Letters in Applied Microbiology ISSN 0266-8254
ORIGINAL ARTICLE
Human monoclonal single-chain antibodies specific to dengue virus envelope protein N. Saokaew1,2, O. Poungpair1, A. Panya1,3, M. Tarasuk1, N. Sawasdee1, T. Limjindaporn4, W. Chaicumpa5 and P. Yenchitsomanus1 1 Division of Molecular Medicine, Department of Research and Development, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand 2 Graduate Program in Immunology, Department of Immunology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand 3 Graduate Program in Biochemistry, Department of Biochemistry, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand 4 Department of Anatomy, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand 5 Laboratory for Research and Technology Development, Department of Parasitology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand
Significance and Impact of the Study: No approved vaccine and specific drug for dengue virus (DENV) infection are available; thus, their developments are urgently required. The human single-chain variable antibody fragments (HuScFv) specific to DENV envelope (E) protein are potential to be developed as therapeutic biomolecules. HuScFv that bound specifically to recombinant full-length DENV E (FL-E) and its domain III (EDIII) were generated and testified for its inhibitory effect in DENV infection. EDIII-specific HuScFv inhibited DENV infection in a dose-dependent manner and has potential to be further developed as a therapeutic biomolecule for DENV infection.
Keywords dengue virus, envelope protein, human single-chain variable fragments, monoclonal antibody, phage display, virus entry. Correspondence Pa-thai Yenchitsomanus, Division of Molecular Medicine, Department of Research and Development, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, Thailand. E-mail: [email protected] 2013/0267: received 13 February 2013, revised 20 October 2013 and accepted 23 October 2013 doi:10.1111/lam.12186
Abstract Dengue virus (DENV) infection is an arthropod-borne disease with increasing prevalence worldwide. Attempts have been made to develop therapeutic molecules for treatment for DENV infection. However, most of potentially therapeutic DENV monoclonal antibody was originated from mouse, which could cause undesirable effects in human recipients. Thus, fully human antibody is preferable for therapeutic development. Human single-chain variable fragments (HuScFv) with inhibitory effect to DENV infection were generated in this study. HuScFv molecules were screened and selected from the human antibody phage display library by using purified recombinant DENV full-length envelope (FL-E) and its domain III (EDIII) proteins as target antigens for biopanning. HuScFv molecules were then tested for their bindings to DENV particles by indirect ELISA and immunofluorescent microscopy. EDIII-specific HuScFv exhibited neutralizing effect to DENV infection in Vero cells in a dose-dependent manner as determined by plaque formation and cell ELISA. Epitope mapping and molecular docking results concordantly revealed interaction of HuScFv to functional loop structure in EDIII of the DENV E protein. The neutralizing HuScFv molecule warrants further development as a therapeutic biomolecule for DENV infection.
Introduction Dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS) are caused by dengue virus (DENV) infection which is transmitted from human to human by Aedes mosquito (Martina et al. 2009; Murrell et al. 2011). 270
More than one-third of the world population is living in endemic areas, and 100 million people are infected yearly (Guzman et al. 2010). Importantly, neither approved vaccine nor antiviral drug is currently available. DENV is a member of Flaviviridae family, and its genome encodes three structural (capsid, premembrane/membrane and
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
N. Saokaew et al.
envelop) and seven nonstructural (NS1, NS2A, NS2B, NS3, NS4A, NS4B and NS5) proteins. The most outer part of the mature virion is covered by envelope (E) homodimer complexes, responsible for receptor-mediated attachment to host cells, and is a major target for neutralizing antibodies, especially at its domain III (EDIII) (Chang 1997; Chen et al. 1997; Hung et al. 1999; Allison et al. 2001). Passive immunization or antibody administration has been well accepted in medical practice. Various types of antibodies against viral and bacterial infection were tested in vitro, in animal models and in clinical trials (Heijtink et al. 2001; Domanski et al. 2005; Taylor et al. 2008). More recently, humanized monoclonal antibody specific to E protein of West Nile virus (WNV) has been used in phase I clinical trial. This trial demonstrated that only single dose of antibody was sufficient for WNV-neutralizing activity with safety and well tolerance in recipients (Beigel et al. 2010). Thus, the aim of this study is to produce HuScFv specific to DENV E protein, an essential molecule for virion assembly and virus entry, with inhibitory effect on DENV infection. Recombinant FL-E and EDIII proteins were generated and used as target antigens for selection of HuScFv molecules from the human antibody phage display library. The inhibitory effect of EDIII-specific HuScFv to cellular infection of DENV2 has been demonstrated. Mimotope searching and computational studies also emphasized interaction of HuScFv and EDIII region on DENV E protein. Results and discussion Production of recombinant DENV E proteins Fl-e and edIII were amplified at sizes of 1341 and 348 bp, respectively (Fig. S1a). Polyhistidine-tagged rFL-E (56 kDa) and rEDIII (17 kDa) were produced from recombinant plasmid-transformed E. coli (Fig. S1b) and purified by affinity chromatography (Fig. S1c). Mouse polyclonal antibodies generated by immunization with these proteins could readily recognize both recombinant and native DENV E protein (data not shown) indicating that the recombinant proteins retain inherent antigenic epitopes suitable for selection of antibodies specific to DENV E in the native form. Selection of phage clones displaying E protein-specific HuScFv After biopanning with rFL-E and rEDIII, there were 74 and 228 huscfv-carrying E. coli clones, respectively (Fig. S2a). Among them, 28 and 79 clones could produce soluble monoclonal HuScFv in different sizes (26–30 kDa)
HuScFv specific to DENV E protein
(Fig. S2b). This result from the variation in the length of immunoglobulin genes during the library construction step (Kulkeaw et al. 2009). HuScFv were tested for their binding activities to native DENV E proteins by indirect ELISA. We selected HuScFv clone numbers 15A for further analysis because it had highest relative binding ratio (although slightly higher than other clones) as screened by indirect ELISA (Fig. 1a), they could interact with native DENV E protein in DENV-infected Vero cells as detected by IFA (Fig. 1b) and they had a high yield of ‘metal chelate’ (not E protein ligand) affinity purification. HuScFv–DENV binding activity Huscfv of clone no. 15A was subcloned into modified pET23b(+) vector for polyhistidine-tagged HuScFv production and purification (Fig. S2c). This HuScFv could interact with native DENV E protein in DENV-infected Vero cells as detected by IFA. Green fluorescence of HuScFv interacting with E protein was visualized throughout cytoplasm of infected cells, while MOCK-infected cells showed no signal (Fig. 1b) indicating that, despite selected against recombinant antigens, the HuScFv readily recognized native E protein. As HuScFv is monoclonal, the fluorescent signal was not expectably as strong as in polyclonal anti-E positive antibody control. This finding also conformed to indirect ELISA result. Neutralizing activity of HuScFv in DENV-infected cells E protein is crucial for DENV binding to host-target cells and mediates the first step of cellular infection. Thus, the ultimate goal of this study is to develop HuScFv specific to E protein and interfere DENV infection. In plaque formation assay (Fig. 2a), which the number of infected cells was observed, DENV2 pre-incubated with HuScFv no. 15A (EDIII-specific) showed significant reduction in plaque number in a dose-dependent manner with the highest inhibition was seen at concentration of 200 ng ll 1. Positive control antibody (4G2)-treated DENV markedly reduced plaque number. After DENV infection, the viral antigen in Vero cells was also determined by cell ELISA using anti-E antibody (4G2) as detecting control (Fig. 2b). ELISA signal from Vero cells infected with HuScFv no.15A-treated DENV2 was significantly lower than signal in cells infected with untreated DENV or controls. HuScFv no. 15A was also demonstrated for its cross-reactivity to DENV1 and DENV4 by cell ELISA (Fig. 2c). At 200 ng ll 1 concentration, HuScFv no. 15A could significantly reduce E antigen in Vero cells infected DENV serotypes 1, 2 and 4 for 40–60% but not in DENV3-infected cells. The explanation for inability of 15A to neutralize DENV-3 is that the experiment was
