Plant-produced Crimean-Congo haemorrhagic fever virus nucleoprotein for use in indirect ELISA

Journal of Virological Methods 236 (2016) 170–177

Contents lists available at ScienceDirect

Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet

Plant-produced Crimean-Congo haemorrhagic fever virus nucleoprotein for use in indirect ELISA Richard Atkinson a , Felicity Burt b , Edward P. Rybicki a,c , Ann E. Meyers a,∗ a

Biopharming Research Unit, Department of Molecular and Cell Biology, University of Cape Town, P Bag X3, Rondebosch 7701, South Africa Department of Medical Microbiology and Virology, National Health Laboratory Service Universitas and Faculty of Health Sciences, University of Free State, P O Box 339, Bloemfontein 9300, South Africa c Institute of Infectious Disease and Molecular Medicine, Faculty of Health Sciences, University of Cape Town, Observatory 7925, South Africa b

a b s t r a c t Article history: Received 19 November 2015 Received in revised form 22 July 2016 Accepted 22 July 2016 Available online 26 July 2016 Keywords: Crimean-Congo haemorrhagic fever Nucleoprotein Recombinant antigen

Crimean-Congo haemorrhagic fever (CCHF) is a disease of serious public concern caused by the CCHF virus (CCHFV). Anti-CCHFV IgG in humans can be detected using ELISA with native antigen prepared from cell cultures which have been infected with virus or from brain tissue of suckling mice which have been inoculated with virus. However, the preparation of these reagents requires high biosafety levels and is expensive. A safer, more cost-effective recombinantly-produced NP reagent is desirable. Recently, plants have been shown to be a cost-effective and safe system for expression of recombinant proteins. This work describes cloning of the CCHFV NP gene into three different plant expression systems and comparison of expression in Nicotiana benthamiana. The highest expressing construct was selected. Expressed NP was purified by ammonium sulphate fractionation prior to histidine affinity chromatography. Purified NP was tested in an indirect ELISA to determine if the recombinant antigen was able to detect anti-CCHFV IgG in sera from convalescent patients. Plant-produced NP detected IgG antibodies against CCHFV in 13/13 serum samples from convalescent patients and 0/13 samples collected from volunteers with no history of CCHFV infection. Results were compared with commercially available immunofluorescent assays and 100% concordance was obtained between the two assays. This suggests that a full evaluation of the plant produced NP for application as a safe recombinant is warranted. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Crimean-Congo haemorrhagic fever virus (CCHFV) is an enveloped single-stranded RNA virus belonging to the Nairovirus genus in the family Bunyaviridae (Whitehouse, 2004). It is the cause of Crimean-Congo haemorrhagic fever (CCHF) in humans which is considered to be a significant public health concern as it has a high fatality rate and propensity to cause nosocomial infections. The virus is maintained predominantly by ixodid ticks belonging to the Hyalomma genus (Hoogstraal, 1979). The virus circulates in an enzootic tick-vertebrate-tick cycle and is transmitted through the eggs of the vector as well as from immature ticks to its subsequent life stages (Bente et al., 2013). The ticks normally feed on

∗ Corresponding author at: Department of Molecular and Cell Biology, University of Cape Town, P Bag X3, Rondebosch, 7700, South Africa. E-mail addresses: [email protected] (R. Atkinson), [email protected] (F. Burt), [email protected] (E.P. Rybicki), [email protected] (A.E. Meyers). http://dx.doi.org/10.1016/j.jviromet.2016.07.025 0166-0934/© 2016 Elsevier B.V. All rights reserved.

a large variety of vertebrates including sheep, goats, cattle, large wild herbivores, hares and hedgehogs which are all susceptible to CCHFV infection. However, most vertebrates do not show symptoms of CCHFV infection but develop sufficient viraemia to support transmission of the virus to uninfected ticks. Humans are generally infected by being bitten by infected ticks, as well as handling crushed infected ticks and by direct contact with infected blood or tissue of animals or humans (Swanepoel et al., 1989). Apart from newborn mice, humans are the only hosts of CCHFV in which disease is manifested. Recently two animal models have been described which will provide a useful tool for further investigations particularly vaccine development and antivirals (Bente et al., 2010; Bereczky et al., 2010). The significance of infection in humans is the rapid development of severe haemorrhagic disease, most often leading to death, as well as the ease with which nosocomial outbreaks seem to occur. The major pathology of CCHFV infection in humans is vascular dysfunction with subsequent haemorrhaging but initial symptoms usually include high

