Published online before print September 22, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Departments of Microbiology and Immunology,
Pathology, and
Pediatrics, Dalhousie University, Halifax, Nova Scotia, Canada
1 Correspondence: Department of Microbiology and Immunology, Dalhousie University, Halifax, Nova Scotia B3H 1X5, Canada. E-mail: robert.anderson{at}dal.ca
ABSTRACT
Severe forms of dengue virus disease, known as dengue hemorrhagic fever and dengue shock syndrome, result from an aberrant immune response involving antibody-dependent enhancement of infection, thrombocytopenia, and a loss of vascular integrity, culminating in hemorrhage, shock, and in some cases, death. Several studies have indicated that dengue virus infection results in the induction of apoptosis of certain cells believed to be contributory players in dengue pathogenesis. However, none have specifically examined the role of antibody enhancement in the context of induction of apoptosis. Here, we show that antibody-enhanced dengue virus infection of the FcR-bearing mast cell/basophil KU812 cell line results in a massive induction of apoptosis. Confocal microscopy and flow cytometry indicate two distinct subpopulations consisting of productively infected cells and apoptotic-uninfected bystanders. Apoptosis was found to be caspase-dependent, involving global caspase activation and cleavage of poly-ADP-ribose polymerase (PARP) and D4-guanosine diphosphate dissociation inhibitor (D4-GDI). Additional FcR-bearing cells, including K562, U937, and human mast cell 1 (HMC-1), were analyzed for apoptosis induction following infection. Although all cells displayed high susceptibility to antibody-enhanced dengue virus infection, only cells of a mast cell phenotype (KU812 and HMC-1) were found to undergo apoptosis. Dengue-induced apoptosis of KU812 cells was shown to require antibody-enhanced dengue virus infection by blockade of Fc
RII. Transfection of KU812 cells with L-SIGN/DC-SIGNR was able to overcome the requirement for antibody enhancement with regard to dengue virus infection and apoptosis.
Key Words: L-SIGN/DC-SIGNR transfection • poly(I:C) • FcR • hemorrhagic fever • shock syndrome
INTRODUCTION
Dengue virus is a ssRNA virus belonging to the flavivirus family and is the causative agent associated with severe disease in humans, referred to as dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS). It is an arthropod-borne virus using a mosquito vector, primarily Aedes aegypti for transmission to its host. Investigation into dengue virus pathogenesis has been hindered as a result of the lack of an acceptable animal model, as humans and nonhuman primates are the only known organisms susceptible to naturally occurring disease [1 , 2 ]. Some recent progress has been made in the use of mouse models of dengue virus infection, which display characteristics of dengue virus infection in humans, including apoptosis [3 ]. Nevertheless, the majority of knowledge about dengue virus pathogenesis has been gained through in vitro studies as well as clinical and epidemiological evidence acquired from patients suffering from dengue virus infection in dengue endemic areas of the world, primarily South-East Asia, South America, and Africa.
Studies have determined that dengue virus has the ability to induce apoptosis in a number of cells important to its pathogenesis or implicated therein {monocytes, lymphocytes, dendritic cells (DC), and endothelial cells [4 5 6 7 8 9 10 ]}. However, the mechanisms responsible have not been completely elucidated. The study of dengue virus disease is confounded further by an intriguing immune-adapted phenomenon known as antibody-dependent enhancement (ADE). This relies on the presence of pre-existing antibodies to previous dengue virus infection(s). This antibody-enhanced infection translates to increased infection of target cells, resulting in a more severe infection, potentially culminating in DHF/DSS in afflicted individuals [11 12 13 ]. Antibody-enhanced dengue virus infection has been linked to heightened production of vasoactive cytokines and chemokines, which have roles in the vascular dysfunction associated with severe dengue disease [14 15 16 ]. Surprisingly, there are no previous studies investigating the effects of ADE on dengue virus-induced apoptosis of any target cell. In the present study, we report a dramatic sensitivity of mast cell-like KU812 and human mast cell 1 (HMC-1) cells to apoptosis triggered by antibody-enhanced dengue virus infection.
MATERIALS AND METHODS
Cell culture
Human KU812 cells [17
] were maintained in RPMI 1640 (Sigma, Oakville, Ontario, Canada), supplemented with 10% FCS (Life Technologies, Grand Island, NY, USA) and 10 mM HEPES. HMC-1 cells [18
] were maintained in IMDM, supplemented with 10% FCS. All cell lines were passaged two to three times per week. K562 cells were grown in RPMI 1640 supplemented with 5% FCS. U937 cells were grown in RPMI 1640, supplemented with 10% FCS.