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
271
N. Saokaew et al.
HuScFv specific to DENV E protein
10·37
1·5
1·0
0·5
0·0
24 3A 4A 5A 7A 9A 11A 13A 17A 21A 25A 26A 33A 35A 36A 39A 2B 3B 7B 9B 11B 12B 13B 14B 15B 17B 18B 86 4A 6A 8A 13A 15A 18A 20A 22A 25A 26A 35A 37A 44A 46A 49A 115A 6B 11B 12B 13B 18B 19B 21B 32B 35B 56B 58B HB2151 PBS mαrFLE
Relative binding ratio
(a) 2·0
Clone number (b)
MPAb
HuScFv15A
MPAb
HuScFv15A
AlexaFluor488
Merged with Hoechst MOCK-infected cells
DENV-infected cells
Figure 1 Interaction of HuScFv with E protein. (a) Monoclonal HuScFv were tested for their binding activity to native E protein on DENV particles in indirect ELISA. Non-HuScFv preparation (HB2151) and mouse polyclonal anti-DENV E protein antibody (marFLE) were used as negative and positive antibody controls, respectively. Relative binding ratio of each particular HuScFv clone was calculated by OD420nm of HuScFv–DENV particle divided by OD420nm of HuScFv-negative antigen control. (b) Purified HuScFv were tested for their interactions to E protein in DENV-infected Vero cells. Mouse polyclonal anti-E protein antibody (1 : 3000) was used as positive antibody control, while MOCK-infected cells stained with identical reagents served as negative control. All slides were visualized at 409 magnification.
carried out with only one dilution, and the concentration may need to be increased and optimized for testing against a different serotype. Further analysis of HuScFv epitope on EDIII of DENV3 may clarify its inability to inhibit DENV3 infection. However, these findings support the results of previous studies which indicate the importance of EDIII in DENV entry and infection (Chin et al. 2007; Abd-Jamil et al. 2008; Watterson et al. 2012). HuScFv–DENV EDIII interaction In epitope mapping using the 12-mer peptide phage display library, consensus peptide sequence aligned with DENV EDIII protein revealed that there are three regions on EDIII responsible for interaction with HuScFv no. 272
15A. These regions include residues 319–332 on the B strand-BC loop, residues 340–358 in C-C’ loop-C’ strand and residues 374–391 on F strand-FG loop (Fig. 3a). In molecular docking, the selected model (anti-SAR ScFv) had 65% sequence identity and 75% similarity to HuScFv no. 15A. The Ramachandran plot revealed that the modelled HuScFv no. 15A carried 18% of residues in disallowed regions. Molecular docking revealed that HuScFv no.15A-EDIII interaction was contributed by CDRs of HuScFv and three mapped regions on EDIII with lowbinding energy ( 9459 kcal mol 1) implying that the interaction could occur spontaneously. CDR1, 2 and three of VH occupied the C-C’ loop (residues 340–346) and FG loop (residues 380–385) of EDIII, whereas CDR1 and two of VL interacted with N-terminal region (residues
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
N. Saokaew et al.
HuScFv specific to DENV E protein
Relative infection (%)
(a) 100 80
**
60
**
40
***
***
20 0
PBS
HuScFv15A
4G2
Irrelevant HuScFv
Relative DENV E protein (%)
(b) 100 80
* *
60 40 *** 20 0 PBS
Relative DENV E protein (%)
(c)
4G2
HuScFv15A
DENV1
Irrelevant HuScFv DENV3
DENV2
DENV4
100 80 ** 60
* ***
***
40
*** ***
***
***
20
PB S el HuS 4 ev c G an Fv 2 t H 15 uS A cF v Irr
PB Irr H S el uS 4 ev c G an Fv 2 t H 15 uS A cF v
PB S el HuS 4 ev c G an Fv 2 t H 15 uS A cF v Irr
Irr
PB S el HuS 4 ev c G an Fv 2 t H 15 uS A cF v
0
Figure 2 HuScFv inhibitory effect to DENV infection. (a) DENV2 was pre-incubated with either 4G2 positive antibody control ( ), HuScFv15A and irrelevant HuScFv at various concentrations, for example, 8 ng ll 1 ( ), 40 ng ll 1 ( ), 200 ng ll 1 ( ) or antibody diluents (PBS, ) before allowed to infect Vero cells and determined per cent of infection by plaque assay. (b) Viral E antigen produced in infected cells after 48 h of infection was also measured by cell ELISA. (c) Antibody diluents ( ), 4G2 positive antibody control ( ), HuScFv15A ( ), or irrelevant HuScFv ( ) was incubated with DENV serotypes 1–4 (DENV1–4) before adding to Vero cell monolayer. Cytoplasmic DENV E antigen in infected cells was determined by cell ELISA after 48 h of infection. Asterisks indicate significant differences from three independent experiments (*P < 01, **P < 001, ***P < 0001).
299–305) and BC loop (residues 327–332) of DENV EDIII, respectively (Fig. 3b,c). The model also revealed that the residues recognized by HuScFv no. 15A locate at the outer boundary and are accessible in the natural dimeric DENV E protein (Fig. 3d). Interestingly, previous studies have
reported BC, CC’ and FG loops as targets of neutralizing antibodies (Sukupolvi-Petty et al. 2007, 2010). Many antibodies with strong DENV-neutralizing activity also recognized epitopes on EDIII suggesting that these regions are critical for antibody-mediated neutralization.
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
273
N. Saokaew et al.
HuScFv specific to DENV E protein
(a)
A-strand 295
(b)
B-strand
N-terminal region
AB loop
BC loop
335 VL
VH C’-strand C-strand C-C’ loop
336
F-strand
D-strand E-strand DE loop
370
G-strand FG loop
371
395
EDIII (c)
N-terminal region
(d)
BC loop FG loop A
C-C’ loop C F
E G B
Figure 3 Mapping HuScFv epitope on EDIII. (a) Sequence homology of HuScFv mimotopes (M1-12) to DENV2 EDIII was analysed using ClustalW program. Epitope of HuScFv clone no. 15A was found to span on 3 regions of EDIII molecule (B strand-BC loop, C-C’ loop-C’ strand and F strand-FG loop). (b) Interaction between HuScFv clone no. 15A (VH coloured in brown and VL in green) and EDIII (coloured in magenta) was performed using the ZDOCK program. (c) Amino acid residues on EDIII recognized by HuScFv clone no. 15A were considered as the epitope highlighted in yellow. (d) The localization of the epitope on EDIII recognized by HuScFv clone no. 15A (displayed as yellow spheres) was revealed and accessible in dimeric E proteins. EDI and EDII were displayed in light green. (PDB entry: 1OAN).
