface markers along with intracellular anti-irf-3, as described in Materials and Methods. The data shown are for PBMC from a single donor (first four columns), as well as for MDDC derived from another donor as described. The top row demonstrates the gating criteria for determination of PDC (HLA- DR, CD123 ), monocytes (CD14 ), T cells (CD3 ), B cells (CD20 ), and MDDC, respectively, for the time 0 IRF-3 determination. Similar matched gates were used for each of the other time points and treatment conditions (not shown). For IRF-3 determination, cells were stained with anti-irf-3 (gray histograms) or matched rabbit IgG isotype control (black histograms) with FITC goat-anti-rabbit as the secondary stain. IRF-3 panels are shown for each cell population when freshly isolated, and after 6 hours mock-stimulation, stimulation with HSV, or stimulation with IFN- at 10,000 IU/ml. Data are representative of 3 similar experiments. a final positive selection of CD4 cells using the Miltenyi DC isolation kit. This procedure yielded PDC that were uniform in their scatter plot (Fig. 8A), were lineage-negative but HLA- DR (Fig. 8B), and were 99% CD123, CD11c (Fig. 8C), a phenotype typical of PDC. The flow profiles shown are for the HSV-stimulated PDC; virtually identical flow profiles were seen for the mock-stimulated cells (data not shown). Both the mock-stimulated and HSV-stimulated PDC expressed IRF-4 Fig. 6. Expression of IRF-3 and IRF-7 in peripheral blood DC populations. (A) PBMC were labeled with a lineage cocktail and anti-hla-dr. HLA-DR. (B) Lineage-negative cells were gated for expression of CD123 on PDC (as described in other figures) or CD11c to define myeloid DC (DC1). (C) IRF-3; (D) IRF-7 expression in DC1; (E) IRF-3; and (F) IRF-7 expression in PDC. Shaded histograms are isotype controls, whereas open histograms are for IRF staining Journal of Leukocyte Biology Volume 74, December

9 Fig. 7. Constitutive high expression of IRF-7 in PDC. PDC and monocytes were enriched by positive selection using anti-bdca-4 and anti-cd14 microbeads, respectively, then surface stained for CD14 (monocytes) and BDCA-2 (PDC). Cells were then permeabilized and stained for IRF-7 or an isotype control followed by an anti-ig FITC and DAPI (to stain the nucleus). Cells were visualized under a fluorescence microscope. (Fig. 8D), with higher levels in the virus-stimulated than unstimulated sample, thus confirming the results in Fig. 2. Expression of IFNA subtypes in virally stimulated subpopulations of IFN-producing cells IFNA subtypes are expressed at different levels in distinct populations of lymphoid cells [56 61]. Because the antiviral activity of all the IFNA subtypes is not identical, the observed difference in the IFNA titers induced in PDC and other cell types may be a reflection of the subtypes produced. Although a number of studies have found functional differences between the different IFN- subtypes, the subtypes expressed in different populations of myeloid and dendritic cells stimulated with different viruses have not been identified. We therefore characterized the profile of IFNA genes expressed in HSV-1 and the Sendai virus-stimulated cell subpopulations analyzed above. The individual IFNA subtypes were identified by sequencing the cloned IFNA cdna fragments as described in Materials and Methods. Figure 9A represents the data derived from subpopulations isolated from the individual donors shown Fig. 8. Expression of IRF-4 mrna in purified PDC. PDC that were enriched by negative selection were stimulated with HSV at an MOI of 1 or mock-stimulated for 5 h. Following stimulation, the PDC were purified by another round of negative selection followed by positive selection with anti-cd4 microbeads as per the Miltenyi DC isolation kit. mrna was extracted as described and was DNase treated. RT-PCR was carried out with the IRF-4 and -actin primers described in Table 1. No bands were seen in conditions without RT. (A) Forward and sidescatter plot for the purified PDC population; (B) Lineage cocktail vs. HLA-DR for the purified PDC; (C): CD123 vs. CD11c for the purified PDC; (D) RT-PCR for IRF-4 and -actin for the HSV-stimulated and mock-stimulated PDC populations. Izaguirre et al. IRF and IFN-alpha expression in human dendritic cells 1133

