Susceptibility Of Human Dendritic Cells

The Journal of Immunology

High Susceptibility of Human Dendritic Cells to Avian Influenza H5N1 Virus Infection and Protection by IFN-␣ and TLR Ligands1 Arunee Thitithanyanont,* Anneke Engering,† Peeraya Ekchariyawat,* Suwimon Wiboon-ut,* Amporn Limsalakpetch,† Kosol Yongvanitchit,† Utaiwan Kum-Arb,† Watcharoot Kanchongkittiphon,*‡ Pongsak Utaisincharoen,* Stitaya Sirisinha,* Pilaipan Puthavathana,§ Mark M. Fukuda,† and Sathit Pichyangkul2† There is worldwide concern that the avian influenza H5N1 virus, with a mortality rate of >50%, might cause the next influenza pandemic. Unlike most other influenza infections, H5N1 infection causes a systemic disease. The underlying mechanisms for this effect are still unclear. In this study, we investigate the interplay between avian influenza H5N1 and human dendritic cells (DC). We showed that H5N1 virus can infect and replicate in monocyte-derived and blood myeloid DC, leading to cell death. These results suggest that H5N1 escapes viral-specific immunity, and could disseminate via DC. In contrast, blood pDC were resistant to infection and produced high amounts of IFN-␣. Addition of this cytokine to monocyte-derived DC or pretreatment with TLR ligands protected against infection and the cytopathic effects of H5N1 virus. The Journal of Immunology, 2007, 179: 5220 –5227.

T

he first documented cases of human infection with avian influenza H5N1 virus occurred in Hong Kong in 1997, when 18 people were reported to be infected, among whom 6 died (1). So far, H5N1 viruses are primarily animal pathogens and mainly circulate in poultry. Most human cases to date have arisen from close contact to infected poultry or their waste products; human-to-human transmission evidently remains very limited (2). Even so, the high case-fatality rate in humans (59%) and the concern that mutant viruses might acquire the capacity for efficient transmission among humans has caused great concern worldwide regarding the potential for a pandemic of H5N1 influenza. Avian influenza H5N1 infection causes a severe disease in humans; patients died within 4 –30 days after the onset of illness (3). Unlike most influenza infections, which are limited to respiratory system, H5N1 infection causes a systemic disease. In infected patients, viral RNA and viable viruses were detected in blood specimens (4, 5), suggesting that H5N1 viruses could spread beyond the primary site of infection in the lungs and cause blood-borne systemic infection. This is in agreement with the few available histopathological reports of deceased patients, which document the *Department of Microbiology, Faculty of Science, Mahidol University, Bangkok, Thailand; †Department of Immunology and Medicine and ‡Department of Enteric Diseases, U.S. Army Medical Component of the Armed Forces Research Institute of the Medical Sciences, Bangkok, Thailand; and §Department of Microbiology, Faculty of Medicine, Siriraj Hospital, Mahidol University, Bangkok, Thailand Received for publication August 9, 2007. Accepted for publication August 10, 2007. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by the National Center for Genetic Engineering and Biotechnology (BIOTEC) Thailand, the Ellison Medical Foundation prime grant, by Thailand Research Fund for Advanced Research Scholar, and by Grant Y1-AI5026-01 from the National Institutes of Health, National Institute of Allergy and Infectious Diseases International Research in Infectious Disease. 2 Address correspondence and reprint requests to Dr. Sathit Pichyangkul, Department of Immunology and Medicine, U.S. Army Medical Component of the Armed Forces Research Institute of the Medical Sciences, 315/6 Rajvithi Road, Bangkok 10400, Thailand. E-mail address: [email protected]

www.jimmunol.org

presence of H5N1 virus in multiple damaged organs (6 –11). Also, lymphoid depletion, leukopenia and lymphopenia have consistently been observed in H5N1-infected patients (5, 7–9). These clinical findings suggest that the H5N1 influenza virus can evade host immunity. We hypothesized that H5N1 virus interacts with immune cells to promote dissemination and to cause pathogenesis, and therefore investigated the interplay between avian influenza H5N1 and human DC.3 DCs play a pivotal role in the initiation of immune responses (12). They are derived from bone marrow and distributed throughout the body to sense and capture invading pathogens through pathogen recognition receptors, including TLRs and C-type lectins (13, 14). After migration to lymph nodes, DCs present processed Ags in association with MHC molecules to initiate T cell responses. As DCs are present just underneath the epithelial layer of the respiratory tract, they could plausibly be the first APCs to contact avian influenza H5N1 virus. In addition, their wide distribution and migratory properties could enhance viral dissemination. The two DC subsets in blood, myeloid DC (mDC) and plasmacytoid DC (pDC), might also play a role in H5N1 immunopathogenesis. Also, pDCs are specialized producers of antiviral type I IFN, and are thus major players in antiviral responses (15). In view of these important properties of DCs, we studied their fate after exposure to H5N1 virus. We also investigated whether pretreatment with IFN-␣ or ligands to TLR alters their fate. As documented below, our results indicate that avian influenza H5N1 virus infects and replicates in monocyte-derived DCs and mDCs, leading to massive cell death. Moreover, DCs could be protected from H5N1 infection after pretreatment with IFN-␣ and TLR ligands.

