Construction and Characterization of the NS3 Protein of Bluetongue Virus as A Bait for Application in Yeast Two-hybrid System

Bluetongue (BT) is an infectious and devastating viral disease of domestic and wild ruminants. The disease poses significant losses to livestock health and the agricultural economy due to high morbidity, mortality, abortion, teratogenicity, reduced performance and restrictions on livestock trade (Sperlova and Zendulkova, 2011; Saminathan et al., 2020). It is primarily an arthropod-transmitted disease, other modes of transmission can occur rarely (Mellor, 1990; Meiswinkel et al., 2008; Belbis et al., 2013; Coetzee et al., 2012). BT has been reported from several parts of the world and has become endemic in Indian subcontinents (Mellor et al., 2008; Sreenivasulu et al., 2004; Rao et al., 2016). The clinical manifestation is characterized by fever, facial edema, hemorrhages, oral mucosal ulceration and coronitis. Although the clinical presentation of BT varies widely among different ruminant hosts, sheep and white-tailed deer show the lethal form of the disease (Modrow et al., 2010; Rojas et al., 2019).
       
The etiological agent of BT is the Bluetongue virus (BTV), a non-enveloped, double-stranded RNA (dsRNA) virus with ten segments. It belongs to the genus Orbivirus within the family Reoviridae. To date, 27 distinct serotypes of BTV are reported worldwide (Alkhamis et al., 2020) along with the two putative serotypes i.e., 28 and 29 (Bumbarov et al., 2020; Yang et al., 2020). BTV genome is approximately 19.2 kb which encodes seven SPs (VP1-VP7) and five NSPs (NS1-NS5) (Belhouchet et al., 2011; Stewart et al., 2015; Ratinier et al., 2011). These SPs and NSPs play a diverse function in virus morphogenesis and pathogenicity (Schwartz-Cornil et al., 2008).
       
NS3 protein of BTV executes multiple functions within-host including virus release by viroporin formation. It is encoded by segment 10 and occurs in two isoforms i.e. NS3 and NS3A (Celma and Roy 2009; Han and Harty 2004). Albeit, the individual role of NS3A has not been identified yet, NS3/NS3A is the only membrane glycoprotein synthesized by BTV (Wu et al., 1992). The NS3 protein sequence is remarkably conserved among the various BTV strains and is considered a potent candidate for diagnostic purposes. Despite these facts, detailed studies are not available on the host-pathogenesis aspect of NS3. Thus the present study is designed to explore the interaction of NS3 protein with host proteins employing the yeast two-hybrid (Y2H) system. Y2H system is a dynamic technique to study the PPI based on a realignment of functional transcription factor (TF) during the interaction of two proteins of interest (Fields and Song 1989). cDNA library from sheep origin was used as prey to investigate the protein-protein interaction (PPI) with BTV NS3 bait.

Cells and virus
 
BTV serotype 10 maintained in the CADRAD, ICAR-Indian Veterinary Research Institute (IVRI) was revived in BHK-21 cells and dsRNA was extracted using phenol-chloroform RNA extraction method and lithium chloride precipitation. The extracted dsRNA was checked on polyacrylamide gel electrophoresis (PAGE) with silver nitrate staining for the presence of all the segments of the virus.
 
Amplification and cloning of NS3
cDNA synthesis was carried out using RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific, USA). Oligonucleotide primers targeting full-length NS3 gene were designed from the available GenBank sequences. Forward primer (GBK_S10_20F: AGATGAATTCATGCTATCCGGGC TGATC) and reverse primer (GBK_S10_709R: GACGTCGA CGTCAGGTTAATGGCATTTC) were flanked with EcoRI and SalI restriction endonucleases at 5' end respectively. Amplification of NS3 was done using DreamTaq DNA Polymerase (Thermo Scientific, USA).  PCR amplification conditions were: initial denaturation at 95°C for 2 min, followed by 35 cycles of denaturation at 95°C for 30 sec, annealing at 55°C for 30 sec, extension at 72°C for 1 min and a final extension at 72°C for 10 min. Agarose gel electrophoresis of amplified product was done and the gel was purified using QIAquick® gel extraction kit (Qiagen, Hilden, Germany). The purified amplicon and pGBKT7-BD vector was double digested with EcoRI and SalI enzymes (NEB, USA). The digested products were ligated and the recombinant plasmid was initially transformed into a competent E. coli DH5α strain. The positive recombinant clones were confirmed by colony PCR, restriction enzyme digestion and nucleotide sequencing analysis.
 
