Abstract

Up-regulation of epidermal growth factor receptor (EGFR) is a hallmark of many solid tumors, and inhibition of EGFR signaling by small molecules and antibodies has clear clinical benefit. Here, we report the isolation and functional characterization of novel camelid single-domain antibodies (sdAbs or VHHs) directed against human EGFR. The source of these VHHs was a llama immunized with cDNA encoding human EGFR ectodomain alone (no protein or cell boost), which is notable in that genetic immunization of large, outbred animals is generally poorly effective. The VHHs targeted multiple sites on the receptor's surface with high affinity (KD range: 1–40 nM), including one epitope overlapping that of cetuximab, several epitopes conserved in the cynomolgus EGFR orthologue, and at least one epitope conserved in the mouse EGFR orthologue. Interestingly, despite their generation against human EGFR expressed from cDNA by llama cells in vivo (presumably in native conformation), the VHHs exhibited wide and epitope-dependent variation in their apparent affinities for native EGFR displayed on tumor cell lines. As fusions to human IgG1 Fc, one of the VHH-Fcs inhibited EGFR signaling induced by EGF binding with a potency similar to that of cetuximab (IC50: ∼30 nM). Thus, DNA immunization elicited high-affinity, functional sdAbs that were vastly superior to those previously isolated by our group through protein immunization.

Introduction

The epidermal growth factor receptor (EGFR) is a receptor tyrosine kinase that is overexpressed and constitutively activated in up to 80% of solid cancers [1]. Following the development of small-molecule inhibitors, naked antibodies (Abs) against EGFR, exemplified by cetuximab, have shown clinical benefit in treating colorectal [2] and head and neck cancers [3]. Other EGFR Abs are under investigation in other indications and as Ab–drug conjugates. Camelid single-domain antibodies (sdAbs or VHHs) have previously been isolated against EGFR using protein immunization [4,5] and whole-cell immunization [6,7], and biparatopic molecules with improved potency have been constructed from these sdAb building blocks [8]. One advantage of sdAb-based biologics for cancer therapy is that they tend to penetrate solid tumors better than full-length IgGs [9].

DNA immunization of large outbred animals is generally recognized as inconsistent and poorly effective in eliciting humoral immune responses [10]. The mechanisms underlying this difficulty are thought to involve low rates of plasmid uptake and antigen expression, potentially relating to concentration effects in peripheral tissue. Six studies have examined DNA immunization of camelid species. In a study published in 2016, Peyrassol et al. [11] immunized four llamas with DNA encoding the G protein-coupled receptor (GPCR) ChemR23 by intradermal injection of plasmid DNA using a Dermojet® device, then boosted the animals with ChemR23-expressing Dubca cells. They observed ChemR23-specific Abs in the sera of only one of four llamas and were able to isolate antagonistic sdAbs with apparent affinities for ChemR23-expressing cells of ∼130–160 nM from phage display libraries constructed using the peripheral B-cell repertoire of this animal. In a subsequent 2018 study, Peyrassol et al. [12] immunized two llamas with DNA encoding the GPCR VPAC1, again by intradermal injection of plasmid DNA followed by boosting with VPAC1-expressing CHO cells. They did not describe the seroconversion of these animals, but were able to isolate sdAbs with nM and µM apparent affinities for VPAC1-expressing cells from peripheral repertoires. Van der Woning et al. [13] immunized four llamas with DNA encoding glucagon receptor (GCGR) using intradermal injection followed by in vivo electroporation, then boosted with GCGR-expressing Dubca cells. The authors claimed that weak serum titers against GCGR were present in all four animals, although the experiment was missing key controls (preimmune and irrelevant immune sera as well as irrelevant antigens). Koch-Nolte et al. [14], Danquah et al. [15] and Fumey et al. [16] immunized llamas with DNA encoding ART2.2, P2X7 and CD38, respectively, by biolistic transfection using a Helios® gene gun system followed by boosting with either recombinant protein or antigen-expressing cells. The off-rates of the anti-CD38 sdAbs were consistent with binding affinities in the low-nM range. However, in these latter three studies, neither serum analyses over the course of immunization nor per-animal success rates were disclosed.

Here, we tested the hypothesis that high-affinity and functional camelid sdAbs could be produced against a model antigen, EGFR, using DNA immunization alone. We describe the comprehensive in vitro characterization of a panel of sdAbs generated in this manner, which were dramatically superior to those we previously isolated using protein immunization [4].

