Appl Microbiol Biotechnol (2014) 98:1547–1555 DOI 10.1007/s00253-013-5351-6
BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING
Preparation and diagnostic use of a novel recombinant single-chain antibody against rabies virus glycoprotein Ruosen Yuan & Xiaoxu Chen & Yan Chen & Tiejun Gu & Hualong Xi & Ye Duan & Bo Sun & Xianghui Yu & Chunlai Jiang Xintao Liu & Chunlai Wu & Wei Kong & Yongge Wu
Received: 6 September 2013 / Revised: 17 October 2013 / Accepted: 19 October 2013 / Published online: 16 November 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract Rabies virus (RABV) causes a fatal infectious disease, but effective protection may be achieved with the use of rabies immunoglobulin and a rabies vaccine. Virus-neutralizing antibodies (VNA), which play an important role in the prevention of rabies, are commonly evaluated by the RABV neutralizing test. For determining serum VNA levels or virus titers during the RABV vaccine manufacturing process, reliability of the assay method is highly important and mainly dependent on the diagnostic antibody. Most diagnostic antibodies are monoclonal antibodies (mAbs) made from hybridoma cell lines and are costly and time consuming to prepare. Thus, production of a cost-effective mAb for determining rabies VNA levels or RABV titers is needed. In this report, we describe the prokaryotic production of a RABV-specific single-chain variable fragment (scFv) protein with a His-tag (scFv98H) from a previously constructed plasmid in a bioreactor, including the purification and refolding process as well as the functional testing of the protein. The antigen-specific binding characteristics, R. Yuan : X. Chen : Y. Chen : T. Gu : H. Xi : Y. Duan : B. Sun : X. Yu : C. Jiang : W. Kong : Y. Wu (*) National Engineering Laboratory for AIDS Vaccine, College of Life Science, Jilin University, Changchun 130012, China e-mail: [email protected] Y. Wu e-mail: [email protected] W. Kong e-mail: [email protected] e-mail: [email protected] X. Liu BCHT Biotechnology Company, Changchun 130012, China C. Wu School of Biosciences and Biotechnology, University of Camerino, Camerino, Italy
affinity, and relative affinity of the purified protein were tested. The scFv98H antibody was compared with a commercial RABV nucleoprotein mAb for assaying the VNA level of anti-rabies serum samples from different sources or testing the growth kinetics of RABV strains for vaccine manufactured in China. The results indicated that scFv98H may be used as a novel diagnostic tool to assay VNA levels or virus titers and may be used as an alternative for the diagnostic antibody presently employed for these purposes. Keywords Rabies virus . Glycoprotein . Monoclonal antibody . Virus-neutralizing antibody . Single-chain Fv fragment
Introduction Human rabies infection is a serious public health problem, as approximately 55,000 people die each year from this disease in many Asian or African countries (Bourhy et al. 2010; Knobel et al. 2005). Although the mortality rate is 100 % after the appearance of clinical symptoms, rabies virus (RABV) infection can be prevented by prompt post-exposure prophylaxis (PEP). PEP includes immediate wound care and administration of both a rabies vaccine and rabies immunoglobulin (RIG). Therefore, accurately determining virus-neutralizing antibody (VNA) levels is an important aspect of monitoring development of immunity from the RABV vaccine. The World Health Organization (WHO) and Office International des Epizooties (OIE) have recommended a VNA titer of ≥0.5 IU/ml as adequate to prevent rabies. VNAs are directed against the RABV glycoprotein (G protein). To measure the level of VNAs in human or animal serum, two virus neutralization tests are used globally, the rapid fluorescent focus inhibition test (RFFIT) (Smith et al. 1973) and
1548
fluorescent antibody virus neutralization assay (FAVN) (Cliquet et al. 1998). These tests are relatively expensive, can be performed only in restricted reference laboratories, and require live RABV and a good diagnostic antibody (Cliquet et al. 1998; Smith et al. 1973). The enzyme-linked immunosorbent assay (ELISA) for detecting VNA levels for RABV is commercially available and more widely accessible than the virus neutralization test (Cliquet et al. 2000; Feyssaguet et al. 2007; Moore and Hanlon 2010). However, the accuracy of results derived from the ELISA is dependent on the coating RABV antigen and assay antibody. RABV titers must be evaluated before and during replication of a vaccine virus in the manufacturing process. After inactivation, the RABV antigen content also needs to be determined. The reliability and accuracy of the VNA test and RABV titers depend on the quality of the monoclonal antibody (mAb). A number of mouse or human mAbs against the RABV glycoprotein have been isolated and characterized (Bakker et al. 2005; Bunschoten et al. 1989; Coulon et al. 1982; Hanlon et al. 2001; Marissen et al. 2005; Ni et al. 1995; Prehaud et al. 1988; Prosniak et al. 2003; Schumacher et al. 1989). However, the use of mAbs has some drawbacks, such as the instability of hybridoma cell lines. The cell culture process is time consuming, and contamination remains a concern. Additionally, the serum-free medium used for protein expression, while convenient, is relatively expensive. For the reasons mentioned above, a cost-effective antirabies antibody that can be produced efficiently is needed. Current recombinant technology allows us to engineer such antigen-specific antibodies with high binding affinity or neutralizing activity. Single-chain variable antibody fragments (scFvs) are polypeptides in which the variable domains of the immunoglobulin heavy (VH) and light (VL) chains are linked through a flexible polypeptide linker (Bird et al. 1988). Escherichia coli is the expression system of choice for these antibody fragments due to the rapid process from antibody gene construction to industrial-scale production and the ability to control batch-to-batch variations. In addition, the ease and speed of generating productive cell lines render the capital costs associated with bacteria protein production relatively low compared with eukaryotic expression systems (Humphreys and Glover 2001). In this study, the expression and subsequent purification and refolding processes of a scFv against RABV glycoprotein, scFv98H, in insoluble inclusion bodies of batch processed fermented bacteria are described. The binding activity, affinity, and relative affinity of this single-chain antibody also were determined. More importantly, scFv98H was evaluated as a diagnostic tool and compared with a commercial RABV nucleoprotein (NP) mAb for assaying the VNA level of antirabies serum samples from different sources and testing the titer of RABV strains.
Appl Microbiol Biotechnol (2014) 98:1547–1555
Materials and methods Fermentation and preparation of bacteria expressing scFv98H in inclusion bodies Expression of the scFv98H plasmid in Escherichia coli BL21 (DE3) cells was carried out as described previously (Gu et al. 2011). The cells were grown in a complex auto-inducing medium, but with some modifications (10 g tryptone, 10 g yeast extract, 8.95 g Na2HPO4, 3.4 g KH2PO4, 0.71 g Na2SO4, 0.49 MgSO4·7H2O, 0.5 g glucose, 2 g lactose, 2.67 g NH4Cl, 5 ml glycerol per liter) (Studier 2005). An overnight seed culture was used to inoculate the 3-L bioreactor New Brunswick Scientific at 2 %. After fermentation, cells were harvested by centrifugation, and the paste was kept at −20 °C if not immediately used. The harvested cells were resuspended in TES buffer (50 mM Tris–HCl, pH 8.0, 5 mM EDTA, 150 mM NaCl) at 20 ml/g per gram (wet weight). Cells were lysed and centrifuged at 10,000×g for 30 min. The pellet containing inclusion bodies was washed with TE buffer (50 mM Tris–HCl, pH 8.0, 5 mM EDTA) containing 1 % Triton X-100, followed by washing with TEN buffer (50 mM Tris–HCl, pH 8.0, 5 mM EDTA, 2 M NaCl). Purification of inclusion bodies The inclusion bodies were dissolved in solubilization buffer (8 M urea, 50 mM Tris–HCl, pH 8.0) at the protein concentration of 5 mg/ml. The denatured proteins were then centrifuged at 10,000×g for 30 min, and the supernatant was filtered through a 0.22-μm membrane before purification. The denatured proteins were loaded at a flow rate of 1.5 ml/min onto a 5 ml His Trap HP Immobilized Metal Affinity Chromatography (IMAC) column (GE Healthcare) preequilibrated with buffer A (8 M urea, 50 mM Tris–HCl, pH 8.0, 0.5 M NaCl, 5 mM imidazole) in a protein purification system (AKTA purifier 100, GE Healthcare). After the protein loading, the column was washed extensively with buffer B (8 M urea, 50 mM Tris–HCl, pH 8.0, 0.5 M NaCl 60 mM imidazole) to remove unbound and nonspecifically bound proteins. The target protein was eluted with buffer C (8 M urea, 50 mM Tris–HCl, pH 9.0, 200 mM imidazole). The elution fractions were pooled and then purified by anion ion exchange chromatography (AIEX). This His-tagged protein was loaded at 2 ml/min and washed at 4 ml/min on an AIEX column (HiPrep™16/10 QXL) in buffer C without imidazole. The target protein was eluted with buffer D (8 M urea, 50 mM Tris–HCl, pH 9.0, 150 mM NaCl). The host protein and other impurities were eluted with buffer E (8 M urea, 50 mM Tris– HCl, pH 9.0, 1 M NaCl). At each step, fractions were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on a 13.5 % polyacrylamide gel and stained with Coomassie Brilliant Blue R-250 (Sigma-Aldrich).
