Biologicals 42 (2014) 339e345

Contents lists available at ScienceDirect

Biologicals journal homepage: www.elsevier.com/locate/biologicals

Production and characterization of single-chain antibody (scFv) against 3ABC non-structural protein in Escherichia coli for sero-diagnosis of Foot and Mouth Disease virus Gaurav K. Sharma, Sonalika Mahajan, Rakesh Matura, Saravanan Subramaniam, Jajati K. Mohapatra, Bramhadev Pattnaik* Project Directorate on Foot and Mouth Disease, Indian Council of Agricultural Research, IVRI Campus, Mukteswar, Uttarakhand 263138, India

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 May 2014 Received in revised form 7 July 2014 Accepted 19 August 2014 Available online 25 October 2014

Differentiation of Foot-and-Mouth Disease infected from vaccinated animals is essential for effective implementation of vaccination based control programme. Detection of antibodies against 3ABC nonstructural protein of FMD virus by immunodiagnostic assays provides reliable indication of FMD infection. Sero-monitoring of FMD in the large country like India is a big task where thousands of serum samples are annually screened. Currently, monoclonal or polyclonal antibodies are widely used in these immunodiagnostic assays. Considering the large population of livestock in the country, an economical and replenishable alternative of these antibodies was required. In this study, specific short chain variable fragment (scFv) antibody against 3B region of 3ABC poly-protein was developed. High level of scFv expression in Escherichia coli system was obtained by careful optimization in four different strains. Two formats of enzyme immunoassays (sandwich and competitive ELISAs) were optimized using scFv with objective to differentiate FMD infected among the vaccinated population. The assays were statistically validated by testing 2150 serum samples. Diagnostic sensitivity/specificity of sandwich and competitive ELISAs were determined by ROC method as 92.2%/95.5% and 89.5%/93.5%, respectively. This study demonstrated that scFv is a suitable alternate for immunodiagnosis of FMD on large scale. © 2014 The International Alliance for Biological Standardization. Published by Elsevier Ltd. All rights reserved.

Keywords: scFv 3ABC Foot-and-mouth disease DIVA ROC

1. Introduction Foot-and-Mouth Disease (FMD) is a highly contagious and economically one of the most important disease of transboundary importance. FMD is caused by infection with FMD virus (FMDV) belonging to the genus Aphthovirus of family Picornaviridae [1]. FMDV has single stranded RNA genome of about 8000 bases size. The viral genome encodes 4 structural and 10 non-structural proteins (NSPs) through a single open reading frame. Individual viral proteins are further processed by the virus encoded proteases [2]. NSPs are not part of inactivated virus vaccine formulations, hence detection of antibodies against the NSPs indicates infection and is basis of differentiation of infected from vaccinated animals (DIVA) [3]. Among all the NSPs of the FMDV, detection of antibodies

* Corresponding author. Tel.: þ91 5942286004; fax: þ91 5942286307. E-mail address: [email protected] (B. Pattnaik).

directed against 3ABC region of the virus has been considered as the most reliable indicator of FMDV infection [3e7]. Furthermore, detection of antibodies against 3B region of the 3ABC polyprotein was shown to differentiate FMD infected from the vaccinated animals with very high specificity, even in the animals received multiple doses of vaccine [8]. Currently, monoclonal and polyclonal antibodies against these NSPs have been widely used for FMD diagnosis. Recombinant antibodies produced in prokaryotic system have emerged as an economical alternate to the conventional monoclonal and polyclonal antibodies. Recombinant antibodies have distinct advantages such as batch uniformity and economical large scale production [9]. In this study, single-chain variable fragment antibody (scFv) against 3B region of 3ABC polyprotein of FMDV was developed in Escherichia coli. Four different E. coli strains were evaluated to obtain high level of scFv expression. Purified scFv was refolded to its native conformation which enhanced the reactivity. The developed scFv was characterized and evaluated for use in DIVA compatible enzyme linked immunosorbent assays (ELISAs).

http://dx.doi.org/10.1016/j.biologicals.2014.08.005 1045-1056/© 2014 The International Alliance for Biological Standardization. Published by Elsevier Ltd. All rights reserved.