Experimental

Antibodies and reagents

Recombinant 6×His-tagged human EGFRvIII was produced by transient transfection of HEK293-6E cells as previously described [17] and purified by immobilized metal affinity chromatography (IMAC) followed by a final size exclusion chromatography polishing step to remove aggregates. Recombinant 6×His-tagged human EGFR ectodomain was from Genscript (Cat. No. Z03194; Piscataway, NJ), recombinant in vivo biotinylated 6×His-tagged human EGFRvIII ectodomain was from ACROBiosystems (Cat. No. EGR-H82E0; Newark, DE) and recombinant streptavidin was from Thermo Fisher Scientific (Waltham, MA). Human EGFR-Fc fusion protein was from Genscript (Cat. No. Z03381), and rhesus and mouse EGFR-Fc fusion proteins were from Sino Biological (Cat. Nos. 90317-K02H and 51091-M02H; Beijing, China). Horseradish peroxidase (HRP)-conjugated goat polyclonal anti-llama IgG was from Cedarlane Laboratories (Cat. No. A160-100P; Burlington, Canada), HRP-conjugated goat polyclonal anti-human IgG was from Sigma–Aldrich (St. Louis, MO) and 3,3′,5,5′-tetramethylbenzidine (TMB) substrate was from Mandel Scientific (Guelph, Canada). Bovine serum albumin (BSA) and Tween-20 were from Sigma–Aldrich, and all cell culture reagents were from Thermo Fisher. Mouse monoclonal anti-c-Myc IgG was from Santa Cruz Biotechnology (clone 9E10, Cat. No. sc-40; Dallas, TX), allophycocyanin (APC)-conjugated goat polyclonal anti-mouse IgG was from Thermo Fisher (Cat. No. A865), Alexa Fluor® 488 (AF488)-conjugated donkey anti-human IgG was from Jackson ImmunoResearch (Cat. No. 709546098; West Grove, PA) and R-phycoerythrin (PE)-conjugated streptavidin was from Thermo Fisher (Cat. No. S866). Erlotinib was from Sigma–Aldrich, recombinant human epidermal growth factor (EGF) and mouse monoclonal Ab against β-actin were from Genscript (Cat. Nos. Z00333 and A00702), rabbit polyclonal Ab against phospho-EGFR (Tyr1068) was from Cell Signaling Technology (Cat. No. 2234; Danvers, MA), mouse monoclonal Ab against human EGFR was a generous gift from Anne Marcil (National Research Council Canada, Montréal, Canada) and cetuximab was a generous gift from Yves Durocher (National Research Council Canada, Montréal, Canada). HRP-conjugated donkey polyclonal anti-mouse IgG was from Jackson ImmunoResearch (Cat. No. 715036150) and HRP-conjugated goat polyclonal anti-rabbit IgG was from Cedarlane Laboratory (Cat. No. CLCC43007). SuperSignal West Pico PLUS chemiluminescent substrate was from Thermo Fisher.

Llama immunization

A male llama (Lama glama) was immunized by biolistic transfection using a Helios® gene gun system (Bio-Rad, Hercules, CA) followed by intradermal injection using a DERMOJET device (AKRA DERMOJET, Pau, France). Two pTT5 vectors [17] encoding either soluble human EGFRvIII (UniProt P00533: residues 1–29/Gly/297–645) or membrane-tethered human EGFRvIII (UniProt P00533: residues 1–29/Gly/297–668) were purified from overnight cultures of Escherichia coli DH5α cells using a QIAGEN® Plasmid Maxi Kit (Qiagen, Hilden, Germany). Briefly, 50 mg of gold particles were coated with 100 µl of 0.05 M spermidine, then vortexed and sonicated. An equimolar mixture of both pTT5 vectors (50 µg each; 100 µg total DNA in 100 µl ultrapure water) was added to the spermidine-coated gold particles, then 100 µl of 1 M CaCl2 was added dropwise to the mixture. After incubating for 10 min at room temperature, the gold particles were pelleted in a microfuge, washed three times with 100% ethanol and resuspended in 6 ml of 100% ethanol containing 0.05 mg/ml polyvinylpyrrolidone. The DNA–gold solution was dried onto the inner walls of two 30-inch lengths of gold-coat tubing under nitrogen flow, then the tubing was cut into 0.5-inch lengths.