Materials and Methods Virus A/open-billed stork/Nakhonsawan/BBD0104F/04, A/open-billed stork/ Suphanburi/TSD0912F/04, A/open-billed stork/Nakhonsawan/BBD1521J/ 3 Abbreviations used in this paper: DC, dendritic cell; mDC, myeloid DC; pDC, plasmacytoid DC; MOI, multiplicity of infection.

The Journal of Immunology 05, A/open-billed stork/Nakhonsawan/BBA3011M/05 (H5N1) were isolated from cloacal swabs of live and dead Asian open-billed storks between 2004 and 2005 and propagated in Madin-Darby canine kidney cells (MDCK) as previously described (16). The titer of virus stock was determined by titration in MDCK and by daily observation for cytopathic effects and was confirmed by a hemagglutination assay. Each virus titer was quantified by plaque assay. In brief, confluent monolayers of MDCK were inoculated with 10-fold dilutions of H5N1 virus and incubated at 37°C for 1 h. The inoculum was removed, and cells were washed and overlaid with MEM containing 1% agarose and 0.2% serum albumin. After 2 days at 37°C, cells were stained with 0.1% crystal violet in 37% formaldehyde solution, and plaque numbers were evaluated. All experiments with H5N1 virus were performed in a BioSafety Level 3 facility by trained researchers. Viral RNA was extracted from virus-stock and reverse transcribed by the One-Step RT-PCR kit (Qiagen) and was used to amplify eight genomic segments as full-length or overlapping fragments by RT-PCR. All PCR products were sequenced by Macrogen. Primer sequences are available upon request. The nucleotide sequences were edited using the BioEdit program and the nucleic sequences and amino acid sequences were aligned for construction of phylogenetic trees using the MEGA 3.1 program. All sequence data described in this report were submitted to GenBank (http://www.ncbi.nlm.nih.gov.lrc1.usuhs.edu/GenBank/index.html) with accession numbers DQ989958-DQ990007.

Cells and reagents PBMC were obtained by centrifugation using Histopaque-1077 (SigmaAldrich). CD14⫹ monocytes were isolated using CD14 MACS MicroBeads (Miltenyi Biotec). DCs were generated by culturing monocytes in RPMI 1640 supplemented with nonessential amino acids, 2 mM L-glutamine, 1 mM sodium pyruvate, 100 ␮g/ml penicillin, and 100 ␮g/ml streptomycin (complete medium; all from Invitrogen Life Technologies) containing 10% FCS, 100 ng/ml recombinant human IL-4 and 100 ng/ml recombinant human GM-CSF (both from R&D Systems) for 5–7 days. Every other day, fresh complete medium containing cytokines was added. Isolation of mDCs and pDCs from T cell-depleted PBMC was performed by sorting CD11cbrightlineage⫺ cells or CD123brightHLA-DR⫹ cells with a FACSVantage (BD Biosciences) as described (17). Based on CD123 expression, purified mDCs were not contaminated with pDCs (data not shown). ␣␤ T cells were sorted as TCR-␣␤-1⫹ cells from enriched T cell populations isolated by rosetting with neuraminidase-treated SRBC. B cells were purified from PBMC using CD20 MACS MicroBeads (Miltenyi Biotec). All cells were cultured in complete medium containing 10% FCS. For viral infection, cells were cocultured with H5N1 virus at multiplicity of infection (MOI) and times indicated. To compare the cytopathic effect induced by H5N1 and H3N2 viruses, DCs from the same donors were infected with either H5N1 or H3N2 (A/Hong Kong/8/68) viruses at an MOI of 1 in serum-free medium (200 ␮l) for 1 h at 37°C and complete medium containing 10% FCS, 200 ng/ml IL-4 and GM-CSF was then added. Cell viability was determined by counting cells with trypan blue. The bronchial/tracheal epithelial cell line NHBE was from Cambrex. All Abs were from BD Pharmingen. Porphyromonas gingivalis LPS, polyinosinic-polycytidylic acid (poly(I:C)), Escherichia coli LPS, and flagellin from Salmonella typhimurium were purchased from InvivoGen. CpG ODN 2216 (ggGGGACGATCGTCgggggG) was a gift from Dr. A. Krieg (Coley Pharmaceutical Group, Wellesley, MA). Recombinant human IFN-␣ B2 was from PBL Biomedical Laboratories, recombinant human TNF-␣/ TNFSF1A, and neutralizing polyclonal Abs against IFN-␣ and TNF-␣ were from R&D Systems.