340

G.K. Sharma et al. / Biologicals 42 (2014) 339e345

2. Materials and methods 2.1. Recombinant 3ABC antigen and monoclonal antibody secreting clones The clone developed earlier [10] for expression of full length mutated 3ABC poly-protein in E. coli was used for the production of 3ABC antigen. Histidine tag fused 3ABC protein was purified under denaturating conditions by immobilized metal affinity chromatography using a commercial kit (Merck, Millipore, NJ, USA). Purified 3ABC protein was lyophilized after addition of 100 mM trehalose and protease cocktail inhibitor set III (Calbiochem, Merck, USA). A hybridoma clone (3B3A5) secreting monoclonal antibody against 3B region of 3ABC poly-protein was selected for the development of scFv. The 3B recombinant protein developed earlier [8] was used to determine the specific reactivity of 3B3A5 monoclonal antibody. 2.2. Isolation of mRNA and cDNA synthesis Monoclonal antibody secreting hybridoma clone (3B3A5) was grown in Hybridoma SFM media (Invitrogen, CA, USA). On appearance of full lawn (after 48 h) cells were harvested by scrapping with cell scrapper followed by pelleting at low speed centrifugation. Media supernatant was discarded and cells were washed twice with 10 ml sterile PBS. Hybridoma cells were reconstituted with 1 ml of PBS. mRNA was extracted from the separated hybridoma cells using Oligotex Direct mRNA extraction kit (Qiagen, Hilden, Germany) as per the recommended protocol. Extracted mRNA was quantified by nanodrop spectrophotometer and approximately 1 mg of mRNA was reverse transcribed using Oligo(dT) primer and Superscript-II reverse transcriptase enzyme (Invitrogen, CA, USA) in 25 ml reaction volume. 2.3. Amplification of variable segments of immunoglobulin genes Commercially available sets of degenerate primers (Progen, Heidelberg, Germany) were used for in-vitro amplification of mouse IgG heavy and light chains. Thirteen polymerase chain reactions were assembled for amplification of mouse monoclonal IgG immunoglobulin variable heavy (VH) chain using 13 forward degenerate primers (1Ae1L) and a common reverse primer (1 M). Similarly, 12 reactions were assembled for amplification of variable kappa light chain (VL) using 12 forward degenerate primers (1Ne1W, 1Y) and a common reverse primer (1X). Briefly, in 20 ml reaction volume, 2 ml of 10x Platinum HiFi Taq DNA polymerase reaction buffer (Invitrogen, CA, USA), 1 ml of 10 mM of each of dNTP's, 1 ml of (10 pmol/ml) of forward and reverse primers, 1 unit of Platinum HiFi Taq DNA polymerase enzyme and 100 ng of cDNA was mixed in thin walled PCR tubes. In vitro amplification was performed in a thermocycler (Applied Biosystems, USA) following thermal profile of initial denaturation of 95  C for 15 min and 35 cycles of denaturation of 95  C for 30 s; annealing at 55  C for 30 s and extension at 72  C for 30 s. Final extension was performed at 72  C for 10 min. Amplified 380 to 400 bp was gel purified using Qiaquick gel extraction kit (Qiagen, Hilden Germany) and subjected to an analogous second PCR using the second set of primers to introduce the restriction endonuclease sites for restriction enzymes (RE) Mlu1/Not1 and Nco1/HindIII for VH and VL genes, respectively. 2.4. Construction of VH-VL-pOPE101 plasmid clone Approximately 1 mg each of amplified VH gene and pOPE101 expression vector (Progen, Heidelberg, Germany) were digested with MluI and NotI restriction enzymes (Thermo Scientific, MA,