The llama was immunized six times (weeks 0, 2, 4, 6, 9 and 12) by biolistic transfection; each immunization consisted of 12 bombardments administered at 600 PSI to shaved sites on the neck and hind limb (10 µg of total DNA per immunization). Thereafter, four additional immunizations (weeks 16, 20, 24 and 28) were administered by intradermal injection of 1 mg (1 mg/ml) of DNA using a DERMOJET device. Serum titration ELISA was conducted as described previously [18,19], and binding was detected using HRP-conjugated polyclonal goat anti-llama IgG. Experiments involving animals were conducted using protocols approved by the National Research Council Canada Animal Care Committee and in accordance with the guidelines set out in the OMAFRA Animals for Research Act, R.S.O. 1990, c. A.22.

Construction and panning of phage-displayed VHH library

A phage-displayed VHH library was constructed from the peripheral blood lymphocytes of the immunized llama as described previously [1820]. Briefly, peripheral blood mononuclear cells were purified by density gradient centrifugation from blood obtained 5 days following the third and the final DERMOJET immunizations. Total RNA was extracted from ∼5 × 107 cells from each time point using a PureLink RNA Mini Kit (Thermo Fisher) and cDNA was reverse transcribed using qScript® cDNA supermix containing random hexamer and olido(dT) primers (Quanta Biosciences, Gaithersburg, MD). VHH genes were amplified using semi-nested PCR and cloned into the phagemid vector pMED1 [20]; the final library had a size of 3 × 107 independent transformants and an insert rate of ∼75%. Phage particles were rescued from the library using M13K07 helper phage and panned against microplate-adsorbed human EGFRvIII for three rounds with triethylamine elution as described previously [1820]. A second independent library selection was carried out in the same manner, except that the target was streptavidin-captured biotinylated EGFRvIII.

Expression of VHHs and VHH-Fc fusions

VHH DNA sequences were cloned into the pSJF2H expression vector and monomeric VHHs tagged C-terminally with c-Myc and 6×His were purified from the periplasm of E. coli TG1 cells by IMAC as previously described [1820]. In addition, in vivo biotinylated monomeric VHHs were produced by co-transformation of E. coli BL21 (DE3) cells with two vectors encoding (i) VHHs C-terminally tagged with a biotin acceptor peptide and 6×His and (ii) the biotin ligase BirA and purified by IMAC [21]. Bivalent VHH-human IgG1 Fc fusions were produced by transient transfection of HEK293-6E cells followed by protein A affinity chromatography as previously described [17,22]. Heterodimeric biparatopic VHH-Fc fusions were produced by co-transfection of HEK293-6E cells with two pTT5 vectors encoding (i) NRC-sdAb032-Fc tagged C-terminally with 6×His and (ii) a second untagged VHH-Fc. The heterodimeric Ab was purified by sequential protein A affinity chromatography and IMAC and eluted using a linear 0 → 0.5 M imidazole gradient over 20 column volumes to separate species bearing one or two 6×His tags. VHHs and VHH-Fcs were dialyzed against or buffer-exchanged into phosphate-buffered saline (PBS), pH 7.4.

ELISA and EGF-competition ELISA

Wells of NUNC® MaxiSorp microtiter plates (Thermo Fisher) were coated overnight at 4°C with 2 µg/ml streptavidin in 100 µl of PBS, pH 7.4. The wells were blocked with 200 µl of PBS containing 1% (w/v) BSA for 1 h at 37°C and then biotinylated VHHs [10 µg/ml in 100 µl of PBS containing 1% BSA and 0.1% (v/v) Tween-20] were captured for 30 min at room temperature. The wells were washed 5× with PBS containing 0.1% Tween-20 and then incubated with human EGFR-Fc (500 ng/ml in 100 µl of PBS containing 1% BSA and 0.05% Tween-20) in the presence or absence of EGF (17 µg/ml) for 1 h at room temperature. The wells were washed 5× again and incubated with HRP-conjugated goat anti-human IgG (1 µg/ml in 100 µl of PBS containing 1% BSA and 0.05% Tween-20) for 1 h at room temperature. After a final wash (5× with PBS containing 0.1% Tween-20), the wells were developed with TMB substrate, stopped with 1 M H2SO4 and the absorbance at 450 nm was measured using a Multiskan FC photometer (Thermo Fisher).