10 Fig. 9. Expression of IFNA subtypes in virally stimulated subpopulations of IFN-producing cells. The IFNA cdna fragments were amplified from RNA of monocytes, MDDC, and PDC infected with the Sendai virus or HSV-1 for 5 h. The mrna was from the same population of cells as described in Figure 2. The primers used correspond to the regions of IFNA genes that are conserved in all subtypes (Table 1). The PCR-amplified DNA fragments were cloned into pbluescript KS II plasmids, cdna insert isolated from randomly selected clones (20 40), and sequenced. The DNA sequences obtained were compared with sequences of individual IFNA genes present in GenBank. The data are expressed as the frequency of the clones representing each specific subtype. For PDC, we analyzed IFNA subtype distribution from both HSV-1 and the Sendai virus-infected cells, whereas for monocytes and MDDC, only the Sendai virus responses were examined since HSV-1 infection resulted in only minimal IFN- induction. (A) Expression of IFNA subtypes in virally stimulated cell subpopulations. Data shown were derived from a separate donor for each of the cell types. (B) Expression of IFNA subtypes in virally infected cell subpopulations isolated from a single donor. in Fig. 2A, whereas the data in Fig. 9B are from cells derived from the single donor whose IRF and IFN expression were shown in Fig. 2B. The data in Fig. 9A show that IFNA1 was the major subtype expressed in the Sendai virus-infected monocytes and MDDC. In addition, the monocytes and MDDC both expressed IFNA2, A6, A8, and A10. The monocytes also expressed IFNA4, and the MDDC expressed IFNA5 and IFNA14. The Sendai virus-stimulated PDC induced a wide range of IFNA subtypes. Although IFNA1 was again the major subtype, these cells also effectively expressed IFNA2, A14, and A16, in addition to lower frequencies of IFNA4, A5, A8, A10, A17, and A21. The HSV-1 stimulated PDC induced 1134 Journal of Leukocyte Biology Volume 74, December

11 IFNA1, A7, A10, A14, A16, A17, and A21, but no IFNA2. Thus, the Sendai virus induced a broader range of IFNA subtypes in PDC than HSV-1. In Fig. 9B, the results of all of the cell types coming from a single donor are shown. It should be noted that fewer clones were derived for the analysis shown in Fig. 9B than in Fig. 9A, but several observations can be made. Interestingly, the cells derived from this donor expressed significantly fewer IFNA subtypes, but again IFNA1 was the major subtype induced in the Sendai virus-infected monocytes and MDDC. Although the profile of IFNA subtypes in monocytes was similar to that observed in the previous donor, the MDDC expressed in addition to IFNA1 only IFNA8 and A10. In infected PDC, Sendai virus induced IFNA5 as the major IFNA subtype in addition to IFNA1, A8, A10 and A14. In HSV-1 infected cells, IFNA1 was the major subtype and IFNA5, A8, A10, A14, and A17 were induced to about the same level. These results indicate that the expression of IFNA subtypes in response to viral infection varies between individual donors, but that, in general, PDC express a wider range of IFNA subtypes. We and others have previously suggested that IFNA2 is effectively produced by stimulation of PBMC with the Sendai virus or HSV [50]. However, IFNA-2 was expressed only as a minor subtype in the samples analyzed in this study. To eliminate the possibility that the universal primers used in the current study were unable to detect IFNA2 transcripts, we used IFNA2-specific primers (Table 1). However, IFNA2-specific primers detected again IFNA1 as the major subtype. In addition to IFNA2, these primers also amplified a number of other IFNA subtypes (data not shown). Thus, even the primers designed to specifically detect IFNA2, detected other IFNA subtypes. It is important to note that levels of IFNA2 transcripts detected by both the universal and IFNA2-specific primers were nearly identical indicating there was not selective amplification or lack of amplification of certain subtypes by the universal primers. DISCUSSION Although many cell types can produce type I IFNs in response to viral infection, PDC are uniquely potent IFN-producing cells. In this study, we have analyzed and compared the expression of IFNA genes in primary human monocytes, MDDC, and PDC and compared these to the expression levels of IRF that are critical for the induction of type I IFN. Previous work regarding IFN subtype distribution has primarily been carried out using tissue culture lines. Primary cell populations were used for the current studies to determine how PDC differ from other type I IFN-producing cells and to develop data parallel to the IRF expression data. Furthermore, using primary cells for this analysis provided guarantees that we are examining the biologically important decisive factors affecting the cell response to viral infection but also introduced genetically based variability into the results. The results of a semiquantitative RT-PCR analysis demonstrated that most of the IRF genes are expressed constitutively in monocytes, MDDC, and PDC. However, the real-time PCR results demonstrated that as compared with monocytes, PDC constitutively express significantly higher levels of IRF-7. IRF-3 and IRF-4 levels were not significantly different between the monocytes and PDC; however, for IRF-5, all but one of the donors expressed higher levels of IRF-5 in the PDC. No constitutive expression of IFNA could be detected in any of the cell populations, whereas in some donors, a low constitutive expression of the IFNB gene could