From this study, mechanisms by which HuScFv no. 15A inhibits DENV infection could possibly be at the step of host cell binding (Lok et al. 2008) or structural transition in the fusion step (Stiasny et al. 2007), which requires further investigations. The neutralizing HuScFv warrants further development as a research reagent or therapeutic biomolecule for DENV-infected patients. Materials and methods
Hawaii), serotype 2 (strain 16681), serotype 3 (strain H87) and serotype 4 (strain H241) were propagated in Aedes albopictus C6/36 cell line (CRL-1660, ATCC, USA) in 28°C incubator. Virus in culture supernatant was titrated by focus formation assay and kept at 80°C for neutralization test. The culture supernatant containing DENV2 viral particles was also subjected to partial purification as previously described (Gromowski and Barrett 2007) for further use in ELISA.
Dengue virus and cell culture
Production of recombinant DENV E proteins
Vero cells (CCL-81, ATCC, USA) were maintained at 37°C in a 5% CO2 incubator. DENV serotype 1 (strain
DENV RNA was extracted from culture supernatant of DENV2-infected C6/36 cells and subjected to cDNA
274
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
N. Saokaew et al.
synthesis. Fl-e and edIII were amplified by PCR using the primers shown in Table S1. After transformation, E. coli carrying verified recombinant plasmids were induced for the production of polyhistidine-tagged rFL-E and rEDIII for further purification using TALONTM Metal Affinity Resin (Clonetech, Mountain View, CA, USA). Phage biopanning DENV E-specific HuScFv were selected from the human antibody phage display library by using purified rFL-E and rEDIII as target antigens (Kulkeaw et al. 2009). The phagemid-transformed bacterial clones were screened for the presence of huscfv by PCR using phagemid-specific primers. Huscfv-positive clones were induced for the production of E-tagged HuScFv as detected by Western blot analysis using rabbit anti-E tag polyclonal antibody (1 : 5000). Indirect ELISA Indirect ELISA was performed to determine binding activity of monoclonal HuScFv to native viral protein as previously described (Poungpair et al. 2009). Partially, purified DENV2 particles (500 ng) and the preparation from MOCK-infected cells were used as antigen and no-antigen control, respectively. DENV2-bound HuScFv was detected with rabbit anti-E tag antibody and HRPconjugated swine anti-rabbit Ig followed by OPD substrate and OD492nm measurement. The relative binding ratio of each HuScFv clone was calculated by dividing the OD492nm of HuScFv interacting with DENV particle by OD492nm of HuScFv interacting with MOCK preparation. Preparation of polyhistidine-tagged HuScFv Huscfv sequences of ELISA-positive clones were subcloned into modified pET23b(+) vector (Poungpair et al. 2010). Transformed BL21 (DE3) E. coli was induced for the production of polyhistidine-tagged HuScFv which was purified by using TALONTM Metal Affinity Resin. Immunofluorescent assay To determine binding specificity of HuScFv to DENV E protein, DENV2-infected Vero cells were fixed, permeabilized and incubated with polyhistidine-tagged HuScFv. DENV E protein-interacting HuScFv was revealed by mouse anti-polyhistidine tag (1 : 5000) and AlexaFluor488-conjugated goat anti-mouse IgG (1 : 1000). Fluorescent signal was visualized by fluorescence Nikon microscope (409 microscopic field).
HuScFv specific to DENV E protein
Plaque assay Dilutions of either purified DENV2 E-specific HuScFv (clone no. 15A), irrelevant HuScFv, mouse monoclonal anti-E (4G2) antibody (positive antibody control at dilution 1 : 1250) or antibody diluents (PBS) were mixed with DENV2 (25 PFU) and incubated at 4°C for 2 h. The mixtures were separately added to Vero cell monolayer and incubated at 37°C for 1 h. Unbound virus was removed, and 12% carboxymethyl cellulose (CMC)-containing growth medium was added. After 7 days, cells were stained with 005% crystal violet, and clear plaques were counted under light microscope. Cell ELISA Either HuScFv at various concentrations or controls was incubated with DENV (1000 PFU) at 4°C for 2 h. The mixtures were added to Vero cell monolayer and incubated at 37°C for 1 h to allow infection; then, unabsorbed virus was removed, and growth medium was added and incubated further for 48 h. Intracellular DENV E protein was detected by mouse monoclonal anti-DENV E protein antibody (4G2) followed by HRP-conjugated rabbit antimouse Ig (1 : 1000). Finally, OPD substrate was added, and OD492nm was determined. Relative neutralizing activity was calculated by comparing OD value from cells infected with HuScFv-treated DENV to those infected with untreated DENV. In cross-neutralization experiment, the same method was used with only a single concentration of HuScFv (200 ng ll 1). Mimotope searching Phage clones displaying peptides that bound to EDIII-specific HuScFv (mimotope) were screened and selected from the PhD-12TM Phage Display Peptide Library (New England Biolabs) as previously described (Poungpair et al. 2009). The 12-mer peptide insert of each phage clone was deduced. Consensus sequences were obtained by multiple alignments of the deduced peptides using ClustalW program. The HuScFv mimotope was determined by alignment of consensus sequences to DENV E protein sequence. Homology modelling and molecular docking The reported structure with high homology to HuScFv, anti-SAR ScFv (PDB entry 2GHW), was identified by BLAST search (DS server) protocol, Discovery Studio 2.5 package (Accelrys Inc., San Diego, CA, USA) against a local sequence database. The multiple sequence alignment between the anti-SAR ScFv and HuScFv was conducted
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
275
N. Saokaew et al.
HuScFv specific to DENV E protein
by ClustalW program. The light and heavy chains of the anti-SAR ScFv were used as a modelled template for the selected HuScFv. Deposited structure of soluble DENV E protein (PDB entry 1OAN) was used for molecular docking. Models of HuScFv and DENV EDIII protein were built by MODELER program and assessed by PROCHECK (Laskowski et al. 1993). HuScFv was docked to DENV EDIII using the rigid body algorithm ZDOCK program. The poses from ZDOCK analysis were further processed to refine docked complex using a CHARM-minimization RDOCK program. The significant intermolecular interactions were rendered by using PyMOL software (The PyMOL Molecular Graphics System, DeLano Scientific, San Carlos, CA, USA). Acknowledgements This work was supported by Mahidol University Research Grant. Ni.S. is supported by Siriraj Graduate Scholarship. P.Y. is a Senior Research Scholar of Thailand Research Fund (TRF). O.P. is recipient of the TRF Young Researcher Grant (No.TRG5480006). A.P. is a recipient of TRF-Royal Golden Jubilee (RGJ) Ph.D. Scholarship. W.C. is supported by the research grant under the National Research University (NRU) Project, Office of Higher Education Commission (OHEC). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors thank Dr. Chanya Putthikhant for providing mouse monoclonal anti-DENV E protein. The human antibody phage display library used in this study is the property of the National Research Council of Thailand (NRCT). Conflict of interest No conflict of interest declared. References Abd-Jamil, J., Cheah, C.Y. and AbuBakar, S. (2008) Dengue virus type 2 envelope protein displayed as recombinant phage attachment protein reveals potential cell binding sites. Protein Eng Des Sel 21, 605–611. Allison, S.L., Schalich, J., Stiasny, K., Mandl, C.W. and Heinz, F.X. (2001) Mutational evidence for an internal fusion peptide in flavivirus envelope protein E. J Virol 75, 4268–4275. Beigel, J.H., Nordstrom, J.L., Pillemer, S.R., Roncal, C., Goldwater, D.R., Li, H., Holland, P.C., Johnson, S. et al. (2010) Safety and pharmacokinetics of single intravenous dose of MGAWN1, a novel monoclonal antibody to West Nile virus. Antimicrob Agents Chemother 54, 2431–2436.