R. Atkinson et al. / Journal of Virological Methods 236 (2016) 170–177

fever, headache, fatigue, muscle aches, abdominal pain, nausea, vomiting and diarrhoea (Ergonul, 2006). The virus is widely distributed and tends to follow the geographical range of the tick vectors. Areas include the western region of the former Soviet Union, south-eastern and south-western Europe, eastern and central Asia, the Middle East and Turkey and Africa (Bente et al., 2013; Leblebicioglu, 2010), where in many areas it is endemic. CCHF incidence has increased in the last few years (Maltezou and Papa, 2011), and is considered an emerging disease due to the emergence or re-emergence of new foci. One of the reasons for this is that climate change is widening the region in which ticks were originally distributed (Kivaria, 2012). This is of particular concern since the virus may spread to previously non-endemic areas. For this reason, surveillance and monitoring of CCHFV in animals and humans is particularly important. In the absence of specific antiviral treatment or a vaccine, treatment of the disease is generally supportive. The CCHFV RNA tripartite genome consists of the small (S), medium (M) and large (L) segments (Hoogstraal, 1979). The S segment encodes a nucleocapsid protein (NP) that associates with the viral RNA in the virion to form a ribonucleocapsid. NP is the most abundant viral protein produced by CCHFV, is highly conserved across CCHFV strains (Deyde et al., 2006) and has been shown to produce a strong specific humoral antibody response (Marriott et al., 1994). It is also an important antigen used for the diagnosis of CCHFV infection (Bente et al., 2013; Vanhomwegen et al., 2012). One of the methods used to confirm a diagnosis of CCHFV infection is by detection of IgG and IgM antibodies in human serum using indirect enzyme-linked immunoassay (ELISA) (Vanhomwegen et al., 2012). This method utilises antigen prepared from mammalian cell cultures that have been infected with live CCHFV or from the brain tissue of suckling mice that have been inoculated with live CCHFV. However, the preparation of this reagent in both cases requires BSL-4 facilities, significantly limiting the number of laboratories that can prepare the antigen. Human and veterinary surveillance is an important part of monitoring and predicting potential outbreaks of the virus. Since CCHFV affects largely developing countries where facilities are not necessarily well equipped, there is a great need for a specific, sensitive and cost-effective assay for testing for CCHFV infection that can be deployed across such countries which do not have the expertise, facilities or funding. To overcome this hurdle, recombinant CCHFV NP would be a suitable option as this circumvents requirements for high biosafety levels and sophisticated facilities. CCHFV NP has previously been expressed in insect and bacterial cell culture. Zhou et al. (Zhou et al., 2011) produced NP in insect cells which formed virus-like particles. Saijo et al. (Saijo et al., 2002) showed that a his-tagged recombinant NP protein cloned into baculovirus and expressed in insect cells was able to detect immunoglobulin G (IgG) antibodies to CCHFV in an ELISA. However, insect cell culture has the drawback of high cost as well as the difficulty of producing and maintaining the required baculovirus clones (Saijo et al., 2002). Samudzi et al. have shown that recombinant NP can be produced in an Escherichia coli expression system and that it is functional as a binding antigen in an indirect ELISA in detecting anti-CCHFV IgG antibodies (Samudzi et al., 2012). However, problems with this expression system are that the protein is often insoluble, which increases downstream processing requirements when scaled up, increasing the cost and time required, to produce a protein ready for use in a diagnostic assay. Plants have been tested for the production of recombinant proteins for many years, and expression is categorised into transgenic or transient strategies. Transient recombinant protein expression by recombinant Agrobacterium tumefaciens-mediated infiltration is

171

rapidly becoming the most viable option as it does not require high levels of sterility, the process is more cost-effective than cell culture, and it is easily scalable (Chen et al., 2013). This work describes the production of a more cost-effective and safer CCHFV NP antigen reagent via transient expression in plants, which has potential for use in an indirect CCHFV ELISA to detect antibodies to the virus in human serum. 2. Material and methods 2.1. Codon optimization and cloning of NP gene The complete nucleotide sequence of the CCHFV NP gene of a South African isolate SPU 415/85 was retrieved from GenBank (Accession no. U88415). Although CCHF isolates from geographically distinct regions are considered antigenically similar and serologically cross reactive, the NP of a South African isolate was selected for the study as serum samples to be tested were from South African survivors. The gene was Nicotiana sp. codonoptimized by GenScript Inc (NJ, USA). Several internal restriction enzyme sites that would have interfered with subcloning were also removed from the sequence, along with 3 different cis-acting elements (PolyA(AATGAA); PolyA(AATGGA) and PolyA(AAAAAAA)). The gene was blunt-end cloned into the standard pUC57 vector(Thermo Fisher Scientific, US) using EcoRI to yield pUC57-NP. pUC57-NP was transformed into E. coli DH5-␣ chemically competent cells and transformants selected for on Luria Bertani (LB) medium containing ampicillin (100 ␮g/ml). Plasmid DNA from a single colony was purified from E. coli using a QIAprep® Spin Miniprep kit (Qiagen) according to the manufacturer’s instructions. The NP gene was subsequently subcloned directly into three different plant expression vectors, pRIC3.0-HT (Regnard et al., 2010), pTRAc-HT (Maclean et al., 2007) and pEAQHT (Sainsbury et al., 2009) using restriction enzymes NcoI and XhoI for the first 2 and XmaI and XhoI for the third. All 3 vectors resulted in the fusion of an in-frame 6 × histidine tag on the N terminus of NP. The recombinant constructs pRIC-NP(HT) and pTRAcNP(HT) were transformed into Agrobacterium tumefaciens GV3101:pMP90RK and pEAQ-NP(HT) into A. tumefaciens LBA4404 cells made electrocompetent using the method described by (Shen and Forde, 1989). Recombinant colonies were recovered according to the method described in Maclean et al. (Maclean et al., 2007) and selected for on LB at 27 ◦ C containing kanamycin (50 ␮g/ml) and rifampicin (50 ␮g/ml) for pEAQ-NP(HT) with the addition of carbenicillin (50 ␮g/ml) for pTRAc-NP(HT) and pRICNP(HT). Positive clones were confirmed by colony PCR of purified plasmid DNA using pRIC3.0/pTRAc- or pEAQ-specific primers (pRIC3.0/pTRAc Forward: 5 CATTTCATTTGGAGAGGACACG 3 ; Reverse: pRIC3.0/pTRAc Reverse 5 GAACTACTCACACATTATTCTGG 3 and pEAQ Forward 5 GACGAACTTGGAGAAAGATTGTTAAGC; pEAQ Reverse 5 GACCGCTCACCAAACATAGAAATG 3 ). Clones were also verified by back-transformation of purified plasmid DNA (QIAprep® Spin Miniprep kit, Qiagen) into competent E. cloni® cells (Lucigen, US) and selection on LB ampicillin (100 ␮g/ml) or kanamycin (50 ␮/ml) for pEAQ clones. 2.2. Small-scale Agrobacterium tumefaciens-mediated infiltration The recombinant Agrobacterium strains harbouring pRICNP(HT) and pTRAc-NP(HT) were infiltrated either separately or in combination with A. tumefaciens LBA4404 containing a silencing suppressor (pBIN-NSs) (provided by Marcel Prins, Laboratory of Virology, Wageningen, The Netherlands) from the Tomato spot-

