Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Harwood, L. J. Articles by McCullough, K. C. Search for Related Content PubMed PubMed Citation Articles by Harwood, L. J. Articles by McCullough, K. C.

 Previous Article  |  Next Article 

Journal of Virology, July 2008, p. 6379-6394, Vol. 82, No. 13
0022-538X/08/$08.00+0     doi:10.1128/JVI.00021-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Dendritic Cell Internalization of Foot-and-Mouth Disease Virus: Influence of Heparan Sulfate Binding on Virus Uptake and Induction of the Immune Response

Lisa J. Harwood,^1* Heidi Gerber,¹ Francisco Sobrino,^2,3 Artur Summerfield,¹ and Kenneth C. McCullough¹

Institute of Virology and Immunoprophylaxis, 3147 Mittelhäusern, Switzerland,¹ Centro de Biología Molecular Severo Ochoa (CSIC-UAM), 28049 Cantoblanco, Madrid, Spain,² Centro de Investigación en Sanidad Animal, INIA, 28130 Valdeolmos, Madrid, Spain³

Received 4 January 2008/ Accepted 2 April 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Dendritic cells (DC), which are essential for inducing and regulating immune defenses and responses, represent the critical target for vaccines against pathogens such as foot-and-mouth disease virus (FMDV). Although it is clear that FMDV enters epithelial cells via integrins, little is known about FMDV interaction with DC. Accordingly, DC internalization of FMDV antigen was analyzed by comparing vaccine virus dominated by heparan sulfate (HS)-binding variants with FMDV lacking HS-binding capacity. The internalization was most efficient with the HS-binding virus, employing diverse endocytic pathways. Moreover, internalization relied primarily on HS binding. Uptake of non-HS-binding virus by DC was considerably less efficient, so much so that it was often difficult to detect virus interacting with the DC. The HS-binding FMDV replicated in DC, albeit transiently, which was demonstrable by its sensitivity to cycloheximide treatment and the short duration of infectious virus production. There was no evidence that the non-HS-binding virus replicated in the DC. These observations on virus replication may be explained by the activities of viral RNA in the DC. When DC were transfected with infectious RNA, only 1% of the translated viral proteins were detected. Nevertheless, the transfected cells, and DC which had internalized live virus, did present antigen to lymphocytes, inducing an FMDV-specific immunoglobulin G response. These results demonstrate that DC internalization of FMDV is most efficient for vaccine virus with HS-binding capacity, but HS binding is not an exclusive requirement. Both non-HS-binding virus and infectious RNA interacting with DC induce specific immune responses, albeit less efficiently than HS-binding virus.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Foot-and-mouth disease (FMD) is a highly contagious disease affecting cloven-hoofed animals, and it is an economically important disease of livestock worldwide. The causative agent of the disease is the FMD virus (FMDV) of the genus Aphthovirus in the family Picornaviridae. FMDV carries a positive single-stranded RNA (ssRNA) genome of 8,500 nucleotides, coding for a single polyprotein that is autolytically processed by viral proteases into a four-virion capsid protein and several nonstructural proteins. During viral infection, FMDV is typical of other viruses in that it must interact with the appropriate cell receptor that is essential for initiating internalization of the virus and leading to its replicative cycle. FMDV can use a number of different cellular receptors and coreceptors, allowing expansion of host range and cell tropism with an increased probability of virus survival (3).

FMDV has been reported to employ particular integrin receptors for interaction with cells, namely, vβ1 (27), vβ3 (10), vβ6 (28), and vβ8 (25). The interaction is mediated through an Arg-Gly-Asp (RGD) motif found on the G-H loop of the FMDV capsid protein VP1, resulting in receptor-mediated entry into cells. Unlike other picornaviruses, e.g., poliovirus (22, 23), FMDV binding to its receptor does not result in a conformational change in structure or inactivation of the virus (7, 19). However, integrins are not the only receptors available to FMDV. Following adaptation to cell culture, as is required for production of vaccine virus, FMDV variants are selected with positively charged amino acids on the capsid surface. This selection dispenses with the requisite for integrin receptors, and the virus becomes capable of entering cells via heparan sulfate (HS) structures on the cell surface (4, 26, 38, 43). Considering the fact that vaccine viruses are produced following several passages in cell culture, such modification of receptor recognition may have important consequences. Certainly, the receptor-binding characteristics of in vitro-passaged FMDV will differ from those of field viruses.

HS is a glycosaminoglycan polymer of highly sulfated disaccharide repeats, and it is therefore negatively charged. Such HS proteoglycans are abundant cell surface structures that are widely distributed in animal tissues as part of the extracellular matrix and integral membrane components. Several groups have undertaken detailed studies investigating the emergence of HS-FMDV variants under tissue culture conditions. The characteristics of the variants may differ from the parental viral clone by displaying a shorter replication cycle in BHK-21 cells, a change in tropism demonstrated by the capacity to infect wild-type Chinese hamster ovary (CHO) and human erythroleukaemia (K562) cells, as well as an enhanced ability to kill cells (4, 5, 48). It has also been noted that tissue culture adaptation of type O FMDV can lead to attenuation of virus virulence for cattle (43).

When considering vaccine viruses, a crucial point is how these viruses interact with the immune system. In this context, dendritic cells (DC) are a critical component. DC are hematopoietic cells specialized in antigen capture and in presentation of the antigen to the immune system for initiation of primary and secondary immune responses. They have a central role in induction and regulation of immunity and are therefore the main target for any vaccine. However, they can also harbor virus infections and act as reservoirs for virus dissemination, potentially carrying viruses from one site to another within the vascular system (2, 30). Consequently, the present work sought to characterize the interaction of FMDV with DC in vitro, focusing in particular on the role played by HS binding when cell-passaged virus such as that employed for vaccine production is used. This role is considered particularly important in the context of vaccine antigens, due to the usage of cell culture passage for the production of current FMDV vaccines. Moreover, for an efficacious vaccine against FMDV, efficient interaction with DC is essential.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Animals, cells, viruses, and plasmids. Swiss White Landrace pigs, kept under specific pathogen-free conditions at the Institute of Virology and Immunoprophylaxis, Mittelhäusern, Switzerland, were used throughout this study.

All cell culture media and additives were obtained from Invitrogen (Basel, Switzerland) unless stated otherwise. Cell lines routinely used in the course of this work were baby hamster kidney 21 (BHK-21) (ATCC CCC-10) and human erythroleukemia (K562) (ATCC CCL 243) cells. BHK-21 cells were grown in Glasgow modified Eagle medium (GMEM) supplemented with 2 mM L-glutamine, 7.5% (wt/vol) bicarbonate, and 5% fetal calf serum (FCS). K562 cells were grown as suspension cultures in RPMI medium in the presence of 10% FCS with 2 mM L-glutamine.

Plasmid pO1K/C-S8c1, a full-length cDNA of FMDV O1K encoding type C capsid protein, was a kind gift from Esteban Domingo (Madrid, Spain), and its construction has been previously described (4, 5). The region transferred to the O1K genetic background spans C-S8c1 genomic positions 1739 to 4066 (the NcoI-HindIII fragment, which corresponds to Ser-33 of VP4 to Lys-62 of nonstructural protein 2B) (the numbering of FMDV C-S8c1 genomic residues is as described in reference 21).

FMDV O₁ Lausanne and FMDV C₁ Oberbayern are high-passage-number viruses routinely used in the laboratory. FMDV C-S8c1 (AJ133357) is a three-times plaque-purified derivative of the European serotype C natural isolate C₁ Santa Pau-Spain 70 (50), a field isolate from diseased swine (Spain, 1970) (18, 50, 51). FMDV MARLS (AF274010) (provided by Esteban Domingo, Madrid, Spain) was derived after 213 passages of FMDV C-S8c1 and is a monoclonal antibody (MAb)-resistant mutant selected with MAb SD6, which includes an L-144S substitution in VP1 (15, 32).

FMDV was grown in BHK-21 cell monolayers as described previously (33). Briefly, BHK-21 cells in serum-free GMEM supplemented with 2 mM L-glutamine and 7.5% (wt/vol) bicarbonate were infected at a multiplicity of infection (MOI) of 0.001 50% tissue culture infective dose (TCID₅₀)/cell and incubated at 37°C with 6% (vol/vol) CO₂ until cytopathic effect (CPE) was observed by light microscopy. Cells were harvested, lysed by freezing at –70°C, and clarified by centrifugation at 3,000 x g for 30 min at 4°C. Virus containing supernatants were stored at –70°C, and the virus titer was calculated from thawed virus stock by end-point titration on BHK-21 cells (35). Mock-cell lysate was prepared from noninfected BHK-21 cells in the same manner.

Generation of DC. Porcine monocyte-derived DC (MoDC) were generated as previously described (13). Briefly, peripheral blood mononuclear cells were obtained from citrated blood of specific pathogen-free pigs by density gradient centrifugation at 1,000 x g, for 25 min, over Ficoll-Paque (1.077 g/liter; Amersham Pharmacia Biotech AG, Dübendorf, Switzerland). Blood monocytes were enriched from peripheral blood mononuclear cells with a purity of 98% by magnetic cell sorting using a MACS system (Miltenyi Biotec GmbH, Bergish Gladbach, Germany) and anti-CD172a MAb 74-22-15 (1) with a positive-selection LS column. The monocytes were cultured for 4 days at 39°C in Dulbecco's MEM (DMEM) supplemented with 10% (vol/vol) porcine serum (PS) and the recombinant cytokines granulocyte-macrophage colony-stimulating factor (150 ng/ml; kindly provided by S. Inumaru, Institute for Animal Health, Ibaraki, Japan) and interleukin-4 (100 U/ml; prepared in our laboratory, as described in reference 13) to allow differentiation of MoDC. At day 3, the cultures were fed with fresh cytokines. MoDC were harvested 24 h later and were contained in the nonadherent cell population. This fraction, based on selection of the nonadherent population with minimal washing with phosphate-buffered saline (PBS)-EDTA, consistently ensures that only MoDC and not macrophages are selected for further studies.

Infection of MoDC with FMDV. Infection of 4-day-old MoDC was performed with either six-well plates at an MOI of 1 TCID₅₀/cell or fibronectin-coated Labteks (Nunc, Wiesbaden, Germany) at an MOI of 60 TCID₅₀/cell. This infection was performed either at 39°C without subsequent washing or at 4°C for 1 h followed by washing eight times, feeding, and culture at 39°C. The medium employed was phenol red-free DMEM-10% PS supplemented with nonessential amino acids and 1 mM sodium pyruvate. Sodium hydrogen phosphate (pH 6; 0.05 M) was employed for removal of virus bound on the surface of the MoDC (42).