Surface plasmon resonance

Prior to surface plasmon resonance (SPR) analyses, monomeric VHHs were purified by preparative size exclusion chromatography using a Superdex 75 10/300 GL column (GE Healthcare, Piscataway, NJ) connected to an ÄKTA FPLC protein purification system (GE Healthcare). In the first SPR experiment, multi-cycle kinetic analyses were performed on a Biacore 3000 instrument (GE Healthcare) at 25°C in HBS-EP buffer [10 mM HEPES, pH 7.4, containing 150 mM NaCl, 3 mM EDTA and 0.005% (w/v) surfactant P20]. Approximately 1304–2158 and 741 resonance units (RUs), respectively, of recombinant human EGFR and EGFRvIII ectodomains were immobilized on a CM5 sensor chip (GE Healthcare) in 10 mM acetate buffer, pH 4.5, using an amine coupling kit (GE Healthcare). An ethanolamine-blocked flow cell served as the reference. Monomeric VHHs at concentrations ranging from 0.1 to 100 nM were injected over the EGFR and EGFRvIII surfaces in HBS-EP buffer at a flow rate of 20 µl/min. For NRC-sdAb032 only, the VHH (212 RUs) was immobilized on a CM5 sensor chip by amine coupling and 0.5–50 nM recombinant human EGFR ectodomain was injected over the VHH surface in HBS-EP buffer at a flow rate of 20 µl/min. Contact times were 180–300 s and dissociation times were 300–600 s. The EGFR, EGFRvIII and NRC-sdAb032 surfaces were regenerated using a 10 s pulse of 10 mM glycine, pH 1.5.

In the second SPR experiment, single-cycle kinetic analyses were performed on a Biacore T200 instrument (GE Healthcare) at 25°C in HBS-EP buffer. Approximately 618, 1051 and 600 RUs of human, rhesus and mouse EGFR-Fc, respectively, were immobilized on three flow cells of a Series S Sensor Chip CM5 (GE Healthcare) in 10 mM acetate buffer, pH 4.5, using an amine coupling kit. An ethanolamine-blocked flow cell served as the reference. Monomeric VHHs at concentrations ranging from 0.6 to 50 nM were injected over the EGFR surfaces in HBS-EP buffer at a flow rate of 40 µl/min. The contact time was 180 s and the dissociation time was 600 s. The EGFR surfaces were regenerated using a 10 s pulse of 10 mM glycine, pH 1.5.

Epitope-binning experiments were performed essentially as described above on a Biacore 3000 instrument, except that 3383 RUs of human EGFR-Fc were immobilized on a CM5 sensor chip. A single VHH (or cetuximab) at a concentration equivalent to 25× KD was injected at 40 µl/min with a contact time of 150 s to saturate the EGFR surface. The second injection consisted of the same VHH (or cetuximab) in the presence of 25× the KD concentration of a second VHH. All data were analyzed by fitting to a 1 : 1 interaction model using BIAevaluation 4.1 software (GE Healthcare).

Flow cytometry and mirrorball® assays

MDA-MB-468 and MCF7 cells were cultured at 37°C in a humidified 5% CO2 atmosphere in T75 flasks containing RPMI-1640 medium supplemented with 10% (v/v) fetal bovine serum, 2 mM glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin and 250 ng/ml amphotericin B. For flow cytometry experiments, cells were grown to 70–80% confluency, dissociated using trypsin–EDTA solution, washed in PBS and then resuspended in PBS containing 1% BSA. Approximately 1 × 105 cells were stained sequentially on ice for 30 min with (i) 20 µg/ml of each VHH, (ii) 5 µg/ml of mouse anti-c-Myc IgG and (iii) 5 µg/ml of APC-conjugated goat anti-mouse IgG. The cells were washed with PBS in between each staining step, and after a final wash, data were acquired on a BD FACSCanto instrument (BD Biosciences, San Jose, CA).

