Advertisement

Human Dendritic Cells as Targets of Dengue Virus Infection

      Dengue virus infections are an emerging global threat. Severe dengue infection is manifested as dengue hemorrhagic fever and dengue shock syndrome, both of which can be fatal complications. Factors predisposing to complicated disease and pathogenesis of severe infections are discussed. Using immunohistochemistry, immunofluorescence, flow cytometry, and ELISA techniques, we studied the cellular targets of dengue virus infection, at both the clinical (in vivo) and the laboratory (in vitro) level. Resident skin dendritic cells are targets of dengue virus infection as demonstrated in a skin biopsy from a dengue vaccine recipient. We show that factors influencing infection of monocytes/macrophages and dendritic cells are different. Immature dendritic cells were found to be the cells most permissive for dengue infection and maybe early targets for infection. Immature dendritic cells exposed to dengue virus produce TNF-α protein. Some of these immature dendritic cells undergo TNF-α mediated maturation as a consequence of exposure to the dengue virus.

      Keyword

      Abbreviations

      ADE
      antibody-dependent enhancement
      DEN
      dengue
      DHF
      dengue hemorrhagic fever
      DSS
      dengue shock syndrome
      DV
      dengue virus
      Disease attributable to the dengue virus (DV) is a major global health burden with an estimated 50 million or more infections occurring each year (
      • World Health Organization
      ). The emergence of dengue as a worldwide public health problem over the past 50 y is, in part, due to increased urbanization and ecologic disruptions, peridomestic mosquito breeding grounds, increased air travel, and the gradual reintroduction of the vector into many areas once considered vector-free (
      • Gubler D.J.
      Dengue and dengue hemorrhagic fever. its history and resurgence as a global public health problem.
      ). Dengue illness can be caused by any one of four antigenically distinct serotypes (DEN 1–4) and unfortunately, no long-term cross-protection is provided by any given serotype. The principal vector, Aedes aegypti, is now firmly re-established in South and Central America, and the Gulfcoast states of the U.S.A., which heightens the threat of potential epidemic dengue transmission within the U.S.A. (
      • Gubler D.J.
      Dengue and dengue hemorrhagic fever. its history and resurgence as a global public health problem.
      ). Four such epidemics have been described in Texas in the past 20 y (DEN-1 in 1980 and 1986, and DEN-2 in 1995 and 1997).
      DV causes a febrile illness with myriad clinical manifestations which range from benign to life threatening. The incubation period averages 2–7 d and symptoms range from a mild febrile illness with rash and muscle and joint pains (“break-bone fever”) to a severe disease called dengue hemorrhagic fever (DHF). Dengue is a reportable disease and specific case definitions are provided by the World Health Organization to facilitate diagnosis and treatment (
      • World Health Organization
      ). Provisional diagnosis of DHF requires (i) fever, (ii) hemorrhagic manifestations, (iii) thrombocytopenia, and (iv) objective evidence of increased capillary permeability. Dengue shock syndrome (DSS) is defined as all four of the above plus hypotension. Without appropriate fluid resuscitation the outcome of DHF/DSS can be fatal.
      The reason for such extremes in the clinical manifestation of dengue infection is still a matter of debate, but there are several possible explanations. Dengue is the prototype virus for the concept of immune-mediated enhancement of infection. This compelling hypothesis, first advanced 30 y ago, attributes severe disease (DHF/DSS) to a process called “antibody-mediated enhancement”, or ADE (
      • Halstead S.
      Observations related to pathogenesis of dengue hemorrhaic fever. VI. Hypotheses and discussion.
      ). This theory is based on the epidemiologic observation that almost all DHF/DSS cases occur in individuals experiencing a second DV infection (
      • Halstead S.
      Observations related to pathogenesis of dengue hemorrhaic fever. VI. Hypotheses and discussion.
      ), and on an in vitro“serologic” phenomenon originally reported 40 y ago describing the paradoxical enhancement of infectivity of arboviruses at subneutralizing concentrations of antibody (
      • Hawkes R.
      Enhancement of the infectivity of arboviruses by specific antisera produced in domestic fowls.
      ). The ADE theory proposes that pre-existing antibodies (to a prior heterotypic DV infection) actually facilitate infection of monocyte/macrophage cells via their abundant Fc receptors. The fact that multiple dengue serotypes usually circulate in areas where DHF occurs, allowing for sequential infections, provides further support of the ADE mechanism. If ADE allows for the in vivo amplification of the virus, which then causes severe dengue disease, one might expect to see higher viremia in severe dengue infections. In fact, recent data published from Thailand provide this in vivo correlate showing higher virus titres (one log higher) in severe dengue infection (
      • Vaughn D.
      • Green S.
      • Kalayanarooj S.
      • et al.
      Dengue viremia titer, antibody response pattern, and virus serotype correlate with disease severity.
      ). Another major hypothesis invokes viral virulence mechanisms. Several recent reviews conclude that a combination of host immunologic and viral factors are involved in the pathogenesis of DHF (
      • Rothman A.
      • Ennis F.
      Immunopathogenesis of dengue hemorrhagic fever.
      ;
      • Vaughn D.
      • Green S.
      • Kalayanarooj S.
      • et al.
      Dengue viremia titer, antibody response pattern, and virus serotype correlate with disease severity.
      ).
      The target cell for DV has been described previously as a mononuclear phagocyte, monocyte, or macrophage. Although some investigators have proposed that DV replication occurs in skin (
      • Taweechaisupspong S.
      • Sriurairatana S.
      • Angsubhakorn S.
      • Yoksan S.
      • Khin M.
      • Sahaphong S.
      • Bhamarapravati N.
      Langerhans cell density and serological changes following intradermal immunisation of mice with dengue 2 virus.
      ), the initial target cell had not been identified. We have recently shown that human skin DC (Langerhans cells) are targets for DV infection (
      • Wu S.J.
      • Grouard-Vogel G.
      • Sun W.
      • et al.
      Human skin Langerhans cells are targets of dengue virus infection.
      ). We observed that immature human dendritic cells (DC), in contrast to other leukocytes, were preferentially permissive for dengue infection. Unlike monocytes/macrophages, DV infection was not enhanced by dengue specific antibody; however, exposure of human DC to DV induces TNF-α-mediated maturation of the DC.

