INTRODUCTION
Nanobodies (Nbs), also called camelid heavy-chain variable domains (VHHs), are single-domain nano-sized antibodies; they are derived from variable fragments of camelid or shark heavy chain-only antibodies (
1,
2). Nbs contain four constant regions, named framework regions (FRs), and three connecting variable regions, called complementarity determining regions (CDRs). FRs are responsible for maintaining the structural integrity of Nbs, while CDRs directly bind to antigen epitopes (
3). On the one hand, because of their nanometer size (∼2.5 nm by 4 nm) and single domain structure, Nbs have the following advantages as antiviral agents: they can be easily expressed for bulk production, they are robust for convenient storage and transportation, and they have good permeability in tissues (
4–6). On the other hand, also because of their small size, Nbs have the following potential limitations as antiviral agents: they may have limited binding affinity for antigens and may be cleared from the body relatively quickly (the upper size limit of proteins for renal clearance is 60 kDa) (
7,
8). Nevertheless, the use of Nbs as antiviral therapeutic agents is gaining more and more clinical acceptance, with the focus on overcoming their potential limitations (
9–11).
Middle East respiratory syndrome coronavirus (MERS-CoV) was first identified in June 2012 (
12) and continues to infect humans: it has led to at least 2,220 confirmed cases and 790 deaths (∼36% fatality rate) in 27 countries (
http://www.who.int/emergencies/mers-cov/en/). Bats and dromedary camels are likely the natural reservoir and transmission hosts, respectively, for MERS-CoV. Whereas camel-to-human transmission of MERS-CoV has accounted for most of the human infections, human-to-human spread of MERS-CoV also occurs sporadically (
13,
14). Currently, no therapeutic agents or vaccines have been approved for human use. Due to the continued threat of MERS-CoV, there is an urgent need to develop highly potent, cost-effective, and broad-spectrum anti-MERS-CoV therapeutics and vaccines with the potential for large-scale industrial production.
Therapeutic antibodies have been shown to be effective antiviral agents (
15,
16). The receptor-binding domain (RBD) of MERS-CoV spike (S) protein is a prime target for therapeutic antibodies. The MERS-CoV S protein guides viral entry into host cells. It first binds to its host receptor dipeptidyl peptidase 4 (DPP4) through the RBD of its S1 subunit and then fuses viral and host membranes through its S2 subunit (
15,
17–22). The RBD contains a receptor-binding motif (RBM) region (residues 484 to 567) that directly interacts with DPP4. We have previously shown that RBD-based vaccines are highly immunogenic and can induce the production of potent anti-MERS-CoV cross-neutralizing antibodies (
23–27). Moreover, we have discovered several RBD-specific monoclonal antibodies (MAbs) with strong neutralizing activities against lethal MERS-CoV infections in human DPP4-transgenic (hDPP4-Tg) mice (
15,
28,
29). These and some other RBD-targeting MAbs are currently being developed as anti-MERS-CoV therapeutics in experimental animal models (
15,
30–36). However, the widespread use of conventional antibodies can be limited by their large size, high production costs, inconvenient storage and transportation, and poor pharmacokinetics (
37), making Nbs attractive alternatives to traditional MAbs to treat MERS-CoV infections. Currently, it has not been shown whether MERS-CoV RBD can reliably trigger the production of Nbs, whether the produced Nbs can overcome the potential limitations (e.g., low binding affinity for the RBD and relatively short half-life in the body), or whether the produced Nbs can demonstrate sufficient therapeutic efficacy to warrant further development in clinical settings.
Here, after immunizing llama with recombinant MERS-CoV RBD protein, we generated a novel neutralizing Nb, NbMS10, and also constructed its human-Fc-fused version, NbMS10-Fc. We further investigated these Nbs for their RBD-binding capabilities, neutralization mechanisms, cross-neutralizing activity against divergent MERS-CoV strains, half-life, and protective efficacy against lethal MERS-CoV infection in an established hDPP4-Tg mouse model (
38). This study reveals that efficacious, robust, and broad-spectrum Nbs can be produced to target MERS-CoV S protein RBD and that they hold great promise as potential anti-MERS-CoV therapeutics.