For mirrorball® experiments, cells were dissociated from flasks using Accutase® solution, washed in Hank's Balanced Salt Solution (HBSS) and then ∼5000 cells in growth medium were plated in each rat tail collagen-coated well of a Nunc MicroWell 96-well optical bottom plate (Thermo Fisher). After incubating at 37°C/5% CO2 for 24 h, the cells were washed with HBSS and Abs [biotinylated VHHs, VHH-Fcs or cetuximab, serially diluted in live cell imaging buffer (LCIB) containing 1% BSA] were added to wells for 2 h at 4°C. The cells were washed with LCIB and secondary detection reagents (40 µg/ml PE-conjugated streptavidin or 30 µg/ml AF488-conjugated donkey anti-human IgG in LCIB containing 1% BSA) were added as appropriate to each well for 1 h at 4°C. The cells were washed with LCIB and stained with 1 µM DRAQ5 for 10 min at 4°C. After a final wash with LCIB, data were acquired on a mirrorball® microplate cytometer (TTP Labtech, Melbourn, U.K.) and analyzed using Cellista software (TTP Labtech).

EGFR phosphorylation assay

Approximately 2 × 105 MDA-MB-468 cells were seeded in wells of 12-well tissue culture plates and then starved in serum-free RPMI-1640 medium overnight. The next day, the medium was replaced with RPMI-1640 containing 1% BSA and various concentrations of erlotinib or Abs (VHH-Fcs or cetuximab). After 30 min at 37°C, EGF was added to a final concentration of 25 nM and incubated for a further 15 min. The cells were cooled immediately on ice, washed twice with PBS and then scraped in 100 µl Laemmli buffer. Cell lysates (5 µl) were electrophoresed on 4–20% Mini-Protean® TGX precast gels (Bio-Rad Laboratories) and transferred to polyvinylidene fluoride membranes using the semi-dry method. Western blotting was performed as previously described [6]. Briefly, membranes were blocked overnight at 4°C with 2% BSA in PBS, then sequentially incubated for 1 h at room temperature with (i) primary Abs (anti-phospho-EGFR, anti-EGFR or anti-β-actin, all diluted 1 : 1500 in PBS containing 1% BSA and 0.1% Tween-20) and (ii) secondary Abs (HRP-conjugated donkey anti-mouse IgG or goat anti-rabbit IgG, diluted 1 : 3000 in PBS containing 1% BSA and 0.1% Tween-20). The membranes were washed extensively with PBS containing 0.1% Tween-20 following incubations with primary and secondary Abs. The blots were developed using enhanced chemiluminescence and imaged using a Molecular Imager® Gel Doc XR+ System (Bio-Rad Laboratories). Band densitometry analysis was conducted using ImageJ version 1.52.

VHH humanization

VHHs were humanized by alignment with human IGHV3-30*01 and IGHJ1-1*01 amino acid sequences. For each VHH, three humanized variants were designed representing a spectrum of increasing homology to the human germline: (i) variant H1, in which all FR sequences were reverted to the human consensus excepting residues located within five positions of an FR-CDR boundary; (ii) variant H2, in which all FR sequences were reverted to the human consensus excepting residues located within two positions of an FR-CDR boundary and (iii) variant H3, in which all FR sequences were fully human. CDR residues as well as FR2 positions 42 and 52 (IMGT numbering) were left unaltered in all variants.

Results and discussion

Llama DNA immunization using gene gun and DERMOJET

We immunized a llama with DNA encoding human EGFRvIII six times every 2–3 weeks by biolistic transfection using a gene gun. No polyclonal serum Ab response was evident in the animal following the six gene gun immunizations, but boosting three times by intradermal injection using a DERMOJET device elicited serum Abs against recombinant EGFR with a half-maximal titer of ∼1 : 5000 (Figure 1). Binding of polyclonal serum Abs to EGFR and EGFRvIII was roughly equivalent and was not improved by further boosting.

Serum titration ELISA of the immunized llama against recombinant human EGFR and human EGFRvIII ectodomains.

Figure 1.
Serum titration ELISA of the immunized llama against recombinant human EGFR and human EGFRvIII ectodomains.

Wells of NUNC® MaxiSorp microtiter plates were coated overnight at 4°C with 100 ng of EGFR or EGFRvIII in 35 µl of PBS. The wells were blocked with 200 µl of PBS containing 2% skim milk for 1 h at 37°C and then sera (serially diluted in PBS containing 1% BSA and 0.1% Tween-20) were added to wells for 2 h at room temperature. The wells were washed 5× with PBS containing 0.1% Tween-20, incubated for 1 h with HRP-conjugated goat anti-llama IgG (diluted 1 : 5000 in PBS containing 1% BSA and 0.1% Tween-20), washed 5× again and then developed with TMB substrate.

Figure 1.
Serum titration ELISA of the immunized llama against recombinant human EGFR and human EGFRvIII ectodomains.