DC isolation and culture
Peripheral blood (50 ml) was obtained from normal donors collected in sodium citrate vacutainer tubes and then diluted 1:1 in RPMI 1640 at room temperature. The mixture was then overlayed on Histopaque (Sigma) and centrifuged at 400 gfor 30 min at room temperature. The PBMC band was collected and washed three times with RPMI 1640. Cells were resuspended in RPMI 1640, supplemented with 10% FCS, and incubated for 2 h in a 75 cm2 flask at 37°C, 5% CO2. Media were removed, and adherent cells were washed twice to ensure removal of nonadherent cells before the addition of 12 ml RPMI 1640, supplemented with 10% FCS, 150 ng/ml GM-CSF, 150 ng/ml IL-4 (Cedarlane Labs, Burlington, Ontario, Canada), and 100 ng/ml polymyxin B sulfate (Sigma). The culture was fed every 2–3 days by replacing it with fresh media until Day 6, when nonadherent cells exhibiting a DC morphology were harvested for use.
Dengue virus propagation
Dengue virus, type 2 (strain 16681) [11
], was propagated in African green monkey kidney Vero cell monolayers, cultured in endotoxin-free RPMI 1640 (Sigma), supplemented with 1% FCS. In some cases, dengue virus was UV-inactivated by exposure to a germicidal bulb at a distance of 10 cm for 10 min. The dsRNA analog polyinosinic:polycytidylic acid [poly(I:C); Sigma] was used from stock solution of 1 mg/ml, dissolved in sterile, distilled water.
Sera
For ADE assays of dengue virus infection, a dengue-immune serum pool was prepared by using nine convalescent-phase sera from patients recovering from a dengue virus serotype 2 infection. The individual dengue-convalescent patient sera were obtained from a collection provided by Dr. Bruce Innis (Armed Forces Research Institute of Medical Science, Bangkok, Thailand) and described briefly in ref. [19
].
Transfection conditions
KU812 cells (5x106) were resuspended in 100 ul Nucleofector (Amaxa, Gaithersburg, MD, USA) and transfected with 5 ug DC-specific ICAM-grabbling nonintegrin receptor (SIGNR)/liver/lymph node (L)-SIGN cDNA [pcDNA3-DC-SIGNR, Cat. #6746, obtained from National Institutes of Health (NIH) AIDS Research and Reference Reagent Program, Bethesda, MD, USA]. Cells were maintained in RPMI 1640, supplemented with 10% FCS plus 1 mg/ml Geneticin (50 mg/ml stock; Invitrogen, Carlsbad, CA, USA) for stable transfection.
Infection conditions
Aliquots of dengue virus were incubated alone or in the presence of dengue-immune human serum for 60 min at 4°C. KU812, DC-SIGNR-transfected KU812, HMC-1, K562, and U937 cells were inoculated with dengue virus (with or without dengue-immune serum) at a multiplicity of infection (MOI) of 1–2 (as assayed on Vero cells). Following 90 min adsorption at 4°C, unadsorbed virus and antibody were removed by washing, and cells were resuspended in medium for incubation at 37°C. Mock infection was performed using Vero cell-conditioned RPMI-1640 media, supplemented with 1% FCS. Where indicated, 5 µg/ml camptothecin (Sigma; 24 h incubation at 37°C) or 1 M sorbitol (Sigma; 2 h at 37°C followed by washout and incubation in normal medium for 4 h) treatment was used as a positive control for apoptosis induction.
Plaque assay
Supernatants from dengue virus-infected cells were harvested, serially diluted, and inoculated onto Vero cell monolayers for plaque assay. At 30 h postinfection, cells were washed once with PBS, fixed with methanol for 10 min, and then placed at 4°C in PBS until immunostaining, which was performed using dengue-specific antibody and goat anti-human alkaline phosphatase (AP) conjugate followed by development with the one-step nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate AP developer (Pierce, Rockford, IL, USA).