Appl Microbiol Biotechnol (2014) 98:1547–1555
Protein refolding The purified protein was dialyzed in four times the volume with refolding buffer (50 mM Tris–HCl, pH 8.0, 10 % glycerin, 1 % glycine, 1 mM EDTA, 50 mM NaCl). The refolding process was carried out at 4 °C for 24 h with replacement of fresh buffer every 4 h without agitation. The refolded proteins were concentrated by ultrafiltration with a 5,000-Da molecular weight cut-off membrane (Millipore). The concentration of the protein was determined by the BCA method with a bovine serum albumin (BSA) standard. Purification and estimation of apparent molecular mass by gel filtration Gel filtration purification and determination of molecular mass were performed on an AKTA system (GE Healthcare) with a HiLoad 16/60 Superdex™ 75 prep grade column (GE Healthcare) pre-equilibrated with refolding buffer. The elution time was calibrated using the protein standards lysozyme (14 kDa), ovalbumin (44 kDa), and BSA (66 kDa). The protein size was analyzed after refolding. Indirect ELISA Ninety-six well ELISA plates were coated overnight at 4 °C with the inactivated RABV (IRABV) aG strain or an unrelated antigen, such as hepatitis B surface antigen (HBsAg), hepatitis A virus (HAV), or varicella-zoster virus (VZV), at the dilution of 1:100 in 100 μL 0.1 M carbonate bicarbonate buffer (pH 9.6). After three washes with PBS containing 0.05 % Tween 20 (PBST), the wells were blocked with 200 μL of PBST/3 % BSA for 2 h at 37 °C and washed as above. The NP mAb (Millipore) or scFv98H serially diluted in PBS was added to the plates in 100 μL volume per well and incubated at 37 °C for 1 h. After washing, a mouse anti-His antibody (Invitrogen) was added to wells (100 μl/well) previously incubated with scFv98H and maintained at 37 °C for 1 h. The plates were washed again, and the bound antibodies were detected using a horseradish peroxidase (HRP)-conjugated goat anti-mouse antibody (Jackson Immunoresearch) in PBST (100 μl/well) at 37 °C for 1 h. The substrate tetramethyl benzidine was subsequently added (100 μl/well) and incubated at 37 °C for 15 min. The reaction was stopped by adding 2 M H2SO4 (50 μl/well), and results were obtained by measuring the absorbance at 450 nm. Immunofluorescence staining Immunofluorescence staining was used to determine whether scFv98H can bind the RABV glycoprotein in cells. BSR cells (a cloned derivative of baby hamster kidney cells) infected with RABV or not (blank control) were added in 200 μL
1549
volume per well to 96-well microtiter plates and grown in DMEM (5 % FBS) at 37 °C in a 5 % CO2 incubator for 24 h. The cells were washed three times in PBS. Some wells were stained with a fluorescein isothiocyanate (FITC)-conjugated NP mAb (Millipore) in 20 μg/ml (50 μl/well). Other wells were incubated with scFv98H in 1 μg/ml (50 μl/well) followed by a mouse anti-His antibody in 1 μg/ml (50 μl/well) for 1 h and Alexa488-conjugated goat anti-mouse antibody in 1 μg/ml (50 μl/well) against the His-tag (Invitrogen) for 45 min in the dark. Subsequently, the cells were washed and fixed in ice-cold 80 % (100 μl/well) acetone for 10 min, followed by DAPI staining of the nuclei. After washing with PBS, the plates were observed under a microscope (Nikon ECLIPST Ti). Relative affinity and affinity assay The scFv98H protein was prepared at 0.01 mg/ml in dilution buffer (PBS, pH 7.5, 1 % BSA) and added to 96-well ELISA plates pre-coated with IRABV (aG strain). The plates were incubated for 1 h at 37 °C and washed with PBST five times. Thereafter, the protein was eluted with NH4SCN in PBS at concentrations ranging from 0 to 5 M (100 μl/well). The elution was performed on a shaker at 500 rpm for 1 h at 37 °C (MacDonald et al. 1988). The NH4SCN was washed from the plate with PBST five times, and the protein bound to IRABV was detected using rabbit anti-scFv98H and HRP-conjugated goat anti-rabbit antibodies (Jackson Immunoresearch). The equilibrium dissociation constant (K d) of scFv98H was measured by a noncompetitive method (Beatty et al. 1987). IRABV was coated on ELISA plates at various dilutions in bicarbonate coat buffer (pH 7.5) at 4 °C overnight, followed by blocking with 3 % BSA for 2 h at 37 °C. The protein was twofold serially diluted at concentrations from 1.7 nM to 3.5 μM. The scFv98H protein that remained bound to the IRABV was detected using rabbit anti-scFv98H and HRP-conjugated goat anti-rabbit antibodies. The K d of this antibody was calculated by previously described methods (Beatty et al. 1987). RABV neutralizing antibody assay According to the WHO, neutralizing antibody levels against RABV can be determined by the RFFIT (Smith et al. 1973). Here, scFv98H was evaluated in this test to determine whether it can be used to assay the neutralizing antibody titer against RABV. The national reference serum with verified potency (21.4 IU/ml) was used as a standard. All serum samples were heated at 56 °C for 30 min before testing in order to inactivate complement. In 96-well microtiter plates, two commercial RIGs, human RIG (HRIG) and equine RIG (ERIG), and two types of RABV antisera from experimental animals, mice (MRIG) and rabbits (RRIG), were threefold serially diluted
1550
in DMEM with 10 % inactivated bovine serum. Thereafter, the RABV challenge virus strain (CVS)-11 titrated to a dose causing 80 % infection of BSR cells was added, and the plates were incubated for 60 min in a humidified incubator at 37 °C with 5 % CO2. Following the incubation, 4×104 BSR cells were added per well, and the plate was incubated at 37 °C for 24 h. Finally, the culture supernatant was removed, and the monolayer cells were fixed with ice-cold 80 % acetone and incubated with a FITC-conjugated NP mAb or with scFv98H followed by a mouse anti-His antibody for 1 h and Alexa488conjugated goat anti-mouse antibody against the His tag for 45 min in the dark. The stained cells were observed under a fluorescence microscope, and the neutralization potency of the test antibodies were calculated against that of the national reference standard serum by the Reed and Muench method (Reed and Muench 1938). Utility of scFv98H for RABV titer determination In order to test whether scFv98H can be used as a diagnostic agent to determine RABV titers, four strains (CVS-11, aG, PM, and ERA) of RABV were used for direct immunofluorescence staining. CVS-11 is used for diagnostics, while the others are used as a vaccine for either humans (aG and PM) or animals (ERA) in China. BSR cells were inoculated in a 96-well plate with RABV serially diluted fivefold in DMEM with 10 % inactivated bovine serum and incubated at 34 °C for 2 days. After discarding the culture medium, the cells were fixed with 80 % acetone and then stained with the FITCconjugated NP mAb or with scFv98H followed by a mouse anti-His antibody for 1 h and Alexa488-conjugated goat antimouse antibody against the His tag for 45 min in the dark. Positive antigen foci were counted under a fluorescence microscope. Titrations were conducted in quadruplicate, and viral titers were calculated as fluorescent focus units per milliliter.
Appl Microbiol Biotechnol (2014) 98:1547–1555
culture medium contained sufficient nutrients to promote cell growth to a final OD600 value of approximately 30. Dissolved oxygen was kept at 30 % by adjusting the agitation and aeration rate. Samples were taken every hour for OD600 measurements. The fermentation biomass was harvested by centrifugation, and the cell pellets were stored at −20 °C. Protein expression was analyzed by SDS-PAGE with Coomassie Blue staining (Fig. 1). The fermentation biomass was lysed by sonication as described in “Materials and methods.” After the inclusion bodies were isolated, washed, and dissolved in 8 M urea buffer, the scFv98H protein was purified by IMAC. SDS-PAGE analysis of the product showed a major band corresponding to the expected molecular mass of scFv98H, which was highly purified (∼90 % purity) after one IMAC step (Fig. 2a). The pooled fractions were loaded onto the AIEX column (HiPrep™16/10 QXL) again, and the main peak fractions pooled from this step were considered as the highly purified (>95 % by host protein ELISA kit, data not shown) form of the unfolded scFv98H protein (Fig. 2b).