      Materials and methods

      Cytokine-driven blood derived DC and macrophages

      Blood DC were generated from PBMC using IL-4 and GM-CSF to drive the CD14+ population to differentiate into DC (
      • Wu S.J.
      • Grouard-Vogel G.
      • Sun W.
      • et al.
      Human skin Langerhans cells are targets of dengue virus infection.
      ). Briefly, positively selected CD14+ monocytes were isolated from PBMC using a MACS cell isolation column (Miltenyi Biotech, Auburn, CA) according to manufacturer's guidelines. Monocytes were seeded in plastic flasks at a density of 1–2 × 106 cells per ml of culture media prepared as follows: RPMI-1640 (Quality Biologics, Gaithersburg, MD) supplemented with 10%-15% heat-inactivated fetal calf serum (PAA Laboratories, Parker Ford PA), 100 U penicillin per ml, 100 µg streptomycin per ml, and 2  L-glutamine (BioWhittaker, Walkersville, MD) containing 1000 U per ml each of recombinant human IL-4 (R&D Systems, Minneapolis, MN) and GM-CSF (Immunex, Seattle, WA). Cytokines were replenished on alternate days and cells were used on day 6 or 7 as immature DC. To produce mature DC, culture media was supplemented with macrophage conditioned media (MCM) and the cells were kept in culture for an additional 2–3 d. The appropriate phenotype of immature and mature DC was confirmed by flow cytometry prior to each experiment (all DC were CD14, CD1a+, HLA-DR+, CD80+, CD86+ only mature DC were CD83+ and DC-LAMP+). For generation of macrophages, PBMC were adhered to IgG-coated flasks in culture media containing 10% heat-inactivated normal human serum and 500 U per ml of recombinant human GM-CSF. After removal of nonadherent cells, macrophages were maintained in culture for 6–7 d.