Caspase inhibition
A number of caspase inhibitors were used at a 50 µM concentration. Broad-spectrum caspase inhibitor [z-Val-Ala-Asp-fluoromethylketone (VAD-fmk)] was from Calbiochem (La Jolla, CA, USA). Caspase 8 inhibitor [z-Ile-Glu-Thr-Asp (IETD)-fmk], caspase 9 inhibitor [z-Leu-Glu-His-Asp (LEHD)-fmk], caspases 3 and 7 inhibitor [z-Asp-Glu-Val-Asp (DEVD)-fmk], and caspase 1 inhibitor [z-Trp-Glu-His-Asp (WEHD)-fmk] were from R&D Systems (Minneapolis, MN, USA).
Cycloheximide inhibition
Following dengue virus adsorption, cells were resuspended in fresh RPMI 1640 containing 1% FCS, with or without 0.1 µg/ml cycloheximide (Sigma) for 24 h.
DNA fragmentation analysis
DNA from KU812 cell cultures was extracted by the Hirt procedure [20
] and subjected to electrophoresis on 1.5% agarose gels. A sorbitol (Sigma) treatment was used as a positive control for apoptosis induction.
FACS analysis
Cells were harvested at indicated times postinfection, fixed in 4% paraformaldehyde, permeabilized using 0.1% saponin, and stained for intracellular expression of dengue envelope (E) protein using mAb 1B7 [21
], nonstructural 1 (NS1) protein using mAb 7E11 [22
], or an IgG2a isotype control antibody. A goat anti-mouse Alexa 488 conjugate (Molecular Probes, San Diego, CA, USA) was used as a secondary antibody. Washed cells were resuspended in 1% paraformaldehyde and subjected to analysis using a FACScalibur flow cytometer (Becton Dickinson, San Jose, CA, USA). Data were analyzed using FCS Express 3 and CellQuest. AnnexinV-Alexa 647 conjugate (Molecular Probes) was used to determine phosphatidylserine (PS) exposure. Propidium iodide (PI; 2 µg/ml) was used to observe late apoptotic/necrotic cells. Analysis for expression of DC-SIGNR/L-SIGN (CD299) or DC-SIGN was performed using anti-DC-SIGNR antibody (#11423; NIH) or anti-DC-SIGN antibody (designated 120507; Abcam, Inc., Cambridge, MA), respectively.
Receptor blockade
Anti-FcRII (CD32) antibody was used as described previously [23
].
Western blot
Cells were harvested at indicated times postinfection, washed in 1x PBS, and resuspended in 60 µl dissociation buffer. The total sample (15 µl) was loaded per well and separated using 15% SDS-PAGE. Protein was then transferred to polyvinylidene difluoride by semidry transfer. Blots were blocked in TBST (0.1%) + 5% nonfat milk, washed, and probed using the specified antibodies. Antibodies against caspases 7–9 were from Cell Signaling Technologies (Danvers, MA, USA). Anti-poly-ADP-ribose polymerase 1 (anti-PARP-1) antibody was from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). Anti-D4-guanosine diphosphate dissociation inhibitor (anti-D4-GDI) antibody was from Imgenex (San Diego, CA, USA).
RESULTS
Antibody-enhanced dengue virus infection of KU812 cells results in DNA fragmentation
We have shown previously that the KU812 prebasophil/mast cell line is susceptible to antibody-enhanced dengue virus infection [15
, 16
]. We now report that antibody-enhanced infection leads to dramatic apoptosis of these cells. This was established initially through analysis of DNA fragmentation following infection by dengue virus. It was found that only cells infected with dengue virus in the presence of subneutralizing concentrations (1:1000 and 1:10,000) of convalescent dengue patient sera resulted in DNA fragmentation (Fig. 1A
). Treatment with 1 M sorbitol is shown as a positive control for apoptosis induction compared with untreated cells (Fig. 1B)
.
 View larger version (73K): [in a new window] |
Figure 1. Antibody-enhanced dengue virus infection of KU812 cells results in DNA fragmentation. (A) KU812 cells were inoculated with dengue virus alone (Den) or dengue virus or UV-inactivated dengue virus with subneutralizing dilutions of human dengue immune serum (1:1000 and 1:10,000 Ab). At 24 h postinfection, cells were harvested, and low molecular weight DNA was extracted using the Hirt method [20]. Samples were then separated by 1.5% agarose gel electrophoresis. Bands were visualized by ethidium bromide. (B) Sorbitol treatment of KU812 cells was used as a positive control for DNA fragmentation compared with untreated cells.