Purification of scFv98H The complete process of expression in auto-inducing medium and purification of scFv98H from bacteria is summarized in Fig. 3a. After the IMAC and AIEX purification steps, 1.2 g of scFv98H was obtained from 2 L of bacterial culture (data not shown). After renaturation, the pooled mixture of monomers, dimer, and aggregates of scFv98H was concentrated by ultrafiltration, and 2 ml of the sample was loaded onto an analytic Superdex 75 column. The size-exclusion chromatography (SEC) purification profile and SDS-PAGE analysis are shown in Fig. 3b. Ultimately, we obtained 540 mg of scFv98H monomers from 2 L of bacterial culture.
Statistical analysis Data were expressed as the mean ± standard deviation (SD) of values obtained from at least three independent experiments. Results were analyzed with the unpaired Student’s t test, and differences were considered to be statistically significant when P 0.05). These results indicate that scFv98H may be used reliably to assay VNA levels. RABV titering To characterize the refolded scFv98H protein, growth kinetics of the four strains of RABV were examined in BSR cells. CVS-11 is a RABV strain used for diagnostics, while the other RABV strains are used for vaccine manufacture in China. As shown in Fig. 5, no significant difference in titer (P >0.05) was observed between scFv98H and the FITC-conjugated NP mAb groups, indicating that scFv98H can be used to monitor replication of different RABV strains.
Discussion The Escherichia coli expression system is among the most popular for recombinant protein production due to its ease of genetic manipulation and scale-up, relatively short duration between transformation and protein purification, as well as its cost effectiveness (Arbabi-Ghahroudi et al. 2005; Demain and Vaishnav 2009; Ohlfest et al. 2012). Although it is preferable to express scFvs in a soluble and active form, the expression usually is low (Kramer et al. 2005; Lombardi et al. 2005; Muller et al. 1997), and most of the antibody fragments tend to be expressed in inclusion bodies (Jiang et al. 2013; Liu et al. 2011; Lombardi et al. 2005). However, expressing proteins in inclusion bodies, even though they are not in an active form, provides some advantages including high protein yield and ease of separating the insoluble form from the soluble impurities of E. coli.
Fig. 5 Comparison of scFv98H and NP mAb for monitoring RABV titers. a To test scFv98H, RABV (CVS-11 strain) was tenfold serially diluted, and the NP mAb was used as a control for immunofluorescence staining. Scale bar=200 μm. b ScFv98H was used to titer the CVS-11, aG, PM, and ERA virus strains during the replication process. The NP mAb was used as a positive control for determining the virus titer. Data are means ± SD from triplicate samples; P >0.05
In this study, scFv98H was expressed in Escherichia coli within insoluble inclusion bodies and therefore was easily separated from the host proteins or other soluble impurities by centrifugation. To remove residual host proteins and DNA, the inclusion bodies were extensively washed by TE buffers with Triton or NaCl. Fortunately, the inclusion bodies could be solubilized entirely in an inexpensive detergent (8 M urea). The recombinant scFv98H protein was produced with a His tag, which allowed purification by using IMAC (Fig. 2a). As the impurities, such as host proteins, lipopolysaccharides, and host DNA could bind tightly to the anion ion exchange medium, they were eluted only at a higher salt concentration (1 M NaCl). Meanwhile, the target protein could be easily eluted with a low salt concentration (150 mM NaCl), and unfolded scFv98H protein can be well separated from these impurities (Fig. 2b). Both IMAC and AIEX, which are highly selective chromatographic processes, were used for polishing steps in the purification of the unfolded scFv98H protein, providing a purity of more than 95 % as determined by the
1554
host protein ELISA kit (data not shown). Such a high purity of the unfolded protein facilitated the refolding process. The refolding of this antibody fragment protein was accomplished by dialysis in refolding buffer. However, it is generally a challenge to increase the yield of refolded proteins from inclusion bodies. Although the rate of recovery of refolded scFv98H was only ∼65 %, the product yield from inclusion bodies was relatively high comparing to other recombinant antibodies (André et al. 2013) at 0.6 g/L with 95 % purity (Fig. 2b). Thus, the high level of protein expression in inclusion bodies could compensate for any loss in recovery during purification (Singh and Panda 2005). Currently, our group is focused on developing an antibody against RABV for immunodetection or therapy, and scFvs against the RABV glycoprotein have been of special interest. Towards this aim, our group previously selected the scFv98H antibody screened from a phage display system (Kramer et al. 2005) and subsequently cloned and expressed it in a bacterial system (Gu et al. 2011). However, this engineered antibody was not characterized, and further development for diagnostic or therapeutic purposes was needed. To characterize this purified refolded scFv98H antibody, its specificity and binding activity were determined in this study. The refolded scFv98H protein was determined to specifically bind to IRABV (aG strain) but not other viruses tested (Fig. 4a). However, scFv98H could not recognize all strains of RABV. The parental antibody CR4098 is known to specifically target the G protein antigenic site III (Bakker et al. 2005). To provide coverage for all or most of the circulating RABVs, combining scFv98H with scFv57, which targets site I of the G protein, may be a good approach (Bakker et al. 2005; Duan et al. 2012). The relative affinity of scFv98H was 2.3 M, which of the disulfide-stabilized scFvs against RABV G protein (Duan et al. 2012) were between 2 and 3 M. The K d of scFv98H was 1×106 M−1, which of the scFvs (Duan et al. 2012) were from 5.65×105 to 1.78×106 M−1. The relative affinity and K d of scFv98H were found to be at the same levels as those of scFvs. However, differences in K d values reflected the reduced avidity of the scFv compared with the parental antibody CR4098, which was expected based on the fact that CR4098 is an IgG with two identical antigen recognition surfaces, while scFv98H retains only one (Yang et al. 2010). The antigen-specific binding of the recombinant scFv98H fragment was tested both in an ELISA and in infected cells. His tag used for purification also facilitated detection of the recombinant protein. Meanwhile, scFv98H was found to bind with RABV (CVS-11 strain) as well as the NP mAb in cells permeabilized by acetone. The RABV CVS-11 strain is a standard virus recommended by the WHO for use in neutralizing antibody assays (FAVN or RFFIT). Results from the current study indicated that scFv98H may be developed as a new diagnostic antibody to test RABV VNA levels in serum
Appl Microbiol Biotechnol (2014) 98:1547–1555
samples from different sources (Table 1). Differences in results between the two types of antibodies were not statistically significant, indicating that scFv98H is a promising candidate antibody for the RABV VNA assay. To further explore whether scFv98H can be used as a diagnostic tool to titer RABV for use in manufacturing vaccines or measuring VNAs, the CVS-11, aG, PM, and ERA strains were compared. Not only was scFv98H found to bind efficiently to RABV, but it also could be used to detect the virus titer during propagation, similar to the NP mAb (Fig. 5). With practical advantages compared with the commercial NP mAb, scFv98H may be used in potential applications as a novel diagnostic agent to test serum neutralizing antibodies, as well as to monitor virus titers during rabies vaccine production. Acknowledgments We gratefully acknowledge Thi Sarkis for editorial support in the preparation of this manuscript.