USA) followed by gel purification as described above. Gel purified VH and pOPE101 vector were ligated in 1:3 molar ratio using T4 DNA Ligation Kit (Novagen, Merck, USA). Ligation mixture was chemically transformed into the competent XL-1 blue cells [11] and transformed colonies were plated on the LB-ampicillin agar plates. The recombinant VH-pOPE101 colonies were screened by amplification of the insert by PCR or by RE digestion. The confirmed VHpOPE101 plasmid and VL amplicon were digested with Nco1 and HindIII restriction enzymes followed by ligation. The VH-VLpOPE101 plasmid construct was transformed into the chemically competent XL-1 blue cells. Colonies positive for VH-VL-pOPE101 plasmid were screened by colony PCR or by RE digestion. 2.5. Sequence analysis of VH and VL genes Nucleotide sequences of the cloned VH and VL genes were determined using VH forward and VL reverse primers in ABI 3130 capillary sequencer. Sequences were analyzed by DNA star software (Lasergene v10.2) and submitted to the NCBI Gene Bank. The VH gene, VL gene, linker and signal sequence were determined by BLAST tool from NCBI. Sequences were analyzed using online abYsis system (http://www.bioinf.org.uk/abysis/) following Chothia numbering scheme for identification of complementary determining and frame work regions along with insertions and presence of un-usual amino acid(s). 2.6. Optimization of soluble scFv expression Four E. coli strains viz; XL-1 blue, BL21(DE3)pLysS, BL21(DE3) pLacI, and Tuner(DE3)pLysS (Novagen, Darmstadt, Germany) were evaluated with aim to enhance the level of expression of scFv. The recombinant VH-VL-pOPE101 plasmid extracted from XL-1 blue cells was transformed into the other three E. coli strains. Positively transformed clones were screened by colony PCR using VH and VL specific primers. Expression of scFv in all the four bacterial cell strains was optimized in different combinations of incubation time, temperature and IPTG concentrations. The optimized conditions were as follows; overnight E. coli transformed cells were inoculated in 1:50 ratio in 2XYT medium containing 100 mM glucose and 50 mg/ml of ampicillin and incubated at 37  C at 220 rpm for approximately 3 h (till the OD reached 0.6). The culture was centrifuged at 3000 g for 10 min and media supernatant was discarded. The bacterial cell pellet was reconstituted in fresh 2XYTA medium containing 0.2 mM IPTG and incubated at 28  C with shaking at 240 rpm for five hours. Post induction samples were collected after 30 min of interval till 8.5 h post induction and the level of expression was analyzed by SDS-PAGE analysis. Expression of scFv was confirmed by western blot analysis using Penta-His HRP conjugates (Qiagen, Hilden, Germany) and with anti-mouse antibodies conjugated with HRPO (Sigma, USA). 2.7. Purification and refolding of expressed scFv Periplasmic, osmotic shock, media, and total protein fractions including inclusion bodies were examined to locate the expressed scFv as described elsewhere [12]. Periplasmic fraction was obtained by reconstituting the E. coli cell pellet in a buffer (25% glucose; 20 mM Tris-HCl pH 7.5; 1 mM EDTA) and incubating on ice for 20 min followed by centrifugation. The osmotic shock fraction was extracted from the remaining cell pellet by addition of MgSO4 and incubating on ice for another 20 min. ScFv was purified from the total protein under native conditions using Ni-magnetic beads (Merck Millipore, NJ, USA). Purified scFv was refolded to its native conformational structure by dialyzing in protein refolding buffer containing 0.1 mM DTT (Merck, NJ, USA) followed by re-dialyzing in

G.K. Sharma et al. / Biologicals 42 (2014) 339e345

protein refolding buffer without DTT. The dialyzed scFv was concentrated by Amicon Ultra-15 centrifugal filter device (Merck, NJ, USA) and purified scFv were preserved by addition of protease cocktail inhibitor set III (Merck Biosciences, NJ, USA) and stored at 20  C till further use. 2.8. Evaluation of scFv for immunodiagnosis of FMD The developed anti-3ABC scFv was evaluated for retrospective FMD diagnosis in two formats of ELISAs (sandwich and competitive). The scFv based ELISAs were standardized and validated by comparing the performance of the assays with the in-house 3ABC mAb based C-ELISA [10]. A panel of serum samples (n ¼ 2150) of known FMDV infection status was constituted for comparison and validation. The panel included 201 serum samples collected from naïve animals, 955 serum samples collected during 7e180 DPI from FDMV infected animals, and samples collected from vaccinated animals on 0 DPV (n ¼ 550) and 28 DPV (n ¼ 444). 2.8.1. Sandwich ELISA Wells of ELISA plates (Nunclone, Denmark) were coated with approximately 100 ng of scFv diluted in carbonate bi-carbonate buffer (pH 9.4). Two wells of the ELISA plates were coated with equal amount of E. coli lysate to act as background control. Wells were washed to remove the unbound scFv and were pre-blocked with 1.5% skimmed milk powder (Merck). Purified 3ABC antigen was allowed to bind with the coated scFv and E. coli lysate in predetermined dilution. Serum samples from the panel were diluted in 1:20 in 3% skimmed milk powder and allowed to react with the captured 3ABC protein. One each of high positive, medium positive and negative serum samples was kept as internal controls to monitor the performance of the assay. The bound bovine antibodies were traced by anti-bovine HRP conjugated secondary antibodies (Dako, Denmark). Color reaction was developed and OD of the color reaction was measured in an ELISA reader. The OD values were normalized in terms of percentage positivity (PP) with reference to the internal positive control as described elsewhere [13]. 2.8.2. Competitive ELISA The wells of ELISA plates were coated by the purified 3ABC protein diluted in carbonate bi-carbonate buffer in pre-determined dilution. Serum samples from the panel were diluted in 1:5 dilution and allowed to bind with the coated antigen. Similar to the sandwich ELISA, one each of high positive, medium positive and negative serum samples was kept as internal controls to monitor the performance of the assay. The unbound/unblocked coated antigens were traced by adding approximately 100 ng of the purified scFv in each well. Bound scFv were detected by anti-mouse Fab specific antibodies conjugated with HRP enzyme (Sigma, USA). Color reaction was developed and OD was measured in an ELISA reader. The OD values were normalized in terms of percentage inhibition (PI) value with reference to the internal negative control as described elsewhere [10].