pGBKT7-NS3 bait construction and its characterization
 
The sequenced confirmed pGBKT7-NS3 plasmid (100 ng) was retransformed to competent Saccharomyces cerevisiae (S. cerevisiae) Y2H Gold cells by lithium acetate transformation method (Gietz et al., 1995). The transformation mixture was diluted to 1/10 and 1/100 dilutions and 100 µl of each dilution was spread on SD/-Trp (SDO), SD/-Trp/ X-α-gal (SDO/X) and SD/-Trp/X-α-gal/AureobasidinA (SDO/X/A) plates (Clontech, USA) and incubated at 30oC for 3-5 days. The transformation was also performed simultaneously for positive (pGBKT7-53 + pGADT7-T), negative (pGBKT7-Lam + pGADT7-T) and bait control (empty pGBKT7-BD vector) and were also plated on SDO, SDO/X, SDO/X/A and SD/-Trp/-Leu/X-α-gal /AureobasidinA (DDO/X/A). The growth of the colonies was recorded after 3-5 days and compared with that of controls to check the presence of toxicity and auto-activation.
       
For expression, a single isolated colony of the pGBKT7-NS3, pGBKT7-53 (positive control) and Y2H Gold cells (negative control) were grown to an OD₆₀₀ 0.4-0.6 with shaking (220 rpm) at 30oC.  The cultures were centrifuged at 1000x g for 5 min at 4oC, washed with 50 ml of ice-cold water and were immediately frozen in liquid nitrogen (LN2). The yeast protein extracts were prepared from the cell pellets following the TCA method according to the Yeast Protocols Handbook (Clontech). The extracted proteins were resolved on 12.5% sodium dodecyl sulfate polyacrylamide (SDS) gel and subsequently transferred to the nitrocellulose membrane (NCM). The membrane was blocked with 3% bovine serum albumin (BSA) in  Tris-buffered saline-Tween (TBST, 50 mM Tris, 150 mM NaCl, 0.05% Tween 20, pH 7.6) overnight at 4oC. Afterward, the membrane was incubated with primary mouse anti-c-myc monoclonal antibody (Clontech, USA) at 1:500 dilution for 1 hour at RT, followed by incubation with horseradish peroxidase- (HRP-) labeled goat anti-mouse IgG antibody (SantaCruz, USA) at 1:1000 dilution. The membrane was washed three times between each step with TBST and finally visualized by developing with a 3,3’-diaminobenzidine (DAB)-substrate solution.
 
Forward library screening
 
The generated pGBKT7-NS3 bait was mated with ovine cDNA library (available at MBL, IVRI) to screen the host protein interaction with NS3 protein of BTV. Briefly, the overnight grown culture of the bait in SD/-Trp medium was centrifuged and re-suspended in SD/-Trp medium to a cell density >1x108 cells/ml. The re-suspended bait culture was mixed with a 1 ml cDNA library in a 2 L sterile flask. 45 ml of 2X YPDA medium was added to it and incubated at 30oC with minimum shaking. Zygote formation was checked after 20 h under a phase-contrast microscope (40X). Thereupon the mated culture was serially diluted to 1/10, 1/100, 1/1000 and 1/10000 dilutions and 100 µl of each dilution were plated on SDO, SD/-Leu (SDO) and SD/-Leu/-Trp (DDO) agar plates. The remainder culture was plated on 150 mm (DDO/X/A)  agar plates (200 µl per plate). Plates were incubated at 30^oC for 3-5 days and mating parameters including the number of independent clones screened, the viability of prey, bait, diploids and efficiency of mating were estimated.

Construction and characterization of bait
 
BHK-21 cells have shown cytopathic effect (CPE) after 72 h of infection with BTV-10. The characteristic CPE of BTV i.e. cell rounding, cell lysis and detachment has been observed (Fig 1) (Clavijo et al., 2000). The dsRNA was extracted and all the 10 segments were visible clearly on PAGE by silver nitrate staining (Fig 2). The cDNA was synthesized and the full-length NS3 coding region of ~690 bp was amplified (Fig 3A). The NS3 amplicon was successfully double digested and cloned into the pGBKT7-BD vector as verified by colony PCR (Fig. 3B), restriction enzyme digestion (Fig 3C) and nucleotide sequencing. Upon confirmation, the positive recombinant plasmid was retransformed into competent S. cerevisiae Y2H Gold cells. The colonies of pGBKT7-NS3 and positive control were equivalent in size indicating the absence of toxicity. Further, pGBKT7-NS3 bait gave only white colonies on both SDO, SDO/X plates and no growth was seen on SDO/X/A plates indicating NS3 does not autonomously activate the reporter gene (Fig 4) (Table 1). The constructed pGBKT7-NS3 bait was well expressed in yeast and a band of expected size ~46 kDa was detected by mouse anti-c-myc monoclonal antibody (Fig 5).
 

