Enhancing dengue virus maturation using a stable furin over-expressing cell line

Abstract

Flaviviruses are positive-stranded RNA viruses that incorporate envelope (E) and premembrane (prM) proteins into the virion. Furin-mediated cleavage of prM defines a required maturation step in the flavivirus lifecycle. Inefficient prM cleavage results in structurally heterogeneous virions with unique antigenic and functional characteristics. Recent studies with dengue virus suggest that viruses produced in tissue culture cells are less mature than those produced in primary cells. In this study, we describe a Vero cell line that ectopically expresses high levels of human furin (Vero-furin) for use in the production of more homogenous mature flavivirus populations. Laboratory-adapted and clinical dengue virus isolates grow efficiently in Vero-furin cells. Biochemical and structural techniques demonstrate efficient prM cleavage in Vero-furin derived virus preparations. These virions also were less sensitive to neutralization by antibodies that bind efficiently to immature virions. This furin-expressing cell line will be of considerable utility for flavivirus neutralization and structural studies.

Keywords: Antibody neutralization; Dengue virus; Flavivirus; Furin; Virion maturation; West Nile virus.

Figure 1

Growth kinetics of DENV in human furin-expressing Vero cells. (A–D) Virus stocks of DENV strains that represent each of the four serotypes, including DENV1 strain Western Pacific, DENV2 strain New Guinea C, DENV3 strain Sleman, DENV4 strain 814, were produced in Aedes albopictus C6/36 cells. Vero and Vero-furin cells were infected at a multiplicity of infection (MOI) of 0.1. Viruses were collected from infected cells at indicated timepoints post-infection and the infectious titer determined by immunofocus assay on Vero cells in duplicate. The growth curves presented are the average titer of three or four independent onestep growth curves. Error bars indicate the standard error of the mean (SEM). (E) The growth kinetics of a clinical DENV2 isolate (strain 172-06) on Vero and Vero-furin cells was determined as above. The results presented are the average of four independent one-step growth curves. Error bars indicate the SEM. (F) Comparisons of growth kinetics of DENV2 strain 172-06 on HEK-293T and HEK-293T-furin cells were performed as detailed above. Data presented reflect the average of three independent experiments; error bars indicate the standard error. In all instances, comparisons of the titer observed at each time point between cell types were performed using a two-way ANOVA are were not found to be significant.

Figure 2

Evaluation of the maturation state of DENV generated in Vero and Vero-furin cells. (A) The prM content of DENV2 virus preparations was evaluated by SDS-PAGE and Western blotting using E- and prM-specific antibodies. prM-specific antibody GTX128092 (Genetex, Irvine, CA) was used for detection of prM levels. (B) Electron microscopic analysis of the morphology of DENV2 stocks produced from Vero and Vero-furin cell lines. M, mature; I, immature; PM, partially mature. (C) Proportion of virions with smooth (mature, M), spiky (I), or irregular morphology (partially mature, PM) in each stock. Statistical analysis was performed using an ANOVA with Tukey’s correction for multiple comparisons. (D) The infectious titer of DENV produced in Vero or Vero-furin cells was measured by infecting Raji-DCSIGNR cells with serial two fold dilutions of virus. Infectivity was measured 48 hours post-infection by immunostaining with AlexaFluor labeled 4G2 antibody. The infectivity of the virus stock is depicted relative to its genomic RNA content, as measured by qRT-PCR.

Figure 3

Impact of furin over-expression on maturation state-sensitive patterns of antibody-mediated neutralization. Neutralization of WNV RVPs produced in Vero and Vero-furin cells by (A) mAb E16 and (B) mAb E53 was measured as detailed previously (Mukherjee et al., 2014b). Serial dilutions of antibody were mixed with virus particles to allow for steady state binding and then added to Raji-DC-SIGNR cells. Infection was measured by flow cytometry two days post infection. Data was analyzed by non-linear regression using GraphPad Prism. Error bars represent the range of two technical replicates. Data are representative of at least three independent antibody-dose response curves. (C-E) Neutralization of infectious DENV2 derived from Vero and Vero-furin cell lines by mAbs DV-96 (C), E60 (D), and prM22 (E) was measured as described previously (Mukherjee et al., 2014b). Virus-antibody mixtures were incubated for 1 h at 37 ºC prior to the addition of Raji-DC-SIGNR cells. Cells were incubated at 37ºC for 48 hours prior to being fixed and stained for intracellular E protein expression. Infection was measured by flow cytometry. Data were analyzed by non-linear regression using GraphPad Prism. Error bars represent the range of two technical replicates. Data are representative of at least three independent antibody-dose response curves.