172

R. Atkinson et al. / Journal of Virological Methods 236 (2016) 170–177

ted wilt virus (TSWV). The influence of this on protein expression was tested, as it has been previously shown that co-expression of silencing suppressors can increase levels of transiently expressed recombinant proteins (Voinnet et al., 2003). Ten ml starter cultures of A. tumefaciens harboring pRIC-NP(HT), pTRAc-NP(HT) or pEAQNP(HT) in LB containing the appropriate antibiotics were incubated in a shaking incubator overnight at 27 ◦ C at 200 rpm. These were inoculated into 1 L flasks containing 100 ml induction medium (LB containing 10 mM MES, 20 ␮M acetosyringone) and appropriate antibiotics as described above which were subsequently incubated overnight at 27 ◦ C at 200 rpm in a shaking incubator (MRC, UK). A 10 ml starter culture of A. tumefaciens LBA4404 harbouring pBIN-NSs in LB containing 50 ␮g/ml rifampicin and 2 mM MgSO4 was concomitantly incubated overnight at 27 ◦ C at 200 rpm and inoculated into 100 ml induction medium containing 50 ␮g/ml rifampicin. Cells were harvested by centrifugation at 1000g for 5 min and resuspended in 1 ml of infiltration medium (10 mM MgCl2 , 10 mM MES, 3% sucrose and 150 ␮M acetosyringone in water, pH 5.6). After incubation at 22 ◦ C for 2 h, the suspensions were further diluted for infiltration: when pRIC-NP(HT), pTRAcNP(HT) and pEAQ-NP(HT) were infiltrated on their own, a dilution to an optical density (OD600 ) of 0.5 was used; when pRIC-NP(HT) and pTRAc-NP(HT) were co-infiltrated with pBIN-NSs, each culture was diluted to an optical density at 600 nm of 1.0 (measured using an Ultrospec 10 spectrophotometer (Amersham Biosciences, UK)) and equal volumes of appropriate culture suspensions subsequently added to each other. Diluted cells were syringe-infiltrated into the abaxial surface of 6-week old Nicotiana benthamiana leaves and the plants subsequently incubated under conditions of 8 h dark/6 h light at 25 ◦ C for up to 7 days. Three leaf-discs were harvested at days 1, 3, 5 and 7 post infiltration using the cap of a 1.5 ml microcentrifuge tube (each disc weighing approximately 10 mg) and immediately flash frozen and ground into a fine powder using liquid nitrogen with a mortar and pestle. Three hundred ␮l of 1 × PBS was added to resuspend the leaf material. The resulting extracts were centrifuged at 13000 × g for 10 min at 4 ◦ C to pellet the insoluble material. The supernatants were retained for further analysis. 2.3. SDS-PAGE analysis and western blotting The supernatant samples were prepared for SDS-PAGE by the addition of 5 × SDS sample buffer (25% (v/v) glycerol, 0.5 M DTT, 5% (w/v) SDS, 0:001% (w/v) bromothymol blue) and incubation at 95 ◦ C for 5 min. Equal amounts of samples were loaded onto a 10% SDS polyacrylamide gel and separated by electrophoresis. The gels were subsequently stained with 0.1% Coomassie blue stain made up in methanol. Expressed NP was initially verified by probing of a western blot with anti-NP (GenScript, US) used at a dilution of 1: 1000 (data not shown). However, western blots probed with anti-6 × his antibody were much cleaner and therefore used thereafter for comparative analyses of expression. Separated proteins on a second gel were electroblotted onto 0.45 ␮m nitrocellulose (AmershamTM ProtranTM ) at 15 V for 1 h using a Trans-blot® SD semi-dry blotter (Bio-Rad). The membrane was blocked in blocking buffer (5% non-fat dairy milk and 1 × PBST (137 mM NaCl, 10 mM Na2 HPO4 , 2.7 mM KCl, 1 mM KH2 HPO4 at pH 7.4 and 0.1% Tween 20) for 1 h and probed overnight with 1:2000 dilution in blocking buffer of 6 × histidine tag antibody (AbD Serotech; Cat. No. MCA1396) at 4 ◦ C. The membrane was washed for 15 min 4 times with blocking buffer (2% skimmed milk in 1 × phosphate-buffered saline) and subsequently incubated in 1: 10 000 dilution in blocking buffer of alkaline phosphatase-labelled goat anti-mouse secondary antibody (SigmaAldrich; Cat. No. A3562) for 2 h at 22 ◦ C. The blot was washed for 15 min 4 times in PBST and protein detected using BCIP/NBT (KPL).