      DV infection

      Cells were exposed to DV at different MOI (ratio of virus:cell) for 2 h at 37°C, washed twice to remove cell free virus, and cultured at 1–2 × 106 cells per ml. Unless otherwise specified, cells were harvested on day 2 after infection and experiments were performed with the prototype strain New Guinea C (NGC), which belongs to DV serotype 2. Cell free supernatants were collected after overnight culture of DV exposed DC for the measurement of cytokines (TNF-α, IL-12, and type I interferons). Cytokines were measured using commercially available kits according to the manufacturer's recommendations (TNF- α and IL-12 R&D, and type I interferon endogen). Select experiments were performed with primary patient strains of DV from serotypes 1, 2, and 4. These primary DV strains were isolated from patient sera. Three IgG1 mouse anti-DV monoclonal antibodies (MoAb) (3H5, 2H2, and 7E11) were used to detect DV antigen on infected cells (
      • Henchal E.A.
      • McCown J.M.
      • Burke D.S.
      • Seguin M.C.
      • Brandt W.E.
      Epitopic analysis of antigenic determinants on the surface of dengue-2 virions using monoclonal antibodies.
      ). 3H5 is specific for DV serotype-2 envelope glycoprotein, 2H2 reacts with an envelope complex epitope conserved on serotypes 1–4, and 7E11 binds to a nonstructural protein 1, which is conserved among DV serotypes (generously donated by R. Putnak). Supernatants from the infected cell cultures collected for a DV plaque assay were centrifuged at 250 g for 10 min at 4°C and DV plaque forming units per ml were measured on BHK21 cells as previously described (
      • Morens D.M.
      • Halstead S.B.
      • Repik P.M.
      • Putvatana R.
      • Raybourne N.
      Simplified plaque reduction neutralization assay for dengue viruses by semimicro methods in BHK-21 cells: comparison of the BHK suspension test with standard plaque reduction neutralization.
      ). For ADE studies, DV (NGC) was preincubated for 2 h at 37°C, at MOI of 0.2 and 2.0, with serial dilutions of dengue immune plasma. Immature blood DC or K562 cells (a transformed continuous cell line permissive for dengue infection and susceptible to antibody-dependent enhancement) were exposed to pretreated virus for 2 h at 37°C and placed into fresh media for 24–48 h. The percentage of cells expressing DV-2 antigen was determined using MoAb 3H5 by both immunofluorescence and flow cytometry.

      Immunofluorescence assay (IFA)

      For immunofluroescence staining, cells were spotted in triplicate on 24 well HTC Super Cured IFA slides (Erie Scientific, Portsmouth, NH) as previously described (
      • Wu S.J.
      • Hayes C.G.
      • Dubois D.R.
      • Windheuser M.G.
      • Kang Y.H.
      • Watts D.M.
      • Sieckmann D.G.
      Evaluation of the severe combined immunodeficient (SCID) mouse as an animal model for dengue viral infection.
      ). The slides were air-dried and then fixed in cold 100% acetone at -20°C for 10 min. Primary anti-DV MoAb were detected with fluorescein isothiocyanate (FITC)-conjugated goat antimouse secondary antibody (Cappel/Organon Teknika, Durham, NC). Preparations were then counterstained with 0.05% Evans blue.

      Flow cytometry

      The DC phenotype was routinely assessed using a panel of directly conjugated MoAb and a FACScan flow cytometer (Becton Dickinson, San Jose, CA). The MoAb used were against HLA-DR, CD80, CD86, CD3, CD14, CD16 (FcγRI), CD20 (Becton Dickinson, San Jose, CA); CD83 (Immunotech, Miami, FL); CD1a, and CD32 (Fc γ RII), CD64 (Fcγ RIII) (PharMingen, San Diego, CA), and isotype matched controls. For the intracellular detection of DV antigens, cells were fixed with 4% paraformaldehyde and permeabilized with 0.5% saponin. In some experiments, directly conjugated 2H2-FITC was used. Alternatively, indirect staining with MoAb 3H5 was detected with an FITC-conjugated goat antimouse antibody.

      Human clinical trial vaccinee

      The volunteer participated in an IRB approved, phase I clinical trial to study a live attenuated tetravalent DV vaccine. The vaccine was composed of all four serotypes of DV that had undergone serial passages in primary dog kidney (PDK) cells and fetal rhesus lung cells: DEN-1 (45AZ5 PDK20), DEN-2 (S16803 PDK50), DEN-3 (CH53489 PDK20), and DEN-4 (341750 PDK20). The vaccine was given subcutaneously and consisted of 1.8 × 106, 3.0 × 106, 3 × 104, and 2.3 × 105 pfu of each serotype, respectively.

      Results

      Hemorrhagic rash in primary dengue infection

      Approximately 50% of patients with dengue fever and most with DHF/DSS experience some skin manifestation of disease. Figure 1 depicts the hemorrhagic nature of a dengue rash in a young child in South-east Asia. The classic distribution is on the extremities and can be petechial or purpuric in nature. Frequently, a more subtle skin manifestation of dengue is an erythematous rash described as “islands of normal skin surrounded by confluence of rash”. Patients with either dengue fever or DHF/DSS can experience hemorrhagic complications; more severe disease is associated with clinically significant hemorrhage and vascular leak syndrome. Most dengue infections in the U.S.A. are imported and inability to recognize the signs of infection can place the individual at unnecessary risk. The rashes associated with primary dengue infection can be biphasic, occuring early in the infection and after defervescence. The critical stage in severe dengue infection is around the time of defervescence. In the setting of DHF/DSS, hemorrhagic complications often begin as the temperature falls to normal and without appropriate fluid management, hypotension, shock, and death may follow. The WHO case definition of DHF/DSS is part of the Global Strategy to reduce the morbidity and mortality due to dengue infection. Current practice, in endemic areas, is to hospitalize patients with disease for observation and, in fact, a liberal admission policy has led to improved outcomes (
      • Vaughn D.
      • Green S.
      • Kalayanarooj S.
      • et al.
      Dengue viremia titer, antibody response pattern, and virus serotype correlate with disease severity.
      ).
      Figure thumbnail gr1
      Figure 1Primary dengue infection in a Thai infant demonstrates the distribution and hemorrhagic nature of the dengue rash.