276
Chang, K.J. (1997) Studies on the serological cross-reaction between dengue and Japanese encephalitis. Zhonghua Min Guo Wei Sheng Wu Ji Mian Yi Xue Za Zhi 30, 207–218. Chen, Y., Maguire, T., Hileman, R.E., Fromm, J.R., Esko, J.D., Linhardt, R.J. and Marks, R.M. (1997) Dengue virus infectivity depends on envelope protein binding to target cell heparan sulfate. Nat Med 3, 866–871. Chin, J.F., Chu, J.J. and Ng, M.L. (2007) The envelope glycoprotein domain III of dengue virus serotypes 1 and 2 inhibit virus entry. Microbes Infect 9, 1–6. Domanski, P.J., Patel, P.R., Bayer, A.S., Zhang, L., Hall, A.E., Syribeys, P.J., Gorovits, E.L., Bryant, D. et al. (2005) Characterization of a humanized monoclonal antibody recognizing clumping factor A expressed by Staphylococcus aureus. Infect Immun 73, 5229–5232. Gromowski, G.D. and Barrett, A.D. (2007) Characterization of an antigenic site that contains a dominant, type-specific neutralization determinant on the envelope protein domain III (ED3) of dengue 2 virus. Virology 30, 349–360. Guzman, M.G., Halstead, S.B., Artsob, H., Buchy, P., Farrar, J., Gubler, D.J., Hunsperger, E., Kroeger, A. et al. (2010) Dengue: a continuing global threat. Nat Rev Microbiol 8, S7–S16. Heijtink, R.A., van Nunen, A.B., van Bergen, P., Ostberg, L., Osterhaus, A.D. and de Man, R.A. (2001) Administration of a human monoclonal antibody (TUVIRUMAB) to chronic hepatitis B patients pre-treated with lamivudine: monitoring of serum TUVIRUMAB in immune complexes. J Med Virol 64, 427–434. Hung, S.L., Lee, P.L., Chen, H.W., Chen, L.K., Kao, C.L. and King, C.C. (1999) Analysis of the steps involved in Dengue virus entry into host cells. Virology 257, 156–167. Kulkeaw, K., Sakolvaree, Y., Srimanote, P., Tongtawe, P., Maneewatch, S., Sookrung, N., Tungtrongchitr, A., Tapchaisri, P. et al. (2009) Human monoclonal ScFv neutralize lethal Thai cobra, Naja kaouthia, neurotoxin. J Proteomics 72, 270–282. Laskowski, R.A., Macarthur, M.W., Moss, D.S. and Thornton, J.M. (1993) PROCHECK: a program to check the stereochemical quality of protein structures. J Appl Cryst 26, 283–290. Lok, S.M., Kostyuchenko, V., Nybakken, G.E., Holdaway, H.A., Battisti, A.J., Sukupolvi-Petty, S., Sedlak, D., Fremont, D.H. et al. (2008) Binding of a neutralizing antibody to dengue virus alters the arrangement of surface glycoproteins. Nat Struct Mol Biol 15, 312–317. Martina, B.E., Koraka, P. and Osterhaus, A.D. (2009) Dengue virus pathogenesis: an integrated view. Clin Microbiol Rev 22, 564–581. Murrell, S., Wu, S.C. and Butler, M. (2011) Review of dengue virus and the development of a vaccine. Biotechnol Adv 29, 239–247. Poungpair, O., Chaicumpa, W., Kulkeaw, K., Maneewatch, S., Thueng-in, K., Srimanote, P., Tongtawe, P., Songserm, T.
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
N. Saokaew et al.
et al. (2009) Human single chain monoclonal antibody that recognizes matrix protein of heterologous influenza A virus subtypes. J Virol Methods 159, 105–111. Poungpair, O., Pootong, A., Maneewatch, S., Srimanote, P., Tongtawe, P., Songserm, T., Tapchaisri, P. and Chaicumpa, W. (2010) A human single chain transbody specific to matrix protein (M1) interferes with the replication of influenza A virus. Bioconjug Chem 21, 1134–1141. Stiasny, K., Brandler, S., Kossl, C. and Heinz, F.X. (2007) Probing the flavivirus membrane fusion mechanism by using monoclonal antibodies. J Virol 81, 11526–11531. Sukupolvi-Petty, S., Austin, S.K., Purtha, W.E., Oliphant, T., Nybakken, G.E., Schlesinger, J.J., Roehrig, J.T., Gromowski, G.D. et al. (2007) Type- and subcomplexspecific neutralizing antibodies against domain III of dengue virus type 2 envelope protein recognize adjacent epitopes. J Virol 81, 12816–12826. Sukupolvi-Petty, S., Austin, S.K., Engle, M., Brien, J.D., Dowd, K.A., Williams, K.L., Johnson, S., Rico-Hesse, R. et al. (2010) Structure and function analysis of therapeutic monoclonal antibodies against dengue virus type 2. J Virol 84, 9227–9239. Taylor, C.P., Tummala, S., Molrine, D., Davidson, L., Farrell, R.J., Lembo, A., Hibberd, P.L., Lowy, I. et al. (2008) Open-label, dose escalation phase I study in healthy volunteers to evaluate the safety and pharmacokinetics of a human monoclonal antibody to Clostridium difficile toxin A. Vaccine 26, 3404–3409.
HuScFv specific to DENV E protein
Watterson, D., Kobe, B. and Young, P.R. (2012) Residues in domain III of the dengue virus envelope glycoprotein involved in cell-surface glycosaminoglycan binding. J Gen Virol 93, 72–82.
Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1. Primers for PCR amplification of fl-e and edIII. DENV cDNA served as template for PCR amplification of fl-e (amino acids 1–446) and edIII (amino acids 294–409) yielding amplicons at size 1341 and 348 bp, respectively. Figure S1. Production of recombinant DENV E proteins. Figure S2. Selection of phage clones displaying HuScFv specific to DENV rFL-E and rEDIII proteins. Figure S3. Peptide mimotopes of EDIII-specific HuScFv were characterized by bio-panning with Ph.D.-12TMphage display peptide library. Figure S4. HuScFvs were tested for their binding activity by indirect ELISA to recombinant full-length E protein (rFL-E) (A), and to recombinant domain III of E protein (rEDIII) (B). Non-HuScFv preparation (HB2151) and mouse polyclonal anti-DENV E protein antibody (marFLE) were used as negative and positive antibody controls, respectively.