2.4. Large-scale Agrobacterium tumefaciens-mediated infiltration A pre-culture of recombinant A. tumefaciens pRIC-NP(HT) was generated by inoculation of 5 ml YEB (5 g/L beef extract, 1 g/L peptone, 5 g/L sucrose, 300 mg/L MgSO4 ) containing the appropriate antibiotics as described above and incubation for 16 h at 27 ◦ C. The pre-culture was inoculated into 250 ml YEB containing the appropriate antibiotics, excluding rifampicin, and incubated at 27 ◦ C for 40 h. The OD600 of the culture was determined and subsequently diluted to an OD600 of 1.0 using 2 × infiltration medium pH 5.6 (100 g/L sucrose, 3.6 g/L). Acetosyringone was added to a final concentration of 200 ␮M. A 500 ml culture of pBIN-NSs was grown up as described in 2.2 and diluted to an OD600 of 1.0 in infiltration medium. Equal volumes of the 2 cultures were added to each other and used for infiltration. Twenty 6-week-old whole N. benthamiana plants were vacuum-infiltrated at −90 kPa with diluted A. tumefaciens pRIC-NP(HT) and incubated for 3 days as described above. Leaves were harvested 3 days post infiltration (dpi), weighed, cut up finely using a sterile pair of scissors and then homogenised on ice in 1 × PBS (phosphate buffered saline pH 7.4) in a ratio of 3:1 (vol:leaf mass). The homogenate was incubated for 30 min at 4 ◦ C with gentle shaking. The samples were then centrifuged at 10 000g at 4 ◦ C for 20 min and filtered through MiraclothTM (22–24 ␮m pore, Millipore, USA).The supernatant was centrifuged twice more at 10 000 g at 4 ◦ C for 20 min to rid the supernatant from any remaining particulate contamination. 2.5. Ammonium sulphate purification The supernatant was treated with a series of increasing ammonium sulphate ((NH4 )2 SO4 ) solutions (0–10%, 10–20%, 20–30%, 30–40%, 40–50%, 50–60%, 60–70%, 70–80%, 80–90% and 90–100%) according to Englard and Seifter (Englard and Seifter, 1990). The amount of solid (NH4 )2 SO4 required was calculated using the calculator on http://www.encorbio.com/protocols/AM-SO4.htm. The (NH4 )2 SO4 was added slowly to the extract and kept at 4 ◦ C with constant stirring for 1 h to equilibrate and for the protein to precipitate. Suspensions were centrifuged at 10 000 × g for 20 min to pellet the extracted protein which was subsequently dissolved in 1 × PBS and stored at 4 ◦ C. The protein content of each fraction was analysed by western blot as described in 2.2 and on a Coomassie-stained gel in order to determine which fractions contained the most NP protein. For scaled-up production of NP, the 40–80% (NH4 )2 SO4 fraction was used for purification by histidine affinity chromatography. 2.6. Purification of NP using histidine affinity chromatography The protein pelleted from the 40 to 80% (NH4 )2 SO4 fraction was resuspended in 1 x PBS and subsequently dialysed in 20 × volume high salt PBS (1 × PBS supplemented with NaCl to a final concentration of 500 mM) overnight at 4 ◦ C to remove residual (NH4 )2 SO4 . The dialysed sample was centrifuged at 15 000g at 4 ◦ C for 20 min to remove any precipitate that may have formed during dialysis. Histidine affinity chromatography was performed using a 1 ml HisTrapTM HP prepacked column (GE Healthcare, USA) on an ÄktaExplorer fast protein liquid chromatography (FPLC) system (GE Healthcare, USA). The buffer flow rate throughout was set at 1 ml/min. The column was equilibrated with 10 column volumes of high salt PBS, supplemented with 20 mM imidazole (buffer A). Twenty ml of dialysed supernatant was injected directly onto the column using a 50 ml SuperLoop and washed with 20 column volumes of buffer A. The protein was eluted in 0.5 ml fractions using a linear gradient of imidazole from 20 mM–500 mM. Fractions were

R. Atkinson et al. / Journal of Virological Methods 236 (2016) 170–177

analysed on a Coomassie-stained SDS gel and by western blotting as described in 2.2. Fractions 5–8 were pooled for quantitation and tested in an indirect ELISA as a binding antigen. 2.7. Quantitation of NP The concentration of purified NP was measured by gel densitometry using a bovine serum albumin (BSA; Roche, Germany) standard curve derived from serially diluted known concentrations of BSA ranging from 1.56 to 25 mg and separated on a SDS-polyacrylamide gel stained with 0.1% Coomassie blue. The concentration of purified NP was calculated using GeneTools V3.07.03 on a Gene Genius Bioimaging System (Syngene). 2.8. Serum samples To determine if the NP antigen was biologically reactive and able to detect IgG antibody against CCHFV, the antigen was reacted against known positive and negative sera in an ELISA format. Serum samples were collected from patients with a previous history of CCHFV infection confirmed at the time of acute illness by the Centre for Emerging and Zoonotic Disease at the National Institute for Communicable Diseases in Johannesburg using RT-PCR, virus isolation, detection of IgM or seroconversion. Samples for this study were collected 5–15 years after acute infection and each serum sample was re-tested for IgG antibody against CCHFV on submission using the Crimean-Congo fever virus Mosaic 2 (IgG) kit (EUROIMMUN AG, Lübeck, Germany). Each slide contains ten fields and each field contains three BIOCHIPS, with either non-transfected cells or coated with cells expressing the CCHFV NP or the CCHFV GP. Sera were tested for anti-CCHFV IgG according to manufacturers’ instructions. The CCHF NP is the most conserved protein and strains of CCHF from geographically distinct regions are antigenically and serologically cross reactive (Rangunwala et al., 2014). Serum samples from 13 volunteers with no history of infection with CCHFV and tested negative with IFA slides were used as a negative control panel. A positive serum sample designated C++ was used as a high positive control. The positivity of this control was confirmed by IFA using the CCHF Mosaic 2 (IgG) kit. Informed consent was obtained from the volunteers (UFS Ethics number 152/06). 2.9. Preparation of mock antigen Mock antigen was prepared by infiltration of leaves with infiltration medium only, and processing leaves of similar mass as used for the scaled-up infiltration process of NP. Crude extract was processed in the same manner as that for NP purification, fractionating the same 40–80% (NH4 )2 SO4 fraction by affinity chromatography and pooling the same fractions collected as for the plant-produced NP. 2.10. Indirect enzyme linked immunosorbent assay (ELISA) Plant-produced NP and mock antigen were used to develop an indirect ELISA for detection of IgG antibody directed against CCHFV. Checkerboard titrations of each reagent were used to optimise the ELISA protocol. Unless stated otherwise, all volumes were 100 ␮l/well, plates were incubated for one hour at 37 ◦ C statically and washed for 3 × 15 s with 0.1% Tween 20 in phosphate-buffered saline (PBS) pH 7.0. Briefly, purified recombinant CCHFV NP antigen (0.172 mg/ml) was diluted 1:400 in PBS, and mock antigen diluted similarly, were used to coat a 96-well PolySorp microtitre plate (Nunc, Germany) overnight at 4 ◦ C. Plates were washed and blocked using 200 ␮l/well 10% skimmed milk/PBS. Serum samples were added at a dilution of 1:100 in 2% skimmed milk/PBS to wells coated with recombinant CCHFV NP and mock antigen.