|
 View larger version (30K): [in a new window] |
Figure 2. Antibody-enhanced dengue virus infection leads to productive infection and apoptosis in two separate subpopulations of KU812 cells. KU812 cells were inoculated with dengue virus and a subneutralizing dilution of human dengue immune serum (1:10,000 Ab). (A) FACS was used to analyze cells for Annexin V binding [fluorescence 4-height (FL4-H)] and expression of dengue virus E protein (FL1-H). SSC, Side-scatter; FSC, forward-scatter. (B) Cells were also observed by confocal microscopy to confirm the result of FACS analysis. Annexin V+ cells are blue, and E protein-expressing cells are green. Camptothecin treatment (5 µg/ml) was used as a positive control for apoptosis induction. Data are representative of at least three separate experiments.
|
RII and exhibit ADE virus infection [15
, 24
, 25
]. Cells were mock-infected or infected with dengue virus and a subneutralizing concentration (1:10,000) of dengue-immune patient sera. Cells were analyzed at 24 h and 48 h postinfection by flow cytometry for apoptosis, as determined by Annexin V binding to externalized PS and detection of dengue virus E protein (Fig. 3 A and B
) or NS1 protein (Fig. 3 C and D)
, indicative of infection. There were no notable mean fluorescence intensity differences between cell types stained for E or NS1 protein. Plaque assays were also performed on supernatants collected at 0, 24, and 48 h postinfection (Table 1
), further confirming active dengue virus infection and the production of infectious virus, despite the differences in the amount of virus produced between cell types. There was no positive E protein staining of cells at the 0 h time-point (Fig. 3E)
, immediately following adsorption, indicating that E protein detected at 24 h and 48 h (Fig. 3 A and B)
is from newly produced E protein and not a remnant of the initial virus inoculum. A noticeable loss of total E protein staining occurred for infected KU812 cells from 24 h to 48 h. This decrease in structural E protein staining may be a result of virus shedding. Interestingly, the NS1 protein staining profile changes from 24 h to 48 h, with more NS1+ cells becoming apoptotic by 48 h postinfection (Fig. 3D)
. This may suggest that at least some apoptotic cells were productively infected. KU812 cells exhibited the most dramatically increased level of apoptosis by 24 h following infection (46.5±17.5 Annexin V+ Den+Ab compared with 6.9±2.2 mock; P<0.05; Fig. 3A
). Apoptosis induction in HMC-1 was not as rapid as that observed in dengue virus-infected KU812 cells, with the greatest increase in Annexin V-positive cells occurring by 48 h postinfection (22.3±0.94 Annexin V+ Den+Ab compared with 16.9±0.48 mock; P<0.01; Fig. 3B
). In contrast, although K562 and U937 cells exhibited a comparable level of infection, they were not observed to undergo any significant level of apoptosis by 48 h postinfection (15.1±7.1 Annexin V+ Den+Ab compared with 15.2±6.5 mock, P>0.05; and 5.4±2.0 Annexin V+ Den+Ab compared with 3.5±0.43 mock, P>0.05, respectively).
 View larger version (59K): [in a new window] |
Figure 3. Dengue-induced apoptosis is dependent on cell type. U937, K562, HMC-1, and KU812 cells were inoculated with dengue virus and a subneutralizing dilution of human dengue immune serum (1:10,000 Ab). Cells were analyzed by FACS for Annexin V binding (FL4) and E protein expression (FL1) at 24 h (A) and 48 h (B) postinfection or NS1 protein expression (FL1) at 24 h (C) and 48 h (D) postinfection. Shown are data from a representative of three separate experiments. (E) Cells were stained for dengue E protein immediately following adsorption. Shaded histograms indicate isotype; open histograms are mAb 1B7 for specific dengue E protein.
|
|
View this table: [in a new window] |
Table 1. Plaque Assay of Infectious Dengue Virus Produced by Dengue-Infected Cells
|
RII is required for dengue infection and apoptosis of KU812 cellsRII antibody (Fig. 4)
. Additionally, UV-inactivated virus with subneutralizing dengue-immune serum did not result in infection or apoptosis.