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
277
ORIGINAL ARTICLE
Human monoclonal single-chain antibodies specific to dengue virus envelope protein N. Saokaew1,2, O. Poungpair1, A. Panya1,3, M. Tarasuk1, N. Sawasdee1, T. Limjindaporn4, W. Chaicumpa5 and P. Yenchitsomanus1 1 Division of Molecular Medicine, Department of Research and Development, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand 2 Graduate Program in Immunology, Department of Immunology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand 3 Graduate Program in Biochemistry, Department of Biochemistry, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand 4 Department of Anatomy, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand 5 Laboratory for Research and Technology Development, Department of Parasitology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand
Significance and Impact of the Study: No approved vaccine and specific drug for dengue virus (DENV) infection are available; thus, their developments are urgently required. The human single-chain variable antibody fragments (HuScFv) specific to DENV envelope (E) protein are potential to be developed as therapeutic biomolecules. HuScFv that bound specifically to recombinant full-length DENV E (FL-E) and its domain III (EDIII) were generated and testified for its inhibitory effect in DENV infection. EDIII-specific HuScFv inhibited DENV infection in a dose-dependent manner and has potential to be further developed as a therapeutic biomolecule for DENV infection.
Keywords dengue virus, envelope protein, human single-chain variable fragments, monoclonal antibody, phage display, virus entry. Correspondence Pa-thai Yenchitsomanus, Division of Molecular Medicine, Department of Research and Development, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, Thailand. E-mail: [email protected] 2013/0267: received 13 February 2013, revised 20 October 2013 and accepted 23 October 2013 doi:10.1111/lam.12186
Abstract Dengue virus (DENV) infection is an arthropod-borne disease with increasing prevalence worldwide. Attempts have been made to develop therapeutic molecules for treatment for DENV infection. However, most of potentially therapeutic DENV monoclonal antibody was originated from mouse, which could cause undesirable effects in human recipients. Thus, fully human antibody is preferable for therapeutic development. Human single-chain variable fragments (HuScFv) with inhibitory effect to DENV infection were generated in this study. HuScFv molecules were screened and selected from the human antibody phage display library by using purified recombinant DENV full-length envelope (FL-E) and its domain III (EDIII) proteins as target antigens for biopanning. HuScFv molecules were then tested for their bindings to DENV particles by indirect ELISA and immunofluorescent microscopy. EDIII-specific HuScFv exhibited neutralizing effect to DENV infection in Vero cells in a dose-dependent manner as determined by plaque formation and cell ELISA. Epitope mapping and molecular docking results concordantly revealed interaction of HuScFv to functional loop structure in EDIII of the DENV E protein. The neutralizing HuScFv molecule warrants further development as a therapeutic biomolecule for DENV infection.
Introduction Dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS) are caused by dengue virus (DENV) infection which is transmitted from human to human by Aedes mosquito (Martina et al. 2009; Murrell et al. 2011). 270
More than one-third of the world population is living in endemic areas, and 100 million people are infected yearly (Guzman et al. 2010). Importantly, neither approved vaccine nor antiviral drug is currently available. DENV is a member of Flaviviridae family, and its genome encodes three structural (capsid, premembrane/membrane and
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
N. Saokaew et al.
envelop) and seven nonstructural (NS1, NS2A, NS2B, NS3, NS4A, NS4B and NS5) proteins. The most outer part of the mature virion is covered by envelope (E) homodimer complexes, responsible for receptor-mediated attachment to host cells, and is a major target for neutralizing antibodies, especially at its domain III (EDIII) (Chang 1997; Chen et al. 1997; Hung et al. 1999; Allison et al. 2001). Passive immunization or antibody administration has been well accepted in medical practice. Various types of antibodies against viral and bacterial infection were tested in vitro, in animal models and in clinical trials (Heijtink et al. 2001; Domanski et al. 2005; Taylor et al. 2008). More recently, humanized monoclonal antibody specific to E protein of West Nile virus (WNV) has been used in phase I clinical trial. This trial demonstrated that only single dose of antibody was sufficient for WNV-neutralizing activity with safety and well tolerance in recipients (Beigel et al. 2010). Thus, the aim of this study is to produce HuScFv specific to DENV E protein, an essential molecule for virion assembly and virus entry, with inhibitory effect on DENV infection. Recombinant FL-E and EDIII proteins were generated and used as target antigens for selection of HuScFv molecules from the human antibody phage display library. The inhibitory effect of EDIII-specific HuScFv to cellular infection of DENV2 has been demonstrated. Mimotope searching and computational studies also emphasized interaction of HuScFv and EDIII region on DENV E protein. Results and discussion Production of recombinant DENV E proteins Fl-e and edIII were amplified at sizes of 1341 and 348 bp, respectively (Fig. S1a). Polyhistidine-tagged rFL-E (56 kDa) and rEDIII (17 kDa) were produced from recombinant plasmid-transformed E. coli (Fig. S1b) and purified by affinity chromatography (Fig. S1c). Mouse polyclonal antibodies generated by immunization with these proteins could readily recognize both recombinant and native DENV E protein (data not shown) indicating that the recombinant proteins retain inherent antigenic epitopes suitable for selection of antibodies specific to DENV E in the native form. Selection of phage clones displaying E protein-specific HuScFv After biopanning with rFL-E and rEDIII, there were 74 and 228 huscfv-carrying E. coli clones, respectively (Fig. S2a). Among them, 28 and 79 clones could produce soluble monoclonal HuScFv in different sizes (26–30 kDa)
HuScFv specific to DENV E protein
(Fig. S2b). This result from the variation in the length of immunoglobulin genes during the library construction step (Kulkeaw et al. 2009). HuScFv were tested for their binding activities to native DENV E proteins by indirect ELISA. We selected HuScFv clone numbers 15A for further analysis because it had highest relative binding ratio (although slightly higher than other clones) as screened by indirect ELISA (Fig. 1a), they could interact with native DENV E protein in DENV-infected Vero cells as detected by IFA (Fig. 1b) and they had a high yield of ‘metal chelate’ (not E protein ligand) affinity purification. HuScFv–DENV binding activity Huscfv of clone no. 15A was subcloned into modified pET23b(+) vector for polyhistidine-tagged HuScFv production and purification (Fig. S2c). This HuScFv could interact with native DENV E protein in DENV-infected Vero cells as detected by IFA. Green fluorescence of HuScFv interacting with E protein was visualized throughout cytoplasm of infected cells, while MOCK-infected cells showed no signal (Fig. 1b) indicating that, despite selected against recombinant antigens, the HuScFv readily recognized native E protein. As HuScFv is monoclonal, the fluorescent signal was not expectably as strong as in polyclonal anti-E positive antibody control. This finding also conformed to indirect ELISA result. Neutralizing activity of HuScFv in DENV-infected cells E protein is crucial for DENV binding to host-target cells and mediates the first step of cellular infection. Thus, the ultimate goal of this study is to develop HuScFv specific to E protein and interfere DENV infection. In plaque formation assay (Fig. 2a), which the number of infected cells was observed, DENV2 pre-incubated with HuScFv no. 15A (EDIII-specific) showed significant reduction in plaque number in a dose-dependent manner with the highest inhibition was seen at concentration of 200 ng ll 1. Positive control antibody (4G2)-treated DENV markedly reduced plaque number. After DENV infection, the viral antigen in Vero cells was also determined by cell ELISA using anti-E antibody (4G2) as detecting control (Fig. 2b). ELISA signal from Vero cells infected with HuScFv no.15A-treated DENV2 was significantly lower than signal in cells infected with untreated DENV or controls. HuScFv no. 15A was also demonstrated for its cross-reactivity to DENV1 and DENV4 by cell ELISA (Fig. 2c). At 200 ng ll 1 concentration, HuScFv no. 15A could significantly reduce E antigen in Vero cells infected DENV serotypes 1, 2 and 4 for 40–60% but not in DENV3-infected cells. The explanation for inability of 15A to neutralize DENV-3 is that the experiment was