173

Positive reactors were detected using anti-human IgG horseradish peroxidase (HRPO) conjugate diluted 1:4000 (Zymed, USA) and 2,2 -Azino di-ethyl-benzothiazoline-sulfonic acid peroxidase substrate (ABTS) (KPL, USA). Plates were incubated for 30 min at room temperature (22–24 ◦ C) in the dark and read after a specified time. Optical density (OD) values were read at 405 nm using a Tecan Sunrise spectrophotometer (Tecan UK) with XReadPlus version 4.3 software. The net OD405 for each serum tested was calculated by subtracting the OD405 reading of the mock antigen from the cognate reading of that obtained using the plant-produced NP to account for any background absorbance in four separate experiments. OD values are absolute measurements that are influenced by variables such as temperature. In order to account for variability, results can be expressed as a function of the reactivity of control samples included in each run. Therefore, mean net OD405 readings were converted to a percentage of high-positive (C++) control serum (PP) value using the following equation: (mean net OD of test sample/mean net OD of C++) × 100. The indirect ELISA cut-off value was determined by calculating the mean net OD plus 2 standard deviations derived from PP values determined using the panel of 13 negative sera and tested in four separate experiments. 3. Results 3.1. NP codon-optimisation and subcloning into plant expression vectors Codon optimisation of NP resulted in the improvement of improve the codon adaptation index (CAI) from 0.37 to 0.87. Sixtyfour percent of the codons fell within the highest usage frequency of N. benthamiana resulting in a reduction of the GC content of the gene from 47.72% to 40.40%. NP was successfully subcloned directly into the plant expression vectors pRIC3.0-HT, pTRAc-HT and pEAQ-HT to generate the constructs pRIC-NP(HT), pTRAc-NP(HT) and pEAQ-NP(HT) and successfully transformed into their respective A. tumefaciens host cells for infiltration. 3.2. Small scale expression of CCHFV NP in N. benthamiana An expression time-trial was set up to compare the expression levels of NP over the course of 7 days between the 3 expression vectors tested. In addition, the effect of co-infiltration of a post translational silencing suppressor with the pRIC-NP(HT) and pTRAc-N(HT) clones (in this case pBIN-NSs (Prins et al., 1996)) on expression levels was compared. The pEAQ-HT vector contains its own p19 silencing suppressor (Sainsbury et al., 2009), therefore co-infiltration with pBIN-NSs was not necessary when infiltrating with pEAQ-N(HT). Levels of NP in crude leaf extracts were assessed qualitatively by the detection of a 55 kDa-sized protein band on a western blot probed with anti–6× his antibody. 3.3. Large-scale expression and purification of CCHFV NP in N. benthamiana Leaves infiltrated with pRIC-NP(HT) showed expression of NP from day 1 through to 7 days post infiltration (dpi) (Fig. 1a) with the highest protein levels being observed on day 3 (lane 4) as depicted by the darkest band. For each day sampled, crude extracts from leaves co-infiltrated with the silencing suppressor showed higher levels of expressed protein than those not co-infiltrated with the silencing suppressor as observed by a darker band. Leaves infiltrated with pTRAc-NP(HT) (Fig. 1b) showed the highest expression levels of NP occurring from 3 to 5 dpi (Lanes 3 to 6). Prior to day 3 and after day 7, expression levels were much lower. For each day

174

R. Atkinson et al. / Journal of Virological Methods 236 (2016) 170–177

pTRAc-NP(HT)

pRIC-NP(HT) dpi NSs

1 -

3 +

-

5 +

-

7 +

-ve

-

M

1

+

-

3 +

-

5 +

-

7 +

-

-ve

M

+

75 kDa

50 kDa

25 kDa

a

b

pEAQ-NP(HT) dpi

M

1

3

5

7

-ve

75 kDa 50 kDa

25 kDa

c Fig. 1. Western blots on crude leaf extracts infiltrated with pRIC-NP(HT) (a), pTRAc-NP(HT) (b) or pEAQ-NP(HT) (c) constructs. Crude extracts from leaves sampled at 1, 3, 5 and 7 days post infiltration (dpi) were separated by PAGE and detected by western blot using a 6 × his antibody. Constructs pRIC-NP(HT) and pTRAc-NP(HT) were infiltrated with (+) or without (−) a silencing suppressor (NSs). Leaves infiltrated with infiltration medium only, served as the negative control (−ve). M − 6 × his tag protein ladder.

