Recombinant lentogenic Newcastle disease virus expressing Ebola virus GP infects cells independently of exogenous trypsin and uses macropinocytosis as the major pathway for cell entry

Recombinant NDV expressing the Zaire EBOV GP (rLa-EBOVGP) was generated by inserting the EBOV GP gene between the P and M genes in the genomic cDNA of the NDV LaSota strain (rLa) (Figure 1A). Expression of the GP gene was confirmed by indirect confocal immunofluorescent staining of rLa-EBOVGP-infected BHK-21 cells. rLa-EBOVGP-infected BHK-21 cells were stained with mouse anti-EBOV GP antiserum, whereas rLa-infected BHK-21 cells were not stained with the antiserum (Figure 1B). NDV antigens and EBOV GP protein colocalized on the surfaces of the BHK-21 cells, confirming the surface expression of the EBOV GP protein in the rLa-EBOVGP-infected BHK-21 cells (Figure 1B).

rLa-EBOVGP showed similar growth properties to those of rLa, with a maximum titer of 10^9.6 × 50% embryo infectious doses (EID₅₀) at 72 h after inoculation in specific-pathogen-free (SPF) chicken eggs. rLa-EBOVGP retained its low pathogenicity, as a lentogenic strain, in eggs and chickens [44], with a mean death time (MDT) > 140 h, an intracerebral pathogenicity index (ICPI) of 0, and an intravenous pathogenicity index (IVPI) of 0. During the three-week observation period after the mice were inoculated either intramuscularly (i.m.) or intranasally (i.n.) with a high dose of rLa-EBOVGP, they showed no signs of sickness or death, and did not differ in weight gain from the mice inoculated with rLa. These results suggest that the expression of EBOV GP does not increase the pathogenicity of the NDV vector in poultry or mice.

Previous studies have reported that the envelope glycoproteins of heterogeneous viruses can be expressed by recombinant NDV and incorporated into the viral vector particles [17, 35]. To investigate whether the EBOV GP expressed by the recombinant NDV is incorporated into the viral vector particles, the viral particles of rLa-EBOVGP and rLa were purified by sucrose gradient centrifugation and subjected to immunoblotting analysis with mouse anti-EBOV GP antiserum as the primary antibody. A clear GP1 band of ~130 kDa was apparent in the rLa-ZEBOVGP sample but not in the rLa sample. This indicates that the GP protein was expressed by rLa-EBOVGP and incorporated into the recombinant NDV particles, which is consistent with a previous report [17].

The cleavage of the F glycoprotein is a prerequisite for the infectivity of NDV. Because it is a lentogenic NDV strain, the infectivity of rLa depends on exogenous trypsin-like extracellular proteases [35, 36]. To investigate whether the incorporation of EBOV GP alters the infectivity of the recombinant NDV, rLa and rLa-EBOVGP were propagated in eggs (rLa-egg and rLa-EBOVGP-egg, respectively) and in BHK-21 cells with or without TPCK-trypsin (Sigma) in the medium (rLa-cell/TPCK, rLa-EBOVGP-cell/TPCK, rLa-cell, and rLa-EBOVGP-cell respectively). These prepared viruses were then used to infect BHK-21 cells at a multiplicity of infection (MOI) of 0.02–0.05 with no exogenous trypsin in the medium. As expected, rLa-cell did not infect any cells (Figure 2, column 2), whereas rLa-egg and rLa-cell/TPCK infected individual cells but did not spread between the cells. However, rLa-EBOVGP-egg, rLa-EBOVGP-cell, and rLa-EBOVGP-cell/TPCK showed similar infectivity and spreadability in the cells. At 120 h postinfection (PI), more than 90% of cells were infected by all the differently prepared rLa-EBOVGP viruses (Figure 2, columns 3 and 4). These results suggest that the expression and incorporation of EBOV GP allow the vector NDV to infect cells and to spread among the cells independently of exogenous trypsin.

The two envelope glycoproteins of NDV, HN and F, are indispensablefor cell entry, which includes receptor binding and virus-cell membrane fusion. This is the first step in infection and a prerequisite for viral replication [26]. EBOV has only one envelope glycoprotein, which functions in receptor binding and membrane fusion [4‐8]. To understand the impact of EBOV GP on the infectivity of the NDV vector, the sensitivities of rLa and rLa-EBOVGP to NDV antiserum and EBOV GP antiserum were evaluated and compared (Figure 3). As expected, both rLa-egg and rLa-cell were resistant to mouse anti-EBOV GP antiserum (diluted 1:10) but were completely neutralized by chicken anti-NDV antiserum (diluted 1:100). However, both anti-NDV antiserum (diluted 1:100) and anti-EBOV GP antiserum (diluted 1:10) only partially neutralized rLa-EBOVGP-cell, rLa-EBOVGP-cell/TPCK, and rLa-EBOVGP-egg (Figure 3, column 2). The anti-NDV antiserum and anti-EBOV GP antiserum reduced the infection with each rLa-EBOVGP virus by about 90% and 60%, respectively (Figure 3, column 3). When the anti-NDV antiserum (diluted 1:100) and anti-EBOV GP antiserum (diluted 1:10) were mixed, infection with rLa-EBOVGP-cell, rLa-EBOVGP-cell/TPCK, or rLa-EBOVGP-cell was completely blocked (Figure 3, column 4). The same dilution of SPF chicken0020serum, naïve mouse serum, or a mixture of SPF chicken serum and naïve mouse serum, used as the controls, showed no neutralization activity against any of the differently prepared rLa or rLa-EBOVGP viruses. These results suggest that the incorporation of EBOV GP into the recombinant viral particle made the NDV vector more resistant to anti-NDV antiserum and more sensitive to anti-EBOV antiserum.