Immunofluorescence assay. A Leica TCS-SL spectral confocal microscope and Leica LCS software (Leica Microsystems AG, Glattbrugg, Switzerland) were employed for immunofluorescence microscopy. The isotype-specific conjugates carried Alexa-488 fluorochrome (Molecular Probes, Leiden, Netherlands). Cells were washed twice with PBS-1% (wt/vol) bovine serum albumin-0.05% (wt/vol) sodium azide and fixed with 4% (wt/vol) paraformaldehyde at different times after infection and were then washed twice with CellWash, followed by a single 0.1% (wt/vol) saponin wash. Primary antibodies were diluted in 0.3% (wt/vol) saponin and incubated for 30 min at 4°C. Cells were then washed three times with 0.1% (wt/vol) saponin and the relevant conjugates applied, diluted in 0.3% (wt/vol) saponin and incubated a further 20 min at 4°C. Finally, cells were washed twice with 0.1% (wt/vol) saponin and twice with CellWash and were mounted in Mowiol prior to the acquisition of confocal data. For each analysis, 10 fields, containing between 20 and 100 cells per field, were analyzed. Representative images were chosen and analyzed further using Imaris imaging software (Bitplane AG).

MAbs. For MoDC phenotyping, hybridomas for MAb CD172a (MAb 74-22-15A) (immunoglobulin G2b [IgG2b]) were kindly donated by A. Saalmüller (BFAV Tübingen, Germany). For FMDV detection, MAb 4C9 recognizing the viral capsid of FMDV O (kindly donated by E. Brocchi, IZS Brescia, Italy) and MAb 5C4 recognizing conformational site D from FMDV C₁ were used. Viral replication was analyzed using MAb 2C2 (anti-3A protease). Cell surface staining was demonstrated with a major histocompatibility complex class II antibody (1F12; Pharmingen). Internal caveolin labeling used a rabbit polyclonal antibody (CAV1; BD Transduction). Double-stranded RNA (dsRNA) was visualized with the dsRNA-specific MAb J2 (English and Scientific Consulting Bt, Szirak, Hungary) to detect the A-helix of double-stranded polyribonucleotide complexes larger than 11 bp (46).

Inhibition of cell metabolism and virus binding to DC. MoDC were treated with inhibitors (all purchased from Sigma) 30 min before addition of FMDV. The following final concentrations were used: 100 nM bafilomycin, 40 µM brefeldin, 5 µM chloroquine, 1 µM chlorpromazine, 100 µg/ml cycloheximide, 10 µM cytochalasin D, 1 µM filipin, and 0.2 µM wortmannin. Subsequent to the treatments, FMDV infection of the DC proceeded as described above. The capacities of the inhibitors to retard endocytosis in MoDC were controlled by the addition of DQ ovalbumin and transferrin-Alexa-488, with uptake being determined by confocal microscopy. Disruption of actin structures after treatment with cytochalasin D was demonstrated by labeling with Alexa-488 phalloidin. In order to prevent virus binding to HS structures, N,N'-bis(1-oxido(1,2,5)oxadiazolo[3,4-d]pyrimidin-7-yl)-3,12-diaza-6,9-diazonia(5,2,5,2)dispirohexadecane dichloride (DSTP) (a kind gift from Michaela Schmidtke, Institute of Virology and Antiviral Therapy, Jena, Germany) was employed at different concentrations. The DSTP was mixed with FMDV and incubated for 1 h on ice, and the mixture was then added to the DC to monitor infection.

Removal of surface-bound FMDV from MoDC. MoDC were infected at an MOI of 1 TCID₅₀/cell on ice for 1 h and then washed eight times to remove unbound virus. Fresh warm medium was added, and the cultures were incubated at 39°C for the time periods given in Results. At these time periods after infection, sodium hydrogen phosphate buffer (0.05 M, pH 6) or 1 mg/ml pronase (Sigma, Buchs, Switzerland) was applied to the cells for 15 min at 4°C or 37°C, respectively, to destroy FMDV on the surface of MoDC (this treatment destroys at least 7 logs of FMDV infectivity). Following the treatment, the cells were washed with cold PBS to return the pH to physiological normal. Cell-associated virus was released by lysing the cells with sterile water for 1 min on ice and then returning to isotonicity with 10x PBS. Virus titers were assessed by titration on BHK-21 cells.

In vitro transcription and labeling of infectious RNA. The cDNA plasmid encoding O1K/C-S8c1 was linearized by restriction digestion with HpaI (New England BioLabs, Ipswich, MA) and purified by using a Qiagen gel extraction kit (Qiagen, Hombrechtikon, Switzerland). Transcription was performed using a Megascript kit (Ambion, Huntingdon, Cambridgeshire, United Kingdom) to generate RNA from the linearized plasmid. In the indicated samples, RNA molecules were labeled with rhodamine by using Label-It reagent (Mirus, Madison, WI) following the manufacturer's instructions (50 µl of RNA at a concentration of 0.1 mg/ml incubated with 50 µl of the labeling reagent for 1 h at 37°C and purified by ethanol precipitation).

Transfection of O1K/C-S8c1 RNA into BHK-21 cells and MoDC. BHK-21 cells at a confluence of 90% were harvested by trypsinization with trypsin-EDTA. MoDC were transfected on day 4 in their immature stage. Cells were resuspended to a final concentration of 2 x 10⁶ cells/ml in serum-free DMEM (MoDC) or GMEM (BHK). RNA (3 µg) diluted in 500 µl of serum-free medium was mixed with 9 µl of the lipofection reagent, TransFast (Promega, Southampton, United Kingdom) at a lipid/mRNA ratio of 3:1. After 15 min of incubation at room temperature, to allow mRNA-lipid complexing, 500 µl of the cell suspension was added to the lipoplexes, and the mixture was incubated for 1 h at 39°C (MoDC) or 37°C (BHK-21 cells). After lipofection, cells were washed twice with prewarmed serum-free medium and resuspended in complete medium (for MoDC, DMEM-10% PS; for BHK-21 cells, GMEM-1% FCS) at a concentration of 10⁶ cells/ml in six-well plates.

FCM analysis. Staining of cell surface molecules was achieved by incubating unfixed cells for 15 min on ice with the relevant primary antibody diluted in CellWash (Becton Dickinson, Basel, Switzerland). Following incubation, the cells were washed with CellWash. Reactivity was detected using fluorescein isothiocyanate or phycoerythrin-isotype-specific goat F(ab')₂ anti-mouse Ig conjugates (Southern Biotechnology Associates, Birmingham, United Kingdom) diluted in CellWash and incubated on ice for 20 min, followed by washing with CellWash. Intracellular antigens were analyzed by fixation and permeabilization using a cell permeabilization and fixation kit (Immunologicals Direct, Loughborough, Leicestershire, United Kingdom). Dead cells were detected by briefly incubating treated MoDC with propidium iodide (Sigma-Aldrich) on ice before flow cytometric (FCM) analysis. FCM was performed using a FACSCalibur analytical FCM (Becton Dickinson, Basel, Switzerland), and the data were analyzed using CellQuest Pro software (Becton Dickinson). Further fluorescence-activated cell sorter (FACS) statistical analysis was undertaken with FlowJo software (Treestar, San Carlos, CA).

Infection of K562 cells with FMDV. Virus was added to K562 cell suspensions containing 1 x 10⁶ cells at an MOI of 0.01 TCID₅₀/cell and allowed to adsorb at 37°C for 1 h. Cells were then washed twice with PBS-EDTA to remove unabsorbed virus and plated in 12-well plates in a volume of 1 ml RPMI medium-10% FCS and further incubated at 37°C for 24 h. Identical cultures were set up with BHK-21 cells and virus at an MOI of 0.01 TCID₅₀/cell in GMEM-1% FCS. Aliquots of media were taken for titration 24 h postinfection (p.i.) to assess infectivity on BHK-21 cell monolayers.

Infectious center assay. Transfected MoDC (250,000 to 0.025 cells) were titrated in tenfold dilutions on semiconfluent BHK-21 cell monolayers in a 96-well microtiter plate (Costar, Cambridge, United Kingdom). The cells were incubated at 39°C with DMEM containing 10% (vol/vol) FCS and 1% (vol/vol) PS and observed every 24 h for the development of CPEs.

Coculture of transfected MoDC with lymphocytes. A coculture system was employed to assess if the small amount of virus produced in transfected MoDC was capable of inducing an FMDV-specific IgG antibody response in CD172a^–-sorted lymphocytes from either an FMDV-vaccinated or a naïve pig. MoDC were transfected with infectious RNA and washed, and different numbers (25,000, 50,000, 100,000, 200,000 and 400,000) were immediately plated with 4 x 10⁶ lymphocytes. MoDC were also infected for 4 h with HS variant virus (C1, MARLS, or O1K/C-S8c1/p5) or non-HS variant virus (C-S8c1 and O1K/C-S8c1/p2). After extensive washing to remove virus not associated with MoDC, cocultures were set up as described above. The cocultures were incubated for 5 days, and the supernatants were harvested and then subjected to an FMDV-IgG-specific sandwich enzyme-linked immunosorbent assay (9, 34). Serum from a hyperimmune pig, diluted 1:1,000, served as a positive control and was included in each test. Antibody titers are expressed as percentages of relative relativity, calculated as [average (test supernatant) – average (background)]/[average (positive control) – average (background)] x 100.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Interaction of MoDC with cell culture-passaged FMDV. Studies with human DC have demonstrated the presence of integrin β1 and β3 (17, 29), but not β6 integrin, on DC. This result has been confirmed with sheep, in which vβ6 integrins are restricted to epithelial tissue (12). Therefore, it is likely that porcine DC do not express detectable levels of the integrin receptor vβ6, which is proposed to be the main receptor for FMDV infection (28, 36). Indeed, antibodies against vβ6 integrin (used to detect vβ6 integrin on bovine epithelial cells) did not stain porcine DC (data not shown). Interaction of the virus with DC would therefore require an alternative receptor. One set of potential candidates would be other integrins, although such interactions may be of low efficiency (19). Antibodies against vβ3, vβ5, or vβ8 integrins did indeed react with porcine DC, but only weakly; antibodies against the β chain, particularly β1 or β3, gave clearly positive results (data not shown).