References André F, Michael H, Thomas S (2013) Expression of recombinant antibodies. Front Immunol 4:217. doi:10.3389/fimmu.2013.00217 Arbabi-Ghahroudi M, Tanha J, MacKenzie R (2005) Prokaryotic expression of antibodies. Cancer Metastasis Rev 24(4):501–519. doi:10. 1007/s10555-005-6193-1 Bakker ABH, Marissen WE, Kramer RA, Rice AB, Weldon WC, Niezgoda M, Hanlon CA, Thijsse S, Backus HHJ, De Kruif J (2005) Novel human monoclonal antibody combination effectively neutralizing natural rabies virus variants and individual in vitro escape mutants. J Virol 79(14):9062–9068. doi:10.1128/JVI.79.14. 9062-9068.2005 Beatty JD, Beatty BG, Vlahos WG (1987) Measurement of monoclonal antibody affinity by non-competitive enzyme immunoassay. J Immunol Methods 100(1–2):173–179 Bird RE, Hardman KD, Jacobson JW, Johnson S, Kaufman BM, Lee SM, Lee T, Pope SH, Riordan GS, Whitlow M (1988) Single-chain antigen-binding proteins. Science 242(4877):423–426. doi:10. 1126/science.3140379, New York, NY Bourhy H, Dautry-Varsat A, Hotez PJ, Salomon J (2010) Rabies, still neglected after 125 years of vaccination. PLoS Negl Trop Dis 4(11): e839. doi:10.1371/journal.pntd.0000839 Bunschoten H, Gore M, Ivo JTM, Uytdehaag FGCM, Dietzschold B, Wunner WH, Osterhaus ADME (1989) Characterization of a new virus-neutralizing epitope that denotes a sequential determinant on the rabies virus glycoprotein. J Gen Virol 70(2):291–298. doi:10. 1099/0022-1317-70-2-291 Cliquet F, Aubert M, Sagne L (1998) Development of a fluorescent antibody virus neutralisation test (FAVN test) for the quantitation of rabies-neutralising antibody. J Immunol Methods 212(1):79–87 Cliquet F, Sagne L, Schereffer J, Aubert M (2000) ELISA test for rabies antibody titration in orally vaccinated foxes sampled in the fields. Vaccine 18(28):3272–3279. doi:10.1016/S0264-410X(00)00127-4 Coulon P, Rollin P, Aubert M, Flamand A (1982) Molecular basis of rabies virus virulence. I. Selection of avirulent mutants of the CVS strain with anti-G monoclonal antibodies. J Gen Virol 61(1):97–100. doi:10.1099/0022-1317-61-1-97 Demain AL, Vaishnav P (2009) Production of recombinant proteins by microbes and higher organisms. Biotechnol Adv 27(3):297–306. doi:10.1016/j.biotechadv.2009.01.008
Appl Microbiol Biotechnol (2014) 98:1547–1555 Duan Y, Gu T, Jiang C, Yuan R, Zhang H, Hou H, Yu X, Chen Y, Zhang Y, Wu Y (2012) A novel disulfide-stabilized single-chain variable antibody fragment against rabies virus G protein with enhanced in vivo neutralizing potency. Mol Immunol 51(2):188–196. doi:10. 1016/j.molimm.2012.03.015 Feyssaguet M, Dacheux L, Audry L, Compoint A, Morize J, Blanchard I, Bourhy H (2007) Multicenter comparative study of a new ELISA, PLATELIA™ RABIES II, for the detection and titration of antirabies glycoprotein antibodies and comparison with the rapid fluorescent focus inhibition test (RFFIT) on human samples from vaccinated and non-vaccinated people. Vaccine 25(12):2244–2251. doi: 10.1016/j.vaccine.2006.12.012 Gu T, Wei W, Duan Y, Jiang C, Chen Y, Yu X, Wu J, Wu Y, Kong W (2011) Identification of binding epitope for anti-rabies virus glycoprotein single-chain Fv fragment FV57. Protein Pept Lett 18(11): 1099–1106 Hanlon C, DeMattos C, DeMattos C, Niezgoda M, Hooper D, Koprowski H, Notkins A, Rupprecht C (2001) Experimental utility of rabies virus-neutralizing human monoclonal antibodies in post-exposure prophylaxis. Vaccine 19(28):3834–3842. doi:10.1016/S0264410X(01)00135-9 Humphreys D, Glover D (2001) Therapeutic antibody production technologies: molecules, applications, expression and purification. Curr Opin Drug Discov Dev 4(2):172–185 Jiang H, Xie Y, Burnette A, Roach J, Giardina SL, Hecht TT, Creekmore SP, Mitra G, Zhu J (2013) Purification of clinical-grade disulfide stabilized antibody fragment variable—Pseudomonas exotoxin conjugate (dsFv-PE38) expressed in Escherichia coli . Appl Microbiol Biotechnol:621–632. doi: 10.1007/s00253-012-4319-2 Knobel DL, Cleaveland S, Coleman PG, Fèvre EM, Meltzer MI, Miranda MEG, Shaw A, Zinsstag J, Meslin FX (2005) Re-evaluating the burden of rabies in Africa and Asia. B World Health Organ 83(5): 360–368. doi:10.1590/S0042-96862005000500012 Kramer RA, Marissen WE, Goudsmit J, Visser TJ, Clijsters-Van der Horst M, Bakker AQ, de Jong M, Jongeneelen M, Thijsse S, Backus HHJ (2005) The human antibody repertoire specific for rabies virus glycoprotein as selected from immune libraries. Eur J Immunol 35(7):2131–2145. doi:10.1002/eji.200526134 Liu X, Lin H, Tang Q, Li C, Yang S, Wang Z, Wang C, He Q, Cao B, Feng Z (2011) Characterization of a human antibody fragment Fab and its calcium phosphate nanoparticles that inhibit rabies virus infection with vaccine. PLoS One 6(5):e19848. doi:10.1371/ journal.pone.0019848 Lombardi A, Sperandei M, Cantale C, Giacomini P, Galeffi P (2005) Functional expression of a single-chain antibody specific for the HER2 human oncogene in a bacterial reducing environment. Protein Expr Purif 44(1):10–15. doi:10.1016/j.pep.2005.05.013 MacDonald RA, Hosking CS, Jones CL (1988) The measurement of relative antibody affinity by ELISA using thiocyanate elution. J
1555 Immunol Methods 106(2):191–194. doi:10.1016/0022-1759(88) 90196-2 Marissen WE, Kramer RA, Rice A, Weldon WC, Niezgoda M, Faber M, Slootstra JW, Meloen RH, Clijsters-van der Horst M, Visser TJ (2005) Novel rabies virus-neutralizing epitope recognized by human monoclonal antibody: fine mapping and escape mutant analysis. J Virol 79(8):4672–4678. doi:10.1128/JVI.79.8.4672-4678.2005 Moore SM, Hanlon CA (2010) Rabies-specific antibodies: measuring surrogates of protection against a fatal disease. PLoS Negl Trop Dis 4(3):e595. doi:10.1371/journal.pntd.0000595 Muller BH, Lafay F, Demangel C, Perrin P, Tordo N, Flamand A, Lafaye P, Guesdon JL (1997) Phage-displayed and soluble mouse scFv fragments neutralize rabies virus. J Virol Methods 67(2):221–233. doi:10.1016/S0166-0934(97)00099-2 Ni Y, Tominaga Y, Honda Y, Morimoto K, Sakamoto S, Kawai A (1995) Mapping and characterization of a sequential epitope on the rabies virus glycoprotein which is recognized by a neutralizing monoclonal antibody, RG719. Med Microbiol Immunol 39(9):693 Ohlfest JR, Zellmer DM, Panyam J, Swaminathan SK, Oh S, Waldron NN, Toma S, Vallera DA (2012) Immunotoxin targeting CD133+ breast carcinoma cells. Drug Deliv and Transl Res: 195–204 Prehaud C, Coulon P, Lafay F, Thiers C, Flamand A (1988) Antigenic site II of the rabies virus glycoprotein: structure and role in viral virulence. J Virol 62(1):1–7 Prosniak M, Faber M, Hanlon CA, Rupprecht CE, Hooper DC, Dietzschold B (2003) Development of a cocktail of recombinantexpressed human rabies virus-neutralizing monoclonal antibodies for postexposure prophylaxis of rabies. J Infect Dis 188(1):53–56. doi:10.1086/375247 Reed LJ, Muench H (1938) A simple method of estimating fifty percent endpoints. Am J Epidemiol 27(3):493–497 Schumacher CL, Dietzschold B, Ertl H, Niu HS, Rupprecht CE, Koprowski H (1989) Use of mouse anti-rabies monoclonal antibodies in postexposure treatment of rabies. J Clin Invest 84(3):971–975. doi:10.1172/JCI114260 Singh SM, Panda AK (2005) Solubilization and refolding of bacterial inclusion body proteins. J Biosci Bioeng 99(4):303–310. doi:10. 1263/jbb.99.303 Smith JS, Yager PA, Baer GM (1973) A rapid reproducible test for determining rabies neutralizing antibody. Bull World Health Organ 48(5):535–541 Studier FW (2005) Protein production by auto-induction in high-density shaking cultures. Protein Expr Purif 41(1):207–234. doi:10.1016/j. pep.2005.01.016 Yang J, Chen R, Wei J, Zhang F, Zhang Y, Jia L, Yan Y, Luo W, Cao Y, Yao L (2010) Production and characterization of a recombinant single-chain antibody against Hantaan virus envelop glycoprotein. Appl Microbiol Biotechnol 86(4):1067–1075. doi:10.1007/s00253009-2379-8
BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING
Preparation and diagnostic use of a novel recombinant single-chain antibody against rabies virus glycoprotein Ruosen Yuan & Xiaoxu Chen & Yan Chen & Tiejun Gu & Hualong Xi & Ye Duan & Bo Sun & Xianghui Yu & Chunlai Jiang Xintao Liu & Chunlai Wu & Wei Kong & Yongge Wu