be detected in monocytes and MDDC, but not in PDC. These results suggest that the production of high levels of IFN- in virus-stimulated PDC is not a result of priming by autocrine IFN. Interestingly, in vitro culturing of monocytes activated low levels of IFNB gene expression, whereas this activation was not observed by in vitro culturing of the PDC. Whereas in PDC, both HSV-1 or the Sendai virus were equally effective inducers of IFNA and IFNB genes and biologically active type I IFN, in monocytes and MDDC, the Sendai virus was more effective than HSV-1. These data demonstrate that IFN- production in response to virus stimulation is determined not only by the DC subset but also by a distinct recognition of some component of the stimulating virus. This failure of MDDC to produce IFN in response to HSV does not result from an inability of HSV to infect MDDC; indeed, our results obtained using a green fluorescent protein expressing HSV construct demonstrate that MDDC but not PDC are productively infected by HSV (data not shown). Rather, these results suggest that distinct pathways for the induction of IFN may be operative in these distinct populations. IRF-3 and IRF-7 have been shown to play critical roles in the induction of type I IFN genes. The IFN-mediated stimulation of IRF-7 expression was suggested to be essential for the expression of a broad spectrum of IFNA genes [37, 52], whereas IRF-3 was found to be sufficient to induce expression of IFNB [51], and IFNA4 in mouse embryo fibroblasts but not in human fibroblasts [37]. Primary mouse fibroblasts derived from genetically modified mice that had IFN- / mediated signaling impaired as a consequence of STAT1 or IRF-9 deletion were not able to express IFN- effectively in response to viral infection [37, 38, 62]. As shown in this study, IRF-7 mrna is expressed constitutively at high levels in PDC in the absence of autocrine production of IFN- /. Moreover, when IRF-7 protein was analyzed by intracellular flow cytometry, the levels detected in PDC were significantly higher than those detected in monocytes, MDDC or peripheral blood DC1, and PDC had much higher levels of IRF-7 than monocytes, as shown by immunofluorescence microscopy. IFN- treatment increased IRF-7 protein levels in all of the cells examined. In contrast, no significant difference was observed in the intracellular levels of IRF-3 proteins expressed in PDC, monocytes, and MDDC, and in agreement with a previous report [34], IFN- did not stimulate expression of IRF-3. These results suggest that high constitutive levels of IRF-7 in PDC that are available for activation by virus may contribute to effective IFN- synthesis in PDC. The contribution of other IRF family members expressed in PDC to the activation of interferon genes remains to be determined. The role of IRF-1 in IFN induction has been questioned, as genetically modified mice lacking IRF-1 can still express type I IFN genes in response to the infection [32]. However, it was shown recently that IRF-1 is a component of Izaguirre et al. IRF and IFN-alpha expression in human dendritic cells 1135

12 the transcriptional enhanceosome assembled both on the IFNA and IFNB promoter region [44, 46, 63] and thus may function in cooperation with IRF-3 or IRF-7. The deletion of IRF-4 transcripts in the highly purified PDC that were enhanced upon virus stimulation is a novel finding. It was previously shown that IRF-4 is expressed in antigen-activated T and B cells and in macrophages [54] but not in IFN-stimulated cells [55]; however, the role of IRF-4 in the induction of IFN genes has not yet been addressed. It should be noted that the expression of some IRF in MDDC, particularly IRF-4, IRF-5, and IRF-8, was donordependent, suggesting the possible role of genotype in the regulation of these genes. Because MDDC were derived by in vitro cultivation of monocytes with IL-4 and GM-CSF, we cannot distinguish at this point whether the genotype affects direct expression of these IRF or the differentiation of monocytes to MDDC. The role of IRF-5 in the induction of IFNA genes was recently demonstrated [36]. Although the relative levels of IRF-5 mrna were similar in PDC, MDDC and monocytes using semiquantitative PCR, the real-time PCR indicates that 4 out of 5 of the donors showed higher levels of IRF-5 expression in the PDC than the monocytes. The levels of IRF-5 protein in these cells still needs to be determined and awaits the development of appropriate anti-irf-5 antibodies. Although both IRF-5 and IRF-7 play important roles in the induction of type I IFN genes [36, 39, 64], recent data have revealed that IRF-5 and IRF-7 may not act in a cooperative manner to enhance IFN- production; instead, these IRF family members appear to compete for binding at the level of IFNA promoters [46]. In cells expressing high levels of IRF-5, IRF-5 but not IRF-7 was the major component of the IFNA enhanceosome, as