 View larger version (49K): [in a new window] |
Figure 4. Antibody-enhanced dengue virus infection and apoptosis are inhibited by blockade of FcRII (CD32). Prior to virus adsorption, KU812 cells were blocked with anti-CD32 for 45 min. Cells were then inoculated with dengue virus or UV-inactivated virus, with or without a subneutralizing dilution of human dengue-immune serum (1:10,000 Ab). At 24 h postinfection, cells were harvested and stained for dengue virus E protein expression or apoptosis by Annexin V and analyzed by FACS. Data are representative of at least three separate experiments.
|
 View larger version (22K): [in a new window] |
Figure 5. Apoptosis induction requires infectious dengue virus. (A) KU812 cells were inoculated with dengue virus and subneutralizing dilutions of human dengue-immune serum (1:1000 and 1:10,000 Ab). Cells were harvested at 24 h postinfection and analyzed by FACS for Annexin V binding and PI exclusion to differentiate early apoptotic cells (AnnV+/PI–) from late apoptotic/necrotic cells (AnnV+/PI+). The dsRNA analog poly(I:C) (pIC) was added to culture medium at concentrations of 1, 10, and 100 µg/ml. Data are representative of three separate experiments. (B) UV treatment does not affect cell binding of dengue virus-antibody complexes. Control or UV-inactivated, 35S-radiolabeled dengue virus was mixed with a 1:10,000 dilution of human dengue-immune sera and added to KU812 cells. Following 90 min adsorption, cells were washed and resuspended in loading buffer. Proteins were separated by 15% SDS-PAGE and visualized by autoradiography. E, E protein; prM, pre-Membrane protein; M, M protein. (C) Cycloheximide (CHX) inhibits induction of apoptosis. KU812 cells were inoculated with dengue virus and a subneutralizing dilution of human dengue immune serum (1:10,000 Ab). Following 90 min adsorption, cells were washed and resuspended in normal medium or in medium containing 0.1 µg/ml cycloheximide for 24 h and then analyzed for Annexin V binding (FL4). Data are representative of at least three separate experiments.
|
Additional experiments performed using combinations of UV-inactivated virus, dengue immune serum, and poly(I:C) showed no effects on cell viability (data not shown). Also, addition of 0.1 µg/ml cycloheximide, an inhibitor of protein synthesis, to KU812 cells immediately following virus adsorption, was able to block apoptosis (Fig. 5C) . When cycloheximide was added at 4 h or 6 h postadsorption, blockade of apoptosis was less effective (data not shown). These results suggest that the presence of native dengue virus-derived dsRNA and/or synthesis of a specific virus protein(s) may be responsible for apoptosis induction.
Antibody-enhanced dengue virus-induced apoptosis of KU812 cells is caspase-dependent
We wished to determine if the inhibition of caspase activity would prevent apoptosis induced by antibody-enhanced dengue virus infection of KU812 cells. Following virus adsorption, cells were treated with a 50 µM concentration of broad or selective inhibitors of caspases including the broad caspase inhibitor (z-VAD), caspase 8 inhibitor (z-IETD), caspase 9 inhibitor (z-LEHD), and caspases 3 and 7 inhibitor (z-DEVD). Effectiveness of the caspase inhibitors was verified using camptothecin-induced apoptosis as a reference control (data not shown) and by Western blotting as indicated below. At 24 h postinfection, cells were harvested and subsequently analyzed for Annexin V binding as well as PI exclusion. Treatment with the broad-spectrum caspase inhibitor (z-VAD) was the most effective, significantly decreasing the level of dengue virus-induced apoptosis (Fig. 6A
). Importantly, treatment with this broad caspase inhibitor had no effect on dengue virus E protein expression (Fig. 6B)
or on infectious virus production as determined by plaque assay (data not shown). The more selective caspase inhibitors resulted in varied amounts of inhibition (Fig. 6A)
.
 View larger version (14K): [in a new window] |
Figure 6. Antibody-enhanced dengue virus infection-induced apoptosis of KU812 cells is caspase-dependent. (A) KU812 cells were inoculated with dengue virus and a subneutralizing dilution of human dengue immune serum (1:1000 Ab). The specific caspase inhibitors for caspase 1 (WEHD), caspases 3 and 7 (DEVD), caspase 8 (IETD), and caspase 9 (LEHD) and the broad-spectrum caspase inhibitor (VAD) were added at 50 µM following adsorption. Cells were harvested at 24 h postinfection, and the level of apoptosis was determined by FACS analysis of Annexin V and PI staining. *, P < 0.05, compared with the D2 + 1:1000 Ab group. (B) Caspase inhibition (50 µM VAD) does not affect the level of dengue virus infection as determined by dengue virus E protein staining (FL1). Data are representative of at least three separate experiments.