Wells of NUNC® MaxiSorp microtiter plates were coated overnight at 4°C with 100 ng of EGFR or EGFRvIII in 35 µl of PBS. The wells were blocked with 200 µl of PBS containing 2% skim milk for 1 h at 37°C and then sera (serially diluted in PBS containing 1% BSA and 0.1% Tween-20) were added to wells for 2 h at room temperature. The wells were washed 5× with PBS containing 0.1% Tween-20, incubated for 1 h with HRP-conjugated goat anti-llama IgG (diluted 1 : 5000 in PBS containing 1% BSA and 0.1% Tween-20), washed 5× again and then developed with TMB substrate.

VHHs elicited by DNA immunization targeted five unique EGFR epitopes including an epitope overlapping that of cetuximab

A phage-displayed VHH library was constructed from the peripheral blood lymphocytes of the immunized llama and VHHs were isolated by panning against either plate-adsorbed EGFRvIII or streptavidin-captured biotinylated EGFRvIII. Ten unique VHHs falling into three sequence families were identified in the panning against plate-adsorbed EGFRvIII (NRC-sdAb021 – NRC-sdAb030; Supplementary Table S1); all of these VHHs as well as two others (NRC-sdAb032 and NRC-sdAb033) were identified by panning against streptavidin-captured biotinylated EGFRvIII. The VHHs had monovalent binding affinities for recombinant human EGFR, ranging from 1 to 40 nM as measured by SPR, and, despite immunization with DNA encoding EGFRvIII, all showed identical binding to EGFR and EGFRvIII (Figure 2A and Table 1). In contrast with a previous report, we measured the EGFR-binding affinity of EG2 (a VHH raised by immunization with the recombinant EGFRvIII ectodomain) as ∼15–20 nM, not 55 nM [4]. The VHHs showed a variety of cross-species reactivity patterns (Table 2 and Supplementary Figure S1): EG2 VHH bound only human EGFR, NRC-sdAb022-family VHHs and NRC-sdAb032 bound human and rhesus EGFR with similar affinities, and NRC-sdAb029-family VHHs and NRC-sdAb033 bound human, rhesus and mouse EGFR with similar affinities. Surprisingly, NRC-sdAb028 bound human and mouse EGFR with similar affinity but did not react with rhesus EGFR. SPR epitope-binning co-injection experiments showed that: (i) EG2 VHH recognized a distinct epitope present only on human EGFR; (ii) NRC-sdAb029-family VHHs recognized a distinct epitope conserved across human, rhesus and mouse EGFR; (iii) NRC-sdAb022-family VHHs and NRC-sdAb028 recognized highly overlapping but distinct epitopes (conserved across human and rhesus EGFR and human, rhesus and mouse EGFR, respectively), while NRC-sdAb033 recognized a partially overlapping but highly conserved epitope, and (iv) NRC-sdAb032 recognized an epitope partially overlapping that of cetuximab (Figure 2B and Supplementary Figure S2). NRC-sdAb032 shared further similarities with cetuximab, in that both Abs cross-reacted with rhesus but not mouse EGFR, neither Ab bound well in SPR to immobilized EGFR but did bind EGFR in solution (data not shown), and both Abs competed with EGF for EGFR binding (Figure 2C).

Binding of VHHs to EGFR by SPR and ELISA.

Figure 2.
Binding of VHHs to EGFR by SPR and ELISA.

(A) SPR sensorgrams showing single-cycle kinetic analysis of VHH binding to human EGFR-Fc. Recombinant EGFR-Fc was immobilized on a Series S Sensor Chip CM5 using amine coupling, then the indicated VHH was flowed over the surface at concentrations ranging from 0.6 to 50 nM (NRC-sdAb022, 1.5–25 nM; NRC-sdAb028, 1.5–25 nM; NRC-sdAb029, 1.5–25 nM; NRC-sdAb032, 3–50 nM; NRC-sdAb033, 0.6–10 nM; EG2, 3–50 nM). Black lines show data and red lines show fits. (B) Summary of epitope binning by SPR. The colors of the circles indicate the cross-species conservation of the epitope (red, human EGFR only; blue, human and rhesus EGFR; yellow, human and mouse EGFR; green, human, rhesus and mouse EGFR). (C) Competitive ELISA showing binding of VHHs to EGFR in the presence or absence of EGF.

Figure 2.
Binding of VHHs to EGFR by SPR and ELISA.