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
271
N. Saokaew et al.
HuScFv specific to DENV E protein
10·37
1·5
1·0
0·5
0·0
24 3A 4A 5A 7A 9A 11A 13A 17A 21A 25A 26A 33A 35A 36A 39A 2B 3B 7B 9B 11B 12B 13B 14B 15B 17B 18B 86 4A 6A 8A 13A 15A 18A 20A 22A 25A 26A 35A 37A 44A 46A 49A 115A 6B 11B 12B 13B 18B 19B 21B 32B 35B 56B 58B HB2151 PBS mαrFLE
Relative binding ratio
(a) 2·0
Clone number (b)
MPAb
HuScFv15A
MPAb
HuScFv15A
AlexaFluor488
Merged with Hoechst MOCK-infected cells
DENV-infected cells
Figure 1 Interaction of HuScFv with E protein. (a) Monoclonal HuScFv were tested for their binding activity to native E protein on DENV particles in indirect ELISA. Non-HuScFv preparation (HB2151) and mouse polyclonal anti-DENV E protein antibody (marFLE) were used as negative and positive antibody controls, respectively. Relative binding ratio of each particular HuScFv clone was calculated by OD420nm of HuScFv–DENV particle divided by OD420nm of HuScFv-negative antigen control. (b) Purified HuScFv were tested for their interactions to E protein in DENV-infected Vero cells. Mouse polyclonal anti-E protein antibody (1 : 3000) was used as positive antibody control, while MOCK-infected cells stained with identical reagents served as negative control. All slides were visualized at 409 magnification.
carried out with only one dilution, and the concentration may need to be increased and optimized for testing against a different serotype. Further analysis of HuScFv epitope on EDIII of DENV3 may clarify its inability to inhibit DENV3 infection. However, these findings support the results of previous studies which indicate the importance of EDIII in DENV entry and infection (Chin et al. 2007; Abd-Jamil et al. 2008; Watterson et al. 2012). HuScFv–DENV EDIII interaction In epitope mapping using the 12-mer peptide phage display library, consensus peptide sequence aligned with DENV EDIII protein revealed that there are three regions on EDIII responsible for interaction with HuScFv no. 272
15A. These regions include residues 319–332 on the B strand-BC loop, residues 340–358 in C-C’ loop-C’ strand and residues 374–391 on F strand-FG loop (Fig. 3a). In molecular docking, the selected model (anti-SAR ScFv) had 65% sequence identity and 75% similarity to HuScFv no. 15A. The Ramachandran plot revealed that the modelled HuScFv no. 15A carried 18% of residues in disallowed regions. Molecular docking revealed that HuScFv no.15A-EDIII interaction was contributed by CDRs of HuScFv and three mapped regions on EDIII with lowbinding energy ( 9459 kcal mol 1) implying that the interaction could occur spontaneously. CDR1, 2 and three of VH occupied the C-C’ loop (residues 340–346) and FG loop (residues 380–385) of EDIII, whereas CDR1 and two of VL interacted with N-terminal region (residues
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
N. Saokaew et al.
HuScFv specific to DENV E protein
Relative infection (%)
(a) 100 80
**
60
**
40
***
***
20 0
PBS
HuScFv15A
4G2
Irrelevant HuScFv
Relative DENV E protein (%)
(b) 100 80
* *
60 40 *** 20 0 PBS
Relative DENV E protein (%)
(c)
4G2
HuScFv15A
DENV1
Irrelevant HuScFv DENV3
DENV2
DENV4
100 80 ** 60
* ***
***
40
*** ***
***
***
20
PB S el HuS 4 ev c G an Fv 2 t H 15 uS A cF v Irr
PB Irr H S el uS 4 ev c G an Fv 2 t H 15 uS A cF v
PB S el HuS 4 ev c G an Fv 2 t H 15 uS A cF v Irr
Irr
PB S el HuS 4 ev c G an Fv 2 t H 15 uS A cF v
0
Figure 2 HuScFv inhibitory effect to DENV infection. (a) DENV2 was pre-incubated with either 4G2 positive antibody control ( ), HuScFv15A and irrelevant HuScFv at various concentrations, for example, 8 ng ll 1 ( ), 40 ng ll 1 ( ), 200 ng ll 1 ( ) or antibody diluents (PBS, ) before allowed to infect Vero cells and determined per cent of infection by plaque assay. (b) Viral E antigen produced in infected cells after 48 h of infection was also measured by cell ELISA. (c) Antibody diluents ( ), 4G2 positive antibody control ( ), HuScFv15A ( ), or irrelevant HuScFv ( ) was incubated with DENV serotypes 1–4 (DENV1–4) before adding to Vero cell monolayer. Cytoplasmic DENV E antigen in infected cells was determined by cell ELISA after 48 h of infection. Asterisks indicate significant differences from three independent experiments (*P < 01, **P < 001, ***P < 0001).
299–305) and BC loop (residues 327–332) of DENV EDIII, respectively (Fig. 3b,c). The model also revealed that the residues recognized by HuScFv no. 15A locate at the outer boundary and are accessible in the natural dimeric DENV E protein (Fig. 3d). Interestingly, previous studies have
reported BC, CC’ and FG loops as targets of neutralizing antibodies (Sukupolvi-Petty et al. 2007, 2010). Many antibodies with strong DENV-neutralizing activity also recognized epitopes on EDIII suggesting that these regions are critical for antibody-mediated neutralization.
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
273
N. Saokaew et al.
HuScFv specific to DENV E protein
(a)
A-strand 295
(b)
B-strand
N-terminal region
AB loop
BC loop
335 VL
VH C’-strand C-strand C-C’ loop
336
F-strand
D-strand E-strand DE loop
370
G-strand FG loop
371
395
EDIII (c)
N-terminal region
(d)
BC loop FG loop A
C-C’ loop C F
E G B
Figure 3 Mapping HuScFv epitope on EDIII. (a) Sequence homology of HuScFv mimotopes (M1-12) to DENV2 EDIII was analysed using ClustalW program. Epitope of HuScFv clone no. 15A was found to span on 3 regions of EDIII molecule (B strand-BC loop, C-C’ loop-C’ strand and F strand-FG loop). (b) Interaction between HuScFv clone no. 15A (VH coloured in brown and VL in green) and EDIII (coloured in magenta) was performed using the ZDOCK program. (c) Amino acid residues on EDIII recognized by HuScFv clone no. 15A were considered as the epitope highlighted in yellow. (d) The localization of the epitope on EDIII recognized by HuScFv clone no. 15A (displayed as yellow spheres) was revealed and accessible in dimeric E proteins. EDI and EDII were displayed in light green. (PDB entry: 1OAN).