Staining of influenza A Cells were allowed to adhere on Superfrost Plus slides (Erie Scientific), fixed in acetone, and virus was stained with Imagen Influenza Virus A (DakoCytomation) according to the manufacturer’s instructions. Viral nucleoprotein and matrix protein were stained with FITC-conjugated Abs (green), whereas cells were counterstained with Evans blue (red). Staining was analyzed with a fluorescent microscope (TE2000-S; Nikon).

Real-time PCR Infected DCs were harvested at various time points for total RNA extraction by use of RNeasy minikit (Qiagen). The cells were washed three times to remove free virions with PBS. Reverse transcription was performed on DNase-treated total RNA as previously described (18). cDNA was synthesized from viral genomic RNA with random hexamer and AMV reverse transcriptase (Promega) according to the protocol provided by the manufacturer; these cDNA samples were used as template. Specific primers were used in a real-time PCR assay, detected by use of a Rotor-Gene 3000 (Corbett Robotics). The standard curve was generated using serial dilution

5221 of plasmids (from 1–107 copies) containing the respective cloned gene targets. Viral M1 gene copies were quantified on the basis of a SYBR green fluorescence signal after PCR (forward primer, 5⬘-CTTCTAACCGAGGT CGAAACG-3⬘andreverseprimer,5⬘-AGGGCATTTTGGACAAA(GT)CG TCTA-3⬘) and were amplified by using HotStarTaq DNA Polymerase (Qiagen). Conditions were as follows: denaturation at 95°C for 15 min, 40 cycles 95°C for 30 s, 45°C for 30 s, and extension at 72°C for 30 s. Dissociation curve analysis was performed after each assay to ensure specific target detection. To standardize results for variability in RNA and cDNA quantity, they are expressed as the number of target gene.

Measurement of cytokines Production of IFN-␣, IL-12p70, and TNF-␣ in supernatants of cultured DCs, mDCs, and pDCs and IL-6 and TNF-␣ of monocytes were measured by ELISA following the manufacturer’s instructions (human IFN-␣ kit from PBL Biomedical Laboratories and human IL-12 Quantikine, human TNF-␣/TNFSF1A Quantikine, or human IL-6 Quantikine, all from R&D Systems).

Results DC as targets for avian influenza virus replication To assess the effect of avian influenza virus on DC, human monocyte-derived DC were exposed to H5N1 virus (A/open-billed stork/Nakhonsawan/BBD0104F/04) at an MOI ranging from 0.008 to 1. This strain was isolated in Thailand from the Asian openbilled stork, and is genetically similar to H5N1 from infected human patients (data not shown). Surprisingly, we observed that 24 h after coculture with H5N1 virus, the majority of DCs had died (Fig. 1A). Even when only one virus particle was present per 125 cells (MOI of 0.008), only 44% (range 39% to 62%) of DC were viable as compared with controls. At lower MOI, cell killing was minimal (data not shown). Coculture of DCs with human influenza virus H3N2 resulted in significantly lower levels of cell death relative to H5N1 virus ( p ⫽ 0.031, n ⫽ 5 donors) (Fig. 1A), suggesting that avian influenza H5N1 is more potent in causing DC killing. More than 25% of DCs were killed by 12 h after infection with the virus at MOI of 1 (Fig. 1A). This cytopathic effect required infectious H5N1 virus because it was abrogated when heatinactivated virus was used (Fig. 1A). In addition, the virus-induced DC killing was not restricted to a single strain of H5N1 virus, but was also observed in other strains isolated at different times from Asian open-billed storks (Fig. 1A). Moreover, viral proteins could readily be detected in a majority of DC by immunofluorescence staining specific to influenza A nucleoprotein and matrix protein at 8 h postinfection (Fig. 1B), as well as by measuring viral RNA in cells by real-time PCR (Fig. 1C). This indicates that the virus is not merely toxic to the DCs, but also infects them. Taken together, our data show that H5N1 virus can infect DC, resulting in cell death, even at very low MOI. Next, we investigated whether DCs support production of H5N1 virus resulting in release of progeny virus. Cells were incubated with H5N1 virus at an MOI of 1 for 2 h, washed extensively, and cultured for various time periods, after which infectious virus in culture supernatant was measured using a plaque assay. As shown in Fig. 1D, infectious viruses were already observed in culture supernatants after 7 h, and the viral titers were markedly increased (by 3 log values) at 24 h postinfection. The level of infectious H5N1 virus particles released was very similar to that observed in primary human bronchial/tracheal epithelial cells NHBE (Fig. 1D) that have been shown previously to be permissive to H5N1 infection (19). To investigate whether the cytopathic effect of avian influenza H5N1 virus occurs early and shuts down subsequent DC functions or, alternatively, whether DC are initially activated by the virus, culture supernatants of infected DCs were collected and cytokine production was measured by ELISA. Noticeably, H5N1-infected