Received: 6 September 2013 / Revised: 17 October 2013 / Accepted: 19 October 2013 / Published online: 16 November 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract Rabies virus (RABV) causes a fatal infectious disease, but effective protection may be achieved with the use of rabies immunoglobulin and a rabies vaccine. Virus-neutralizing antibodies (VNA), which play an important role in the prevention of rabies, are commonly evaluated by the RABV neutralizing test. For determining serum VNA levels or virus titers during the RABV vaccine manufacturing process, reliability of the assay method is highly important and mainly dependent on the diagnostic antibody. Most diagnostic antibodies are monoclonal antibodies (mAbs) made from hybridoma cell lines and are costly and time consuming to prepare. Thus, production of a cost-effective mAb for determining rabies VNA levels or RABV titers is needed. In this report, we describe the prokaryotic production of a RABV-specific single-chain variable fragment (scFv) protein with a His-tag (scFv98H) from a previously constructed plasmid in a bioreactor, including the purification and refolding process as well as the functional testing of the protein. The antigen-specific binding characteristics, R. Yuan : X. Chen : Y. Chen : T. Gu : H. Xi : Y. Duan : B. Sun : X. Yu : C. Jiang : W. Kong : Y. Wu (*) National Engineering Laboratory for AIDS Vaccine, College of Life Science, Jilin University, Changchun 130012, China e-mail: [email protected] Y. Wu e-mail: [email protected] W. Kong e-mail: [email protected] e-mail: [email protected] X. Liu BCHT Biotechnology Company, Changchun 130012, China C. Wu School of Biosciences and Biotechnology, University of Camerino, Camerino, Italy
affinity, and relative affinity of the purified protein were tested. The scFv98H antibody was compared with a commercial RABV nucleoprotein mAb for assaying the VNA level of anti-rabies serum samples from different sources or testing the growth kinetics of RABV strains for vaccine manufactured in China. The results indicated that scFv98H may be used as a novel diagnostic tool to assay VNA levels or virus titers and may be used as an alternative for the diagnostic antibody presently employed for these purposes. Keywords Rabies virus . Glycoprotein . Monoclonal antibody . Virus-neutralizing antibody . Single-chain Fv fragment
Introduction Human rabies infection is a serious public health problem, as approximately 55,000 people die each year from this disease in many Asian or African countries (Bourhy et al. 2010; Knobel et al. 2005). Although the mortality rate is 100 % after the appearance of clinical symptoms, rabies virus (RABV) infection can be prevented by prompt post-exposure prophylaxis (PEP). PEP includes immediate wound care and administration of both a rabies vaccine and rabies immunoglobulin (RIG). Therefore, accurately determining virus-neutralizing antibody (VNA) levels is an important aspect of monitoring development of immunity from the RABV vaccine. The World Health Organization (WHO) and Office International des Epizooties (OIE) have recommended a VNA titer of ≥0.5 IU/ml as adequate to prevent rabies. VNAs are directed against the RABV glycoprotein (G protein). To measure the level of VNAs in human or animal serum, two virus neutralization tests are used globally, the rapid fluorescent focus inhibition test (RFFIT) (Smith et al. 1973) and
1548
fluorescent antibody virus neutralization assay (FAVN) (Cliquet et al. 1998). These tests are relatively expensive, can be performed only in restricted reference laboratories, and require live RABV and a good diagnostic antibody (Cliquet et al. 1998; Smith et al. 1973). The enzyme-linked immunosorbent assay (ELISA) for detecting VNA levels for RABV is commercially available and more widely accessible than the virus neutralization test (Cliquet et al. 2000; Feyssaguet et al. 2007; Moore and Hanlon 2010). However, the accuracy of results derived from the ELISA is dependent on the coating RABV antigen and assay antibody. RABV titers must be evaluated before and during replication of a vaccine virus in the manufacturing process. After inactivation, the RABV antigen content also needs to be determined. The reliability and accuracy of the VNA test and RABV titers depend on the quality of the monoclonal antibody (mAb). A number of mouse or human mAbs against the RABV glycoprotein have been isolated and characterized (Bakker et al. 2005; Bunschoten et al. 1989; Coulon et al. 1982; Hanlon et al. 2001; Marissen et al. 2005; Ni et al. 1995; Prehaud et al. 1988; Prosniak et al. 2003; Schumacher et al. 1989). However, the use of mAbs has some drawbacks, such as the instability of hybridoma cell lines. The cell culture process is time consuming, and contamination remains a concern. Additionally, the serum-free medium used for protein expression, while convenient, is relatively expensive. For the reasons mentioned above, a cost-effective antirabies antibody that can be produced efficiently is needed. Current recombinant technology allows us to engineer such antigen-specific antibodies with high binding affinity or neutralizing activity. Single-chain variable antibody fragments (scFvs) are polypeptides in which the variable domains of the immunoglobulin heavy (VH) and light (VL) chains are linked through a flexible polypeptide linker (Bird et al. 1988). Escherichia coli is the expression system of choice for these antibody fragments due to the rapid process from antibody gene construction to industrial-scale production and the ability to control batch-to-batch variations. In addition, the ease and speed of generating productive cell lines render the capital costs associated with bacteria protein production relatively low compared with eukaryotic expression systems (Humphreys and Glover 2001). In this study, the expression and subsequent purification and refolding processes of a scFv against RABV glycoprotein, scFv98H, in insoluble inclusion bodies of batch processed fermented bacteria are described. The binding activity, affinity, and relative affinity of this single-chain antibody also were determined. More importantly, scFv98H was evaluated as a diagnostic tool and compared with a commercial RABV nucleoprotein (NP) mAb for assaying the VNA level of antirabies serum samples from different sources and testing the titer of RABV strains.
Appl Microbiol Biotechnol (2014) 98:1547–1555
Materials and methods Fermentation and preparation of bacteria expressing scFv98H in inclusion bodies Expression of the scFv98H plasmid in Escherichia coli BL21 (DE3) cells was carried out as described previously (Gu et al. 2011). The cells were grown in a complex auto-inducing medium, but with some modifications (10 g tryptone, 10 g yeast extract, 8.95 g Na2HPO4, 3.4 g KH2PO4, 0.71 g Na2SO4, 0.49 MgSO4·7H2O, 0.5 g glucose, 2 g lactose, 2.67 g NH4Cl, 5 ml glycerol per liter) (Studier 2005). An overnight seed culture was used to inoculate the 3-L bioreactor New Brunswick Scientific at 2 %. After fermentation, cells were harvested by centrifugation, and the paste was kept at −20 °C if not immediately used. The harvested cells were resuspended in TES buffer (50 mM Tris–HCl, pH 8.0, 5 mM EDTA, 150 mM NaCl) at 20 ml/g per gram (wet weight). Cells were lysed and centrifuged at 10,000×g for 30 min. The pellet containing inclusion bodies was washed with TE buffer (50 mM Tris–HCl, pH 8.0, 5 mM EDTA) containing 1 % Triton X-100, followed by washing with TEN buffer (50 mM Tris–HCl, pH 8.0, 5 mM EDTA, 2 M NaCl). Purification of inclusion bodies The inclusion bodies were dissolved in solubilization buffer (8 M urea, 50 mM Tris–HCl, pH 8.0) at the protein concentration of 5 mg/ml. The denatured proteins were then centrifuged at 10,000×g for 30 min, and the supernatant was filtered through a 0.22-μm membrane before purification. The denatured proteins were loaded at a flow rate of 1.5 ml/min onto a 5 ml His Trap HP Immobilized Metal Affinity Chromatography (IMAC) column (GE Healthcare) preequilibrated with buffer A (8 M urea, 50 mM Tris–HCl, pH 8.0, 0.5 M NaCl, 5 mM imidazole) in a protein purification system (AKTA purifier 100, GE Healthcare). After the protein loading, the column was washed extensively with buffer B (8 M urea, 50 mM Tris–HCl, pH 8.0, 0.5 M NaCl 60 mM imidazole) to remove unbound and nonspecifically bound proteins. The target protein was eluted with buffer C (8 M urea, 50 mM Tris–HCl, pH 9.0, 200 mM imidazole). The elution fractions were pooled and then purified by anion ion exchange chromatography (AIEX). This His-tagged protein was loaded at 2 ml/min and washed at 4 ml/min on an AIEX column (HiPrep™16/10 QXL) in buffer C without imidazole. The target protein was eluted with buffer D (8 M urea, 50 mM Tris–HCl, pH 9.0, 150 mM NaCl). The host protein and other impurities were eluted with buffer E (8 M urea, 50 mM Tris– HCl, pH 9.0, 1 M NaCl). At each step, fractions were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on a 13.5 % polyacrylamide gel and stained with Coomassie Brilliant Blue R-250 (Sigma-Aldrich).