From this study, mechanisms by which HuScFv no. 15A inhibits DENV infection could possibly be at the step of host cell binding (Lok et al. 2008) or structural transition in the fusion step (Stiasny et al. 2007), which requires further investigations. The neutralizing HuScFv warrants further development as a research reagent or therapeutic biomolecule for DENV-infected patients. Materials and methods
Hawaii), serotype 2 (strain 16681), serotype 3 (strain H87) and serotype 4 (strain H241) were propagated in Aedes albopictus C6/36 cell line (CRL-1660, ATCC, USA) in 28°C incubator. Virus in culture supernatant was titrated by focus formation assay and kept at 80°C for neutralization test. The culture supernatant containing DENV2 viral particles was also subjected to partial purification as previously described (Gromowski and Barrett 2007) for further use in ELISA.
Dengue virus and cell culture
Production of recombinant DENV E proteins
Vero cells (CCL-81, ATCC, USA) were maintained at 37°C in a 5% CO2 incubator. DENV serotype 1 (strain
DENV RNA was extracted from culture supernatant of DENV2-infected C6/36 cells and subjected to cDNA
274
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
N. Saokaew et al.
synthesis. Fl-e and edIII were amplified by PCR using the primers shown in Table S1. After transformation, E. coli carrying verified recombinant plasmids were induced for the production of polyhistidine-tagged rFL-E and rEDIII for further purification using TALONTM Metal Affinity Resin (Clonetech, Mountain View, CA, USA). Phage biopanning DENV E-specific HuScFv were selected from the human antibody phage display library by using purified rFL-E and rEDIII as target antigens (Kulkeaw et al. 2009). The phagemid-transformed bacterial clones were screened for the presence of huscfv by PCR using phagemid-specific primers. Huscfv-positive clones were induced for the production of E-tagged HuScFv as detected by Western blot analysis using rabbit anti-E tag polyclonal antibody (1 : 5000). Indirect ELISA Indirect ELISA was performed to determine binding activity of monoclonal HuScFv to native viral protein as previously described (Poungpair et al. 2009). Partially, purified DENV2 particles (500 ng) and the preparation from MOCK-infected cells were used as antigen and no-antigen control, respectively. DENV2-bound HuScFv was detected with rabbit anti-E tag antibody and HRPconjugated swine anti-rabbit Ig followed by OPD substrate and OD492nm measurement. The relative binding ratio of each HuScFv clone was calculated by dividing the OD492nm of HuScFv interacting with DENV particle by OD492nm of HuScFv interacting with MOCK preparation. Preparation of polyhistidine-tagged HuScFv Huscfv sequences of ELISA-positive clones were subcloned into modified pET23b(+) vector (Poungpair et al. 2010). Transformed BL21 (DE3) E. coli was induced for the production of polyhistidine-tagged HuScFv which was purified by using TALONTM Metal Affinity Resin. Immunofluorescent assay To determine binding specificity of HuScFv to DENV E protein, DENV2-infected Vero cells were fixed, permeabilized and incubated with polyhistidine-tagged HuScFv. DENV E protein-interacting HuScFv was revealed by mouse anti-polyhistidine tag (1 : 5000) and AlexaFluor488-conjugated goat anti-mouse IgG (1 : 1000). Fluorescent signal was visualized by fluorescence Nikon microscope (409 microscopic field).
HuScFv specific to DENV E protein
Plaque assay Dilutions of either purified DENV2 E-specific HuScFv (clone no. 15A), irrelevant HuScFv, mouse monoclonal anti-E (4G2) antibody (positive antibody control at dilution 1 : 1250) or antibody diluents (PBS) were mixed with DENV2 (25 PFU) and incubated at 4°C for 2 h. The mixtures were separately added to Vero cell monolayer and incubated at 37°C for 1 h. Unbound virus was removed, and 12% carboxymethyl cellulose (CMC)-containing growth medium was added. After 7 days, cells were stained with 005% crystal violet, and clear plaques were counted under light microscope. Cell ELISA Either HuScFv at various concentrations or controls was incubated with DENV (1000 PFU) at 4°C for 2 h. The mixtures were added to Vero cell monolayer and incubated at 37°C for 1 h to allow infection; then, unabsorbed virus was removed, and growth medium was added and incubated further for 48 h. Intracellular DENV E protein was detected by mouse monoclonal anti-DENV E protein antibody (4G2) followed by HRP-conjugated rabbit antimouse Ig (1 : 1000). Finally, OPD substrate was added, and OD492nm was determined. Relative neutralizing activity was calculated by comparing OD value from cells infected with HuScFv-treated DENV to those infected with untreated DENV. In cross-neutralization experiment, the same method was used with only a single concentration of HuScFv (200 ng ll 1). Mimotope searching Phage clones displaying peptides that bound to EDIII-specific HuScFv (mimotope) were screened and selected from the PhD-12TM Phage Display Peptide Library (New England Biolabs) as previously described (Poungpair et al. 2009). The 12-mer peptide insert of each phage clone was deduced. Consensus sequences were obtained by multiple alignments of the deduced peptides using ClustalW program. The HuScFv mimotope was determined by alignment of consensus sequences to DENV E protein sequence. Homology modelling and molecular docking The reported structure with high homology to HuScFv, anti-SAR ScFv (PDB entry 2GHW), was identified by BLAST search (DS server) protocol, Discovery Studio 2.5 package (Accelrys Inc., San Diego, CA, USA) against a local sequence database. The multiple sequence alignment between the anti-SAR ScFv and HuScFv was conducted
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
275
N. Saokaew et al.
HuScFv specific to DENV E protein
by ClustalW program. The light and heavy chains of the anti-SAR ScFv were used as a modelled template for the selected HuScFv. Deposited structure of soluble DENV E protein (PDB entry 1OAN) was used for molecular docking. Models of HuScFv and DENV EDIII protein were built by MODELER program and assessed by PROCHECK (Laskowski et al. 1993). HuScFv was docked to DENV EDIII using the rigid body algorithm ZDOCK program. The poses from ZDOCK analysis were further processed to refine docked complex using a CHARM-minimization RDOCK program. The significant intermolecular interactions were rendered by using PyMOL software (The PyMOL Molecular Graphics System, DeLano Scientific, San Carlos, CA, USA). Acknowledgements This work was supported by Mahidol University Research Grant. Ni.S. is supported by Siriraj Graduate Scholarship. P.Y. is a Senior Research Scholar of Thailand Research Fund (TRF). O.P. is recipient of the TRF Young Researcher Grant (No.TRG5480006). A.P. is a recipient of TRF-Royal Golden Jubilee (RGJ) Ph.D. Scholarship. W.C. is supported by the research grant under the National Research University (NRU) Project, Office of Higher Education Commission (OHEC). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors thank Dr. Chanya Putthikhant for providing mouse monoclonal anti-DENV E protein. The human antibody phage display library used in this study is the property of the National Research Council of Thailand (NRCT). Conflict of interest No conflict of interest declared. References Abd-Jamil, J., Cheah, C.Y. and AbuBakar, S. (2008) Dengue virus type 2 envelope protein displayed as recombinant phage attachment protein reveals potential cell binding sites. Protein Eng Des Sel 21, 605–611. Allison, S.L., Schalich, J., Stiasny, K., Mandl, C.W. and Heinz, F.X. (2001) Mutational evidence for an internal fusion peptide in flavivirus envelope protein E. J Virol 75, 4268–4275. Beigel, J.H., Nordstrom, J.L., Pillemer, S.R., Roncal, C., Goldwater, D.R., Li, H., Holland, P.C., Johnson, S. et al. (2010) Safety and pharmacokinetics of single intravenous dose of MGAWN1, a novel monoclonal antibody to West Nile virus. Antimicrob Agents Chemother 54, 2431–2436.