A second candidate for FMDV interaction with DC would be the HS structures on the cell surface. Accordingly, viruses that had received several passages in cell culture—both a serotype O₁ and a serotype C₁ virus—were chosen due to the reported capacity of cell culture-adapted virus to bind with the HS structures on target cells (26). This was the same virus shown previously to interact with porcine macrophages (42). The kinetics of binding to and internalization by MoDC were analyzed for this cell culture-passaged FMDV (Fig. 1), measured between 10 min and 24 h p.i. The kinetic analysis was performed at 39°C without any washing of the cells to remove unadsorbed virus. This allowed identification of the time required for detectable levels of the virus to interact with the DC. Accordingly, time zero is the time at which the virus was added to the cultures.

View larger version (56K): [in this window] [in a new window]

FIG. 1. HS-variant FMDV interacts, internalizes, and transiently replicates in MoDC. Four-day-old MoDC were infected with FMDV O1 (green) and incubated for (A) 10 min or (B) 4 h. Costaining of FMDV (red) with an internal anti-caveolin marker (green) at 10 min p.i. (C) and major histocompatibility complex class II surface-staining (green) at 4 h p.i (D). (E) FMDV (green) uptake after 24 h at an MOI of 60 TCID50/cell. (F) Mock-infected cells. At the different times p.i., the cells were fixed and analyzed by confocal microscopy using either a type O- or type C-specific MAb (4C9 or 5C4, respectively). (G) pH resistance of DC-associated FMDV was measured by infecting 4-day-old MoDC with FMDV O1 at an MOI of 1 TCID50/cell for 1 h at 4°C. Cells were then washed eight times to remove unbound virus followed by the addition of prewarmed medium, and the temperature was then shifted to 39°C. At the indicated time points, half the cells were treated with 0.05 M sodium hydrogen phosphate (pH 6) to destroy virus on the cell surface. Cells were then lysed to measure the titers of cell-associated virus by titration on BHK-21 cells. (H) Replication of FMDV O1 was determined by pretreating cells with cycloheximide and measuring infectious cell-associated virus by titration on BHK-21 cells. The results shown are representative of results of three independent experiments.

The first time point at which virus could be visualized on the cell surface was 10 min p.i.; this was observed similarly for serotypes O₁ (Fig. 1A) and C₁ (results not shown) compared with mock-infected cells (Fig. 1F). By 4 h, the virus was consistently detectable within the cell in all experiments, although some virus did appear to be at or near the cell surface (Fig. 1B). The presence of the virus on the cell surface at 10 min and internalization by 4 h was confirmed using unfixed cells. By comparing the images obtained with fixed and unfixed cells, as well as dual labeling with caveolin (Fig. 1C) and major histocompatibility complex class II (Fig. 1D), virus was seen to be primarily surface bound at 10 min p.i. and internalized by 4 h p.i. Additional experiments at 4°C confirmed the presence of virus on the cell surface within 10 min p.i. (data not shown).

These results imply that the majority of detectable virus antigen followed a gradual rather than a rapid uptake by the DC, although the results also suggest that a certain amount of the virus was internalized more rapidly. From 8 h p.i. to 24 h p.i., the antigen was mainly cytoplasmic but was becoming reduced in quantity. There was no evidence for efficient virus replication; the ability to detect virus antigen was declining by 24 h p.i. (Fig. 1E), eventually becoming undetectable. Moreover, images of inactivated virus (data not shown) were similar to those with the live virus as shown in Fig. 1A to F. These kinetics also relate to the reported observations on FMDV interaction with macrophages (42).

In certain experiments, the virus could be detected as early as 5 min p.i., but this result was inconsistent and probably related to the efficiency with which the method could detect virus on the DC surface. Certainly, the number of antigen-positive cells detectable at 10 min p.i. did not reflect the high MOI used, nor did the number of positive cells that were found with apparently internalized antigen at 4 h p.i. It was presumed that this result reflected inefficiency in the antigen detection process as well as the transient nature of the uptake. A minimum number of virus particles had to be associated with the DC before they could be detected by the microscopic method employed, meaning that a diffuse distribution of viral antigen may not be detectable.

Fate of FMDV following interaction with DC. The results shown in Fig. 1A to F, along with those obtained using inactivated virus, would suggest that FMDV is unable to replicate in DC. In order to confirm this hypothesis, infectious virus was measured at different times following infection of the DC, by lysing the cells to release both surface and internalized virus. Prior to lysis, the DC cultures were either untreated, to detect total virus, or treated with a pH 6.0 buffer as described in Materials and Methods, to remove surface-bound infectivity and to measure only internalized virus. This method allowed a comparison of the relative internalization rates for the virus at various time p.i. In these experiments, infectious titers were analyzed rather than virus antigen production, due to the sensitivity of virus infectivity at pH 6.0. Moreover, the ability to detect virus infectivity is a more sensitive assay than microscopic detection of antigen. Unfortunately, it does not facilitate observations of individual cells.

Accordingly, the MoDC were allowed to interact with the virus for 1 h at 4°C before a wash with prewarmed (37°C) medium and a shift of the incubation temperature to 39°C. This technique ensured that the virus being monitored was primarily that which had bound to the DC during the 1-h adsorption period at 4°C. At different times after this shift in temperature from 4°C to 39°C (Fig. 1G), the cells were treated with sodium hydrogen phosphate buffer (pH 6) for 15 min to destroy infectious virus on the cell surface. Any infectious titer remaining was taken as signifying the presence of pH-resistant virus that was most likely internalized by the MoDC, similar to observations made with macrophages (42). These analyses confirmed that the virus was localized on the cell surface (pH labile) at 10 min p.i. (Fig. 1G). Interestingly, about 5% of the infectivity was pH resistant after the 1-h adsorption at 4°C (Fig. 1G, 0 min). It seems most likely that this result was due to membrane ruffling occurring when the cells were shifted to 39°C. Nonetheless, within 10 min of the shift to 39°C, all virus had returned to a pH-labile condition.

During the first 60 min after infection, all detectable infectivity remained on the cell surface, although there was a reduction in titer. This result may reflect a dissociation of virus from the DC surface or a rapid internalization together with destruction or uncoating of the virus in early/sorting endosomes. By 2 h p.i., the detectable titer of infectious virus associated with the DC had somewhat stabilized. A proportion of this infectivity was seen to have converted into a pH-resistant form, indicating internalization of the virus. The pH-resistant titers represented, on average, 30% of the total infectious titers obtained in the absence of pH 6 buffer treatment.

Between 2 h and 4 h p.i., there was an increase in the relative proportion of the virus that had become resistant to the pH treatment; more virus was pH resistant than pH sensitive at 4 h p.i., in contrast to the situation at 2 h p.i. These results suggest further internalization of the virus which had bound to the DC surface. By 8 h p.i., the situation was reversed in that the majority of virus had become pH sensitive. This finding may relate to the observations for macrophages, suggesting a recycling of the virus to the cell surface (42). By 24 h p.i., infectious virus was no longer recoverable, a result probably reflecting degradation of the capsid as would occur following virion uncoating or processing of antigen by the DC.

Potential for FMDV replication in DC. Considering the decrease in the total infectious titer during the first 60 min p.i., together with the observed increase in internalized antigen demonstrated in Fig. 1, it is possible that the early events reflected virus uncoating. This result raised the question of whether the virus internalized during the initial 60 min was initiating replication, however transient. If replication were occurring, the FMDV antigen levels would be different from those observed with a susceptible cell line such as BHK-21 cells (42), in which levels of FMDV antigen continually increased until cell death.

In order to examine the question concerning possible transient replication in DC, infectious virus titers were measured after infection of DC treated with cycloheximide (Fig. 1H). Cycloheximide was used to inhibit protein synthesis and therefore prevent the formation of new infectious particles. Figure 1H shows that similar levels of virus were found after the adsorption at 4°C at time zero with both cycloheximide-treated and nontreated cells. By 30 min p.i., there was a 1-log decrease in the amount of infectious virus associated with the DC when the cells had been treated with cycloheximide. At 1 h, no infectious virus associated with the cycloheximide-treated cells was detectable. After this time point, the amount of infectious virus increased in untreated cells, but no titers were found with the treated DC. These results suggest that a transient replication of the virus does occur between 2 and 8 h, leading to the observed increase in infectious virus titers at this time (Fig. 1G). Such a replication is considered to be transient due to the observation that by 24 h p.i., infectious virus titers were no longer detectable, presumably resulting from the degradative processes of the DC ultimately dominating over virus replication. One possible explanation for this apparent "shutdown" of replication would be the production of type I interferon (IFN-I). Accordingly, culture supernatants were collected from the infected DC at all time points, and IFN- production was detected by enzyme-linked immunosorbent assay. At 2 h p.i., IFN- could be detected at low levels (approximately 20 U/ml), but this result was inconsistent, measurable in only 50% of the experiments. The production of IFN- by MoDC was investigated in a previous study (24), in which significant production of this cytokine prior to viral infection could not be detected.

Endocytic activity and endosomal processing during FMDV interaction with MoDC. Considering the importance of endosomal function and membrane recycling for both virus uncoating and antigen processing by DC, the role of endosomes in the activities shown in Fig. 1 was analyzed. Bafilomycin was used for its ability to prevent both endosomal acidification and membrane recycling. Treatment with bafilomycin did not prevent the virus from interacting with the cell surface, and the virus even became internalized in certain cells (Fig. 2A). Chloroquine also inhibits endosomal acidification by its interaction with protons. DC treated with this inhibitor gave results similar to those seen with bafilomycin (Fig. 2B). Comparing these results with those obtained in the absence of inhibitors (Fig. 2G) and for the mock control (Fig. 2H), there appeared to be reduced antigen intensity in the drug-treated cells. Moreover, the distribution of antigen after treatment of DC with bafilomycin and chloroquine appeared to be similar to that found for CHO-K1 cells (31), in that the antigen was seen as a diffuse staining throughout the cytoplasm, but a concentrated punctuate/vesicle signal close to the nucleus was rarely noted.

View larger version (69K): [in this window] [in a new window]

FIG. 2. Influence of metabolic inhibitors on HS-variant FMDV binding and internalization by MoDC. Four-day-old MoDC were pretreated with metabolic inhibitors for 30 min at 39°C and then infected for 4 h with FMDV O1 at an MOI of 60 TCID50/cell (also at 39°C) in the presence or absence of metabolic inhibitors. Cells were then fixed and FMDV antigen was detected using MAb 4C9 and goat anti-mouse IgG conjugated with Alexa-488. Arrows indicate the presence of virus.