DISCUSSION
MERS-CoV continues to infect humans with a high fatality rate. Because camels likely serve as the transmission hosts for MERS-CoV and also because humans have contact with camels, the constant and continuing transmissions of MERS-CoV from camels to humans make it difficult to eradicate MERS-CoV from the human population. Thus, efficacious, cost-effective, and broad-spectrum anti-MERS-CoV therapeutic agents are needed to prevent and treat MERS-CoV infections in both humans and camels. Nbs have been gaining acceptance as antiviral agents because of their small size, good tissue permeability, and cost-effective production, storage, and transportation. However, their small size may also lead to relative low antigen-binding affinity and quick clearance from the host body. In this study, we have developed a novel MERS-CoV-targeting Nb, NbMS10, and its Fc-fused version, NbMS10-Fc, both of which demonstrate great promise as anti-MERS-CoV therapeutic agents.
NbMS10 and NbMS10-Fc present superior characteristics common to other Nbs. They target the MERS-CoV RBD, which plays an essential role in cell entry of MERS-CoV by binding to its receptor hDPP4. Both Nbs can be expressed in yeast cells with high purity and yields and are soluble in solutions. All of these properties suggest cost-effective production, easy storage, and convenient transportation of these Nbs in potential commercial applications.
The MERS-CoV RBD-targeting Nbs developed also demonstrate good qualities comparable to previously reported MERS-CoV RBD-specific conventional IgGs. First, the Nbs bind to MERS-CoV RBD with high affinities. The
Kd values for NbMS10 and NbMS10-Fc to bind MERS-CoV RBD were 8.71 × 10
−10 M and 3.46 × 10
−10 M, respectively. The
Kd values for RBD-targeting conventional IgGs to bind MERS-CoV RBD range from 7.12 × 10
−8 M to 4.47 × 10
−11 M (
29,
35,
36). Moreover, the ND
50 values for NbMS10 and NbMS10-Fc to neutralize MERS-CoV (EMC2012 strain) infection in cultured cells were 3.52 and 2.33 μg/ml, respectively. The ND
50 values for RBD-specific conventional IgGs to neutralize various MERS-CoV strains ranged from micrograms/ml to nanograms/ml (
30,
32,
35,
39,
40). Thus, the Nbs developed in this study and conventional IgGs reported previously have comparable MERS-CoV RBD-binding affinities and MERS-CoV-neutralizing activities. Structural comparisons of conventional IgGs and Nbs have shown that the antigen-binding site of IgGs consists of paired heavy-chain and light-chain variable (VH-VL) domains, whereas Nbs lack the light chain and hence cannot form the paired VH-VL domains (
8,
41). Instead, Nbs have an extended CDR3 region (>16 amino acid residues), longer than that of the VHs of conventional IgGs (average length 12 amino acid residues) (
42–44). Moreover, the Nbs developed here contain a 22-amino-acid CDR3; the extended CDR3 enables the Nbs to bind to the antigens with higher affinity (
37). Furthermore, although the single-domain Nb (i.e., NbMS10) is small and can be cleared from the serum relatively quickly, the Fc-fused Nb (i.e., NbMS10-Fc) with relatively increased size demonstrates extended
in vivo half-life. Therefore, the potential short half-life of Nbs can be overcome by adding the appropriate tag to the Nbs to increase their half-life. Overall, the present study has shown the feasibility of overcoming the potential limitations of Nbs.