|
 View larger version (56K): [in a new window] |
Figure 7. Antibody-enhanced dengue virus infection of KU812 cells triggers cleavage of caspases 8 and 9 (C8 and C9, respectively). (A) KU812 cells were mock-inoculated or inoculated with dengue virus and a subneutralizing dilution of human dengue immune serum (1:10,000 Ab). At 6, 12, 18, and 24 h postinfection (Hours PI), cells were harvested, and cell extracts were resolved by 15% SDS-PAGE and Western blotted for detection of full-length and cleaved fragments of caspases 8 and 9. (B) Caspase inhibitors decrease cleavage of caspases 8 and 9 by antibody-enhanced dengue virus infection of KU812 cells. Caspase inhibitors for caspases 3 and 7 (DEVD), caspase 8 (IETD), and caspase 9 (LEHD) and the broad caspase inhibitor (VAD) were added at 50 µM following adsorption, and cells were harvested at 24 h postinfection. β-Actin was used as a loading control. Data are representative of at least three separate experiments.
|
Effector caspase activation and cleavage of caspase substrates following dengue virus infection are prevented by caspase inhibition
Cleavage and activation of the effector caspase 7, in addition to cleavage of the well-known caspases 3 and 7 substrates PARP and D4-GDI, were also analyzed by Western blot. The greatest inhibition of caspase 7 cleavage was observed with the broad-spectrum caspase inhibitor (z-VAD), almost completely preventing the appearance of the p20 fragment. Cleavage of PARP and D4-GDI was found to occur, demonstrating functional caspases 3 and 7 activity following antibody-enhanced dengue virus infection (Fig. 8
). Treatment of cells with camptothecin, an inducer of apoptosis, provided a positive control for cleavage of D4-GDI. Addition of the broad-spectrum caspase inhibitor (z-VAD) was sufficient to prevent cleavage of PARP and D4-GDI. The specific inhibitors for caspases 3 and 7 (z-DEVD) and caspase 8 (z-IETD) were able to decrease the cleavage of PARP and D4-GDI compared with untreated cells but to a lesser extent than the broad inhibitor (z-VAD). This was expected, as the broad caspase inhibitor z-VAD blocks the activation of the upstream protease cascade during the apoptotic process, whereas z-DEVD acts specifically downstream as a competitive inhibitor for the caspase 3 cleavage site.
 View larger version (46K): [in a new window] |
Figure 8. Caspase 7 activation and cleavage of caspase substrates PARP and D4-GDI following antibody-enhanced dengue virus infection of KU812 cells, which were inoculated with dengue virus and a subneutralizing dilution of human dengue immune serum (1:10,000 Ab). Caspase inhibitors for caspases 3 and 7 (DEVD), caspase 8 (IETD), and the broad caspase inhibitor (VAD) were added at 50 µM following adsorption. Cells were harvested at 24 h postinfection and were probed for full-length and cleaved fragments of caspase 7, PARP, and the cleaved fragment of D4-GDI. Camptothecin (5 µg/ml) treatment was used as a positive control for D4-GDI cleavage. Data are representative of at least three separate experiments.
|
RII-blocking antibody. Thus, L-SIGN/DC-SIGNR is able to substitute for FcRII in mediating dengue infection and apoptosis of KU812 cells.
 View larger version (54K): [in a new window] |
Figure 9. Transfection of KU812 cells with L-SIGN/DC-SIGNR removes the requirement for ADE in dengue virus infection and apoptosis. (A) L-SIGN/DC-SIGNR-transfected KU812 cells were compared against normal KU812 cells (negative) for surface expression. Additionally, normal KU812 cells and monocyte-derived DC (positive) were stained for surface expression of DC-SIGN. The empty curves are isotype controls; filled curves are anti-DC-SIGNR/L-SIGN or anti-DC-SIGN. (B) Normal KU812 cells and L-SIGN/DC-SIGNR-transfected KU812 cells were treated with dengue virus, with or without a subneutralizing dilution of dengue immune serum (1:10,000) or mock. At 24 h postinfection, cells were harvested, stained for expression of dengue E protein and apoptosis by Annexin V binding, and analyzed by FACS. Data are representative of three separate experiments.