Additional analyses looked at the endocytic processing in more depth, employing chlorpromazine, cytochalasin, filipin and wortmannin. Akin to the observations with bafilomycin and chloroquine, chlorpromazine, an inhibitor of clathrin-dependent endocytosis (54), did not prevent the entry of virus into the cell (Fig. 2C) but did lead to a more punctate distribution of the antigen in the cytoplasm without a clear perinuclear distribution. Moreover, an apparent reduction in the antigen signal was noted compared with untreated cells (Fig. 2G). With cytochalasin D, which prevents actin polymerization and therefore transport between endocytic compartments (20), internalization of antigen was again observed, with areas containing perinuclear staining still being observable (Fig. 2D). Filipin also had no apparent influence on antigen uptake (Fig. 2E) and did not prevent the appearance of perinuclear staining; in fact, the number of cells with the perinuclear staining appeared to increase compared to that seen in the absence of any inhibitor (Fig. 2G). The final inhibitor tested was wortmannin, an inhibitor of PI3 kinase and therefore completion of macropinocytosis (55). While wortmannin did not completely prevent the virus from interacting with and being internalized by the DC (Fig. 2F), the uptake of virus appeared to have been retarded. As with bafilomycin and chlorpromazine, there was a more-diffuse staining of cytoplasmic vesicles containing antigen, along with a reduced level of antigen detectable in the cells compared with levels for untreated DC.

Control of the inhibition of endocytic activities. The above results suggested that certain of the metabolic inhibitors did not influence FMDV interaction with the DC, while others impaired internalization rather than binding. Accordingly, it was necessary to ensure that the inhibitors were functionally active. The agents were therefore tested for their capacity to inhibit the uptake and processing of a model antigen, DQ ovalbumin (Fig. 3A to G). Therein, the "DQ" component is an overconjugation of acid-insensitive fluorescein, such that fluorescence is obtained only when the DQ ovalbumin enters acidifying early endosomes containing active esterases that are sensitive to bafilomycin and chloroquine (Fig. 3A and C). Interference with endosomal acidification was more complete with bafilomycin (Fig. 3A) than with chloroquine (Fig. 3C). This result is also characteristic of the interaction of African swine fever virus with porcine monocytic cells (37). In the latter case, the effect was shown to be due to the lower efficiency with which chloroquine impairs endosomal acidification. Brefeldin partially blocked the degradation of DQ ovalbumin (Fig. 3B), resulting in a reduced fluorescence signal. This was considered to reflect the capacity of the drug to interfere with membrane recycling (endosomal tubulation), which occurs prior to the total collapse of the endosomal system following prolonged incubation with brefeldin.

View larger version (108K): [in this window] [in a new window]

FIG. 3. Assessment of metabolic inhibitors in porcine MoDC. (A to F) Four-day-old MoDC were pretreated with metabolic inhibitors, or left untreated, for 30 min at 39°C and then 1 mg/ml DQ ovalbumin was applied. After 4 h incubation, cells were fixed, and the processing of DQ ovalbumin was assessed by confocal microscopy. The effect of cytochalasin (G) was visualized by pretreating the cells for 30 min and then staining for phalloidin and comparing the result with that seen in the absence of cytochalasin (H). Chlorpromazine activity (I) was demonstrated by inhibition of transferrin uptake compared to that for control cells (J).

Internalization of the DQ ovalbumin by MoDC employed macropinocytosis and caveolar uptake, which are sensitive to filipin and wortmannin, respectively (Fig. 3D and E). The apparently greater influence of filipin on DQ ovalbumin uptake/processing may reflect the influence of this drug on lipid raft formation, in contrast to wortmannin, which inhibits PI3 kinase type I activity. Cytochalasin D also impaired DQ ovalbumin uptake, but the activity of the drug was controlled by staining the actin filaments with phalloidin (16). The concentrations of cytochalasin D used in these experiments have been demonstrated to remove stress fibers, upsetting the supramolecular organization of the actin filaments (44). In Fig. 3G, it can clearly be seen that the microtubules have collapsed after treatment with cytochalasin D compared with the untreated control (Fig. 3H).

Chlorpromazine has less effect on DQ ovalbumin uptake due to its inhibition of clathrin-mediated endocytosis. Accordingly, the activity of chlorpromazine was assessed using fluorescent-labeled transferrin uptake (52). Results shown in Fig, 3I, compared with those for the untreated control (Fig. 3J), demonstrate that the chlorpromazine indeed impaired transferrin uptake.

Pronase sensitivity of FMDV interaction with MoDC. The observations presented in Fig. 2 suggested that certain of the inhibitors forced the virus to remain at or near the cell surface. While microscopic analysis allows detailed observations on individual cells, it does not necessarily reflect how the culture as a whole is behaving. Moreover, as mentioned above, a minimum quantity of virus must interact with the cells to allow its detection by microscopy. Therefore, MoDC were treated (at 39°C) in the same manner as shown for Fig. 2. The temperature was then reduced to 4°C, and the cells were infected with FMDV at an MOI of 1 TCID₅₀/cell for 1 h at 4°C. After this adsorption period, the cultures were washed, the temperature was shifted back to 39°C (this time point is "0 min"), and the cultures were incubated for the times shown in Fig. 4. Certain of the replicates were then treated with pronase to remove bound virus which had not yet been internalized. Pronase was used in preference to acid treatment, because interaction with cell receptors would have stabilized the virus and protected it from the effects of the acidic buffer. Measurement of cell-associated infectivity was employed, due to its high sensitivity compared with antigen detection.

View larger version (17K): [in this window] [in a new window]

FIG. 4. Effect of metabolic inhibitors on binding and internalization of FMDV by DC with and without pronase treatment. Four-day-old MoDC were pretreated for 30 min with metabolic inhibitors at 39°C, as in Fig. 2, or were left untreated (No inhibitor). The cultures were then placed at 4°C and infected with FMDV O1 at an MOI of 1 TCID50/cell for 1 h at 4°C. Cells were washed eight times to remove any unbound virus followed by the addition of prewarmed medium and were then shifted back to 39°C. At 30 min (A) and 4 h (B) p.i., cells were harvested and treated with pronase to destroy bound virus. Cells were then lysed, and titers of cell-associated virus were assessed by titration on BHK-21 cells. (C) Percentage of bound virus internalized (remaining after pronase treatment) by the MoDC. The results shown are representative of results of three independent experiments.

At 30 min p.i. in the absence of metabolic inhibitors, virus was found in both pronase-sensitive and insensitive forms (Fig. 4A). This relates to the results with the acid treatment (Fig. 1G), suggesting that some virus was already sufficiently internalized. Nevertheless, this proportion represented approximately only 5% of the total amount of virus associated with the cells. In the presence of bafilomycin, brefeldin, cytochalasin, filipin, and wortmannin, the titers of virus associated with the cells (without pronase) were similar to those obtained in the absence of inhibitor. In contrast, more of this virus was in a pronase-sensitive state; the exception was with the bafilomycin-treated cells, which had a profile similar to that of the untreated DC. Chloroquine and chlorpromazine appeared to be inhibiting both the binding of the virus to the cell (without pronase) and its internalization (pronase-resistant titers).

At 4 h p.i., all cultures treated with the metabolic inhibitors had similar titers of infectious virus associated with the cells (Fig. 4B). These titers were similar to, or in certain cases slightly greater than, the titers obtained with the untreated cells. It was now evident that only the chloroquine, chlorpromazine, and cytochalasin clearly reduced the titers of virus that became pronase resistant, compared with the untreated control.

Considering the above observations, it was important to compare the percentage of pronase-resistant virus, relative to the total amount bound, for each of the inhibitors. The aim was to clarify if the inhibitors were interfering with the binding of the virus or with its internalization. Although the inhibitors employed interfere with different elements of the endocytic processing system in DC, they all have the potential to impede virus binding due to their indirect effect on membrane recycling. The analysis clarified that at 30 min p.i., the percentage of bound virus entering the cells with bafilomycin- and chlorpromazine-treated cells was similar to that seen with untreated DC (Fig. 4C). This result implies that the influence of chlorpromazine seen in Fig. 4A was more on the binding of the virus to the cells. In contrast, brefeldin, cytochalasin, filipin, and wortmannin clearly impaired internalization of the virus (Fig. 4C). Chloroquine was interesting in that it appeared to increase virus internalization at 30 min p.i. but reduce internalization when cells were viewed at 4 h p.i. (Fig. 4C). This was an exception, because at 4 h p.i., all inhibitors impaired virus internalization compared with results for untreated DC.

The data shown in Fig. 4C could also offer an explanation for the low titers of total virus (the no-pronase bars) seen at 30 min p.i. with the chloroquine- and chlorpromazine-treated cultures (Fig. 4A) compared to the levels obtained at 4 h p.i. By interfering more with binding of the virus than with internalization at 30 min p.i., these treatments might have led to a reduced stability of the virus binding to the cell surface. This in turn would have increased the likelihood of virus detachment from the cell surface, leading to results such as those shown in Fig. 4A. Any detached virus would retain the ability for rebinding to the cell surface, giving results as seen in Fig. 4B. Indeed, between 30 min and 4 h p.i., an increase in cell-free virus titers was observed (data not shown).

Blocking of HS-variant FMDV interaction with MoDC. The above results demonstrated that the cell culture-passaged virus could bind to the DC even when endocytic activity was inhibited, suggesting that the virus was using receptor-mediated binding for entry into the DC. Candidates for this interaction of FMDV with DC surfaces are the HS structures with which cell culture-passaged FMDV is reported to interact (26). In order to test the hypothesis, dispirotripiperazine (DSTP) was tested for its ability to prevent virus infection. DSTP binds specifically to heparin and is more efficient than heparin or pentosan polysulfate at blocking virus interaction with HS structures on cell surfaces (45). Consequently, FMDV was added to MoDC in the presence or absence of different DSTP concentrations, and the mixture was incubated for either 4 h or 20 h. In the absence of DSTP (Fig. 5A), the virus uptake is seen as the typical punctuate staining observed in the previous figures. When DSTP was present, no FMDV was detectable in association with the MoDC internally or on the cell surface (Fig. 5A), and similar results were obtained at 4 h and 20 h p.i. (results not shown). The interaction of FMDV with MoDC was sensitive to all concentrations of DSTP employed (Fig. 5A), which did not alter the cell viability (data not shown).

View larger version (68K): [in this window] [in a new window]

FIG. 5. DSTP inhibits the infection of DC by HS-binding virus. (A) Four-day-old MoDC were infected for 4 h with FMDV O1 at an MOI of 60 TCID50/cell in the presence of different amounts of the HS-binding inhibitor DSTP (25 µg, 100 µg, and 400 µg). This infection was performed at 39°C without a washing step, to give the virus the maximum chance for interacting with the cells. The 4-h infection was repeated with DSTP (C and E, +DSTP) and without 100 µg DSTP (B and D, –DSTP) and with the addition of bafilomycin (B and C) or chloroquine (D and E). Cells were fixed, and FMDV antigen was detected using MAb 4C9 and goat anti-mouse IgG conjugated to Alexa-488. Arrows indicate the presence of virus.