The MERS-CoV RBD-targeting Nbs potently neutralize MERS-CoV entry into host cells. The
Kd values between the Nbs and MERS-CoV RBD are significantly lower than that between MERS-CoV RBD and hDPP4 receptor. As a result, the Nbs can outcompete hDPP4 for the binding of MERS-CoV RBD, thereby blocking the binding of MERS-CoV to DPP4, as well as MERS-CoV entry into host cells. It is worth noting that the RBD on the MERS-CoV S trimer frequently undergoes conformational changes, switching between a lying down, receptor-inaccessible conformation and a standing-up, receptor-accessible conformation. Hence, in the context of the virus particles where the RBD is part of the S protein, the Nbs would need to bind the RBD when the RBD is in the standing-up conformation (
45). Importantly, the Nbs demonstrate strong cross-neutralizing activities against various MERS-CoV strains isolated from different hosts (humans and camels) and from different time points during MERS-CoV circulation in humans (from years 2012 to 2015). NbMS10 had a relatively high ND
50 against the AGV08584/2012 strain containing a V534A mutation, which is consistent with the slightly reduced binding affinity between NbMS10 and MERS-CoV RBD containing the V534A mutation (
Fig. 4A). The broad neutralizing spectrum of the Nbs results from the binding site of the Nbs on MERS-CoV RBD, which is located in the Asp539-containing region that plays a critical role in DPP4 binding. Interestingly, several MERS-CoV RBD-specific conventional IgGs also bind to the same epitope (
39,
46), suggesting that this region is a hot spot for immune recognition. Although mutations in this region can eliminate the binding of the Nbs to MERS-CoV RBD and hence lead to viral immune evasion, they also reduce the binding of MERS-CoV RBD to receptor DPP4 and hence decrease the efficiency of viral entry. Thus, viral immune evasion from the inhibition of the Nbs through mutations can be costly to MERS-CoV itself. Indeed, residue Asp539 in S protein RBD is highly conserved in almost all of the natural MERS-CoV strains published to date (
Fig. 8). Therefore, the MERS-CoV-specific Nbs can potentially be developed into broad-spectrum anti-MERS-CoV therapeutic agents. Despite the above analysis, this study did not examine all possible mutations in the Nb-binding region (since the atomic structures of MERS-CoV RBD complexed with the Nbs are still unknown), and thus it is possible that future escape mutations may occur to residues that this study did not cover. In that case, a combination of the current Nbs and other antibodies targeting other S regions or various RBD epitopes may be helpful in battling the emergence of immune escape MERS-CoV strains.
In sum, the MERS-CoV-specific Nbs developed in the present study possess superior qualities common to all Nbs such as their small size and cost-effective production. They also overcome potential limitations of other Nbs by maintaining a high binding affinity for their target MERS-CoV RBD and an optimized half-life. Moreover, they recognize a functionally important region on MERS-CoV RBD, rendering viral immune evasion costly and at the same time making themselves good candidates as broad-spectrum anti-MERS-CoV therapeutics. We have confirmed the effectiveness of the Nbs by showing that the Fc-fused Nb completely protected animal models from lethal MERS-CoV challenge. Thus, the Nbs can potentially be used in both humans and camels to prevent and treat MERS-CoV infections in either of these hosts and also block the camel-to-human transmission of MERS-CoV. Overall, our study proves the feasibility of developing highly effective Nbs as anti-MERS-CoV therapeutic agents and points out strategies to preserve the advantages of Nbs, as well as to overcome the potential limitations of Nbs.
MATERIALS AND METHODS
Ethics statement.
The animal studies were carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the State Key Laboratory of Pathogen and Biosecurity at the Beijing Institute of Microbiology and Epidemiology of China and the National Institutes of Health (NIH). The animal protocols were approved by the IACUC of the State Key Laboratory of Pathogen and Biosecurity, Beijing Institute of Microbiology and Epidemiology (permit BIME 2015-0024) and by the Committee on the Ethics of Animal Experiments of the New York Blood Center (approval 194.18).
Construction of VHH library and screening for MERS-CoV-RBD-specific Nbs.