5222

AVIAN INFLUENZA INFECTION OF HUMAN DC

FIGURE 1. Avian influenza H5N1 can infect and replicate in DC. A, Cell viability of human monocyte-derived DC (f) and CD14⫹ monocytes (u) compared with untreated control. The ability of avian influenza H5N1 to induce DC death was compared with human influenza H3N2 at an MOI of 1 for 24 h. Kinetics of DC survival after infection with H5N1 were assessed. Cells were incubated with different strains of H5N1 (right), live or heat-inactivated (60°C for 30 min), at the indicated MOI for 24 h. Data are shown as mean percentage ⫾ SEM from at least three different donors. B, Staining of influenza A (green) in DCs infected with H5N1 at MOI 1 for 8 h. Cells appear in red. C, Real-time PCR of H5N1 M gene in DCs after infection with H5N1 as in A. ND, Not detected. Data shown are from one representative donor of four. D, Data are plaque-forming units in supernatant of DCs and monocytes from two donors, and bronchial/tracheal epithelial cells (two experiments) infected with H5N1 at MOI 1 for 2 h, washed, and incubated for indicated times.

DC were found to produce the antiviral cytokine IFN-␣ at all MOI tested (Fig. 2A). The amounts of IFN-␣ secreted, especially at low MOI, were high in comparison to activation of DC with poly(I:C), a ligand for TLR3 that is the most potent inducer of IFN-␣ among a variety of TLR ligands tested (data not shown). In addition, DC infected with H5N1 virus produced low amounts of TNF-␣, another antiviral cytokine, whereas no production of bioactive IL12p70 could be detected (Fig. 2A). An evaluation of the kinetics of cytokine production indicated that secretion of both IFN-␣ and TNF-␣ was detected at 12 h after infection and further increased at

24 h (Fig. 2B). Production of IFN-␤ was low to undetectable in H5N1-infected DC (data not shown). Cytokine production required live H5N1 virus because no IFN-␣ and TNF-␣ secretion was detected upon incubation with heat-killed virus (data not shown). Thus, H5N1 infection of DCs resulted in both viral replication and cell killing, and in cell activation with production of antiviral cytokines. Similarly to DC, H5N1 virus could infect peripheral blood monocytes (precursor cells of DC) resulting in massive cell death (Fig. 1A). However, no replication and production of infectious

The Journal of Immunology

5223

FIGURE 2. Avian influenza H5N1 infection induces selective IFN-␣ production by DCs and low cytokine production by monocytes. Production of IFN-␣, IL-12p70, and TNF-␣ by DCs at 24 h (A), kinetics of IFN-␣ and TNF-␣ production by DCs at MOI 1 (B), and production of TNF-␣ and IL-6 by monocytes at 24 h (C) after infection with H5N1 at indicated MOI. Poly(I:C) (100 ␮g/ml) and E. coli LPS (5 ␮g/ml) were used as positive control for DCs and monocytes, respectively. Data are shown as mean cytokine secreted ⫾ SEM from three different donors. ND, Not detected.

virus was observed in monocytes (Fig. 1D), suggestive of a postentry block associated with self-destruction of these cells. This observation is in agreement with negligible production of the cytokines TNF-␣ and IL-6 by monocytes after H5N1 infection (Fig. 2C). The pDC, but not mDC, is resistant to avian influenza H5N1 infection H5N1 patients carry infectious virus particles in blood, a feature rarely observed in other human influenza infections (4, 5). Therefore, we studied the relative infectivity of H5N1 virus in several blood cell types, namely mDCs, pDCs, and other circulating cells. Highly purified CD11cbrightlineage⫺ mDCs, CD123brightHLADR⫹ pDCs, CD20⫹ B cells, and ␣␤⫹ T cells were cocultured with H5N1 virus at an MOI of 1 for 24 h. Immunofluorescence staining for viral proteins showed that whereas almost all mDCs were in-

fected with H5N1 virus, pDCs were resistant to infection as were B cells and ␣␤ T cells (Fig. 3A). Also, mDCs were killed 24 h postinfection (data not shown), but produced IFN-␣ and TNF-␣ and no IL-12p70 (Fig. 3B), similarly to monocyte-derived DCs. IFN-␣ and TNF-␣ secretion required infectious virus because no cytokines were produced from mDCs using heat-killed H5N1 (data not shown). Interestingly, pDCs, also known as type I IFN producers, secreted very high amounts of IFN-␣ after coculture with H5N1 virus, up to five times more than after stimulation with the potent TLR ligand CpG ODN 2216 and 25 times higher as compared with infected monocyte-derived DCs and mDCs (Fig. 3B). Also, pDCs produced TNF-␣, but no IL-12p70 (Fig. 3B). We therefore hypothesize that the protection from H5N1 infection observed in pDCs might be due to their production of high levels of IFN-␣. Indeed, this mechanism has already been described for other viruses (15).