The results in Fig. 5A demonstrate that FMDV was indeed interacting with the DC via the HS structures on the cell surface. Nevertheless, there was still the possibility that a small amount of virus was entering the cells by another pathway. If this were clathrin-mediated endocytosis employing a different receptor, the rapid endosomal acidification associated with this process could have disrupted the virus structure such that the detecting antibody was no longer effective. In order to ascertain if such a possibility existed, the experiment with DSTP was repeated in the presence of chloroquine or bafilomycin to block endosomal acidification. The effect of these drugs would allow virus to accumulate in the cell, if it were entering via a receptor or pathway independent of the HS structures, by involving rapid endosomal acidification. Figure 5B to E shows that no detectable virus antigen was associated with the DSTP-treated MoDC when chloroquine or bafilomycin was employed to inhibit endosomal acidification.

Generation of non-HS-binding progeny. Considering the results showing the major role played by the HS structures on DC in the interaction with cell culture-passaged FMDV, the question of whether non-HS-binding FMDV could interact with DC in a similar manner was raised. Accordingly, FMDV variants of the C serotype virus (C-S8c1) lacking HS-binding ability were analyzed in terms of interaction with DC. Propagation of C-S8c1 in BHK cells is known to generate HS-binding virus, (5) for which reason a cDNA plasmid encoding FMDV O1K/C-S8c1 was employed to produce non-HS-binding virus. This plasmid encodes the full-length sequence of FMDV O1K with the capsid sequence replaced by that of FMDV C-S8c1, to ensure the production of non-HS-binding virus. It was not possible to use infectious clones of the entire serotype C genome, because construction of these clones has been problematic (5). In order to employ the same transfecting agent for both the cell culture lines and the DC, infectious RNA was generated as described in Materials and Methods due to the poor translation capacity of plasmid DNA in DC (14, 53).

Transfecting BHK-21 cells with infectious RNA resulted in efficient translation, measured in terms of viral nonstructural protein 3A (Fig. 6A). When the transfected cells were tested for infectious virus, a titer of 10^5.25 TCID₅₀/ml was obtained by titration on BHK-21 cells (Table 1, O1K/c-s8c1/p1). O1K/c-s8c1/p1 was confirmed to be non-HS binding by its inability to replicate in K562 cells (no infectious virus could be detected after titration of cell lysates on BHK-21 cells). The K562 human bone marrow-derived, chronic myelogenous leukemia cell line lacks integrin receptors, such that only HS-binding FMDV can infect these cells. Infection of BHK-21 cells by O1K/C-S8c1 was also resistant to inhibition by DSTP (a shift similar to that shown in Fig. 6A was seen in an analysis using FCM in the presence or absence of DSTP; results not shown). These characteristics were mimicked with the cloned FMDV C-S8c1, confirming that this virus was also non-HS binding.

View larger version (14K): [in this window] [in a new window]

FIG. 6. Influence of circumventing the HS structures on MoDC with respect to viral RNA translation and infectious virus association with the cells. (A) Production of FMDV protease 3A following transfection of BHK-21 cells with infectious RNA (O1K/C-S8c1). The FMDV protease 3A protein production was detected by FACS analysis at 48 h posttransfection. (B) Comparison of HS-binding and non-HS-binding virus interaction with MoDC in terms of the presence of infectious virus. Four-day-old MoDC were infected with FMDV O1 and FMDV C-S8c1 at an MOI of 1 TCID50/cell for 1 h at 4°C. Cells were then shifted to 39°C for the duration of the experiment. At each of the time points shown on the x axis, the cells were washed eight times to remove any unbound virus. The cells were then lysed after the final centrifugation to release cell-associated virus, and the infectious virus titers were measured by titration on BHK-21 cells. The results shown are representative of results of three independent experiments.

View this table: [in this window] [in a new window]

TABLE 1. Phenotypic traits of FMDV variants with different passage histories in BHK-21 cell culture

The non-HS-binding characteristic of FMDV is known to be unstable after passage on BHK-21 cells. In order to confirm that the O1K/C-S8c1 virus preserved these characteristics, the "stability" of its non-HS-binding phenotype was determined by serial passage of the virus progeny from transfected BHK-21 cells at an MOI of 0.001 TCID₅₀/cell. Virus preparations resulting from these passages (O1K/C-s8c1/p2 to O1K/C-s8c1/p5) were titrated on both BHK-21 and K562 cells to identify HS-binding capacity. As the passage of the virus increased, there was no significant increase in the titer on BHK-21 cells (Table 1). In contrast, by the third passage, the virus had acquired the capacity to infect and replicate in the K562 cells, and by the fourth passage, CPEs were more rapidly induced in BHK-21 cells. The ability to replicate in K562 cells was taken as indicative of HS-binding capacity. Although the titers in the K562 cells were lower than those obtained for the BHK-21 cells, the ratio of K562 cell titers to BHK-21 cell titers was higher than that for known HS-binding viruses FMDV O and FMDV MARLS. Moreover, results with the cloned FMDV C-S8c1, a low-passage non-HS-binding virus (4), confirmed that the K562 cells would not accommodate such viruses (Table 1).

Infection of MoDC with HS- and non-HS-variant FMDV. Having confirmed the HS-binding characteristics of the above viruses, we then analyzed the kinetics of infection for the HS and non-HS variants in MoDC. The HS-binding FMDV O₁ and FMDV C₁, as well as the non-HS-binding C-S8c1 (see Table 1), were used to infect MoDC. Bafilomycin was also employed to prevent endosomal acidification and to allow detection of virus before it entered the acidic degradation pathway. In contrast to results shown in Fig. 1 and 2, no virus antigen was detectable when the DC were "fed" FMDV C-S8c1 (data not shown). The presence of bafilomycin did not alter the image, showing that the virus was not "disappearing" due to rapid degradation in the endosomal system.

The inability to detect this non-HS-binding virus may have reflected an absence of the appropriate integrins on the DC. Nevertheless, it was possible that low levels of virus were binding, although less efficiently than virus binding to epithelial cells with the appropriate integrins or HS-binding variant to the HS structures. Consequently, the experiment was repeated using virus infectivity as the readout, rather than antigen detectable by microscopy. This method would also permit the identification of the time point at which the virus encountered the acidifying environment of the DC endosomal system, losing its infectious nature. Such analyses relate to the observations shown in Fig. 2 and 3, determining the characteristics of virus interaction with DC with respect to endosomal activity.

Figure 6B confirms the relatively high efficiency of HS-binding virus interaction with the MoDC. The time point "0 min" shown in Fig. 6 refers to the end of the adsorption period, which was 1 h at 4°C. At this point, the cultures were shifted to 39°C for the duration of the experiment. In contrast to the previous experiments, the cultures were incubated for each of the time points shown on the x axis before being washed eight times to remove unbound virus. In this manner, if the virus required longer than the initial 1 h at 4°C to interact with the DC, this interaction would become possible as the time p.i. increased. The analyses focused on the cell-associated infectious virus titers which had interacted with the DC at the different times of incubation. Incubation of the cultures at 39°C demonstrated that the maximum level of infectious HS virus binding occurred during the 1-h adsorption period (Fig. 6B). As with the results shown in Fig. 1G and H, there was again a decrease in cell-associated virus titers between 1 and 2 h p.i.—in this case, the titers actually became undetectable—with an increase again at 4 h p.i. and a final disappearance of infectivity by 24 h p.i. In contrast, the non-HS-binding virus did not apparently infect the MoDC, even when the virus was given several hours to interact with the DC (Fig. 6B, FMDV C-S8c1).

Progeny virus production in MoDC. The above results showed that HS-binding FMDV, but not non-HS-binding FMDV, interacts with DC, with detectable virus infectivity remaining for up to 8 h p.i. This result is intriguing considering the active endosomal system possessed by DC, and it contrasts with results for BHK-21 cells, wherein virus infectivity is lost within minutes due to the activity of early endosomes. This finding raised the question of whether a low level of virus replication, masked by the levels of detectable virus antigen and infectivity, might have occurred following infection of the DC. Indeed, results for the cycloheximide treatment (Fig. 1H) demonstrated that the increase in virus infectivity at 4 h p.i. was due to de novo synthesis of presumably progeny virus. This result would imply that the observed loss of virus antigen between 1 h and 2 h p.i. in DC reflected virion uncoating, as occurs during the replicative cycle of FMDV. Nevertheless, this explanation did not take into consideration the specialized capacity of the DC to process antigen rather than to facilitate the initiation of virus replication.

Accordingly, experiments were performed to determine if the FMDV genome were indeed translated in the DC and whether it could in fact replicate. The procedure employed was designed to facilitate the identification of progeny virus production unclouded by residual infectivity from any input infectious virus. For these reasons, the MoDC were transfected with in vitro transcribed O1K/C-S8c1 RNA. Importantly, this transfection and the presence of the RNA in the cell cytoplasm did not alter the FCM light scatter characteristics of the cells (data not shown). Propidium iodide staining for cell viability gave similar values for both mock- and O1K/C-S8c1-transfected cells (8% and 12%, respectively); this result also compared favorably with that for the control nontransfected cells, which had a 5% reduction in viability.

The readout for RNA activity focused on the generation of viable virus progeny, as assessed by end-point titration on BHK-21 cells. At 24 h, low titers of viable extracellular virus (ECV) were detected (10^2.5 TCID₅₀/ml) (Fig. 7A). This finding was unexpected, because no virus-specific CPEs were apparent in the MoDC culture; lysis of the cells is generally the mode of FMDV release from its host cell, although cellular autophagy may also facilitate virus release or recycling of receptor-bound virus to the cell surface. Despite this clear indication of RNA translation, it was a particularly transient affair. Both prior to and after this time point, no infectious virus titers were obtained, and no cell-associated virus titers were obtained at any time point.

View larger version (42K): [in this window] [in a new window]

FIG. 7. Transfection of MoDC with infectious RNA results in abortive replication. (A) Following transfection of MoDC with O1K/C-S8c1 infectious RNA, the presence of ECV and cell-associated virus (CAV) was determined at different times posttransfection by titrating on BHK-21 cells. (B) Four-day-old MoDC transfected with rhodamine-labeled O1K/C-S8c1 RNA (red). At 4 h posttransfection, the MoDC were washed eight times. Replicate cultures were either prepared immediately for confocal microscopy or incubated further at 39°C for the indicated times. The preparation for microscopy entailed fixation and permeabilization of the cells as described in Materials and Methods, followed by staining with an Alexa-labeled anti-FMDV capsid antibody (green). Arrows denote presence of virus. (C) FACS analysis of viral protease 3A production in the MoDC transfected with O1K/C-S8c1 infectious RNA, using MAb against FMDV protease 3A following fixation and permeabilization of the cells. (D) Detection of dsRNA after transfection of the MoDC with O1K/C-S8c1 RNA. The dsRNA was detected using MAb J2 and confocal microscopy at 24 h, 48 h, and 72 h posttransfection.