276
Chang, K.J. (1997) Studies on the serological cross-reaction between dengue and Japanese encephalitis. Zhonghua Min Guo Wei Sheng Wu Ji Mian Yi Xue Za Zhi 30, 207–218. Chen, Y., Maguire, T., Hileman, R.E., Fromm, J.R., Esko, J.D., Linhardt, R.J. and Marks, R.M. (1997) Dengue virus infectivity depends on envelope protein binding to target cell heparan sulfate. Nat Med 3, 866–871. Chin, J.F., Chu, J.J. and Ng, M.L. (2007) The envelope glycoprotein domain III of dengue virus serotypes 1 and 2 inhibit virus entry. Microbes Infect 9, 1–6. Domanski, P.J., Patel, P.R., Bayer, A.S., Zhang, L., Hall, A.E., Syribeys, P.J., Gorovits, E.L., Bryant, D. et al. (2005) Characterization of a humanized monoclonal antibody recognizing clumping factor A expressed by Staphylococcus aureus. Infect Immun 73, 5229–5232. Gromowski, G.D. and Barrett, A.D. (2007) Characterization of an antigenic site that contains a dominant, type-specific neutralization determinant on the envelope protein domain III (ED3) of dengue 2 virus. Virology 30, 349–360. Guzman, M.G., Halstead, S.B., Artsob, H., Buchy, P., Farrar, J., Gubler, D.J., Hunsperger, E., Kroeger, A. et al. (2010) Dengue: a continuing global threat. Nat Rev Microbiol 8, S7–S16. Heijtink, R.A., van Nunen, A.B., van Bergen, P., Ostberg, L., Osterhaus, A.D. and de Man, R.A. (2001) Administration of a human monoclonal antibody (TUVIRUMAB) to chronic hepatitis B patients pre-treated with lamivudine: monitoring of serum TUVIRUMAB in immune complexes. J Med Virol 64, 427–434. Hung, S.L., Lee, P.L., Chen, H.W., Chen, L.K., Kao, C.L. and King, C.C. (1999) Analysis of the steps involved in Dengue virus entry into host cells. Virology 257, 156–167. Kulkeaw, K., Sakolvaree, Y., Srimanote, P., Tongtawe, P., Maneewatch, S., Sookrung, N., Tungtrongchitr, A., Tapchaisri, P. et al. (2009) Human monoclonal ScFv neutralize lethal Thai cobra, Naja kaouthia, neurotoxin. J Proteomics 72, 270–282. Laskowski, R.A., Macarthur, M.W., Moss, D.S. and Thornton, J.M. (1993) PROCHECK: a program to check the stereochemical quality of protein structures. J Appl Cryst 26, 283–290. Lok, S.M., Kostyuchenko, V., Nybakken, G.E., Holdaway, H.A., Battisti, A.J., Sukupolvi-Petty, S., Sedlak, D., Fremont, D.H. et al. (2008) Binding of a neutralizing antibody to dengue virus alters the arrangement of surface glycoproteins. Nat Struct Mol Biol 15, 312–317. Martina, B.E., Koraka, P. and Osterhaus, A.D. (2009) Dengue virus pathogenesis: an integrated view. Clin Microbiol Rev 22, 564–581. Murrell, S., Wu, S.C. and Butler, M. (2011) Review of dengue virus and the development of a vaccine. Biotechnol Adv 29, 239–247. Poungpair, O., Chaicumpa, W., Kulkeaw, K., Maneewatch, S., Thueng-in, K., Srimanote, P., Tongtawe, P., Songserm, T.
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
N. Saokaew et al.
et al. (2009) Human single chain monoclonal antibody that recognizes matrix protein of heterologous influenza A virus subtypes. J Virol Methods 159, 105–111. Poungpair, O., Pootong, A., Maneewatch, S., Srimanote, P., Tongtawe, P., Songserm, T., Tapchaisri, P. and Chaicumpa, W. (2010) A human single chain transbody specific to matrix protein (M1) interferes with the replication of influenza A virus. Bioconjug Chem 21, 1134–1141. Stiasny, K., Brandler, S., Kossl, C. and Heinz, F.X. (2007) Probing the flavivirus membrane fusion mechanism by using monoclonal antibodies. J Virol 81, 11526–11531. Sukupolvi-Petty, S., Austin, S.K., Purtha, W.E., Oliphant, T., Nybakken, G.E., Schlesinger, J.J., Roehrig, J.T., Gromowski, G.D. et al. (2007) Type- and subcomplexspecific neutralizing antibodies against domain III of dengue virus type 2 envelope protein recognize adjacent epitopes. J Virol 81, 12816–12826. Sukupolvi-Petty, S., Austin, S.K., Engle, M., Brien, J.D., Dowd, K.A., Williams, K.L., Johnson, S., Rico-Hesse, R. et al. (2010) Structure and function analysis of therapeutic monoclonal antibodies against dengue virus type 2. J Virol 84, 9227–9239. Taylor, C.P., Tummala, S., Molrine, D., Davidson, L., Farrell, R.J., Lembo, A., Hibberd, P.L., Lowy, I. et al. (2008) Open-label, dose escalation phase I study in healthy volunteers to evaluate the safety and pharmacokinetics of a human monoclonal antibody to Clostridium difficile toxin A. Vaccine 26, 3404–3409.
HuScFv specific to DENV E protein
Watterson, D., Kobe, B. and Young, P.R. (2012) Residues in domain III of the dengue virus envelope glycoprotein involved in cell-surface glycosaminoglycan binding. J Gen Virol 93, 72–82.
Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1. Primers for PCR amplification of fl-e and edIII. DENV cDNA served as template for PCR amplification of fl-e (amino acids 1–446) and edIII (amino acids 294–409) yielding amplicons at size 1341 and 348 bp, respectively. Figure S1. Production of recombinant DENV E proteins. Figure S2. Selection of phage clones displaying HuScFv specific to DENV rFL-E and rEDIII proteins. Figure S3. Peptide mimotopes of EDIII-specific HuScFv were characterized by bio-panning with Ph.D.-12TMphage display peptide library. Figure S4. HuScFvs were tested for their binding activity by indirect ELISA to recombinant full-length E protein (rFL-E) (A), and to recombinant domain III of E protein (rEDIII) (B). Non-HuScFv preparation (HB2151) and mouse polyclonal anti-DENV E protein antibody (marFLE) were used as negative and positive antibody controls, respectively.
Letters in Applied Microbiology 58, 270--277 © 2013 The Society for Applied Microbiology
277
