Species-independent bioassay for sensitive quantification of antiviral type I interferons

Using our reverse genetics system [21], we generated the recombinant virus RVFV-Ren with the IFN antagonist NSs replaced by Renilla luciferase (see Materials and Methods). Luciferase expression by this virus was correlating with viral replication and remained stable over several passages (data not shown). We then tested the IFN sensitivity of RVFV-Ren in human A549 cells. Cells were seeded in 96-well microtiter plates and pre-treated with serial dilutions of standard IFN for 7 hours to allow the establishment of an antiviral state. Afterwards, cells were infected with RVFV-Ren at an MOI of 1 for 16 hours and Renilla luciferase activity was measured in cell lysates (Fig. 1A). To determine the linear range of the assay, IFN dilutions from 0.5 U/ml up to 100 U/ml were tested. A significant reduction of luciferase activity was observed using IFN concentrations from 1 U/ml on (Fig. 1B). The highest concentration of IFN in the linear range was 50 U/ml.

We used the RVFV-Ren assay to quantify IFN induction by two recombinant viruses which had been generated by our group. Wild-type (wt) La Crosse virus is able to suppress IFN induction, whereas a mutant virus upregulates the IFN-β gene [22]. Supernatants were taken from cells infected with these viruses and sterilized with β-propiolactone to destroy infectiousness. After removal of the disinfectant, undiluted and serially diluted samples of supernatants were subjected to the RVFV-Ren assay in parallel to the serial dilutions of standard IFN (see Fig. 1B). The IFN dilutions were used to create a standard curve by regression analysis (Fig. 2A) which, in turn, served to calculate IFN concentrations of the La Crosse virus supernatants. For cells infected with wt La Crosse virus, the activity of the undiluted supernatants was taken to determine the amount of type I IFN. For the mutant virus, we had to use the 1:10 dilution as the basis for the calculations since undiluted supernatants would have been out of the linear range of the assay. As expected, only little IFN was present in supernatants of cells infected with wt La Crosse virus, but substantial amounts were measured in the supernatant of mutant virus-infected cells (Fig 2B).

Since RVFV is able to infect a broad range of vertebrate hosts, we wanted to know whether the assay which was established for human IFN could be adapted to measure IFN from other species. To this aim, we infected L929 mouse cells and chicken embryo fibroblasts (CEFs), and tested the anti-RVFV-Ren activity of murine and avian IFN, respectively. As shown in Figure 3, RVFV-Ren was able to replicate in both cell lines and viral growth was inhibited in a dose-dependent manner by the species-compatible IFN. The sensitivity of the assay on L929 (Fig. 3A) as well as on CEFs (Fig. 3B) was comparable to human cells, with a linear range from 1 to 50 U/ml and from 1 to 25 U/ml, respectively.

Type I IFNs are the main, but not the only, cytokines produced during an innate immune response. We wanted to know whether other antiviral molecules, e.g. IFN-λ, could disturb our bioassay. To investigate this, we employed embryo fibroblasts from knockout mice lacking the receptor for type I IFNs (IFNAR-/- MEFs). Supernatants containing antiviral mouse cytokines were obtained by infecting L929 cells with the RVFV strain clone 13 [15, 16], a virus mutant which in our experience is one of the strongest inducers of antiviral cytokines. Clone 13 upregulated an innate immune response including IFN-β, as expected (Fig. 4A). When we performed the bioassay on IFNAR-/- MEFs, the indicator virus RVFV-Ren was not inhibited by IFN-α, as expected, but also not by supernatants (Fig. 4B). This strongly indicates that it is type I IFNs which are causing the antiviral effect against RVFV-Ren.

BHK-21, Vero E6 (ATCC; CRL-1586), 293T (ATCC; CRL-11268), human A549 (ATCC; CCL-185) cells, and murine L929 cells (ATCC; CCL-1) and IFNAR -/- MEFs (kindly provided from Jovan Pavlovic, University of Zurich, Switzerland) were cultivated in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS; Biochrom AG) and antibiotics. Chicken embryo fibroblasts (CEF) were prepared from 10-day-old chicken embryos and maintained in DMEM with 2% chicken serum (Invitrogen), 8% FCS and antibiotics. Viruses used in this study were recombinant Rift Valley fever virus (RVFV) expressing Renilla luciferase (see below), RVFV Clone 13 [15, 16], and recombinant La Crosse viruses expressing (wt) or lacking expression (mutant) of the NSs gene [22]. Virus titers were determined by standard plaque assay on Vero E6 cells.

Recombinant pan-species IFN-α (IFN-α B/D Bgl II) was purchased from PBL Biomedical Laboratories, and Multiferon, a mix of natural human IFN-α subtypes, was from Viragen. Recombinant chicken interferon (chIFN) was purified from E. coli and calibrated as described previously [23, 24].

We used our plasmid-based rescue system for generation of recombinant RVFV[21]. Expression of Renilla luciferase by RVFV-Ren was achieved by replacing the non-structural NSs gene on the genomic S segment with the Renilla gene. The corresponding S segment rescue plasmid pHH21-RVFV-vN_Ren was generated by inserting the Renilla luciferase open reading frame, amplified from plasmid pRL-SV40 (Promega), into the cloning site of pHH21-RVFV-vN_TCS [21]. RVFV-Ren was rescued by transfecting cocultures of 293T and BHK-21 cells in six-well plates with 0.5 μg of helper plasmids (pI.18-RVFV-L and pI.18-RVFV-N), together with 1 μg each of pHH21-RVFV-vL, pHH21-RVFV-vM, and pHH21-RVFV-vN_Ren using Nanofectin transfection reagent (PAA Laboratories). Supernatants containing recombinant viruses were collected 5 days post transfection and used to grow virus stock on Vero E6 cells. All RVFV rescues were performed under biosafety level (BSL) 3 conditions.

Prior to measurement of IFN-containing samples, remaining virus was inactivated using β-propiolactone (Acros Organics) [11, 25]. Briefly, supernatants were first incubated in the presence of 0.05% β-propiolactone in plastic dishes overnight at 4°C, and then at 37°C for 2 hours for hydrolysis of β-propiolactone.

Approximately 10,000 cells were seeded into each well of a 96-well microtiter plate and incubated overnight in a humified incubator at 5% CO₂ and 37°C. Cells were then treated either with different dilutions of standard IFN or serial ten-fold dilutions of IFN-containing samples in 100 μl of growth medium for 7 hours. Subsequently, cell culture medium was removed and 10,000 plaque forming units of RVFV-Ren in 100 μl of infection medium (DMEM with 2% FCS and 20 mM HEPES, pH 7.3) were added per well. After 16 hours of further incubation, supernatants were removed and cells lysed in 50 μl of 1 × Renilla lysis buffer (Promega). Luciferase activity in 10 μl of cell lysate was measured using the Renilla luciferase assay system (Promega), according to the manufacturer's instructions.