sampled, there was not much difference between the intensity of the 55 kDa bands. NP protein expression using pEAQ-NP(HT) was barely discernible with very faint bands visible in lanes 2–4 (Fig. 1c). In the case of all 3 expression constructs screened, the negative control (leaves infiltrated with buffer only) showed no NP-specific protein expressed. Repeat experiments comparing pRIC-NP(HT) and pTRAc-NP(HT) expression together with silencing suppressors showed that pRICNP(HT) co-infiltrated with pBIN-NSs and leaves harvested at 3 dpi consistently produced the higher levels of N protein (data not shown) and consequently this construct together with the silencing suppressor constructs and conditions were used for scaled-up production and purification of NP. Leaves from twenty whole N. benthamiana plants (296 g) vacuum-infiltrated with pRIC-NP(HT) and pBIN-NSs. Ammonium sulphate precipitation of crude leaf extracts resulted in the majority of NP being detected in the fractions ranging from 40 to 80% (NH4 )2 SO4 as seen in a western blot probed with anti 6× his antibody (Fig. 2 – boxed region). NP was successfully purified from the 40 to 80% (NH4 )2 SO4 precipitated fraction on an affinity column with the elution profile showing a 280 nm protein absorbance peak across fractions 5–8 (Fig. 3). Under optimal buffer conditions the NP adsorbed as expected to the affinity column, as determined by a lack of NP detected on a western blot probed with anti- 6 × his antibody in the flow through fraction (Fig. 4a; lane 3–flow through) compared to the total amount loaded onto the column (Fig. 4a; lane 2–40-80%). The majority of the NP protein was eluted over a concentration of 150–300 mM imidazole in 4 fractions spanning the 280 nm protein peak (Fig. 4a; fractions 5–8). A cognate Coomassie-stained gel (Fig. 4b) showed that the protein across the peak was very pure compared with what was originally loaded onto the column.

Fig. 2. Western blot of (NH4 )2 SO4 -precipitated fractions. The boxed region indicates the 40–80% fractions which were used for affinity purification. M − 6 × his tag protein ladder.

3.4. Protein quantification The concentration of pooled affinity-purified NP (fractions 5–8) was calculated by gel densitometry and the final concentration of purified NP calculated to be 0.17 mg/ml. This was calculated to be a final yield of 1.6 mg NP/kg of fresh leaf material harvested. 3.5. Detection of antibody against plant-produced NP in human sera A total of 13/13 serum samples collected from survivors were confirmed IgG positive and 13/13 sera from volunteers were confirmed IgG negative against CCHFV using the commercial IFA. The

R. Atkinson et al. / Journal of Virological Methods 236 (2016) 170–177

175

Fraction 8

Fraction 7

Fraction 6

Fraction 5

Fraction 4

Fraction 3

Flow through

M

40-80%

Fraction 8

Fraction 7

Fraction 6

Fraction 5

Fraction 4

Fraction 3

Flow through

M

40-80%

Fig. 3. Purification of NP using affinity chromatography. The chromatographic trace shows NP elution from the HisTrapTM HP prepacked column detected as absorbance at 280 nm (primary y axis) with increasing imidazole concentration on the secondary y axis (20–500 mM).

100 kDa 70 kDa 55 kDa 40 kDa 35 kDa

a

b

Fig. 4. Detection of fractionated NP by western blot and cognate Coomassie-stained SDS polyacrylamide gel. a) Western blot of the 40–80% (NH4 )2 SO4 -precipitated fraction loaded onto the affinity column compared with unbound (flowthrough) protein and fractions 3–8. NP protein was detected with 6 × his antibody. b) Cognate Coomassie-stained polyacrylamide gel. Boxed region indicates purified NP which was pooled for quantitation and testing in ELISA.

Fig. 5. Detection of anti-CCHFV IgG antibody in human sera by ELISA using plant-produced NP. Samples 1–13 represent PP values calculated for sera from positive reactors and 14–26 represent PP values calculated for sera from negative reactors. Error bars indicate ± mean standard deviation calculated for each PP value.

plant-produced affinity-purified NP (0.17 mg/ml) was used as an antigen in an indirect ELISA assay to determine its ability to detect anti-CCHFV IgG antibodies in human serum from patients previously infected with CCHFV. Checker-board titrations were used to optimise the ELISA and a 1/400 dilution of NP was found to be opti-

mal (ie the point at which negative values remain below 0.2 but positive values do not become saturated after 30 min incubation with ABTS) and all further work was carried out at this concentration. The recombinant antigen was able to detect IgG antibody in 13/13 sera from convalescent patients. The results of indirect ELISA

176

R. Atkinson et al. / Journal of Virological Methods 236 (2016) 170–177

of 13 sera from previously infected patients and 13 anti-CCHFV IgG negative serum samples are shown in Fig. 5 as mean PP values for each serum. The plant-produced NP was able to detect anti-CCHFV IgG in all positive serum samples () as indicated by PP values above those obtained for the negative serum samples. A cut-off value of 25.03% was calculated to separate positive results from negative results. All negative serum samples (䊏) had PP values below the cut-off value.