Appl Microbiol Biotechnol (2014) 98:1547–1555
Protein refolding The purified protein was dialyzed in four times the volume with refolding buffer (50 mM Tris–HCl, pH 8.0, 10 % glycerin, 1 % glycine, 1 mM EDTA, 50 mM NaCl). The refolding process was carried out at 4 °C for 24 h with replacement of fresh buffer every 4 h without agitation. The refolded proteins were concentrated by ultrafiltration with a 5,000-Da molecular weight cut-off membrane (Millipore). The concentration of the protein was determined by the BCA method with a bovine serum albumin (BSA) standard. Purification and estimation of apparent molecular mass by gel filtration Gel filtration purification and determination of molecular mass were performed on an AKTA system (GE Healthcare) with a HiLoad 16/60 Superdex™ 75 prep grade column (GE Healthcare) pre-equilibrated with refolding buffer. The elution time was calibrated using the protein standards lysozyme (14 kDa), ovalbumin (44 kDa), and BSA (66 kDa). The protein size was analyzed after refolding. Indirect ELISA Ninety-six well ELISA plates were coated overnight at 4 °C with the inactivated RABV (IRABV) aG strain or an unrelated antigen, such as hepatitis B surface antigen (HBsAg), hepatitis A virus (HAV), or varicella-zoster virus (VZV), at the dilution of 1:100 in 100 μL 0.1 M carbonate bicarbonate buffer (pH 9.6). After three washes with PBS containing 0.05 % Tween 20 (PBST), the wells were blocked with 200 μL of PBST/3 % BSA for 2 h at 37 °C and washed as above. The NP mAb (Millipore) or scFv98H serially diluted in PBS was added to the plates in 100 μL volume per well and incubated at 37 °C for 1 h. After washing, a mouse anti-His antibody (Invitrogen) was added to wells (100 μl/well) previously incubated with scFv98H and maintained at 37 °C for 1 h. The plates were washed again, and the bound antibodies were detected using a horseradish peroxidase (HRP)-conjugated goat anti-mouse antibody (Jackson Immunoresearch) in PBST (100 μl/well) at 37 °C for 1 h. The substrate tetramethyl benzidine was subsequently added (100 μl/well) and incubated at 37 °C for 15 min. The reaction was stopped by adding 2 M H2SO4 (50 μl/well), and results were obtained by measuring the absorbance at 450 nm. Immunofluorescence staining Immunofluorescence staining was used to determine whether scFv98H can bind the RABV glycoprotein in cells. BSR cells (a cloned derivative of baby hamster kidney cells) infected with RABV or not (blank control) were added in 200 μL
1549
volume per well to 96-well microtiter plates and grown in DMEM (5 % FBS) at 37 °C in a 5 % CO2 incubator for 24 h. The cells were washed three times in PBS. Some wells were stained with a fluorescein isothiocyanate (FITC)-conjugated NP mAb (Millipore) in 20 μg/ml (50 μl/well). Other wells were incubated with scFv98H in 1 μg/ml (50 μl/well) followed by a mouse anti-His antibody in 1 μg/ml (50 μl/well) for 1 h and Alexa488-conjugated goat anti-mouse antibody in 1 μg/ml (50 μl/well) against the His-tag (Invitrogen) for 45 min in the dark. Subsequently, the cells were washed and fixed in ice-cold 80 % (100 μl/well) acetone for 10 min, followed by DAPI staining of the nuclei. After washing with PBS, the plates were observed under a microscope (Nikon ECLIPST Ti). Relative affinity and affinity assay The scFv98H protein was prepared at 0.01 mg/ml in dilution buffer (PBS, pH 7.5, 1 % BSA) and added to 96-well ELISA plates pre-coated with IRABV (aG strain). The plates were incubated for 1 h at 37 °C and washed with PBST five times. Thereafter, the protein was eluted with NH4SCN in PBS at concentrations ranging from 0 to 5 M (100 μl/well). The elution was performed on a shaker at 500 rpm for 1 h at 37 °C (MacDonald et al. 1988). The NH4SCN was washed from the plate with PBST five times, and the protein bound to IRABV was detected using rabbit anti-scFv98H and HRP-conjugated goat anti-rabbit antibodies (Jackson Immunoresearch). The equilibrium dissociation constant (K d) of scFv98H was measured by a noncompetitive method (Beatty et al. 1987). IRABV was coated on ELISA plates at various dilutions in bicarbonate coat buffer (pH 7.5) at 4 °C overnight, followed by blocking with 3 % BSA for 2 h at 37 °C. The protein was twofold serially diluted at concentrations from 1.7 nM to 3.5 μM. The scFv98H protein that remained bound to the IRABV was detected using rabbit anti-scFv98H and HRP-conjugated goat anti-rabbit antibodies. The K d of this antibody was calculated by previously described methods (Beatty et al. 1987). RABV neutralizing antibody assay According to the WHO, neutralizing antibody levels against RABV can be determined by the RFFIT (Smith et al. 1973). Here, scFv98H was evaluated in this test to determine whether it can be used to assay the neutralizing antibody titer against RABV. The national reference serum with verified potency (21.4 IU/ml) was used as a standard. All serum samples were heated at 56 °C for 30 min before testing in order to inactivate complement. In 96-well microtiter plates, two commercial RIGs, human RIG (HRIG) and equine RIG (ERIG), and two types of RABV antisera from experimental animals, mice (MRIG) and rabbits (RRIG), were threefold serially diluted
1550
in DMEM with 10 % inactivated bovine serum. Thereafter, the RABV challenge virus strain (CVS)-11 titrated to a dose causing 80 % infection of BSR cells was added, and the plates were incubated for 60 min in a humidified incubator at 37 °C with 5 % CO2. Following the incubation, 4×104 BSR cells were added per well, and the plate was incubated at 37 °C for 24 h. Finally, the culture supernatant was removed, and the monolayer cells were fixed with ice-cold 80 % acetone and incubated with a FITC-conjugated NP mAb or with scFv98H followed by a mouse anti-His antibody for 1 h and Alexa488conjugated goat anti-mouse antibody against the His tag for 45 min in the dark. The stained cells were observed under a fluorescence microscope, and the neutralization potency of the test antibodies were calculated against that of the national reference standard serum by the Reed and Muench method (Reed and Muench 1938). Utility of scFv98H for RABV titer determination In order to test whether scFv98H can be used as a diagnostic agent to determine RABV titers, four strains (CVS-11, aG, PM, and ERA) of RABV were used for direct immunofluorescence staining. CVS-11 is used for diagnostics, while the others are used as a vaccine for either humans (aG and PM) or animals (ERA) in China. BSR cells were inoculated in a 96-well plate with RABV serially diluted fivefold in DMEM with 10 % inactivated bovine serum and incubated at 34 °C for 2 days. After discarding the culture medium, the cells were fixed with 80 % acetone and then stained with the FITCconjugated NP mAb or with scFv98H followed by a mouse anti-His antibody for 1 h and Alexa488-conjugated goat antimouse antibody against the His tag for 45 min in the dark. Positive antigen foci were counted under a fluorescence microscope. Titrations were conducted in quadruplicate, and viral titers were calculated as fluorescent focus units per milliliter.
Appl Microbiol Biotechnol (2014) 98:1547–1555
culture medium contained sufficient nutrients to promote cell growth to a final OD600 value of approximately 30. Dissolved oxygen was kept at 30 % by adjusting the agitation and aeration rate. Samples were taken every hour for OD600 measurements. The fermentation biomass was harvested by centrifugation, and the cell pellets were stored at −20 °C. Protein expression was analyzed by SDS-PAGE with Coomassie Blue staining (Fig. 1). The fermentation biomass was lysed by sonication as described in “Materials and methods.” After the inclusion bodies were isolated, washed, and dissolved in 8 M urea buffer, the scFv98H protein was purified by IMAC. SDS-PAGE analysis of the product showed a major band corresponding to the expected molecular mass of scFv98H, which was highly purified (∼90 % purity) after one IMAC step (Fig. 2a). The pooled fractions were loaded onto the AIEX column (HiPrep™16/10 QXL) again, and the main peak fractions pooled from this step were considered as the highly purified (>95 % by host protein ELISA kit, data not shown) form of the unfolded scFv98H protein (Fig. 2b).