Fate of FMDV RNA in DC. The poor yield of infectious progeny following FMDV RNA transfection of DC may have reflected unsuccessful transfection. Therefore, the efficiency of transfection was controlled by confocal microscopy using rhodamine-labeled RNA. The results indicated that approximately 30% of cells were successfully transfected (Fig. 7B), with the transfected RNA remaining detectable for 72 h. A similar transfection rate was obtained with a "mock" RNA transfection control.

Accordingly, the focus turned to the translation of the transfected RNA. Viral capsid protein production was analyzed by confocal microscopy using an anti-FMDV capsid antibody. Only 1 in 30 of the successfully transfected cells showed evidence of translation, in terms of detectable viral capsid protein, during the first 24 h posttransfection (Fig. 7B). FCM analysis confirmed the low percentage of antigen-positive cells detected by microscopy. This result was seen for both FMDV capsid and nonstructural proteins (Fig. 7C shows the results for the nonstructural 3A protein). Only at 24 h p.i. was a detectable shift seen in the mean fluorescence intensity observed with FCM (Fig. 7C). At 72 h posttransfection, the antigenic signal was lost (Fig. 7B).

The results in Fig. 6B correspond to the low virus titers shown in Fig. 7A at 24 h posttransfection, suggesting that FMDV RNA was inefficient at translating viral proteins in the majority of MoDC. Alternatively, this might reflect a minor subpopulation of DC that are capable of supporting FMDV replication. Accordingly, serial dilutions of the transfected MoDC were plated on to BHK-21 cells in an infectious center assay. The plates were then incubated at 39°C until the appearance of CPEs, which should relate to ECV released from the MoDC. It took 7 days of incubation before any CPE was confirmed (data not shown). A minimum of 2.5 x 10⁵ transfected MoDC were required to produce this effect. That is, approximately 1 in 10,000 transfected MoDC released virus infectious for the BHK-21 cell detection system. This percentage is clearly much lower than the 1% of cells translating viral proteins. Moreover, considering that the titer of the released virus in these MoDC cultures was 10^2.5 TCID₅₀/ml (Fig. 7A), the quantity of infectious virus released per cell is likely to be in the single figures.

The apparently low level of RNA translation may have been due to "misdelivery" of the transfected RNA. Considering the fact that the lipofection involves targeting of RNA-liposome complexes to cell endosomes, it was possible that endosomal processing was destroying the RNA, or the viral proteins were rapidly translocated into the degradative endolysosomal system. Accordingly, analyses were performed for detecting the formation of replicative forms of the transfected RNA, which is essential for the replicative process to continue. Using an antibody against dsRNA, confocal microscopy demonstrated that dsRNA was indeed formed in the cell and was detectable between 24 and 72 h posttransfection (Fig. 7D). The number of dsRNA⁺ cells was approximately 1%, suggesting a relationship with the number of cells producing virus antigen. Such results imply that only 1 in 30 of the transfected DC can accommodate RNA replication and translation, but viral maturation is even less efficient. It is not possible to perform a double-labeling experiment to compare the dsRNA replicative forms of FMDV against virus antigen production, because both antibodies have the same isotype.

Induction of lymphocyte responses after MoDC interaction with FMDV or FMDV RNA. The difficulties in detecting viral antigen following transfection of the DC may have been due in part to the DC rapidly processing the antigen. Moreover, the virus antigen seen in the DC following interaction with HS-binding virus would not necessarily enter into the antigen-processing pathway, which is the primary function of DC. Consequently, it was necessary to determine the relative capacity of the FMDV RNA to induce antigen-specific lymphocyte responses. When DC were transfected with FMDV RNA prior to coculture with FMDV-specific lymphocytes from a vaccinated pig, the induction of anti-FMDV antibody was detected, expressed as percent relative reactivity compared with a hyperimmune serum (Fig. 8A; the mock RNA employed for transfection was an RNA encoding the NS3/NS4 protein from classical swine fever virus). This in vitro property of the DC was dependent on the ratio of DC to lymphocytes employed, typical of DC activity when processing antigen for stimulating antigen-specific lymphocytes in culture. Interestingly, when the transfected MoDC were cocultured with lymphocytes from a naïve pig, a low level of specific antibody was induced when high DC/lymphocyte ratios were again employed (Fig. 8B). Overall, it is clear that transfection of MoDC with infectious RNA does produce levels of viral antigen that are sufficient to induce a specific lymphocyte response. Moreover, it was also noted that DC which had interacted with either the HS-binding or the non-HS-binding virus were also capable of stimulating the antigen-specific autologous lymphocytes in culture (Fig. 8C). Taken together, these results demonstrate that the DC are capable of employing even the low levels of antigen produced by transfected RNA to activated antigen-specific lymphocytes.

View larger version (14K): [in this window] [in a new window]

FIG. 8. MoDC transfected with infectious RNA and virus induce lymphocyte recall responses in vitro. A coculture system of CD172a– cells (containing T and B lymphocytes) and O1K/C-S8c1 RNA-transfected MoDC were used to measure T- and B-lymphocyte-dependent anti-FMDV IgG production. MoDC were transfected with O1K/C-S8c1 and cocultured with CD172a– cells from a vaccinated pig (A) and a naïve pig (B). MoDC from a vaccinated pig were infected for 4 h with HS-variant FMDV (C1, MARLS, and O1K/C-S8c1/p5) and non-HS variant (C-S8c1 and O1K/C-S8c1/p2), after which the cells were extensively washed. Cocultures of CD172a– were set up, and T- and B-lymphocyte-dependent anti-FMDV IgG production was measured (C). These experiments are representative of results of three separate experiments.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
When vaccine virus interacts with the immune system, a critical step is the binding of the virus to DC. Therefore, the present work has demonstrated an important role for HS structures on the surface of DC for uptake of FMDV vaccine virus. The differences between HS-binding and non-HS-binding FMDV variants in terms of their interaction with DC are important to understand how the initial steps in the induction of immune defenses can proceed. DC are efficient at processing viral antigen but are most effective when HS-binding virus is present. Although non-HS-binding virus is endocytosed by DC, as witnessed by the ability of such cells to promote a recall immune response in vitro, the kinetics of interaction with the DC is less efficient than that of HS-binding virus. Nevertheless, both types of virus are adequately processed by DC for stimulating the immune system. This result demonstrates that once antigen is delivered to DC, only small quantities of antigen are required by these cells to promote adaptive immune defense development. Increased efficiency in the delivery of vaccine to the DC would therefore enable these cells to perform their immunopromotional tasks with lower vaccine antigen or with alternatives, such as RNA vaccines.

The limited expression or absence of detectable vβ3 and vβ6 integrins on DC demonstrates that FMDV interaction with DC employs alternative receptors, e.g., HS proteoglycans (26); such structures are involved in the binding of human papillomavirus 16 virus-like particles to DC (11). With FMDV type O₁ and C₁ (possessing HS-binding capacity due to their cell culture passage), the virus was seen binding to the DC surface as early as 10 min p.i., the initial time point at which the virus was detectable varied from experiment to experiment. The virus was gradually internalized by the cell over a period of 4 to 8 h, similar to reports on internalization of FMDV by skin DC after 1 h (6) and the internalization of FMDV by macrophages (42).

Previous studies have reported an absence of detectable FMDV replication in DC (6, 24, 40) and that only early steps—production of nonstructural proteins—could be found in macrophage cell lines (8). These results suggested that DC can internalize the virus but are unable to support the replication of FMDV, which would be similar to the situation with primary macrophages (42). However, the experiments with cycloheximide reported in the present paper demonstrate that there is in fact a clear but transient replication, which is ultimately aborted. One hypothesis for the inability of FMDV to replicate in macrophages and DC implicated the induction of IFN-I (24). Although IFN- production could be detected at 2 h p.i., the levels were low (approximately 20 U/ml) and inconsistent between experiments. Such characteristics have important implications for the innate immune response activity. During the early stages of virus replication, ssRNA and dsRNA intermediates are formed; these intermediates possess the capacity to stimulate the DC via TLR7/8 and TLR3, as well as cytosolic helicases, such as RIG-I. These intermediates would have clear modulatory influences on the activities of macrophages, DC, and NK cells (47, 49) and, ultimately, on the development of adaptive immune defenses.

The ability of FMDV to interact with the DC in the presence of the metabolic inhibitors, without an increase in infectivity, raised the question of the mode of virus uptake. It is reported that HS-binding FMDV enters epithelial cells by a clathrin-mediated pathway that is sensitive to chlorpromazine (39). However, HS structures can also mediate internalization of ligands via caveolin-mediated endocytosis (41). The binding of FMDV to DC was not influenced by most inhibitors of the different endocytic processes active with DC, nor by the vacuolar H⁺-ATPase inhibitor bafilomycin. Chlorpromazine did reduce virus binding to the DC, but only slightly. This compound is a cationic amphiphilic drug that interferes with the clathrin AP-2 subunit. The outcome is prevention of clathrin binding to the plasma membrane, loss of clathrin-coated pits at the cell surface, and the appearance of clathrin coat deposition on endosomal membranes. These effects would have an influence more on receptor recycling than on membrane recycling, implying that the slight effect on FMDV binding to DC may have resulted from modulated receptor recycling. Indeed, the absence of an influence by bafilomycin and brefeldin, which would affect membrane recycling, supports the suggestion that membrane recycling did not influence the binding of FMDV to the DC.

The metabolic inhibitors had a greater influence on the initial internalization of the virus by DC, primarily with filipin, wortmannin, cytochalasin D, and brefeldin. This result would imply that the initial entry of the virus was dependent on lipid raft activity (filipin), PI3 kinase type I activity (wortmannin), actin polymerization (cytochalasin D), and endosomal recycling (brefeldin). In contrast, chlorpromazine-dependent events, such as those involving clathrin activities, as well as impairment of vacuolar H⁺-ATPase by bafilomycin, were less influential on the initial internalization of FMDV by DC. This points to a role for macropinocytosis, although filipin is also known to impair caveolar uptake. These are not acidic vesicles until they have interacted with acidic endosomes or lysosomes. The lack of impaired virus internalization of FMDV by bafilomycin and chloroquine during the first 30 min after infection confirms that this period of virus entry was not dependent on rapid endosomal acidification. Interestingly, chloroquine actually increased the level of detectable internalized pronase-resistant virus. Chloroquine impedes vacuolar acidification by sequestering protons, in contrast to bafilomycin, which directly inhibits the vacuolar H⁺-ATPase but does not neutralize the acidifying capacity of already-formed protons. Accordingly, the results imply that initial internalization of FMDV by DC most likely involves macropinosomes or caveolae that interact with existing acidic endosomal structures but that it is not yet dependent on vacuolar H⁺-ATPase activity to promote further endosomal acidification and antigen processing.