4. Discussion ELISA is a frequently used method for detecting anti-CCHFV IgG in humans for diagnosis and in animals for surveillance purposes (Vanhomwegen et al., 2012). This method utilises CCHFV antigen that is prepared from cell cultures which have been infected with live CCHFV or from brain tissue of suckling mice which have been inoculated with live CCHFV. However, the preparation of these reagents requires a Biosafety level 4 (BSL-4) facility, which severely limits the number of laboratories worldwide that can do this work. In developing countries where CCHFV is prevalent, and where monitoring and diagnosis of CCHFV is therefore very important, both surveillance and rapid diagnosis are hindered by the lack of sophisticated facilities to carry out these tests. The availability of a safer recombinant NP would facilitate diagnosis and surveillance in such countries. In addition, recombinant systems are usually cheaper for producing protein than virus cultures. There are a few examples of recombinant NP production in E. coli and insect cells (Saijo et al., 2010; Samudzi et al., 2012; Zhou et al., 2011), but these are still reputed to more costly processes of production, and stand the risk of contamination. Plants have been indicated as being a more cost-effective and scalable platform in which to produce such proteins (Chen et al., 2013) which run the risk of lower possibilities of contamination and are favourable as they are not composed of animal-derived products. In this work, we reported on the use of an Agrobacteriummediated transient expression system in N. benthamiana to produce recombinant NP protein. It was then purified and tested to determine whether it could be used to detect anti-CCHFV IgG antibodies in human serum. The expression of NP was compared using three different plant expression vectors to determine which one would yield the most NP. It has been shown in previous work in our laboratory that the levels of recombinant protein expression in plants vary depending on the plant expression vector used, and are influenced by the type of protein itself; therefore, expression levels have to be empirically determined for each protein-vector combination. We showed that expression using the pRIC3.0-HT vector was slightly better for NP production in that expression was visible over 7 dpi, compared to the pTRAc-HT vector where expression was most predominantly visible over 3–5 dpi. Expression using the pEAQ-HT vector was barely discernible when harvested at any time post infiltration. The presence of a silencing suppressor when using pRIC-NP(HT) qualitatively increased the levels of NP protein slightly but did not seem to make a difference when co-expressed with pTRAc-NP(HT). This is not unexpected as silencing suppressors should minimise the degradation of ‘foreign’ nucleic acid within the host cell, thereby increasing the potential for recombinant protein expression. Thus, the combination of pRIC-NP(HT) co-infiltrated with pBIN-NSs was selected for use in the scaled-up production of NP. Ammonium sulphate precipitation served as a very efficient method for partial purification of NP, particularly as it allowed separation of NP from the RuBisCO component of the crude plant extract (in the 0–40% (NH4 )2 SO4 fraction), which is a common contaminant of plant-produced proteins. Nickel affinity chromatography of the 40–80% (NH4 )2 SO4 fraction was very successful in purifying NP as

there were very few contaminating protein bands detected on a Coomassie-stained gel in the fractions spanning the protein peak. The yield of protein was calculated to be 1.6 mg/kg fresh leaf weight. The purified NP was able to bind IgG antibodies specific for CCHFV in sera from convalescent patients and distinguish between sera that were positive and negative for anti-CCHFV IgG antibodies. Within the cohort of samples tested, 13/13 were IgG antibody positive with no false positive reactions or false negative reactions determined. However a limitation of this study is the small sample number and the lack of cross reactivity studies against related viruses. The sample number tested was very small for assessment of the sensitivity of NP for diagnostic use and a larger panel of samples, including samples collected during the acute phase of illness, will have to be used for its validation as a diagnostic tool. In addition, specificity, which would require testing the NP antigen for cross reactivity against IgG antibody against related viruses will have to be validated if the plant-produced NP is to be used for diagnostic purposes or as a surveillance tool. Diagnosis of CCHFV infections during the acute phase is based on isolation of the virus or amplification of viral RNA. Patients that survive the infection all have a demonstrable IgG and IgM antibody response from days 7 to 9 after the onset of illness and in some instances it can be detected earlier (Burt et al., 1994; Shepherd et al., 1988). Antibody detection is usually performed by ELISA or immunofluorescent assays. Recombinant protein technology is a useful and cost effective platform for preparation of safe antigens and hence the plant recombinant NP could have application in diagnosis or surveillance. The recombinant antigen was able to bind IgG antibodies in all sera tested by ELISA and further studies using larger panels of both convalescent and negative samples to obtain validation data are justified. However, these results are preliminary and show a trend that is promising for use in diagnosis and surveillance. 5. Conclusions This study confirmed that recombinant CCHFV NP can be produced in plants, and purified in sufficient amounts to use in an ELISA. We showed that plant-produced NP is functionally active and was able to bind anti-CCHFV antibodies. Further work would involve testing of the NP using a larger panel of positive and negative sera, and developing an ELISA to determine whether the NP can detect IgM antibodies and thus have potential as an early diagnostic tool. Acknowledgements The authors would like to thank Brandon Weber (EMU, UCT) for assistance with operating the ÄktaExplorer chromatography system, George Lomonossoff (John Innes Centre, Norwich, UK) for providing the pEAQ-HT vector, Rainer Fischer for providing the pTRAc-HT vector and Marcel Prins (Wageningen) for providing pBIN-NSs. This research was supported by the Poliomyelitis Research Foundation (PRF), South Africa, Grant No.12/15. RA was supported by funding from the National Research Foundation, the Poliomyelitis Research Foundation (PRF) and the UCT Postgraduate Funding Office. The authors of this manuscript have no conflicts of interest. References Bente, D.A., Alimonti, J.B., Shieh, W.J., Camus, G., Stroher, U., Zaki, S., Jones, S.M., 2010. Pathogenesis and immune response of crimean-Congo hemorrhagic fever virus in a STAT-1 knockout mouse model. J. Virol. 84, 11089–11100. Bente, D.A., Forrester, N.L., Watts, D.M., McAuley, A.J., Whitehouse, C.A., Bray, M., 2013. Crimean-Congo hemorrhagic fever: history, epidemiology, pathogenesis, clinical syndrome and genetic diversity. Antiviral Res. 100, 159–189.