5224

AVIAN INFLUENZA INFECTION OF HUMAN DC

FIGURE 3. Avian influenza H5N1 replicates in blood mDCs, but not in pDCs, B cells, or T cells. A, Staining of influenza A in mDCs, pDCs, B cells, and ␣␤ T cells exposed to H5N1 at an MOI of 1 for 24 h. B, ELISA of IFN-␣, IL-12p70, and TNF-␣ produced by mDCs and pDCs after infection with H5N1 at MOI 1 for 24 h. Data from donor 1 (f) and donor 2 (u) are shown. ND: not detected.

TLR ligands and antiviral cytokines protect DCs from avian influenza infection and cytopathic effects DCs express a variety of pathogen receptors on their cell surface, including TLRs that recognize pathogen-associated molecular patterns resulting in activation of DCs to combat such pathogens. We reasoned that activation of DCs via TLRs might reduce their susceptibility to H5N1 infection, as previously shown for other viruses (20 –22). Thus, DCs were incubated with well-characterized ligands for TLR2 (LPS from P. gingivalis), TLR3 (poly(I:C)), TLR4 (LPS from E. coli), or TLR5 (flagellin from S. typhimurium) for 24 h, and subsequently infected with H5N1 virus. As depicted in Fig. 4A, TLR triggering abolished the cytopathic effect of H5N1, irrespective of the TLR ligand used. TLR treatment also reduced infection as measured by the number of viral RNA copies per cell, with 1.5 log reduction for TLR2 and TLR4 ligands and over 2 log reduction for TLR3 and TLR5 ligands (Fig. 4B). Addition of neutralizing Abs against IFN-␣, TNF-␣, or both into the TLR ligand-treated DCs had no significant effect on H5N1induced cell killing (Fig. 4C), suggesting that these cytokines were not involved in protection against H5N1 upon TLR acti-

vation. A slight decrease in cell viability was observed when cells were treated with anti-IFN-␣ as compared with untreated DCs, but the difference did not reach statistical significance ( p values ranged from 0.063 to 0.125). Next, we studied whether pretreatment with antiviral cytokines could have an effect on H5N1 infection. DCs were preincubated with different concentrations of IFN-␣ or TNF-␣ for 24 h, and subsequently H5N1 virus at an MOI of 1 was added. The cytopathic effect of H5N1 was reduced after treatment with 50 or 500 ng/ml TNF-␣ and 1000 U/ml IFN-␣, whereas 10,000 U/ml IFN-␣ completely abolished H5N1-induced cell killing (Fig. 5A). Viral RNA was only reduced 1–1.5 log for both concentrations of TNF-␣ (Fig. 5B). In marked contrast, pretreatment with IFN-␣ resulted in an over 3 log reduction of viral RNA, with residual levels as low as 4.6 copies (range 3–9 copies) per 106 DCs (Fig. 5B). Thus, whereas TLR ligands and TNF-␣ rendered DCs partially insensitive to H5N1 virus, IFN-␣, especially at a high dose, can completely protect DCs from infection, replication, and the cytopathic effect of avian influenza.

The Journal of Immunology

5225

FIGURE 4. Ligation of TLRs on DCs confers resistance to avian influenza H5N1. Cell viability (A) and real-time PCR of H5N1 M gene (B) of DCs preincubated with ligands to TLR2 (100 ␮g/ml, P. gingivalis LPS), TLR3 (100 ␮g/ml, poly(I:C)), TLR4 (5 ␮g/ml, E. coli LPS), and TLR5 (10 ␮g/ml, flagellin from S. typhimurium) for 24 h, followed by infection with H5N1 at an MOI of 0.04. C, DCs were stimulated with TLR ligands in the absence or presence of neutralizing Abs against IFN-␣ (2500 U/ml), TNF-␣ (10 ␮g/ml), or in combination for 24 h, and then infected with H5N1 at an MOI of 0.04. Mean percentage of cell viability ⫾ SEM was assessed at 24 h postinfection. Data shown are from four different donors (A), from one representative donor of four (B), or from five different donors (C).

Discussion

FIGURE 5. DCs treated with IFN-␣ and TNF-␣ are resistant to avian influenza H5N1. Cell viability (A) and real-time PCR of H5N1 M gene (B) of DCs preincubated with IFN-␣ and TNF-␣ at concentrations indicated for 24 h, followed by infection with H5N1 at an MOI of 1. Data are the mean ⫾ SEM from four different donors (A) or from one representative donor of four (B) for each experiment.