NVD envelope glycoproteins HN and F usually bind to receptor and induce virus-cell membrane fusion at neutral pH. A previous study reported that NDV can also partially enter the host cells by caveolae-mediated endocytosis [45], whereas EBOV mainly enters cells via macropinocytosis [46‐48] in a GP-dependent manner [46, 49]. Because EBOV GP is incorporated into the vector NDV particles and independently mediates the cell entry of rLa-EBOVGP, it is necessary to determine whether the incorporation of the GP protein alters the endocytosis pattern of rLa-EBOVGP during infection. To address this question, we used two chemicals, IPA3 and methyl β-cyclodextrin (MβCD) to treat BHK-21 cells before their infection with the rLa or rLa-EBOV virus. IPA3 inhibits the activation of PAK1 kinase, which is required for macropinocytosis [50, 51] and MβCD sequesters cholesterol from the cell membrane, thus inhibiting clathrin- and caveolae-mediated endocytosis [52, 53]. As shown in Figure 4, the infection of IPA3-pretreated BHK-21 cells with rLa-EBOVGP-cell, rLa-EBOVGP-cell/TPCK, or rLa-EBOVGP-egg was greatly reduced, whereas IPA3 had no inhibitory effect on the infectivity of rLa-egg or rLa-cell/TPCK. Fewer than 10% of IPA3-pretreated cells were infected with rLa-EBOVGP-cell, rLa-EBOVGP-cell/TPCK, or rLa-EBOVGP-egg compared with the untreated cells. Infection by rLa-EBOVGP-cell, rLa-EBOVGP-cell/TPCK, or rLa-EBOVGP-egg was not inhibited in cells pretreated with MβCD, but their infection by rLa-cell/TPCK and rLa-egg was moderately reduced. Pretreatment with a combination of IPA3 and MβCD almost completely blocked the infection of cells by rLa-EBOVGP-cell, rLa-EBOVGP-cell/TPCK, and rLa-EBOVGP-egg, and also mildly inhibited the infection of cells by rLa-cell/TPCK and rLa-egg. To further testify the role of EBOV GP on the internalization of the recombinant virus, we use 800 mU/ml bacterial neuraminidase (NA, Sigma N2876) to treat cells with chemical inhibitors together with IPA3 or alone before infection. Our preliminary data showed use 800 mU/ml NA to treat cells could block over 90% of NDV infection, thus in this assay we used NA to exclude the influence of HN protein in the internalizaiton of rLa-EBOVGP-egg and rLa-EBOVGP-cell/TPCK. Also shown in Figure 4 (right panel), NA treatment reduced over 90% of rLa-egg and rLa-cell/TPCK infection, while it had almost no reduction on the infection of rLa-EBOVGP-egg, rLa-EBOVGP-cell/TPCK and rLa-EBOVGP-cell. The NA + IPA3 treatment could block 90% of the recombinant viruses from infection, while NA + MβCD had no inhibition on these viruses. These results futher explicated the role of GP on the macropinocytotic internalization of the recombinant viruses. In conclusion, Our results indicate that recombinant rLa-EBOVGP uses macropinocytosis as its major internalization pathway for cell entry, rather than the direct fusion at the cell plasma membrane at neutral pH like NDV.

The BSR-T7/5 cells for virus rescue and the BHK-21 cells for virus growth and titration were maintained in complete Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS). The NDV strains were propagated and titrated in 9-day-old SPF embryonated chicken eggs [44] or BHK-21 cells in the presence or absence of TPCK-trypsin (Sigma).

To construct a full-length recombinant genomic cDNA, the cDNA of the GP gene of Zaire EBOV was amplified from synthesized cDNA (GenBank accession no. AF086833.2) using the following primers: 5′-GACTGTTTAAACttagaaaaaaTacgggtagaaCgccaccATGGGCGTTACAGGAATATTGCAG-3′ and 5′-CTGAGTTTAAACGCTAAAAGACAAATTTGCATATACAG-3′. The GP gene was flanked by the restriction site for endonuclease PmeI (boldface letters); the NDV gene start sequence (GS, 5′-acgggtagaa-3′) and gene end sequence (GE, 5′-ttagaaaaaa-3′) are included before the optimal Kozak sequence (italic lowercase letters) and the GP sequence (italic uppercase letters). The amplified fragment was digested with PmeI and then inserted into the P–M intergenic region at nucleotide position 3165 of the NDV genome, as described previously [32]. The resultant plasmid was used for recombinant NDV rescue, as described previously [32]. The expression of EBOV GP was confirmed with an immunofluorescence assay (IFA) and western blotting. The resultant recombinant virus was designated “rLa-EBOVGP”.