      DC are preferential targets of dengue infection

      Historically, the macrophage has played a prominent role in dengue pathogenesis (
      • Halstead S.B.
      • O'Rourke E.J.
      • Allison A.J.
      Dengue viruses and mononuclear phagocytes II. Identity of blood and tissue leukocytes supporting in vitro infection.
      ). Many investigators have demonstrated the infectibility of mononuclear phagocytes, and in particular the enhancement of that infection in the presence of immune antibodies. We set out to determine if DC could be infected with DV, since the disease is transmitted through the skin (a relatively DC-rich site) and previous work has suggested that dengue replication may occur within the skin. Initial experiments compared infection rates between monocytes/macrophages and DC derived from the same donors. Figure 2(a, b) shows that approximately 10 times as many DC were infected when compared with macrophages after exposure to DV (MOI = 0.2). This relationship held true even when the cells were exposed to higher amounts of virus (MOI 2); with maximal infection of macrophages and DC being 5% and 35%, respectively. As early experiments involved “mixed” DC populations containing both immature and mature DC, we next asked whether either subpopulation of DC was more susceptible to infection. Separate populations of immature and mature DC were compared for infectivity and for evidence of productive infection, as determined by expression of dengue antigens, 3H5 (surface glycoprotein) and 7E11 (nonstructural protein indicating active replication), respectively, using flow cytometry. Figure 2(c) is a representative experiment (one of four) showing that immature DC were preferentially infected when compared with mature DC from the same donor under identical conditions. Experiments performed with primary isolates from DEN-1, -2, and -4 serotypes demonstrated similar results with expression of DV antigens restricted to the immature DC population, as shown in a representative experiment in Figure 2(d).
      Figure thumbnail gr2
      Figure 2Immunofluorescence and flow cytometry of DV-infected DC. Immunofluorescence staining of (a) macrophages and (b) DC infected with DEN-2 (NGC) virus at an MOI of 0.2 with MoAb 3H5 and an FITC-conjugated secondary antibody (green). Evan's blue was used as a counterstain (red). (c) Flow cytometry of DEN-2 (NGC) exposed immature and mature DC populations stained with 3H5 (structural protein) and 7E11 (nonstructural protein indicative of replication) MoAb. (d) Flow cytometry of immature DC exposed to three different serotypes of primary dengue isolates. Dengue antigens were detected on day 2. Scale bar: 20 µm.

      Infection of FcR-bearing immature DC is unaffected by immune serum

      The next set of experiments set out to determine if the ADE phenomenon could be observed in immature DC, as these cells are reported to express Fcγ receptors, the putative receptor involved in ADE. Figure 3(a) is a comparison of K562 cells (a human neoplastic cell line standardized for in vitro DV-antibody dependent enhancement assays) and DC that were exposed to virus that had been incubated with immune serum. Antibody titration experiments were performed. At the lowest dilutions tested, the immune serum neutralized DV infection of both K562 cells and DC; however, when the serum was diluted to 1:2560, the number of K562 cells expressing DV antigen increased from 5% to 40%. In contrast, there was no enhancement of the infection of DC at any serum dilution tested. Interestingly, both the immature DC and the K562 cells expressed similar levels of Fcγ receptors. Figure 3(b) shows the mean percentage (mean ± SD) of cells binding to anti-FcR antibodies (CD16, CD32, and CD64) on both DC and K562 cells using by flow cytometry. We found that 69% ± 18.5% of immature DC and 30% ± 13.5% of K562 cells expressed Fcγ RII in a minimum of three experiments. Furthermore, the mean fluorescence intensity (MFI ± SD) of the FcRII expression on immature DC was greater than that of the K562 controls (125.4 ± 64 and 38.3 ± 24.2, respectively). Much lower levels of Fcγ RI and Fcγ RIII were found on both of these cell types. Similar to K562 cells, primary monocytes under test conditions for enhancement, increased DV antigen expression from 4% to 24% in the presence of immune serum (data not shown).
      Figure thumbnail gr3
      Figure 3Flow cytometry application of the ADE assay using immune serum. (a) ADE assay comparing K562 cells (positive control cells) and immature DC after exposure to DV that had been incubated with various dilutions of immune serum. Abscissa indicates intensity infection as determined by binding of 2H2-FITC (structural protein) and ordinate is the side scatter (SSC) in each dot plot. (b) Fc receptor expression on immature DC and K562 cells as determined by binding of specific MoAb of each receptor and analyzed using flow cytometry. Graphs show the mean percentage of cells staining positive for the respective Fc receptor of at least three independent experiments (mean ± SD).