Avian influenza is a deadly disease that is widely feared as the cause of the next influenza pandemic. Improved insight into the immunopathogenesis of the H5N1 virus is crucial for development of effective preventive and therapeutic strategies. We have studied, and are the first to report as best we can determine, the interaction between human DCs and avian influenza. We used avian influenza H5N1 viruses isolated from Asian open-billed storks that cluster phylogenetically with highly pathogenic H5N1 viruses isolated from chickens, cats, tigers, and humans. In addition, the NS1 protein of the strains used contains a deletion of 5 aa at position 80 – 84 and alanine at position 149; both strongly influence viral pathogenesis and resistance to antiviral cytokines (23, 24). Our data demonstrate that H5N1 infects and replicates in monocytederived DCs and blood mDCs, and causes severe cytopathic effects in these cell types. Preliminary experiments showed that DCs died by apoptosis; we are currently investigating the mechanism of cell death by H5N1. Our results imply that DCs would represent novel targets of H5N1 virus, contributing to production and dissemination of new virions. Interestingly, pretreatment of DCs with the antiviral cytokines TNF-␣ or type I IFN and with TLR ligands could protect DCs from H5N1 in varying degrees. Upon coculture of H5N1 with DCs for 24 h, the majority of DCs died even at very low MOI. This strong cytopathic effect was not observed with other viruses, such as human influenza virus or even the lethal Ebola virus (22, 25, 26). Several studies have shown that immature and mature DCs differ in their susceptibility to viral infection (20 –22). Consistent with this research, we observed that maturation of DCs by TLR ligands increased their resistance to H5N1 infection. Although DCs produce antiviral cytokines upon TLR ligation, this could not fully explain protection against H5N1 because the levels of IFN-␣ and TNF-␣ secreted were much lower

5226 than those required for in vitro protection (Fig. 5). In addition, neutralizing Abs to IFN-␣, TNF-␣, or both cytokines did not significantly abolish the protective effects of TLR ligands, indicating that TLR ligands may trigger other pathways for viral resistance. The mechanisms underlying this resistance are unclear, although for human influenza the cytoplasmic antiviral protein MxA was shown to play a role in protection against cytopathic effects and infection (22). Further research on TLR ligand-induced resistance could well prove useful for further development of therapeutic agents against avian influenza. In the lung, hemagglutinin of avian influenza H5N1 virus preferentially binds to sialic acid linked to galactose by an ␣2,3 linkage, SA␣2,3Gal, which is found on epithelial cells in the lower respiratory tract. In contrast, human influenza hemagglutinin binds to sialic acid linked to galactose by an ␣2,6 linkage, SA␣2,6Gal, which is predominantly expressed on epithelial cells in the upper respiratory tract (27). This might partially explain the low humanto-human transmission of H5N1 virus as compared with human influenza. DCs were found to express SA␣2,3Gal as well as SA␣2,6Gal (A. Engering, unpublished observations), but other receptors on DCs may also be involved in binding and entry of H5N1. Our preliminary results (data not shown) suggest that the C-type lectin DC-SIGN, which has been described to bind a range of viruses (14), may be involved in binding of H5N1 virus. DCSIGN binding might enhance both infection and replication in DCs and might also be involved in viral dissemination into lymph nodes, as shown previously for HIV (28). In contrast to blood mDCs, pDCs were resistant to H5N1 infection and replication, and produced very high amounts of IFN-␣. Such resistance to viral infection and IFN-␣ production independent of infection has previously been described for human influenza (29) and other viruses (15) and suggests that these cells are important in resisting viral infection. This could have physiological relevance in protection of bystander cells by pDC-derived IFN-␣. Interestingly, pretreatment of monocyte-derived DCs with IFN-␣ abolished H5N1 infection completely. The protective effect of IFN-␣ was also observed for other human primary cells (data not shown). Our observations are inconsistent with a previous study in which infection of porcine lung epithelial cells with H5N1/97 was not affected by pretreatment with IFN-␣ (24). This discrepancy was not due to dose differences. Also, the NS1 protein of the H5N1 viral strains used in our experiments contains a 5 aa deletion at position 80 – 84 and alanine at position 149, both of which are associated with resistance to IFN-␣ (23, 24). Future studies should reveal whether there are inherent differences between epithelial cells and DCs, or between cells from porcine and human origin, in suppressor activity of NS1 and/or resistance to H5N1 virus by type I IFNs. Based on our findings, we hypothesize that tissue and blood DCs are targets for avian influenza. Future studies are required to determine whether lung DCs behave similarly as monocytederived DCs and indeed represent the secondary target cells that are infected upon release of H5N1 virus from primary target lung epithelial cells. Infected DCs produce new viral progeny that might use the migratory routes of DCs to gain access to the lymphatic systems and bloodstream and subsequently to other tissues. The viral titers in DC culture supernatants were indeed comparable to those detected in human bronchial/tracheal epithelial cell culture supernatants. After entering the bloodstream, H5N1 virus could further infect several circulating cell types. CD14⫹ monocytes were highly susceptible to H5N1 virus-induced cell killing, but neither produced cytokines nor supported viral replication. A high proportion of blood mDCs were infected by H5N1 virus and this subset of blood cells could rep-

AVIAN INFLUENZA INFECTION OF HUMAN DC resent another site of viral production leading to increased and sustained viremia. Indeed, in blood of patients, large amounts of viable H5N1 viruses were detected (4, 5). The pantropic nature of H5N1 virus would thus support a systemic disease. Although pDCs were not infected, the H5N1-induced IFN-␣ and TNF-␣ from pDCs and mDCs may potentiate the severity of disease and is in agreement with increased levels of both cytokines in the blood of H5N1 patients (3, 5). In addition to infection and dissemination, the H5N1-induced cell killing of DCs will impair the induction of virus-specific immune responses leading to immune escape. These and other events could contribute to the observed rapid multiorgan dysfunction and death.