Construction of the Nb (i.e., VHH) library and screening of MERS-CoV-RBD-specific Nbs were performed as previously described (
47). Briefly, male and female alpacas (llama pacos, 1 year) were subcutaneously immunized with recombinant RBD-Fc (260 μg/alpaca) (
48) plus Freund complete adjuvant, and boosted three times with the same immunogen plus Freund incomplete adjuvant (InvivoGen). Blood was collected 10 days after the last immunization, and then PBMCs were isolated using Ficoll-Paque gradient centrifugation (GE Healthcare). Total RNA was extracted with TRIzol reagent (Invitrogen). cDNA was synthesized by reverse transcription-PCR (RT-PCR) using a TransScript cDNA Synthesis SuperMix (TransGen Biotech, China), followed by PCR amplification of the N-terminal IgG heavy-chain fragment (∼700 bp), using the forward primer VHH-L-F (5′-GGTGGTCCTGGCTGC-3′) and the reverse primer CH2-R (5′-GGTACGTGCTGTTGAACTGTTCC-3′). The VHH gene (∼300 to 450 bp) was further amplified using the above DNA fragment as the template and the forward primer VHH-FR1-D-F (5′-TTTCTATTACTA
GGCCCAGCCGGCCGAGTCTGGAGGRRGCTTGGTGCA-3′) and the reverse primer VHH-FR4-D-R (5′-AAACCGTT
GGCCATAATGGCCTGAGGAGACGRTGACSTSGGTC-3′) (the SfiI restriction site is underlined). The SfiI-digested VHH DNA fragment was then inserted into phagemid vector pCANTAB5e (Bio-View Shine Biotechnology, China) to construct the VHH phage display library (
49). Phage particles were analyzed by ELISA using recombinant MERS-CoV RBD-Fc and Fc of human IgG1 proteins as the positive and negative target proteins, respectively, to screen for RBD-specific Nbs. After four rounds of bio-panning, one of five positive clones, CAb10, with the highest binding to MERS-CoV RBD, was selected for further analyses (
Fig. 1).
Expression of MERS-CoV-RBD-specific Nbs in yeast cells.
NbMS10 and NbMS10-Fc Nbs containing a C-terminal His
6 and Fc of human IgG1, respectively, were constructed based on the aforementioned CAb10 VHH. The DNA sequences encoding NbAb10 and NbAb10-Fc were synthesized (GenScript) and inserted into the
Pichia pastoris secretory expression vector, pPICZαA (Invitrogen) (
Fig. 1). The recombinant NbMS10 and NbMS-Fc were expressed in
Pichia pastoris GS115 cells and purified using a Ni-NTA column (for NbMS10; GE Healthcare) and a protein A Sepharose 4 Fast Flow column (for NbMS10-Fc; GE Healthcare), respectively.
SDS-PAGE and Western blotting.
The purified anti-MERS-CoV-RBD Nbs were analyzed using SDS-PAGE and Western blotting (
23,
48). Briefly, Nbs (3 μg) were loaded onto 10% Tris-glycine SDS-PAGE gels and stained using Coomassie brilliant blue or transferred to nitrocellulose membranes. After being blocked overnight at 4°C with 5% nonfat milk/phosphate-buffered saline–Tween 20 (5% PBST), the membranes were incubated sequentially with goat anti-llama IgG (1:3,000; Abcam) and horseradish peroxidase (HRP)-conjugated anti-goat IgG (1:1,000; R&D Systems) antibodies for 1 h at room temperature and then with ECL Western blot substrate reagents. Finally, the membranes were visualized using Amersham Hyperfilm (GE Healthcare). A SARS-CoV-RBD-specific MAb, 33G4 (
50), was used as a control.
ELISA.