|
In this study, we provide the first evidence for induction of apoptosis in antibody-enhanced dengue virus infection in mast cell/basophil-like KU812 and HMC-1 cells. KU812 and HMC-1 as well as cord blood-derived mast cells have been shown recently to support antibody-enhanced dengue virus infection mediated primarily through FcRII [23
]. Mast cells are prominent in the skin (the initial site of dengue infection) [29
, 30
] as well as surrounding blood vessels and are capable of producing vasoactive mediators, such as histamine and inflammatory factors {including IL-1β, IL-6, CCL5 (RANTES), CXCL10 (IFN-induced protein-10), CCL3/CCL4 (MIP-1
/β) [15
, 16
]}. Although it is well-known that mast cell degranulation results in the release of a variety of vasoactive (i.e., histamine) and cytotoxic granules (i.e., granzyme B), it is important to note that in our studies of antibody-enhanced dengue virus infection of mast cell-like cells, we do not observe any degranulation as determined by β-hexosaminidase release assay but rather release of a specific profile of secreted cytokines and chemokines [16
]. Taken together with other evidence for mast cell involvement in dengue pathogenesis [31
32
33
34
35
36
37
], the results of the present study indicate multiple mechanisms of mast cell dysregulation by dengue virus.
Dengue virus-induced apoptosis has been shown in vivo, specifically in CD8+ T cells, as well as in liver hepatocytes and Kupffer cells from dengue-infected patients [6 , 38 ]. Apoptosis has also been documented in in vitro dengue virus-infected cells, including human hepatoma cells, neuroblastoma cells, DC, endothelial cells, and monocytes [4 , 7 , 9 , 39 ]; however, a clear-cut, cellular mechanism for dengue virus-induced apoptosis remains to be determined.
Our finding that dengue-mediated apoptosis of KU812 cells primarily affects bystander cells is intriguing. We propose two possible mechanisms for this observation. The first mechanism involves induction of apoptosis early in virus infection with rapid shut-down of further viral gene expression. The second mechanism involves virus-triggered release of soluble mediators of apoptosis, which then act on uninfected neighboring cells. We have eliminated several candidate soluble mediators of apoptosis, e.g., Fas ligand, TNF-, and lymphotoxin- (data not shown). In addition, KU812 cells are reported to be resistant to TRAIL-mediated apoptosis [40
]. Although TRAIL has been shown to be up-regulated in dengue virus-infected hepatoma cells as well as in infected HUVECs, TRAIL induces apoptosis only in the infected hepatoma cells. A possible explanation is that the receptor repertoire may vary between these cell types. There exist a number of receptors for TRAIL, including decoys, which would affect the availability of TRAIL to initiate a proapoptotic signal. Additionally, the intracellular action of pro- and antiapoptotic proteins can greatly influence sensitivity to a proapoptotic ligand such as TRAIL. The mechanism of bystander apoptosis in KU812 cells is an important subject for ongoing investigation.
In our study, promiscuous caspase activation was found to be a prominent feature of antibody-enhanced dengue infection of KU812 cells. Of these, the inhibitor for the effector caspases 3 and 7 (z-DEVD), the hub of caspase activation, and execution responsible for the downstream actions of the caspase cascade provided the greatest level of protection from apoptosis. Inhibition of either of the initiator caspases 8 (extrinsic) and 9 (intrinsic), although able to decrease the level of apoptosis significantly, was not sufficient to prevent apoptosis. Moreover, caspases 8 and 9 underwent cleavage at similar time-points. These results indicate that antibody-enhanced, dengue virus-induced apoptosis is a caspase-dependent process with roles for the intracellular and extracellular arms of the caspase pathway. This stands in contrast to examples of antibody-independent dengue infection of cells such as neuroblastoma [41 ] and Hela cells [42 , 43 ], which implicated specific caspases as major apoptotic mediators. For example, caspases 6 and 9 were found to play a dominant role in dengue-mediated apoptosis of neuroblastoma cells [41 ]. In contrast, a role for caspase 3 but not caspase 9 has been inferred from overexpressed dengue virus M protein-mediated apoptosis of HeLa cells [42 , 43 ]. Furthermore, studies with another flavivirus, West Nile virus (WNV), have implicated cleavage and activation of caspase 8 in apoptosis mediated by virus protease (NS3) [44 ]. NS3 has also been implicated in dengue virus-mediated apoptosis of Vero cells [45 ].