NDV infection was detected in cells with IFA with chicken anti-NDV antiserum, as previously described [32]. For the confocal assays, BHK-21 cells were grown in 24-well plates or plated on cover slips in dishes (35 mm diameter) and infected with rLa or rLa-EBOVGP. At 24 h after infection, the cells were fixed in prechilled 3% paraformaldehyde in phosphate-buffered saline (PBS) for 15 min at room temperature, washed three times with PBS, and then blocked with PBS containing 1% (wt/vol) bovine serum albumin at room temperature for 1 h. The cells were then incubated with mouse anti-EBOV GP antiserum (the antiserum was prepared in mice immunized with two doses of recombinant VSV expressing EBOV GP, which was generated in our laboratory) or chicken anti-NDV antiserum for 1 h at room temperature. The cells were then washed three times with PBS containing 0.05% Tween 20 and stained with a fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse antibody (Sigma) or a tetramethylrhodamine isothiocyanate-conjugated rabbit anti-chicken antibody (Sigma) for 30 min. The cells were washed three times with PBS, stained with DAPI, and then analyzed with fluorescence microscopy or confocal laser microscopy. The images were acquired with a Zeiss (Thornwood, NY) Axioskop microscope equipped for epifluorescence with a Sensys charge-coupled device camera (Photometrics, Tucson, AZ) using the IPLab software (Scanalytics, Vienna, VA).

Egg-propagated rLa and rLa-EBOVGP were purified by sucrose gradient ultracentrifugation. The total protein (5 μg) of each purified virus was subjected to SDS–PAGE under denaturing conditions. After the proteins were transferred from the gel to nitrocellulose membrane, the target band (s) were detected with chicken anti-NDV antiserum or mouse anti-EBOV GP antiserum as the primary antibody and the corresponding horseradish-peroxidase-conjugated goat anti-chicken or goat anti-mouse immunoglobulin G as the secondary antibody. The bands were visualized with ECL Plus Western Blotting Detection Reagents (GE Health Science) on Kodak X-ray film.

Each recombinant virus (5 × 10² TCID₅₀) was mixed with chicken anti-NDV antiserum (diluted 1:100) or mouse anti-EBOV GP antiserum (diluted 1:10) or with a mixture of chicken and mouse antisera (at the same dilutions). The viruses were also mixed with a mixture of SPF chicken serum and naïve mouse serum as the mock-treated group. The virus–serum mixtures were incubated at 37°C for 1 h and then used to infect monolayers of BHK-21 cells in a 12-well plate. At 1 h after infection, the supernatants were discarded and the cells were washed three times with DMEM. At 48 h after infection, the cells were fixed, and IFA was performed with chicken anti-NDV antiserum as the primary antibody.

To determine the pathogenicity of rLa-EBOVGP in poultry, MDT, ICPI, and IVPI were determined according to the OIE Manual [44]. To assess the pathogenicity of rLa-EBOVGP in mammalian cells, two groups of 10 six-week-old female Balb/c mice (Vital River, Beijing, China) were inoculated i.m. with 10⁸ EID₅₀ of rLa-EBOVGP or rLa, and simultaneously i.n. with 3 × 10⁷ EID₅₀ of rLa-EBOVGP or rLa. The third group of 10 mice was inoculated i.m. with 0.1 mL and i.n. with 0.03 mL of PBS as the mock-infection control. The mice were monitored daily to detect any weight changes, signs of illness, or death.

BHK-21 cells were pretreated with either the inhibitor 10 μM IPA3 (Sigma) or 10 mM MβCD (Sigma) alone or together at 37°C. To exclude the influence of HN of vector NDV virus on the internalization of the recombinant viruses, another plate of BHK-21 cells were pretreated with 800 mU/ml bacterial neuraminadase (NA, Sigma N2876) alone or together with the inhibitors at 37°C. At 1 h after treatment, the cells were washed three times and infected with 5 × 10² TCID₅₀ of rLa or rLa-EBOVGP at 4°C for 1 h in the presence of the inhibitor(s). The cells were then washed three times on ice. DMEM containing 10% FBS was added and the samples were incubated in 37°C for 7 h. After incubation, the cells were fixed with 3% paraformaldehyde and permeabilized with 0.1% saponin. Cells infected with rLa or rLa-EBOVGP were detected with immunofluorescence assay, as described above. The mean number of flurorecent possitive cells of a minimum of 5 fields of view were counted.The data was expressed as means and standard deviations.

The present study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of the People's Republic of China. The protocol was approved by the Animal Research Ethics Committee of Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences (approval numbers 20132085 for chickens and 20132138 for mice).