      Exposure of DC to DV induces TNF-α-mediated maturation

      The following experiments describe the consequences of DV and DC interactions. Our previous work had suggested a maturational effect on immature DC upon exposure to DV with induction and expression of intracellular DC-LAMP (
      • Wu S.J.
      • Grouard-Vogel G.
      • Sun W.
      • et al.
      Human skin Langerhans cells are targets of dengue virus infection.
      ), a DC-specific maturational marker (
      • de Saint-Vis B.
      • Vincent J.
      • Vandenabeele S.
      • et al.
      A novel lysosome-associated membrane glycoprotein, DC-LAMP, induced upon DC maturation, is transiently expressed in MHC class II compartment.
      ). We and colleagues (
      • Ho L.J.
      • Wang J.J.
      • Shaio M.F.
      • Kao C.L.
      • Chang D.M.
      • Han S.W.
      • Lai J.H.
      Infection of human dendritic cells by dengue virus causes cell maturation and cytokine production.
      ) have found that DC secrete TNF-α protein after exposure to DV. We hypothesized that as TNF-α itself can mature DC – this may, in part, be a mechanism for maturation. Figure 4(a) shows the concentration-dependent induction of TNF-α measured as secreted protein in culture supernatants after overnight incubation (18–24 h) in one representative experiment of four. As shown in Figure 4(b), in the presence of a specific monoclonal antibody that blocks both soluble and membrane-bound TNF-α induction of CD83 was markedly inhibited, whereas the control antibody (IgG1) allowed maturation. Heat killed DV did not infect the DC, nor did it induce any appreciable maturation (not shown). We looked for, but did not find, IL-12 or other soluble mediators of DC maturation that logically might be induced in the setting of viral infection, i.e., type I interferons (not shown). Data from our laboratory also suggest that DV infection induces functional maturity in DC (manuscript in preparation).
      Figure thumbnail gr4
      Figure 4TNF-α production by dengue-exposed DC induces DC maturation. (a) TNF-α protein detected (ELISA) in culture supernatants 24 h after DV exposure. Virus concentration indicated on abscissa and TNF-α in pg per ml plotted on ordinate. (b) CD83 expression on DC exposed to DEN-2 (black line on histogram) in the presence of 20 µg per ml anti-TNF-α antibody or IgG control antibody. Green line is the mock infected (media alone) DC.

      Evidence for in vivo infection of skin DC in dengue vaccine recipient

      A 26-y-old Asian female volunteer in a phase I clinical trial of a tetravalent live, attenuated dengue vaccine developed a maculopapular rash 12 d after the subcutaneous injection of the vaccine. A skin punch biopsy of the rash was taken from the lower extremity. Figure 5(a–d) shows stained skin biopsies from the rash. Figure 5(a) is a formalin-fixed parafin-embedded biopsy stained with hematoxylin and eosin, revealing a mild superficial, dermal, perivascular lymphocytic infiltrate. Figure 5(b), a serial section of part (a), stained with anti-CD1a (a Langerhans' cell marker), delineates enlarged bodies and extended processes of epidermal Langerhans' cells. Figure 5(c) and (d) are frozen sections. Figure 5(c) is stained for DV envelope glycoprotein (2H2) in red and Figure 5(d) demonstrates double staining of both DV antigens (red) and CD1a expression on Langerhans' cells (blue).
      Figure thumbnail gr5
      Figure 5Infection of skin DC in DV vaccine recipient. (a–d) Skin biopsies from the lower extremity rash area. (a) Hematoxylin and eosin staining of formalin fixed, paraffin-embedded sections showing a mild, superficial, dermal perivascular lymphocytic infiltrate. (b) CD1a immunohistochemistry staining of a serial section of that in (a) showing enlarged cell bodies and processes of intraepithelial Langerhans' cells. (c) Frozen section stained with 2H2, a MoAb specific for envelope glycoprotein, shows two red cells containing DV glycoprotein. (d) Double labeling of frozen section using a Langerhans' cell marker, CD1a (blue) and a DV antigen, 2H2 (red). Scale bars: (a) 10 µm, (b) 15 µm, (c, d) 5 µm.