As the time p.i. progressed, there was an increase in the percentage of internalized virus becoming more sensitive to all metabolic inhibitors. Such an image could suggest that the virus enters by various modes of endocytosis. However, it seems more likely that the observed effect is caused by the ultimate influence which all these inhibitors have on membrane recycling. As time progresses after treatment with such inhibitors of endocytic and endosomal processes, the availability of membrane for virus internalization would become more limiting due to the block in endosome maturation and therefore membrane recycling.

A blocker of HS binding, DSTP (45), was employed to determine if the HS structures were involved in FMDV binding to DC. The DSTP treatment inhibited uptake of the cell culture-passaged FMDV by the DC. Although this result demonstrated the involvement of HS structures on the DC surface for binding the majority of the FMDV, a minor quantity of virus may have entered by another pathway. Such a scenario would be difficult to observe if the process involved led to rapid acidification of the vesicles carrying the virus. Pretreatment of the DC with bafilomycin or chloroquine did not alter the image seen with DSTP, implying that little if any of the cell culture-passaged virus was interacting with DC through surface entities other than the cellular HS structures.

The above results may be pertinent to cell culture-passaged vaccine virus dominated by HS-binding virus. For comparative purposes, analyses were extended to consider the interaction of non-HS-binding virus, typifying field virus, with DC. No evidence of non-HS-binding virus antigen interaction with the DC was observed. This was not due to the virus interacting rapidly with early endosomes, because application of endosomal inhibitors did not alter the image. In certain aspects, this result was not surprising: the non-HS-binding virus had to use a receptor other than the HS structures, the interaction of which may be less efficient than that of HS binding. Consequently, the kinetics of virus infectivity associated with the DC was measured, but there was no observable increase in the titers of the HS-binding virus associated with the DC as the time of adsorption was extended.

Although the above analyses were performed with live virus, similar results for antigen uptake were obtained with inactivated virus, implying that the viral genome may have a problem initiating replication and translation. Accordingly, DC were transfected with infectious RNA, thus bypassing the mode of virus uptake. The aim was to determine if the viral RNA was infectious, as is the case for BHK-21 cells. Low levels of antigen and extracellular viable virus were detected in the transfected DC, but this was only a transient event. The low viral titers also corresponded to low levels of dsRNA, which were detected in only a minority of the cells. A similar phenomenon of abortive replication has been seen in plasmacytoid cells after uptake of FMDV in complexes with antibody (24) and in murine DC (40).

Although only 1 in 30 of the successfully transfected DC translated detectable viral protein, this is not an insignificant number for DC to function as antigen-presenting cells. Indeed, despite the low level of FMDV antigen associated with RNA-transfected DC, there was a clear capacity of these DC to induce FMDV-specific immune responses. Moreover, the use of non-HS-binding virus, with which it was difficult to detect antigen associated with the DC, could also induce an immune response. Either the small amounts of antigen associated with the DC were sufficient for such responses to be induced, or there was a rapid processing such that the antigenic material could not be detected. Nevertheless, the induction of the B cell responses would have required antigen that should have been detectable, implying that only low levels of antigen are necessary. Consequently, the lower levels of non-HS-binding virus associating with the DC, compared with HS-binding virus, may well be adequate for stimulating immune defenses. Of course, the higher antigen content associating with DC when the HS-binding virus is employed would suggest that such viruses have an advantage when it comes to vaccine production. Such results are also interesting in the context of RNA vaccine development. If the delivery of RNA could be improved to stabilize its cytosolic translocation, it may prove interesting in the future as an alternative vaccine. Nevertheless, the most efficient delivery of antigen to DC is by HS-binding virus, even though such high levels of antigen are not essential for efficient immune response development.

 
This work was supported by State Secretariat for Education and Research grant 02.0063 within the EU framework 5 project QLK2-CT2002-0719 and by State Secretariat for Education and Research grant 03.0519 within the EU project FMD_Improcon FP6 503603.

We are grateful to Esteban Domingo for the pO1K/C-S8c1 plasmid and FMDV MARLS strain and Michaela Schmidtke for the kind gift of DSTP. We also thank the animal handlers, Daniel Brechbühl, Andreas Michel, and Hans-Peter Lüthi, for taking care of the blood donor pigs and for routine blood sampling.

 
* Corresponding author. Mailing address: Institute of Virology and Immunoprophylaxis, CH-3147 Mittelhäusern, Switzerland. Phone: 41-31-848-9319. Fax: 41-31-848-9222. E-mail: Lisa.Harwood{at}ivi.admin.ch