 

 

 

 

 

       
Toxic and self-activating bait yield a high frequency of false-positive interactions. Similarly, non-expressing or over-expressing bait also results in false-positive and non-specific interactions (Serebriiskii and Golemis, 2001). Qualification of all these parameters by bait is of at most importance to minimalize the chances of false-positive interactions with prey proteins (Brückner et al., 2009). The results imply that the constructed bait has qualified all the parameters and can be used as bait in Y2H to screen its interacting host proteins.
 
Library screening for interacting proteins
 
The mating between pGBKT7-NS3 bait and prey library was performed and grew on SDO (both SD/-Trp and SD/-Leu), DDO and DDO/X/A plates. The viability of prey, bait and mating diploids was calculated from the colonies developed on SDO and DDO (Table 2). Approximately two million clones were screened and a mating efficiency of 3% was achieved. However, no blue colonies grew on DDO/X/A plates indicating the absence of an interacting proteins. The experiment was repeated but, similar results were obtained. As there are only two variables, bait and prey, to verify which one failed to perform, the mating was conducted using BTV NS4 bait since changing the bait is much easier and cost effective than changing the prey library (Brückner et al., 2009). The mating between BTV NS4 bait and the same library resulted in several interaction hits, indicating that the problem was with the NS3 bait only. In Y2H some full-length proteins do not yield interactions despite good expression as bait or prey. Full-length proteins fail to perform might be due to incorrect folding in yeast, precluding interaction with its partner, or due to their biology (Galletta and Rusan, 2015). On thorough analysis, we observed the presence of two transmembrane domains (TM) in our NS3 sequence. The position of TM domains were located from amino acids 122 to 139 and amino acids 166 to 182 using Phobius (Käll et al., 2004) and TMPred software (Hofmann and Stoffel, 1993) (Fig 6).
 

 

         
Previously, it has been reported that the expression of full-length NS3 protein is difficult in the prokaryotic system. Mutation analysis study has confirmed that the TM domains are responsible to produce this cytotoxicity in prokaryotic hosts (Chacko et al., 2015; Mohanty et al., 2016). The membrane association and glycosylation of NS3 protein of Ibaraki virus, a member of orbivirus genus have also been documented to have a cytotoxic effect (Urata et al., 2016). In addition to this, TM domain carrying NS3 protein of African horse sickness virus (AHSV), another orbivirus found to produce cytotoxic effect on protein expression in the baculovirus system (van Staden et al., 1995; van Niekerk et al., 2001). NS3 protein from orbiviruses has been reported to produce cytotoxicity in mammalian and insects cells too (van Staden et al., 1995; Wirblich et al., 2006). For producing cytotoxicity NS3 requires the membrane association as well as the presence of the transmembrane domains (Han and Harty, 2004). Literature suggests that the structure and distribution of the TM domains within NS3 protein control the expression of cytotoxicity it produce. Thus it can be hypothesized that the TM domains in the full-length NS3 protein were interfering with PPI studies in the Y2H system. Mutations causing alterations in the TM domains or deletion of these TM domains reduce the cytotoxicity of the protein (van Niekerk et al., 2001; Huismans et al., 2004; Mohanty et al., 2016). The studies available on BTV NS3-host PPI using Y2H system has reported the p11, a light chain component of the Calpactin complex and the serine/threonine-protein kinase B33 Raf (BRAF) as an interacting cellular factors (Beaton et al., 2002; Kundlacz et al., 2019). Among these, Beaton et al., (2002) has mentioned the use of the truncated portions of NS3 only for Y2H study. The presence of TM domains in our bait sequence may also be producing some cytotoxicity during interaction with prey proteins and may be the reason for not getting the interactions upon mating. Thus, we redesigned our experiment to generate N-terminus and C-terminus truncated NS3 excluding the TM domains and we are in the process of performing it.

We here report the construction and characterization of pGBKT7-NS3 bait to study the BTV NS3 PPI. pGBKT7-NS3 bait was found to be well expressed and suitable to screen host-virus interaction. However, we could not found any interaction with the full-length NS3 protein owing to the presence of TM domains. Thus we hypothesize that the presence of TM domains interfere with the transcription of reporters in the Y2H system. For the future, we redesigned our study to include truncated NS3 as a bait, eliminating TM domains to screen host-virus PPI.

The authors are grateful to the Indian Council of Agricultural Research and SERB for providing funding.
 
Funding information
 
This work was funded by the Science and Engineering Research Board (SERB) under the project ECR/2017/001106.

The authors declare no conflict of interest.