R. Atkinson et al. / Journal of Virological Methods 236 (2016) 170–177 Bereczky, S., Lindegren, G., Karlberg, H., Akerstrom, S., Klingstrom, J., Mirazimi, A., 2010. Crimean-Congo hemorrhagic fever virus infection is lethal for adult type I interferon receptor-knockout mice. J. Gen. Virol. 91, 1473–1477. Burt, F.J., Leman, P.A., Abbott, J.C., Swanepoel, R., 1994. Serodiagnosis of Crimean-Congo haemorrhagic fever. Epidemiol. Infect. 113, 551–562. Chen, Q., Lai, H., Hurtado, J., Stahnke, J., Leuzinger, K., Dent, M., 2013. Agroinfiltration as an effective and scalable strategy of gene delivery for production of pharmaceutical proteins. Adv. Tech. Biol. Med. 1, 103. Deyde, V.M., Khristova, M.L., Rollin, P.E., Ksiazek, T.G., Nichol, S.T., 2006. Crimean-Congo hemorrhagic fever virus genomics and global diversity. J. Virol. 80, 8834–8842. Englard, S., Seifter, S., 1990. 22] precipitation techniques. In: Murray, P.D. (Ed.), Methods in Enzymology. Academic Press, pp. 285–300. Ergonul, O., 2006. Crimean-Congo haemorrhagic fever. Lancet Infect. Dis. 6, 203–214. Hoogstraal, H., 1979. The epidemiology of tick-borne Crimean-Congo hemorrhagic fever in Asia, Europe and Africa. J. Med. Entomol. 15, 307–417. Kivaria, F.M., 2012. The shifting landscape of tick-borne diseases and the effects of global change. Vet. J. 193, 311–312. Leblebicioglu, H., 2010. Crimean-Congo haemorrhagic fever in Eurasia. Int. J. Antimicrob. Agents 36, S43–S46. Maclean, J., Koekemoer, M., Olivier, A.J., Stewart, D., Hitzeroth, I.I., Rademacher, T., Fischer, R., Williamson, A.L., Rybicki, E.P., 2007. Optimization of human papillomavirus type 16 (HPV-16) L1 expression in plants: comparison of the suitability of different HPV-16 L1 gene variants and different cell-compartment localization. J. Gen. Virol. 88, 1460–1469. Maltezou, H.C., Papa, A., 2011. Crimean-Congo hemorrhagic fever: epidemiological trends and controversies in treatment. BMC Med. 9. Marriott, A.C., Polyzoni, T., Antoniadis, A., Nuttall, P.A., 1994. Detection of human-antibodies to Crimean-Congo hemorrhagic-fever virus using expressed viral nucleocapsid protein. J. Gen. Virol. 75, 2157–2161. Prins, M., Resende Rde, O., Anker, C., van Schepen, A., de Haan, P., Goldbach, R., 1996. Engineered RNA-mediated resistance to tomato spotted wilt virus is sequence specific. Mol. Plant Microbe Interact. 9, 416–418. Rangunwala, A., Samudzi, R.R., Burt, F.J., 2014. Detection of IgG antibody against Crimean-Congo haemorrhagic fever virus using ELISA with recombinant nucleoprotein antigens from genetically diverse strains. Epidemiol. Infect. 142, 2147–2154. Regnard, G.L., Halley-Stott, R.P., Tanzer, F.L., Hitzeroth, I.I., Rybicki, E.P., 2010. High level protein expression in plants through the use of a novel autonomously replicating geminivirus shuttle vector. Plant Biotechnol. J. 8, 38–46.

177

Saijo, M., Qing, T., Niikura, M., Maeda, A., Ikegami, T., Prehaud, C., Kurane, I., Morikawa, S., 2002. Recombinant nucleoprotein-based enzyme-linked immunosorbent assay for detection of immunoglobulin G antibodies to Crimean-Congo hemorrhagic fever virus. J. Clin. Microbiol. 40, 1587–1591. Saijo, H., Mori, Y., Fujihara, H., Kanno, Y., Chikazu, D., Ohkubo, K., Hikiji, H., Iino, M., Yonehara, Y., Takato, T., 2010. Evaluation and analysis of formation of bone at the palate in patients with cleft lip and palate after palatoplasty based on computed tomograms and three-dimensional data. J. Plast. Surg. Hand Surg. 44, 21–25. Sainsbury, F., Thuenemann, E.C., Lomonossoff, G.P., 2009. pEAQ: versatile expression vectors for easy and quick transient expression of heterologous proteins in plants. Plant Biotechnol. 7, 682–693. Samudzi, R.R., Leman, P.A., Paweska, J.T., Swanepoel, R., Burt, F.J., 2012. Bacterial expression of Crimean-Congo hemorrhagic fever virus nucleoprotein and its evaluation as a diagnostic reagent in an indirect ELISA. J. Virol. Methods 179, 70–76. Shen, W.J., Forde, B.G., 1989. Efficient transformation of agrobacterium spp by high-voltage electroporation. Nucleic Acids Res. 17, 8385–8385. Shepherd, A.J., Swanepoel, R., Gill, D.E., 1988. Evaluation of enzyme-linked immunosorbent assay and reversed passive hemagglutination for detection of Crimean-Congo haemorrhagic fever virus antigen. J. Clin. Microbiol. 26, 347–353. Swanepoel, R., Gill, D.E., Shepherd, A.J., Leman, P.A., Mynhardt, J.H., Harvey, S., 1989. The clinical pathology of Crimean-Congo hemorrhagic-fever. Rev. Infect. Dis. 11, S794–S800. Vanhomwegen, J., Alves, M.J., Zupanc, T.A., Bino, S., Chinikar, S., Karlberg, H., Korukluoglu, G., Korva, M., Mardani, M., Mirazimi, A., Mousavi, M., Papa, A., Saksida, A., Sharifi-Mood, B., Sidira, P., Tsergouli, K., Wolfel, R., Zeller, H., Dubois, P., 2012. Diagnostic assays for Crimean-Congo hemorrhagic fever. Emerg. Infect. Dis. 18, 1958–1965. Voinnet, O., Rivas, S., Mestre, P., Baulcombe, D., 2003. An enhanced transient expression system in plants based on suppression of gene silencing by the p19 protein of tomato bushy stunt virus (Retracted Article). Plant J. 33, 949–956. Whitehouse, C.A., 2004. Crimean-Congo hemorrhagic fever. Antiviral Res. 64, 145–160. Zhou, Z.-r., Wang, M.-l., Deng, F., Li, T.-x., Hu, Z.-h., Wang, H.-l., 2011. Production of CCHF virus-like particle by a baculovirus-insect cell expression system. Virol. Sin. 26, 338–346.