Purification of scFv98H The complete process of expression in auto-inducing medium and purification of scFv98H from bacteria is summarized in Fig. 3a. After the IMAC and AIEX purification steps, 1.2 g of scFv98H was obtained from 2 L of bacterial culture (data not shown). After renaturation, the pooled mixture of monomers, dimer, and aggregates of scFv98H was concentrated by ultrafiltration, and 2 ml of the sample was loaded onto an analytic Superdex 75 column. The size-exclusion chromatography (SEC) purification profile and SDS-PAGE analysis are shown in Fig. 3b. Ultimately, we obtained 540 mg of scFv98H monomers from 2 L of bacterial culture.
Statistical analysis Data were expressed as the mean ± standard deviation (SD) of values obtained from at least three independent experiments. Results were analyzed with the unpaired Student’s t test, and differences were considered to be statistically significant when P 0.05). These results indicate that scFv98H may be used reliably to assay VNA levels. RABV titering To characterize the refolded scFv98H protein, growth kinetics of the four strains of RABV were examined in BSR cells. CVS-11 is a RABV strain used for diagnostics, while the other RABV strains are used for vaccine manufacture in China. As shown in Fig. 5, no significant difference in titer (P >0.05) was observed between scFv98H and the FITC-conjugated NP mAb groups, indicating that scFv98H can be used to monitor replication of different RABV strains.
Discussion The Escherichia coli expression system is among the most popular for recombinant protein production due to its ease of genetic manipulation and scale-up, relatively short duration between transformation and protein purification, as well as its cost effectiveness (Arbabi-Ghahroudi et al. 2005; Demain and Vaishnav 2009; Ohlfest et al. 2012). Although it is preferable to express scFvs in a soluble and active form, the expression usually is low (Kramer et al. 2005; Lombardi et al. 2005; Muller et al. 1997), and most of the antibody fragments tend to be expressed in inclusion bodies (Jiang et al. 2013; Liu et al. 2011; Lombardi et al. 2005). However, expressing proteins in inclusion bodies, even though they are not in an active form, provides some advantages including high protein yield and ease of separating the insoluble form from the soluble impurities of E. coli.
Fig. 5 Comparison of scFv98H and NP mAb for monitoring RABV titers. a To test scFv98H, RABV (CVS-11 strain) was tenfold serially diluted, and the NP mAb was used as a control for immunofluorescence staining. Scale bar=200 μm. b ScFv98H was used to titer the CVS-11, aG, PM, and ERA virus strains during the replication process. The NP mAb was used as a positive control for determining the virus titer. Data are means ± SD from triplicate samples; P >0.05
In this study, scFv98H was expressed in Escherichia coli within insoluble inclusion bodies and therefore was easily separated from the host proteins or other soluble impurities by centrifugation. To remove residual host proteins and DNA, the inclusion bodies were extensively washed by TE buffers with Triton or NaCl. Fortunately, the inclusion bodies could be solubilized entirely in an inexpensive detergent (8 M urea). The recombinant scFv98H protein was produced with a His tag, which allowed purification by using IMAC (Fig. 2a). As the impurities, such as host proteins, lipopolysaccharides, and host DNA could bind tightly to the anion ion exchange medium, they were eluted only at a higher salt concentration (1 M NaCl). Meanwhile, the target protein could be easily eluted with a low salt concentration (150 mM NaCl), and unfolded scFv98H protein can be well separated from these impurities (Fig. 2b). Both IMAC and AIEX, which are highly selective chromatographic processes, were used for polishing steps in the purification of the unfolded scFv98H protein, providing a purity of more than 95 % as determined by the
1554
host protein ELISA kit (data not shown). Such a high purity of the unfolded protein facilitated the refolding process. The refolding of this antibody fragment protein was accomplished by dialysis in refolding buffer. However, it is generally a challenge to increase the yield of refolded proteins from inclusion bodies. Although the rate of recovery of refolded scFv98H was only ∼65 %, the product yield from inclusion bodies was relatively high comparing to other recombinant antibodies (André et al. 2013) at 0.6 g/L with 95 % purity (Fig. 2b). Thus, the high level of protein expression in inclusion bodies could compensate for any loss in recovery during purification (Singh and Panda 2005). Currently, our group is focused on developing an antibody against RABV for immunodetection or therapy, and scFvs against the RABV glycoprotein have been of special interest. Towards this aim, our group previously selected the scFv98H antibody screened from a phage display system (Kramer et al. 2005) and subsequently cloned and expressed it in a bacterial system (Gu et al. 2011). However, this engineered antibody was not characterized, and further development for diagnostic or therapeutic purposes was needed. To characterize this purified refolded scFv98H antibody, its specificity and binding activity were determined in this study. The refolded scFv98H protein was determined to specifically bind to IRABV (aG strain) but not other viruses tested (Fig. 4a). However, scFv98H could not recognize all strains of RABV. The parental antibody CR4098 is known to specifically target the G protein antigenic site III (Bakker et al. 2005). To provide coverage for all or most of the circulating RABVs, combining scFv98H with scFv57, which targets site I of the G protein, may be a good approach (Bakker et al. 2005; Duan et al. 2012). The relative affinity of scFv98H was 2.3 M, which of the disulfide-stabilized scFvs against RABV G protein (Duan et al. 2012) were between 2 and 3 M. The K d of scFv98H was 1×106 M−1, which of the scFvs (Duan et al. 2012) were from 5.65×105 to 1.78×106 M−1. The relative affinity and K d of scFv98H were found to be at the same levels as those of scFvs. However, differences in K d values reflected the reduced avidity of the scFv compared with the parental antibody CR4098, which was expected based on the fact that CR4098 is an IgG with two identical antigen recognition surfaces, while scFv98H retains only one (Yang et al. 2010). The antigen-specific binding of the recombinant scFv98H fragment was tested both in an ELISA and in infected cells. His tag used for purification also facilitated detection of the recombinant protein. Meanwhile, scFv98H was found to bind with RABV (CVS-11 strain) as well as the NP mAb in cells permeabilized by acetone. The RABV CVS-11 strain is a standard virus recommended by the WHO for use in neutralizing antibody assays (FAVN or RFFIT). Results from the current study indicated that scFv98H may be developed as a new diagnostic antibody to test RABV VNA levels in serum
Appl Microbiol Biotechnol (2014) 98:1547–1555
samples from different sources (Table 1). Differences in results between the two types of antibodies were not statistically significant, indicating that scFv98H is a promising candidate antibody for the RABV VNA assay. To further explore whether scFv98H can be used as a diagnostic tool to titer RABV for use in manufacturing vaccines or measuring VNAs, the CVS-11, aG, PM, and ERA strains were compared. Not only was scFv98H found to bind efficiently to RABV, but it also could be used to detect the virus titer during propagation, similar to the NP mAb (Fig. 5). With practical advantages compared with the commercial NP mAb, scFv98H may be used in potential applications as a novel diagnostic agent to test serum neutralizing antibodies, as well as to monitor virus titers during rabies vaccine production. Acknowledgments We gratefully acknowledge Thi Sarkis for editorial support in the preparation of this manuscript.