      Discussion

      We have reviewed the clinical aspects of dengue infection, skin manifestations, and complications, and emphasized the enhanced pathogenicity of dengue under certain conditions. DHF/DSS, once a geographically restricted phenomenon (South-east Asia), is now reported in the Americas and considered an emerging infectious disease. Certain conditions need to exist for complicated dengue infections to occur within a population. These conditions include the need for the vector (Aedes aegyptii, now widespread in the tropics and in the south-eastern parts of the U.S.A.) and multiple circulating serotypes of dengue, allowing for sequential infections. These criteria are now present in many areas of the tropics and may explain the increasing impact of dengue disease in spite of heightened surveillance and global strategies.
      The in vivo evidence for ADE to explain DHF/DSS pathogenesis is compelling and gaining momentum, but still remains an area of controversy. The interested reader is referred to several excellent reviews of dengue and ADE (
      • Halstead S.
      Observations related to pathogenesis of dengue hemorrhaic fever. VI. Hypotheses and discussion.
      ;
      • Burke D.
      • Nisalak A.
      • Johnson D.
      • Scott R.
      A prospective study of dengue infections in Bangkok.
      ;
      • Morens D.M.
      Antibody-dependent enhancement of infection and the pathogenesis of viral disease.
      ). The corollary of ADE, higher viremia in more severe dengue infections, has been published recently in a prospective trial in 168 Thai children with acute dengue infection (
      • Vaughn D.
      • Green S.
      • Kalayanarooj S.
      • et al.
      Dengue viremia titer, antibody response pattern, and virus serotype correlate with disease severity.
      ). In addition to higher virus titres correlating with more severe disease, two other factors emerged from that study. First, those patients with DEN-2 infection experienced more severe disease than those infected with other serotypes. Second, more than 80% of the children studied experienced a secondary DV infection that was also associated with more severe disease. The association of DEN-2 with severe disease has been born out in many trials with the particular sequence of DEN-x followed by DEN-2 being a distinct predisposing risk factor for severe disease (
      • Morens D.M.
      Antibody-dependent enhancement of infection and the pathogenesis of viral disease.
      ).
      We have extended our previous work (
      • Wu S.J.
      • Grouard-Vogel G.
      • Sun W.
      • et al.
      Human skin Langerhans cells are targets of dengue virus infection.
      ) that showed that immature skin DC can be infected by DV and may be early targets for DV infection. Our data suggest that the FcRII that is involved in ADE is apparently not involved in the entry of DV into DC, or at least is not susceptible to immune enhancement phenomenon. DC do not manifest ADE, whereas the K562 cells display much higher levels of infectivity at subneutralizing antibody concentrations. Interestingly, both the immature DC and the K562 cells expressed moderate levels of FcRII. In fact, a greater percentage of the immature DC expressed FcRII and at greater intensity when compared with K562 cells. It is recognized, however, that although DC may bear Fc receptors, there may be variability in the expression of subunits of these receptors and/or variability in receptor function (
      • de la Salle H.
      • Haegel-Kronenberger H.
      • Bausinger H.
      • et al.
      Functions of Fc receptors on human dendritic Langerhans cells.
      ). Our results are in agreement with earlier work supporting the idea of two different entry mechanisms for DV, one being Fc-dependent and the other being Fc-independent (
      • Daughaday C.
      • Brandt W.
      • McCown J.
      • Russell P.
      Evidence for two mechanisms of dengue virus infection of adherent human monocytes: trypsin-sensitive virus receptors and trypsin-resistant immune complex receptors.