Published ahead of print on 30 April 2008.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Alvarez, B., C. Sanchez, R. Bullido, A. Marina, J. Lunney, F. Alonso, A. Ezquerra, and J. Dominguez. 2000. A porcine cell surface receptor identified by monoclonal antibodies to SWC3 is a member of the signal regulatory protein family and associates with protein-tyrosine phosphatase SHP-1. Tissue Antigens 55:342-351.[CrossRef][Medline]
2
2. Banchereau, J., and R. M. Steinman. 1998. Dendritic cells and the control of immunity. Nature 392:245-252.[CrossRef][Medline]
3
3. Baranowski, E., N. Molina, J. I. Nunez, F. Sobrino, and M. Saiz. 2003. Recovery of infectious foot-and-mouth disease virus from suckling mice after direct inoculation with in vitro-transcribed RNA. J.Virol. 77:11290-11295.[Abstract/Free Full Text]
4
4. Baranowski, E., C. M. Ruiz-Jarabo, N. Sevilla, D. Andreu, E. Beck, and E. Domingo. 2000. Cell recognition by foot-and-mouth disease virus that lacks the RGD integrin-binding motif: flexibility in aphthovirus receptor usage. J. Virol. 74:1641-1647.[Abstract/Free Full Text]
5
5. Baranowski, E., N. Sevilla, N. Verdaguer, C. M. Ruiz-Jarabo, E. Beck, and E. Domingo. 1998. Multiple virulence determinants of foot-and-mouth disease virus in cell culture. J. Virol. 72:6362-6372.[Abstract/Free Full Text]
6
6. Bautista, E. M., G. S. Ferman, D. Gregg, M. C. Brum, M. J. Grubman, and W. T. Golde. 2005. Constitutive expression of alpha interferon by skin dendritic cells confers resistance to infection by foot-and-mouth disease virus. J. Virol. 79:4838-4847.[Abstract/Free Full Text]
7
7. Baxt, B., and H. L. Bachrach. 1980. Early interactions of foot-and-mouth disease virus with cultured cells. Virology 104:42-55.[CrossRef][Medline]
8
8. Baxt, B., and P. W. Mason. 1995. Foot-and-mouth disease virus undergoes restricted replication in macrophage cell cultures following Fc receptor-mediated adsorption. Virology 207:503-509.[CrossRef][Medline]
9
9. Bergamin, F., L. Saurer, V. Neuhaus, K. C. McCullough, and A. Summerfield. 2007. Porcine B-cell activating factor promotes anti-FMDV antibodies in vitro but not in vivo after DNA vaccination of pigs. Vet. Immunol. Immunopathol. 120:115-123.[CrossRef][Medline]
10
10. Berinstein, A., M. Roivainen, T. Hovi, P. W. Mason, and B. Baxt. 1995. Antibodies to the vitronectin receptor (integrin _Vβ₃) inhibit binding and infection of foot-and-mouth disease virus to cultured cells. J. Virol. 69:2664-2666.[Abstract]
11
11. Bousarghin, L., P. Hubert, E. Franzen, N. Jacobs, J. Boniver, and P. Delvenne. 2005. Human papillomavirus 16 virus-like particles use heparan sulfates to bind dendritic cells and colocalize with langerin in Langerhans cells. J. Gen. Virol. 86:1297-1305.[Abstract/Free Full Text]
12
12. Brown, J. K., S. M. McAleese, E. M. Thornton, J. A. Pate, A. Schock, A. I. Macrae, P. R. Scott, H. R. P. Miller, and D. D. S. Collie. 2006. Integrin-_vβ₆, a putative receptor for foot-and-mouth disease virus, is constitutively expressed in ruminant airways. J. Histochem. Cytochem. 54:807-816.[Abstract/Free Full Text]
13
13. Carrasco, C. P., R. C. Rigden, R. Schaffner, H. Gerber, V. Neuhaus, S. Inumaru, H. Takamatsu, G. Bertoni, K. C. McCullough, and A. Summerfield. 2001. Porcine dendritic cells generated in vitro: morphological, phenotypic and functional properties. Immunology 104:175-184.[CrossRef][Medline]
14
14. Ceppi, M., N. Ruggli, V. Tache, H. Gerber, K. C. McCullough, and A. Summerfield. 2005. Double-stranded secondary structures on mRNA induce type I interferon (IFN alpha/beta) production and maturation of mRNA-transfected monocyte-derived dendritic cells. J. Gene Med. 7:452-465.[CrossRef][Medline]
15
15. Charpentier, N., M. Davila, E. Domingo, and C. Escarmis. 1996. Long-term, large-population passage of aphthovirus can generate and amplify defective noninterfering particles deleted in the leader protease gene. Virology 223:10-18.[CrossRef][Medline]
16
16. Cooper, J. A. 1987. Effects of cytochalasin and phalloidin on actin. J. Cell Biol. 105:1473-1478.[Free Full Text]
17
17. D'Amico, G., G. Bianchi, S. Bernasconi, L. Bersani, L. Piemonti, S. Sozzani, A. Mantovani, and P. Allavena. 1998. Adhesion, transendothelial migration, and reverse transmigration of in vitro cultured dendritic cells. Blood 92:207-214.[Abstract/Free Full Text]
18
18. Domingo, E., M. Davila, and J. Ortin. 1980. Nucleotide sequence heterogeneity of the RNA from a natural population of foot-and-mouth-disease virus. Gene 11:333-346.[CrossRef][Medline]
19
19. Duque, H., M. LaRocco, W. T. Golde, and B. Baxt. 2004. Interactions of foot-and-mouth disease virus with soluble bovine _Vβ₃ and _Vβ₆ integrins. J. Virol. 78:9773-9781.[Abstract/Free Full Text]
20
20. Durrbach, A., D. Louvard, and E. Coudrier. 1996. Actin filaments facilitate two steps of endocytosis. J. Cell Sci. 109:457-465.[Abstract]
21
21. Escarmis, C., M. Davila, N. Charpentier, A. Bracho, A. Moya, and E. Domingo. 1996. Genetic lesions associated with Muller's ratchet in an RNA virus. J. Mol. Biol. 264:255-267.[CrossRef][Medline]
22
22. Fricks, C. E., and J. M. Hogle. 1990. Cell-induced conformational change in poliovirus: externalization of the amino terminus of VP1 is responsible for liposome binding. J. Virol. 64:1934-1945.[Abstract/Free Full Text]
23
23. Gómez Yafal, A., G. Kaplan, V. R. Racaniello, and J. M. Hogle. 1993. Characterization of poliovirus conformational alteration mediated by soluble cell receptors. Virology 197:501-505.[CrossRef][Medline]
24
24. Guzylack-Piriou, L., F. Bergamin, M. Gerber, K. C. McCullough, and A. Summerfield. 2006. Plasmacytoid dendritic cell activation by foot-and-mouth disease virus requires immune complexes. Eur. J. Immunol. 36:1674-1683.[CrossRef][Medline]
25
25. Jackson, T., S. Clark, S. Berryman, A. Burman, S. Cambier, D. Mu, S. Nishimura, and A. M. Q. King. 2004. Integrin _Vβ₈ functions as a receptor for foot-and-mouth disease virus: role of the β-chain cytodomain in integrin-mediated infection. J. Virol. 78:4533-4540.[Abstract/Free Full Text]
26
26. Jackson, T., F. M. Ellard, R. Abu Ghazaleh, S. M. Brookes, W. E. Blakemore, A. H. Corteyn, D. I. Stuart, J. W. I. Newman, and A. M. Q. King. 1996. Efficient infection of cells in culture by type O foot-and-mouth disease virus requires binding to cell surface heparan sulfate. J. Virol. 70:5282-5287.[Abstract/Free Full Text]
27
27. Jackson, T., A. P. Mould, D. Sheppard, and A. M. Q. King. 2002. Integrin _Vβ₁ is a receptor for foot-and-mouth disease virus. J. Virol. 76:935-941.[Abstract/Free Full Text]
28
28. Jackson, T., D. Sheppard, M. Denyer, W. Blakemore, and A. M. Q. King. 2000. The epithelial integrin vβ6 is a receptor for foot-and-mouth disease virus. J. Virol. 74:4949-4956.[Abstract/Free Full Text]
29
29. Jancic, C., H. E. Chuluyan, A. Morelli, A. Larregina, E. Kolkowski, M. Saracco, M. Barboza, W. S. Leiva, and L. Fainboim. 1998. Interactions of dendritic cells with fibronectin and endothelial cells. Immunology 95:283-290.[CrossRef][Medline]
30
30. Klagge, I. M., and S. Schneider-Schaulies. 1999. Virus interactions with dendritic cells. J. Gen. Virol. 80:823-833.[Medline]
31
31. Knox, C., K. Moffat, S. Ali, M. Ryan, and T. Wileman. 2005. Foot-and-mouth disease virus replication sites form next to the nucleus and close to the Golgi apparatus, but exclude marker proteins associated with host membrane compartments. J. Gen. Virol. 86:687-696.[Abstract/Free Full Text]
32
32. Mateu, M. G., M. A. Martinez, L. Capucci, D. Andreu, E. Giralt, F. Sobrino, E. Brocchi, and E. Domingo. 1990. A single amino acid substitution affects multiple overlapping epitopes in the major antigenic site of foot-and-mouth disease virus of serotype C. J. Gen. Virol. 71:629-637.[Abstract/Free Full Text]
33
33. McCullough, K. C., and R. Butcher. 1982. Monoclonal antibodies against foot-and-mouth disease virus 146S and 12S particles. Arch. Virol. 74:1-9.[CrossRef][Medline]
34
34. McCullough, K. C., F. De Simone, E. Brocchi, L. Capucci, J. R. Crowther, and U. Kihm. 1992. Protective immune response against foot-and-mouth disease. J. Virol. 66:1835-1840.[Abstract/Free Full Text]
35
35. McCullough, K. C., D. Parkinson, and J. R. Crowther. 1988. Opsonization-enhanced phagocytosis of foot-and-mouth disease virus. Immunology 65:187-191.[Medline]
36
36. Miller, L. C., W. Blakemore, D. Sheppard, A. Atakilit, A. M. Q. King, and T. Jackson. 2001. Role of the cytoplasmic domain of the β-subunit of integrin _vβ₆ in infection by foot-and-mouth disease virus. J. Virol. 75:4158-4164.[Abstract/Free Full Text]
37
37. Natale, V. A., and K. C. McCullough. 1998. Macrophage cytoplasmic vesicle pH gradients and vacuolar H+-ATPase activities relative to virus infection. J. Leukoc. Biol. 64:302-310.[Abstract]
38
38. Neff, S., D. Sa-Carvalho, E. Rieder, P. W. Mason, S. D. Blystone, E. J. Brown, and B. Baxt. 1998. Foot-and-mouth disease virus virulent for cattle utilizes the integrin _vβ₃ as its receptor. J. Virol. 72:3587-3594.[Abstract/Free Full Text]
39
39. O'Donnell, V., M. LaRocco, H. Duque, and B. Baxt. 2005. Analysis of foot-and-mouth disease virus internalization events in cultured cells. J. Virol. 79:8506-8518.[Abstract/Free Full Text]
40
40. Ostrowski, M., M. Vermeulen, O. Zabal, J. R. Geffner, A. M. Sadir, and O. J. Lopez. 2005. Impairment of thymus-dependent responses by murine dendritic cells infected with foot-and-mouth disease virus. J. Immunol. 175:3971-3979.[Abstract/Free Full Text]
41
41. Reilly, J. F., E. Mizukoshi, and P. A. Maher. 2004. Ligand dependent and independent internalization and nuclear translocation of fibroblast growth factor (FGF) receptor 1. DNA Cell Biol. 23:538-548.[CrossRef][Medline]
42
42. Rigden, R. C., C. P. Carrasco, A. Summerfield, and K. C. McCullough. 2002. Macrophage phagocytosis of foot-and-mouth disease virus may create infectious carriers. Immunology 106:537-548.[CrossRef][Medline]
43
43. Sa-Carvalho, D., E. Rieder, B. Baxt, R. Rodarte, A. Tanuri, and P. W. Mason. 1997. Tissue culture adaptation of foot-and-mouth disease virus selects viruses that bind to heparin and are attenuated in cattle. J. Virol. 71:5115-5123.[Abstract]
44
44. Schliwa, M. 1982. Action of cytochalasin D on cytoskeletal networks. J. Cell Biol. 92:79-91.[Abstract/Free Full Text]
45
45. Schmidtke, M., A. Karger, A. Meerbach, R. Egerer, A. Stelzner, and V. Makarov. 2003. Binding of a N,N'-bisheteryl derivative of dispirotripiperazine to heparan sulfate residues on the cell surface specifically prevents infection of viruses from different families. Virology 311:134-143.[CrossRef][Medline]
46
46. Schönborn, J., J. Oberstrass, E. Breyel, J. Tittgen, J. Schumacher, and N. Lukacs. 1991. Monoclonal antibodies to double-stranded RNA as probes of RNA structure in crude nucleic acid extracts. Nucleic Acids Res. 19:2993-3000.[Abstract/Free Full Text]
47
47. Schröder, M., and A. G. Bowie. 2005. TLR3 in antiviral immunity: key player or bystander? Trends Immunol. 26:462-468.[CrossRef][Medline]
48
48. Sevilla, N., and E. Domingo. 1996. Evolution of a persistent aphthovirus in cytolytic infections: partial reversion of phenotypic traits accompanied by genetic diversification. J. Virol. 70:6617-6624.[Abstract/Free Full Text]
49
49. Sioud, M. 2005. Induction of inflammatory cytokines and interferon responses by double-stranded and single-stranded siRNAs is sequence-dependent and requires endosomal localization. J. Mol. Biol. 348:1079-1090.[CrossRef][Medline]
50
50. Sobrino, F., M. Davila, J. Ortin, and E. Domingo. 1983. Multiple genetic variants arise in the course of replication of foot-and-mouth disease virus in cell culture. Virology 128:310-318.[CrossRef][Medline]
51
51. Sobrino, F., E. L. Palma, E. Beck, M. Davila, J. C. de la Torre, P. Negro, N. Villanueva, J. Ortin, and E. Domingo. 1986. Fixation of mutations in the viral genome during an outbreak of foot-and-mouth disease: heterogeneity and rate variations. Gene 50:149-159.[CrossRef][Medline]
52
52. Sun, X., V. K. Yau, B. J. Briggs, and G. R. Whittaker. 2005. Role of clathrin-mediated endocytosis during vesicular stomatitis virus entry into host cells. Virology. 338:53-60.[CrossRef][Medline]
53
53. Van Tendeloo, V. F., H. W. Snoeck, F. Lardon, G. L. Vanham, G. Nijs, M. Lenjou, L. Hendriks, C. Van Broeckhoven, A. Moulijn, I. Rodrigus, P. Verdonk, D. R. Van Bockstaele, and Z. N. Berneman. 1998. Nonviral transfection of distinct types of human dendritic cells: high-efficiency gene transfer by electroporation into hematopoietic progenitor- but not monocyte-derived dendritic cells. Gene Ther. 5:700-707.[CrossRef][Medline]
54
54. Wang, L. H., K. G. Rothberg, and R. G. Anderson. 1993. Mis-assembly of clathrin lattices on endosomes reveals a regulatory switch for coated pit formation. J. Cell Biol. 123:1107-1117.[Abstract/Free Full Text]
55
55. Xu, J., D. Ziemnicka, G. S. Merz, and L. Kotula. 2000. Human spectrin Src homology 3 domain binding protein 1 regulates macropinocytosis in NIH 3T3 cells. J. Cell Sci. 113:3805-3814.[Abstract]

---

Journal of Virology, July 2008, p. 6379-6394, Vol. 82, No. 13
0022-538X/08/$08.00+0     doi:10.1128/JVI.00021-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Harwood, L. J. Articles by McCullough, K. C. Search for Related Content PubMed PubMed Citation Articles by Harwood, L. J. Articles by McCullough, K. C.