determined by the ChIP assay. In contrast, when the relative levels of IRF-7 were higher than IRF-5, IRF-7 bound to the IFNA promoters, yet binding of IRF-5 was not completely excluded. It is important, however, to note that analysis of IRF bound to the endogenous IFNA promoters yields an overall picture of IRF binding to the majority of IFNA promoters. This assay does not allow one to distinguish quantitatively IRF binding to the promoters of individual endogenous genes. The possibility that binding of IRF-5 and IRF-7 to endogenous IFNA promoters represents binding to distinct IFNA promoters cannot yet be excluded. Furthermore, these data were generated in an in vitro cellular system where very high levels of IRF-5 and IRF-7 protein expression were obtained. The question of whether these two factors cooperate in an endogenous cellular system, where expression levels vary such as in a PDC population, to induce high levels of type I IFN after virus stimulation remains to be addressed. Altogether, these results suggest that the high levels of IRF-7 protein in PDC contributes to the high production of IFN- in these cells. Takauji et al. recently demonstrated that IRF-7 is expressed in PDC and is regulated in a MAP-kinase dependent manner [65]. These authors postulate that CpG- DNA can lead to the increased transcription of IRF-7 that is necessary for the induction of IFN-. However, our data clearly indicate that IRF-7 is selectively expressed at very high levels in PDC compared with other cell populations, thus making the PDC immediately able to respond vigorously to virus stimulation. The high levels of biologically active IFN- synthesized in PDC could be a reflection of a broader spectrum of IFNA subtypes synthesized in these cells when compared with monocytes or MDDC. The profile of IFNA subtypes expressed in the Sendai virus-stimulated monocytes, MDDC, and PDC from different donors indicate that the IFNA1 gene was the major subtype expressed, which is in agreement with published data in other cell types [39]. In addition to IFNA1, the PDC expressed about 5 8 other IFNA subtypes in response to HSV. Moreover, the profile of IFNA subtypes induced by the Sendai virus in PDC was generally broader than in monocytes or MDDC. An exception to this observation was the finding that in one donor, the Sendai virus stimulated fewer IFNA genes in all cell types examined, and IFNA5 was the major subtype expressed in PDC. In contrast, for the same donor s PDC population, HSV-1 stimulated IFNA1 as the major subtype expressed. These data indicate that in the human population there may be a significant variation in the IFNA genes expressed in response to viral infection. This is not a completely unexpected finding since different inbred strains of mice show large differences in their ability to produce IFN- and IFN-, and several virus-specific loci (Ifn) that regulate the interferon synthesis have been identified [66, 67]. Strain-specific expression of IFNA genes has also been observed [68, 69]. Thus, a broad baseline needs to be established before any generalization is made about the role of individual IFN- subtypes in innate responses to viral infection. However, it is known that although the different IFN- subtypes signal through the same receptor, the biological activity of the subtypes is not identical with some IFN- subtypes showing higher antiviral activity than others [70 73] or differential immunoregulation [74]. Thus, it is possible that the main function of the IFN production by PDC may not be to inhibit viral replication but, rather, to initiate cell-mediated immune responses [30, 75]. In conclusion, this is the first comprehensive analysis of IRF expression in primary human monocytes, MDDC, and high IFN- -producing PDC. The results show that (1) most of the IRF genes are expressed constitutively in all three cell types, but PDC have extremely high constitutive expression of IRF-7 message and protein; (2) there is a significant difference between MDDC, monocytes, and PDC in their response to virus stimulation, with PDC expressing not only much higher levels of IFN- protein, but also a broader profile of IFNA genes; (3) in extension of our previous study [14], the PDC respond to a broader range of viruses than monocytes or MDDC. We conclude that by virtue of the their high expression of IRF-7, the broad range of IFN- subtypes expressed, and their response to a wide range of viral and nonviral stimuli, PDC are uniquely suited as vital type I IFN-producing cells. ACKNOWLEDGMENTS This study was supported by NIAID grants RO1 AI (to PFB) and RO1 AI and R21AI (to PMP). B.B. was supported by the AntiCancer DrugDevelopment Pharmacology-Oncology training grant and A.I. was supported by a predoctoral fellowship from the UMDNJ-GSBS and a DHHS Institutional Training Grant T32 CA09665 from the National 1136 Journal of Leukocyte Biology Volume 74, December