341

3. Results 3.1. Expression and purification of soluble scFv Mouse hybridoma clone 3B3A5 was selected for production of scFv against 3ABC protein of the FMD virus. The selected clone secretes mAb of IgG1 isotype against the 3B region of 3ABC protein, as determined by specific reactivity towards recombinant 3B protein (data not presented). mRNA isolated from the hybridoma cells was reverse transcribed with Oligo(dT) primer and used as template for amplification of VL and VH gene segments of 380 bp and 400 bp, respectively (Fig. 1). A set of degenerate primers was used and amplification was obtained in four and five primers, respectively for VL and VH gene. The pOPE101-VH-VL was obtained by sequential cloning of gel purified VH and VL genes. In four E. coli strains, induction parameters such as time, temperature, and IPTG concentrations were optimized to obtain high scFv expression. Expression level in the transformed XL-1 blue cells was very low which could not be improved further even in different combinations of time, temperature and IPTG concentrations (Fig. 2). Whereas, level of scFv expression in another three expression host (BL21(DE3)pLysS, BL21(DE3)pLacI, and Tuner(DE3)pLysS) were significantly high as determined by SDS-PAGE analysis. The level of expression was highest in BL21(DE3)pLysS and Tuner(DE3)pLysS strains (Fig. 2). Reactivity of scFv with anti-His mAb-conjugated with HRP and anti-mouse antibodies conjugated with HRP could be confirmed in western blot analysis (Fig. 3). Purified scFv had low reactivity with 3ABC antigen in ELISA even though the scFv was purified under native conditions. However, after refolding of scFv in presence of 0.1 mM DTT, the reactivity in ELISA significantly improved. 3.2. Sequence analysis Cloning of VH and VL in pOPE101 vector was confirmed by nucleotide sequencing (Gene Bank accession nos. KF906356 and KF906357). A database search of light chain sequence using the BLAST algorithm [16] against the Protein Data Bank (PDB) and Kabat database [17] revealed 91% identity to the Ig kappa chain V-III region PC 6684 of mouse immunoglobulin [18] and 85% identity with Ig heavy chain V region VH558 A1/A4 [19]. The antibody gene was preceded by a pelB bacterial leader sequence for secretion into the

2.9. Comparisons with in-house ELISA Results obtained for the serum samples of panel (n ¼ 2150) tested by scFv based ELISAs and mAb based C-ELISA were compared by Spearman's rank of correlation coefficient. Diagnostic sensitivity (DSn) and specificity (DSp) of the two scFv assays based ELISAs were estimated by Receiver Operating Curve (ROC) method [14] following DeLong et al. [15] procedure in SAS 9.3 statistical software.

Fig. 1. In-vitro amplification of variable light and heavy chains of IgG against 3ABC (3B3A5 clone). 1. 100 bp DNA ladder (Fermentas). 2. Variable Light chain (VL) of IgG of 400 bp. 3. Variable Heavy chain (VH) of IgG of 380 bp.

342

G.K. Sharma et al. / Biologicals 42 (2014) 339e345

acids, respectively (Table 2). There were no unusual amino acids observed in the light chain except threonine at position 19 (Table 2) which is rarely observed in that position in the framework I region. The scFv possessed a heavy chain that was classified as a member of the subgroup I [21]. 3.3. Sandwich ELISA The scFv coated on wells of ELISA plates could specifically capture the 3ABC protein which was evident by high OD (>1.0) in the positive controls, whereas OD values in all the negative samples and in background controls was

Production and characterization of single-chain antibody (scFv) against 3ABC non-structural protein in Escherichia coli for sero-diagnosis of Foot and Mouth Disease virus.

Differentiation of Foot-and-Mouth Disease infected from vaccinated animals is essential for effective implementation of vaccination based control prog...
788KB Sizes 0 Downloads 7 Views