Acknowledgment We thank Dr. A. Krieg for CpG ODN 2216.

Disclosures The authors have no financial conflict of interest.

References 1. Claas, E. C., A. D. Osterhaus, R. van Beek, J. C. De Jong, G. F. Rimmelzwaan, D. A. Senne, S. Krauss, K. F. Shortridge, and R. G. Webster. 1998. Human influenza A H5N1 virus related to a highly pathogenic avian influenza virus. Lancet 351: 472– 477. 2. Fauci, A. S. 2006. Pandemic influenza threat and preparedness. Emerg. Infect. Dis. 12: 73–77. 3. Beigel, J. H., J. Farrar, A. M. Han, F. G. Hayden, R. Hyer, M. D. de Jong, S. Lochindarat, T. K. Nguyen, T. H. Nguyen, T. H. Tran, et al. 2005. Avian influenza A (H5N1) infection in humans. N. Engl. J. Med. 353: 1374 –1385. 4. Chutinimitkul, S., P. Bhattarakosol, S. Srisuratanon, A. Eiamudomkan, K. Kongsomboon, S. Damrongwatanapokin, A. Chaisingh, K. Suwannakarn, T. Chieochansin, A. Theamboonlers, and Y. Poovorawan. 2006. H5N1 influenza A virus and infected human plasma. Emerg. Infect. Dis. 12: 1041–1043. 5. de Jong, M. D., C. P. Simmons, T. T. Thanh, V. M. Hien, G. J. Smith, T. N. Chau, D. M. Hoang, N. Van Vinh Chau, T. H. Khanh, V. C. Dong, et al. 2006. Fatal outcome of human influenza A (H5N1) is associated with high viral load and hypercytokinemia. Nat. Med. 12: 1203–1207. 6. Yuen, K. Y., P. K. Chan, M. Peiris, D. N. Tsang, T. L. Que, K. F. Shortridge, P. T. Cheung, W. K. To, E. T. Ho, R. Sung, and A. F. Cheng. 1998. Clinical features and rapid viral diagnosis of human disease associated with avian influenza A H5N1 virus. Lancet 351: 467– 471. 7. Chotpitayasunondh, T., K. Ungchusak, W. Hanshaoworakul, S. Chunsuthiwat, P. Sawanpanyalert, R. Kijphati, S. Lochindarat, P. Srisan, P. Suwan, Y. Osotthanakorn, et al. 2005. Human disease from influenza A (H5N1), Thailand, 2004. Emerg. Infect. Dis. 11: 201–209. 8. To, K. F., P. K. Chan, K. F. Chan, W. K. Lee, W. Y. Lam, K. F. Wong, N. L. Tang, D. N. Tsang, R. Y. Sung, T. A. Buckley, et al. 2001. Pathology of fatal human infection associated with avian influenza A H5N1 virus. J. Med. Virol. 63: 242–246. 9. Peiris, J. S., W. C. Yu, C. W. Leung, C. Y. Cheung, W. F. Ng, J. M. Nicholls, T. K. Ng, K. H. Chan, S. T. Lai, W. L. Lim, et al. 2004. Re-emergence of fatal human influenza A subtype H5N1 disease. Lancet 363: 617– 619. 10. Uiprasertkul, M., P. Puthavathana, K. Sangsiriwut, P. Pooruk, K. Srisook, M. Peiris, J. M. Nicholls, K. Chokephaibulkit, N. Vanprapar, and P. Auewarakul. 2005. Influenza A H5N1 replication sites in humans. Emerg. Infect. Dis. 11: 1036 –1041. 11. de Jong, M. D., V. C. Bach, T. Q. Phan, M. H. Vo, T. T. Tran, B. H. Nguyen, M. Beld, T. P. Le, H. K. Truong, V. V. Nguyen, et al. 2005. Fatal avian influenza A (H5N1) in a child presenting with diarrhea followed by coma. N. Engl. J. Med. 352: 686 – 691. 12. Banchereau, J., and R. M. Steinman. 1998. Dendritic cells and the control of immunity. Nature 392: 245–252. 13. Akira, S., S. Uematsu, and O. Takeuchi. 2006. Pathogen recognition and innate immunity. Cell 124: 783– 801. 14. Geijtenbeek, T. B., S. J. van Vliet, A. Engering, B. A. t Hart, and Y. van Kooyk. 2004. Self- and nonself-recognition by C-type lectins on dendritic cells. Annu. Rev. Immunol. 22: 33–54. 15. Asselin-Paturel, C., and G. Trinchieri. 2005. Production of type I interferons: plasmacytoid dendritic cells and beyond. J. Exp. Med. 202: 461– 465. 16. Puthavathana, P., P. Auewarakul, P. C. Charoenying, K. Sangsiriwut, P. Pooruk, K. Boonnak, R. Khanyok, P. Thawachsupa, R. Kijphati, and P. Sawanpanyalert. 2005. Molecular characterization of the complete genome of human influenza H5N1 virus isolates from Thailand. J. Gen. Virol. 86: 423– 433. 17. Pichyangkul, S., K. Yongvanitchit, U. Kum-arb, H. Hemmi, S. Akira, A. M. Krieg, D. G. Heppner, V. A. Stewart, H. Hasegawa, S. Looareesuwan, et al. 2004. Malaria blood stage parasites activate human plasmacytoid dendritic cells and murine dendritic cells through a Toll-like receptor 9-dependent pathway. J. Immunol. 172: 4926 – 4933. 18. Quinlivan, M., A. Cullinane, M. Nelly, K. Van Maanen, J. Heldens, and S. Arkins. 2004. Comparison of sensitivities of virus isolation, antigen detection,