The study of the mechanisms involved in virus-induced apoptosis is complicated as a result of the intricate relationship between virus infection and the host innate immune response. Apoptosis may result from the action of proapoptotic viral/virus-induced proteins and/or triggering of innate pathogen recognition receptors for products of virus replication (ssRNA or dsRNA) including TLR3, -7, or -8, dsRNA-activated protein kinase receptor (PKR), retinoic acid-inducible gene I (RIG-I), and melanoma differentiation-associated gene-5 (MDA-5), leading to antiviral responses, including type I IFN production [26 , 27 , 46 47 48 ]. Investigation of WNV infection in a mouse model found that detection of dsRNA by TLR3 led to increased inflammation and breakdown of the blood-brain barrier in a TNFRI-dependent manner, indicative of apoptosis [49 ]. Bovine viral diarrhea virus, another flavivirus, has been shown to induce apoptosis of cells via PKR [50 ]. Recently, the flaviviruses have been shown to up-regulate mRNA expression for RIG-I and MDA-5 [51 ]. It has also been demonstrated that flavivirus infection triggers IFN production via a RIG-I-dependent mechanism [52 ]. Whether these responses play a role in dengue virus-induced apoptosis remains to be resolved.
Treatment of KU812 cells with the dsRNA analog poly(I:C) was found not to induce apoptosis in our study, arguing against a role for external or endosomal dsRNA as an apoptotic trigger. This is in agreement with other studies using poly(I:C) in the context of antiviral responses mediated by TLR3 [53
, 54
]. Mast cells can detect extracellular dsRNA via TLR3, leading to a transcriptional increase and release of IFN-/β. Importantly, responses to external or endosomal poly(I:C) do not result in mast cell degranulation or changes in cell viability [53
]. It is unknown whether intracellular dsRNA sensors (RIG-I, MDA-5, PKR) triggered by interaction with de novo-produced viral dsRNA during infection, transfected poly(I:C), or native dsRNA sequences could interface with apoptotic machinery, resulting in cell death. Recent evidence suggests this may be the case [55
].
Our observation that L-SIGN/DC-SIGNR can replace FcRII and dengue-specific antibody for dengue virus infection and apoptosis of KU812 cells illustrates the idea that apoptosis induction is not absolutely dependent on a specific virus receptor. Nevertheless, for normal KU812 cells, the absence of L-SIGN/DC-SIGNR coupled with the presence of FcRII suggests a role for ADE in dengue virus infection and associated apoptosis. Moreover, our report is the first to show that cells lacking known dengue virus receptors DC-SIGN or L-SIGN/DC-SIGNR but supporting infection via FcR are susceptible to dengue virus-mediated apoptosis.
ACKNOWLEDGEMENTS
This research was supported by the Canadian Institutes of Health Research. The L-SIGN/DC-SIGNR expression plasmid #6746 and anti-DC-SIGNR antibody #11423 were obtained through the NIH AIDS Research and Reference Reagent Program, Division of AIDS: pcDNA3-DC-SIGNR, from Drs. S. Pöhlmann, F. Baribaud, F. Kirchhoff, and R. W. Doms. The anti-NS1 antibody mAb 7E11 was kindly provided by Drs. Tim Endy and Robert Putnak (Walter Reed Army Institute of Research, Washington, DC, USA). We are grateful to David Conrad, Ian Haidl, and Pat Colp for helpful expertise, advice, and discussion.
Received March 7, 2008; revised August 20, 2008; accepted August 27, 2008.
REFERENCES
caused by infection of cultured human monocytes with dengue virus Am. J. Trop. Med. Hyg. 68,48-53
B which in turn induces apoptotic cell death J. Virol. 71,3244-3249RII in antibody-enhanced dengue virus infection of human mast cells and associated CCL5 release J. Leukoc. Biol. 80,1242-1250B are sequentially involved J. Virol. 74,8680-8691-stimulated neutrophils and monocytes release a soluble form of TNF-related apoptosis-inducing ligand (TRAIL/Apo-2 ligand) displaying apoptotic activity on leukemic cells Blood 103,3837-3844B pathway and the innate immune response by viruses Oncogene 25,6844-6867[CrossRef][Medline]B activation Microbes Infect. 8,157-171[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  Y.-W. Lin, S.-W. Wang, Y.-Y. Tung, and S.-H. Chen Enterovirus 71 Infection of Human Dendritic Cells Exp Biol Med, October 1, 2009; 234(10): 1166 - 1173. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