ELISA was performed to detect the binding between Nbs and MERS-CoV S1 or RBD proteins (
23,
51). Briefly, ELISA plates were coated overnight at 4°C, respectively, with recombinant MERS-CoV S1-His (
48), RBD-Fc (
48), RBD-Fd (
51), or one of the mutant RBDs containing a C-terminal human Fc tag (
28). After being blocked with 2% PBST for 2 h at 37°C, the plates were further incubated sequentially with serially diluted Nbs (containing a C-terminal His
6 or Fc tag), either goat anti-llama (1:5,000) or mouse anti-His (1:3,000) antibody (Sigma) and either HRP-conjugated anti-goat IgG (1:3,000) or HRP-conjugated anti-mouse IgG (1:5,000) antibody (GE Healthcare) for 1 h at 37°C. ELISA substrate (3,3′,5,5′-tetramethylbenzidine [TMB]; Invitrogen) was added to the plates, and the reactions were stopped with 1 N H
2SO
4. The absorbance at 450 nm (
A450) was measured using a Tecan Infinite 200 Pro microplate reader (Tecan).
To detect the binding between Nbs and denatured MERS-CoV RBD protein, ELISA plates were coated with RBD-Fd protein (2 μg/ml) overnight at 4°C and then sequentially incubated with DTT (10 mM) and iodoacetamide (50 mM) (Sigma) for 1 h at 37°C (
28). After three washes using PBST, ELISA was performed as described above.
Inhibition of the binding between MERS-CoV RBD and hDPP4 proteins by Nbs was performed using ELISA as described above, except that recombinant hDPP4 protein (2 μg/ml; R&D Systems), and serially diluted Nbs were added simultaneously to the RBD-Fc-coated plates. The binding between RBD and DPP4 was detected using goat anti-hDPP4 antibody (1:1,000; R&D Systems) and HRP-conjugated anti-goat IgG (1:3,000). The percent inhibition was calculated based on the A450 values of RBD-hDPP4 binding in the presence or absence of Nbs. SARS-CoV 33G4 MAb was used as a negative control to Nbs.
Surface plasmon resonance.
The binding between Nbs and MERS-CoV S1 or RBD protein was detected using a BiacoreS200 instrument (GE Healthcare) as previously described (
29). Briefly, recombinant Fc-fused MERS-CoV RBD-Fc protein or NbMS10-Fc Nb (5 μg/ml) was captured using a Sensor Chip protein A (GE Healthcare), and recombinant His
6-tagged MERS-CoV S1-His protein or NbMS10 Nb at various concentrations was flown over the chip surface in a running buffer containing 10 mM HEPES (pH 7.4), 150 mM NaCl, 3 mM EDTA, and 0.05% surfactant P20. The sensorgram was analyzed using Biacore S200 software, and the data were fitted to a 1:1 binding model.
Flow cytometry.
This assay was performed to detect the inhibition of the binding between MERS-CoV RBD and cell surface hDPP4 by Nbs (
28). Briefly, Huh-7 cells expressing hDPP4 were incubated with MERS-CoV RBD-Fc protein (20 μg/ml) for 30 min at room temperature in the absence or presence of Nbs at various concentrations. Cells were incubated with fluorescein isothiocyanate-labeled anti-human IgG antibody (1:50, Sigma) for 30 min and then analyzed by flow cytometry. The percent inhibition was calculated based on the fluorescence intensity of RBD-Huh-7 cell binding in the presence or absence of Nbs.
MERS pseudovirus neutralization assay.
Neutralization of MERS pseudovirus entry by Nbs was performed as previously described (
23,
52). Briefly, 293T cells were cotransfected with a plasmid encoding Env-defective, luciferase-expressing HIV-1 genome (pNL4-3.luc.RE) and a plasmid encoding MERS-CoV S protein. The MERS pseudoviruses were harvested from supernatants at 72 h posttransfection and then incubated with Nbs at 37°C for 1 h before being added to Huh-7 cells. After 72 h, the cells were lysed in cell lysis buffer (Promega), incubated with luciferase substrate (Promega), and assayed for relative luciferase activity using Tecan Infinite 200 Pro Luminator (Tecan). The ND
50 of the Nbs was calculated as previously described (
53).