The Journal of Immunology

19.

20.

21.

22.

23.

and nucleic acid amplification for detection of equine influenza virus. J. Clin. Microbiol. 42: 759 –763. Chan, M. C., C. Y. Cheung, W. H. Chui, S. W. Tsao, J. M. Nicholls, Y. O. Chan, R. W. Chan, H. T. Long, L. L. Poon, Y. Guan, and J. S. Peiris. 2005. Proinflammatory cytokine responses induced by influenza A (H5N1) viruses in primary human alveolar and bronchial epithelial cells. Respir. Res. 6: 135–148. Granelli-Piperno, A., E. Delgado, V. Finkel, W. Paxton, and R. M. Steinman. 1998. Immature dendritic cells selectively replicate macrophagetropic (M-tropic) human immunodeficiency virus type 1, while mature cells efficiently transmit both M- and T-tropic virus to T cells. J. Virol. 72: 2733–2737. Spiegel, M., K. Schneider, F. Weber, M. Weidmann, and F. T. Hufert. 2006. Interaction of severe acute respiratory syndrome-associated coronavirus with dendritic cells. J. Gen. Virol. 87: 1953–1960. Cella, M., M. Salio, Y. Sakakibara, H. Langen, I. Julkunen, and A. Lanzavecchia. 1999. Maturation, activation, and protection of dendritic cells induced by doublestranded RNA. J. Exp. Med. 189: 821– 829. Li, Z., Y. Jiang, P. Jiao, A. Wang, F. Zhao, G. Tian, X. Wang, K. Yu, Z. Bu, and H. Chen. 2006. The NS1 gene contributes to the virulence of H5N1 avian influenza viruses. J. Virol. 80: 11115–11123.

5227 24. Seo, S. H., E. Hoffmann, and R. G. Webster. 2002. Lethal H5N1 influenza viruses escape host anti-viral cytokine responses. Nat. Med. 8: 950 –954. 25. Bhardwaj, N., A. Bender, N. Gonzalez, L. K. Bui, M. C. Garrett, and R. M. Steinman. 1994. Influenza virus-infected dendritic cells stimulate strong proliferative and cytolytic responses from human CD8⫹ T cells. J. Clin. Invest. 94: 797– 807. 26. Mahanty, S., K. Hutchinson, S. Agarwal, M. McRae, P. E. Rollin, and B. Pulendran. 2003. Cutting edge: impairment of dendritic cells and adaptive immunity by Ebola and Lassa viruses. J. Immunol. 170: 2797–2801. 27. Shinya, K., M. Ebina, S. Yamada, M. Ono, N. Kasai, and Y. Kawaoka. 2006. Avian flu: influenza virus receptors in the human airway. Nature 440: 435– 436. 28. Geijtenbeek, T. B., D. S. Kwon, R. Torensma, S. J. van Vliet, G. C. van Duijnhoven, J. Middel, I. L. Cornelissen, H. S. Nottet, V. N. KewalRamani, D. R. Littman, C. G. Figdor, and Y. van Kooyk. 2000. DC-SIGN, a dendritic cell-specific HIV-1-binding protein that enhances trans-infection of T cells. Cell 100: 587–597. 29. Fonteneau, J. F., M. Gilliet, M. Larsson, I. Dasilva, C. Munz, Y. J. Liu, and N. Bhardwaj. 2003. Activation of influenza virus-specific CD4⫹ and CD8⫹ T cells: a new role for plasmacytoid dendritic cells in adaptive immunity. Blood 101: 3520 –3526.