      ).
      In addition, we have extended our previous work by proposing a mechanism for the dengue-induced maturation of DC exposed to the virus. We and colleagues (
      • Ho L.J.
      • Wang J.J.
      • Shaio M.F.
      • Kao C.L.
      • Chang D.M.
      • Han S.W.
      • Lai J.H.
      Infection of human dendritic cells by dengue virus causes cell maturation and cytokine production.
      ) have shown that exposure of immature DC to DV induces TNF-α production. Here we show TNF-α-blocking experiments to demonstrate the inhibition of maturation (as defined by CD83 expression) in the presence of anti-TNF antibody that blocks soluble and membrane-bound TNF.
      Our data extend previous reports in primates and mice that suggested that DV could replicate in the skin and supported the concept that Langerhans cells/DC migrated from the skin (i.e., the site of antigenic challenge) and traveled to the draining lymph node by demonstration of antigen-labeled veiled cells in the lymphatics of sensitized animals (
      • Taweechaisupspong S.
      • Sriurairatana S.
      • Angsubhakorn S.
      • Yoksan S.
      • Khin M.
      • Sahaphong S.
      • Bhamarapravati N.
      Langerhans cell density and serological changes following intradermal immunisation of mice with dengue 2 virus.
      ,
      • Taweechaisupspong S.
      • Sriurairatana S.
      • Angsubhakorn S.
      • Yoksan S.
      • Bhamarapravati N.
      In vivo and in vitro studies on the morhphological change in the monkey epidermal langerhans cells following exposure to dengue 2 (16681) virus.
      ). The following scenario seems possible: the Langerhans cell is infected shortly after the bite of the mosquito and local replication of DV ensues. Some of these infected Langerhans cells/DC migrate to the draining lymph node where the virus can spread or be transferred to other cells in the context of a dengue-bearing DC. Once the virus has been amplified, and if the appropriate level of subneutralizing antibody is present or generated in the serum, the virus gains access to the monocyte/macrophage cellular compartment and DHF/DSS may develop. The triggering of an inflammatory cytokine cascade, including the in vivo induction of TNF- α (
      • Green S.
      • Vaughn D.
      • Klayanarooj S.
      • et al.
      Early immune activation in acute dengue illness is related to development of plasma leakage and disease severity.
      ), may play a role in dengue pathogenesis. Both monocytes/macrophages and DC produce TNF-α. One report indicates that whereas DV-infected monocytes produce TNF-α in the presence of enhancing antibodies, more IL-1 β is induced (Shaio et al, 1995) - both cytokines being early elements of the cytokine cascade. Other factors involved in severe disease are hemorrhagic events; some recent work implicates the virus itself as being able to activate plasminogen and subsequent fibrinolytic processes, supporting virus virulence mechanisms at work (
      • Monroy V.
      • Ruiz B.
      Participation of the Dengue virus in the fibrinolytic process.
      ).
      Currently there is no licensed dengue vaccine. Historically, we have had our best successes against viral pathogens using live attenuated vaccines. The complexities of DV, the four distinct serotypes, and the likelihood that immune enhancement is playing a role in disease pathogenesis pose unique vaccine development challenges. Current vaccine design employs tetravalent candidates with the hopes of eliciting a protective immune response against all four serotypes.
      The understanding of the cellular targets of dengue infection and the downstream consequences of DV infection at the cellular and molecular level, and of the viral virulence factors in the context of dengue pathogenesis, is of fundamental importance for the advancement of safe and effective strategies. Continued active research focused on understanding these phenomenon may advance vaccine development as it continues into the twenty-first century.