References André F, Michael H, Thomas S (2013) Expression of recombinant antibodies. Front Immunol 4:217. doi:10.3389/fimmu.2013.00217 Arbabi-Ghahroudi M, Tanha J, MacKenzie R (2005) Prokaryotic expression of antibodies. Cancer Metastasis Rev 24(4):501–519. doi:10. 1007/s10555-005-6193-1 Bakker ABH, Marissen WE, Kramer RA, Rice AB, Weldon WC, Niezgoda M, Hanlon CA, Thijsse S, Backus HHJ, De Kruif J (2005) Novel human monoclonal antibody combination effectively neutralizing natural rabies virus variants and individual in vitro escape mutants. J Virol 79(14):9062–9068. doi:10.1128/JVI.79.14. 9062-9068.2005 Beatty JD, Beatty BG, Vlahos WG (1987) Measurement of monoclonal antibody affinity by non-competitive enzyme immunoassay. J Immunol Methods 100(1–2):173–179 Bird RE, Hardman KD, Jacobson JW, Johnson S, Kaufman BM, Lee SM, Lee T, Pope SH, Riordan GS, Whitlow M (1988) Single-chain antigen-binding proteins. Science 242(4877):423–426. doi:10. 1126/science.3140379, New York, NY Bourhy H, Dautry-Varsat A, Hotez PJ, Salomon J (2010) Rabies, still neglected after 125 years of vaccination. PLoS Negl Trop Dis 4(11): e839. doi:10.1371/journal.pntd.0000839 Bunschoten H, Gore M, Ivo JTM, Uytdehaag FGCM, Dietzschold B, Wunner WH, Osterhaus ADME (1989) Characterization of a new virus-neutralizing epitope that denotes a sequential determinant on the rabies virus glycoprotein. J Gen Virol 70(2):291–298. doi:10. 1099/0022-1317-70-2-291 Cliquet F, Aubert M, Sagne L (1998) Development of a fluorescent antibody virus neutralisation test (FAVN test) for the quantitation of rabies-neutralising antibody. J Immunol Methods 212(1):79–87 Cliquet F, Sagne L, Schereffer J, Aubert M (2000) ELISA test for rabies antibody titration in orally vaccinated foxes sampled in the fields. Vaccine 18(28):3272–3279. doi:10.1016/S0264-410X(00)00127-4 Coulon P, Rollin P, Aubert M, Flamand A (1982) Molecular basis of rabies virus virulence. I. Selection of avirulent mutants of the CVS strain with anti-G monoclonal antibodies. J Gen Virol 61(1):97–100. doi:10.1099/0022-1317-61-1-97 Demain AL, Vaishnav P (2009) Production of recombinant proteins by microbes and higher organisms. Biotechnol Adv 27(3):297–306. doi:10.1016/j.biotechadv.2009.01.008
Appl Microbiol Biotechnol (2014) 98:1547–1555 Duan Y, Gu T, Jiang C, Yuan R, Zhang H, Hou H, Yu X, Chen Y, Zhang Y, Wu Y (2012) A novel disulfide-stabilized single-chain variable antibody fragment against rabies virus G protein with enhanced in vivo neutralizing potency. Mol Immunol 51(2):188–196. doi:10. 1016/j.molimm.2012.03.015 Feyssaguet M, Dacheux L, Audry L, Compoint A, Morize J, Blanchard I, Bourhy H (2007) Multicenter comparative study of a new ELISA, PLATELIA™ RABIES II, for the detection and titration of antirabies glycoprotein antibodies and comparison with the rapid fluorescent focus inhibition test (RFFIT) on human samples from vaccinated and non-vaccinated people. Vaccine 25(12):2244–2251. doi: 10.1016/j.vaccine.2006.12.012 Gu T, Wei W, Duan Y, Jiang C, Chen Y, Yu X, Wu J, Wu Y, Kong W (2011) Identification of binding epitope for anti-rabies virus glycoprotein single-chain Fv fragment FV57. Protein Pept Lett 18(11): 1099–1106 Hanlon C, DeMattos C, DeMattos C, Niezgoda M, Hooper D, Koprowski H, Notkins A, Rupprecht C (2001) Experimental utility of rabies virus-neutralizing human monoclonal antibodies in post-exposure prophylaxis. Vaccine 19(28):3834–3842. doi:10.1016/S0264410X(01)00135-9 Humphreys D, Glover D (2001) Therapeutic antibody production technologies: molecules, applications, expression and purification. Curr Opin Drug Discov Dev 4(2):172–185 Jiang H, Xie Y, Burnette A, Roach J, Giardina SL, Hecht TT, Creekmore SP, Mitra G, Zhu J (2013) Purification of clinical-grade disulfide stabilized antibody fragment variable—Pseudomonas exotoxin conjugate (dsFv-PE38) expressed in Escherichia coli . Appl Microbiol Biotechnol:621–632. doi: 10.1007/s00253-012-4319-2 Knobel DL, Cleaveland S, Coleman PG, Fèvre EM, Meltzer MI, Miranda MEG, Shaw A, Zinsstag J, Meslin FX (2005) Re-evaluating the burden of rabies in Africa and Asia. B World Health Organ 83(5): 360–368. doi:10.1590/S0042-96862005000500012 Kramer RA, Marissen WE, Goudsmit J, Visser TJ, Clijsters-Van der Horst M, Bakker AQ, de Jong M, Jongeneelen M, Thijsse S, Backus HHJ (2005) The human antibody repertoire specific for rabies virus glycoprotein as selected from immune libraries. Eur J Immunol 35(7):2131–2145. doi:10.1002/eji.200526134 Liu X, Lin H, Tang Q, Li C, Yang S, Wang Z, Wang C, He Q, Cao B, Feng Z (2011) Characterization of a human antibody fragment Fab and its calcium phosphate nanoparticles that inhibit rabies virus infection with vaccine. PLoS One 6(5):e19848. doi:10.1371/ journal.pone.0019848 Lombardi A, Sperandei M, Cantale C, Giacomini P, Galeffi P (2005) Functional expression of a single-chain antibody specific for the HER2 human oncogene in a bacterial reducing environment. Protein Expr Purif 44(1):10–15. doi:10.1016/j.pep.2005.05.013 MacDonald RA, Hosking CS, Jones CL (1988) The measurement of relative antibody affinity by ELISA using thiocyanate elution. J
1555 Immunol Methods 106(2):191–194. doi:10.1016/0022-1759(88) 90196-2 Marissen WE, Kramer RA, Rice A, Weldon WC, Niezgoda M, Faber M, Slootstra JW, Meloen RH, Clijsters-van der Horst M, Visser TJ (2005) Novel rabies virus-neutralizing epitope recognized by human monoclonal antibody: fine mapping and escape mutant analysis. J Virol 79(8):4672–4678. doi:10.1128/JVI.79.8.4672-4678.2005 Moore SM, Hanlon CA (2010) Rabies-specific antibodies: measuring surrogates of protection against a fatal disease. PLoS Negl Trop Dis 4(3):e595. doi:10.1371/journal.pntd.0000595 Muller BH, Lafay F, Demangel C, Perrin P, Tordo N, Flamand A, Lafaye P, Guesdon JL (1997) Phage-displayed and soluble mouse scFv fragments neutralize rabies virus. J Virol Methods 67(2):221–233. doi:10.1016/S0166-0934(97)00099-2 Ni Y, Tominaga Y, Honda Y, Morimoto K, Sakamoto S, Kawai A (1995) Mapping and characterization of a sequential epitope on the rabies virus glycoprotein which is recognized by a neutralizing monoclonal antibody, RG719. Med Microbiol Immunol 39(9):693 Ohlfest JR, Zellmer DM, Panyam J, Swaminathan SK, Oh S, Waldron NN, Toma S, Vallera DA (2012) Immunotoxin targeting CD133+ breast carcinoma cells. Drug Deliv and Transl Res: 195–204 Prehaud C, Coulon P, Lafay F, Thiers C, Flamand A (1988) Antigenic site II of the rabies virus glycoprotein: structure and role in viral virulence. J Virol 62(1):1–7 Prosniak M, Faber M, Hanlon CA, Rupprecht CE, Hooper DC, Dietzschold B (2003) Development of a cocktail of recombinantexpressed human rabies virus-neutralizing monoclonal antibodies for postexposure prophylaxis of rabies. J Infect Dis 188(1):53–56. doi:10.1086/375247 Reed LJ, Muench H (1938) A simple method of estimating fifty percent endpoints. Am J Epidemiol 27(3):493–497 Schumacher CL, Dietzschold B, Ertl H, Niu HS, Rupprecht CE, Koprowski H (1989) Use of mouse anti-rabies monoclonal antibodies in postexposure treatment of rabies. J Clin Invest 84(3):971–975. doi:10.1172/JCI114260 Singh SM, Panda AK (2005) Solubilization and refolding of bacterial inclusion body proteins. J Biosci Bioeng 99(4):303–310. doi:10. 1263/jbb.99.303 Smith JS, Yager PA, Baer GM (1973) A rapid reproducible test for determining rabies neutralizing antibody. Bull World Health Organ 48(5):535–541 Studier FW (2005) Protein production by auto-induction in high-density shaking cultures. Protein Expr Purif 41(1):207–234. doi:10.1016/j. pep.2005.01.016 Yang J, Chen R, Wei J, Zhang F, Zhang Y, Jia L, Yan Y, Luo W, Cao Y, Yao L (2010) Production and characterization of a recombinant single-chain antibody against Hantaan virus envelop glycoprotein. Appl Microbiol Biotechnol 86(4):1067–1075. doi:10.1007/s00253009-2379-8