MERS-CoV microneutralization assay.
Neutralization of MERS-CoV infection by Nbs was performed as previously described (
28,
54). Briefly, MERS-CoV (EMC2012 strain) at an amount equal to 100 median tissue culture infective doses (TCID
50) was incubated with Nbs at different concentrations for 1 h at 37°C. The Nb-virus mixture was then incubated with Vero E6 cells for 72 h at 37°C in the presence of 5% CO
2. The cytopathic effect (CPE) was observed daily. The neutralizing activity of Nbs was reported as the ND
50. The Reed-Muench method was used to calculate the ND
50 value for each Nb (
55).
Measurement of half-life of Nbs.
Male and female C57BL/6 mice (6 to 8 weeks old) were intravenously injected with Nbs (50 μg in 200 μl per mouse) into the tail vein. Sera were collected at different time points (30 min, 2 h, 6 h, 1 day, 5 days, and 10 days postinjection). The concentrations of Nbs in the sera were detected by ELISA, as described above. Briefly, MERS-CoV S1-His protein (2 μg/ml) was used to coat ELISA plates, and then sera, goat anti-llama antibodies (1:5,000), and HRP-conjugated anti-goat IgG antibodies (1:3,000) were sequentially added for ELISA reactions.
Evaluation of protective efficacy of NbMS10-Fc Nb.
The prophylactic and therapeutic efficacy of NbMS10-Fc was evaluated in hDPP4-Tg mice as previously described (
29). Briefly, male and female mice (8 to 10 weeks old) were intraperitoneally anesthetized with sodium pentobarbital (5 mg/kg of body weight) before being intranasally inoculated with lethal dose of MERS-CoV (EMC2012 strain, 10
5.3 TCID
50) in 20 μl of Dulbecco modified Eagle medium. Either 3 days preinfection or 1 or 3 days postinfection, the mice were intraperitoneally injected with NbMS10-Fc (10 mg/kg). Trastuzumab MAb was used as a control to the Nb. The infected mice were observed daily for 14 days, and their body weights and survivals were recorded.
Statistical analysis.
Statistical analysis was performed using GraphPad Prism version 5.01. To compare the binding of Nbs to MERS-CoV S1 or RBD protein, as well as the RBDs with or without D539A mutation to hDPP4 receptor, a two-tailed Student t test was used. One-way analysis of variance was used to compare the inhibition of Nbs to RBD-hDPP4 binding. Statistical significance between survival curves was analyzed using Kaplan-Meier survival analysis with a log-rank test. P values lower than 0.05 were considered statistically significant. In the figures, “*,” “**,” and “***” indicate P < 0.05, P < 0.01, and P < 0.001, respectively.
Data availability.
All data needed to evaluate the conclusions presented here are included. Additional data related to this study may be requested from the authors.
ACKNOWLEDGMENTS
This study was supported by the National Key Plan for Scientific Research and Development of China (2016YFD0500306, NSFC81571983); the Technology Innovation Fund in China (grant 3407049), the State Key Laboratory of Pathogen and Biosecurity (grant SKLPBS1704 [to G.Z. and Y.Z.]); NIH grants R01AI137472, R21AI109094, and R21AI128311 (to S.J. and L.D.); NIH grants R01AI089728 and R01AI110700 (to F.L.); and NIH grant R01AI139092 (to S.J., F.L., and L.D.). The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
The authors declare no competing interests.
G.Z., L.D., and Y.Z. designed the study. G.Z., L.H., S.S., H.Q., W.T., J.C., J.L., Y.C., Y.G., Y.W, K.J., R.F., and E.D. performed the experiments. G.Z., W.T., S.J., L.D. and Y.Z. summarized and analyzed the data. J.S. and F.L. performed the structural analysis. G.Z., F.L., L.D., and Y.Z. wrote the manuscript. S.J., F.L., L.D., and Y.Z. revised the manuscript.