      ACKNOWLEDGMENTS

      We thank Douglas Walsh, Armed Forces Research Institute Medical Sciences, Bangkok, Thailand for the courtesy of the photograph in Figure 1.

      References

        • Burke D.
        • Nisalak A.
        • Johnson D.
        • Scott R.
        A prospective study of dengue infections in Bangkok.
        Am J Trop Med Hyg. 1988; 38: 172-180
        • Chang D.M.
        • Shaio M.F.
        Production of interleukin-1 (IL-1) and IL-1 inhibitor by human monocytes exposed to dengue virus.
        J Infect Dis. 1994; 170: 811-817
        • Daughaday C.
        • Brandt W.
        • McCown J.
        • Russell P.
        Evidence for two mechanisms of dengue virus infection of adherent human monocytes: trypsin-sensitive virus receptors and trypsin-resistant immune complex receptors.
        Infect Immun. 1981; 32: 469-473
        • Green S.
        • Vaughn D.
        • Klayanarooj S.
        • et al.
        Early immune activation in acute dengue illness is related to development of plasma leakage and disease severity.
        J Infect Dis. 1999; 179: 755-762
        • Gubler D.J.
        Dengue and dengue hemorrhagic fever. its history and resurgence as a global public health problem.
        in: Gubler D.J. Kuno G. Dengue and Dengue Hemorhagic Fever. Wallingford, CAB International1997: 1-22
        • Halstead S.
        Observations related to pathogenesis of dengue hemorrhaic fever. VI. Hypotheses and discussion.
        Yale J Biol Med. 1970; 42: 350-362
        • Halstead S.B.
        • O'Rourke E.J.
        • Allison A.J.
        Dengue viruses and mononuclear phagocytes II. Identity of blood and tissue leukocytes supporting in vitro infection.
        J Exp Med. 1977; 146: 218-229
        • Hawkes R.
        Enhancement of the infectivity of arboviruses by specific antisera produced in domestic fowls.
        Aust J Exp Biol Med Sci. 1964; 42: 465-482
        • Henchal E.A.
        • McCown J.M.
        • Burke D.S.
        • Seguin M.C.
        • Brandt W.E.
        Epitopic analysis of antigenic determinants on the surface of dengue-2 virions using monoclonal antibodies.
        Am J Trop Med Hyg. 1985; 34: 162-169
        • Ho L.J.
        • Wang J.J.
        • Shaio M.F.
        • Kao C.L.
        • Chang D.M.
        • Han S.W.
        • Lai J.H.
        Infection of human dendritic cells by dengue virus causes cell maturation and cytokine production.
        J Immunol. 2001; 166: 1499-1506
        • Kouri G.
        • Guzman M.
        • Bravo J.R.
        • Triana C.
        Dengue hemorrhagic fever/dengue shock syndrome: lessons from the Cuban epidemic, 1981.
        Bull World Health Organ. 1989; 67: 375-380
        • Monroy V.
        • Ruiz B.
        Participation of the Dengue virus in the fibrinolytic process.
        Virus Genes. 2000; 21: 197-208
        • Morens D.M.
        Antibody-dependent enhancement of infection and the pathogenesis of viral disease.
        Clin Infect Dis. 1994; 19: 500-512
        • Morens D.M.
        • Halstead S.B.
        • Repik P.M.
        • Putvatana R.
        • Raybourne N.
        Simplified plaque reduction neutralization assay for dengue viruses by semimicro methods in BHK-21 cells: comparison of the BHK suspension test with standard plaque reduction neutralization.
        J Clin Microbiol. 1985; 22: 250-254
        • O'Sullivan M.
        • Killen H.
        The differentiation state of monocytic cells affects their susceptibility to infection and the effects of infection by dengue virus.
        J Gen Virol. 1994; 75: 2387-2392
        • Rothman A.
        • Ennis F.
        Immunopathogenesis of dengue hemorrhagic fever.
        Virology. 1999; 257 (10.1006/viro.1999.9656): 1-6
        • de Saint-Vis B.
        • Vincent J.
        • Vandenabeele S.
        • et al.
        A novel lysosome-associated membrane glycoprotein, DC-LAMP, induced upon DC maturation, is transiently expressed in MHC class II compartment.
        Immunity. 1998; 9: 325-336
        • de la Salle H.
        • Haegel-Kronenberger H.
        • Bausinger H.
        • et al.
        Functions of Fc receptors on human dendritic Langerhans cells.
        Int Rev Immunol. 1997; 16: 187-203
        • Taweechaisupspong S.
        • Sriurairatana S.
        • Angsubhakorn S.
        • Yoksan S.
        • Khin M.
        • Sahaphong S.
        • Bhamarapravati N.
        Langerhans cell density and serological changes following intradermal immunisation of mice with dengue 2 virus.
        J Med Microbiol. 1996; 45: 138-145
        • Taweechaisupspong S.
        • Sriurairatana S.
        • Angsubhakorn S.
        • Yoksan S.
        • Bhamarapravati N.
        In vivo and in vitro studies on the morhphological change in the monkey epidermal langerhans cells following exposure to dengue 2 (16681) virus.
        Southeast Asian J Trop Med Public Health. 1996; 27: 664-672
        • Vaughn D.
        • Green S.
        • Kalayanarooj S.
        • et al.
        Dengue in the early febrile phase: viremia and antibody responses.
        J Infect Dis. 1997; 176: 322-330
        • Vaughn D.
        • Green S.
        • Kalayanarooj S.
        • et al.
        Dengue viremia titer, antibody response pattern, and virus serotype correlate with disease severity.
        J Infect Dis. 2000; 181: 2-9
        • World Health Organization
        Dengue Haemorrhagic Fever: Diagnosis, Treatment, Prevention and Control. 2nd edn. Geneva, WHO1997
        • World Health Organization
        Strengthening the Implementation of the Global Strategy for Dengue Fever/Dengue Hemorrhagic Fever Prevention and Control. Geneva, WHO2000
        • Wu S.J.
        • Hayes C.G.
        • Dubois D.R.
        • Windheuser M.G.
        • Kang Y.H.
        • Watts D.M.
        • Sieckmann D.G.
        Evaluation of the severe combined immunodeficient (SCID) mouse as an animal model for dengue viral infection.
        Am J Trop Med Hyg. 1995; 52: 468-476
        • Wu S.J.
        • Grouard-Vogel G.
        • Sun W.
        • et al.
        Human skin Langerhans cells are targets of dengue virus infection.
        Nat Med. 2000; 6: 816-820