(A) SPR sensorgrams showing single-cycle kinetic analysis of VHH binding to human EGFR-Fc. Recombinant EGFR-Fc was immobilized on a Series S Sensor Chip CM5 using amine coupling, then the indicated VHH was flowed over the surface at concentrations ranging from 0.6 to 50 nM (NRC-sdAb022, 1.5–25 nM; NRC-sdAb028, 1.5–25 nM; NRC-sdAb029, 1.5–25 nM; NRC-sdAb032, 3–50 nM; NRC-sdAb033, 0.6–10 nM; EG2, 3–50 nM). Black lines show data and red lines show fits. (B) Summary of epitope binning by SPR. The colors of the circles indicate the cross-species conservation of the epitope (red, human EGFR only; blue, human and rhesus EGFR; yellow, human and mouse EGFR; green, human, rhesus and mouse EGFR). (C) Competitive ELISA showing binding of VHHs to EGFR in the presence or absence of EGF.

Table 1
Monovalent affinities and kinetics of the interactions between VHHs and recombinant human EGFR and EGFRvIII extracellular domains (pH 7.4, 25°C)
VHEGFR EGFRvIII 
kon (M−1 s−1koff (s−1KD (nM) kon (M−1 s−1koff (s−1KD (nM) 
EG2 1.1 × 106 2.8 × 10−2 19.1 1.2 × 106 1.7 × 10−2 14.6 
NRC-sdAb021 3.1 × 105 1.2 × 10−2 38.5 2.4 × 105 9.8 × 10−3 40.4 
NRC-sdAb022 4.0 × 105 6.7 × 10−4 1.7 2.9 × 105 7.7 × 10−4 2.6 
NRC-sdAb023 4.6 × 105 5.7 × 10−3 12.6 3.4 × 105 6.7 × 10−3 19.4 
NRC-sdAb024 2.5 × 105 7.3 × 10−4 2.9 1.8 × 105 9.0 × 10−4 4.9 
NRC-sdAb025 3.6 × 105 5.9 × 10−4 1.7 2.6 × 105 7.1 × 10−4 2.8 
NRC-sdAb026 3.6 × 105 5.6 × 10−3 15.4 2.7 × 105 6.8 × 10−3 25.0 
NRC-sdAb027 5.1 × 105 7.9 × 10−4 1.6 3.8 × 105 8.6 × 10−4 2.3 
NRC-sdAb028 8.4 × 104 7.5 × 10−4 9.0 9.8 × 104 8.4 × 10−4 8.6 
NRC-sdAb029 4.0 × 105 4.0 × 10−3 9.9 4.4 × 105 4.4 × 10−3 10.2 
NRC-sdAb030 2.7 × 105 5.9 × 10−3 21.4 2.9 × 105 6.6 × 10−3 22.7 
NRC-sdAb032 1.0 × 105 1.3 × 10−3 12.91 n.d.2 n.d.2 n.d.2 
NRC-sdAb033 1.5 × 105 1.7 × 10−4 1.1 n.d.2 n.d.2 n.d.2 
VHEGFR EGFRvIII 
kon (M−1 s−1koff (s−1KD (nM) kon (M−1 s−1koff (s−1KD (nM) 
EG2 1.1 × 106 2.8 × 10−2 19.1 1.2 × 106 1.7 × 10−2 14.6 
NRC-sdAb021 3.1 × 105 1.2 × 10−2 38.5 2.4 × 105 9.8 × 10−3 40.4 
NRC-sdAb022 4.0 × 105 6.7 × 10−4 1.7 2.9 × 105 7.7 × 10−4 2.6 
NRC-sdAb023 4.6 × 105 5.7 × 10−3 12.6 3.4 × 105 6.7 × 10−3 19.4 
NRC-sdAb024 2.5 × 105 7.3 × 10−4 2.9 1.8 × 105 9.0 × 10−4 4.9 
NRC-sdAb025 3.6 × 105 5.9 × 10−4 1.7 2.6 × 105 7.1 × 10−4 2.8 
NRC-sdAb026 3.6 × 105 5.6 × 10−3 15.4 2.7 × 105 6.8 × 10−3 25.0 
NRC-sdAb027 5.1 × 105 7.9 × 10−4 1.6 3.8 × 105 8.6 × 10−4 2.3 
NRC-sdAb028 8.4 × 104 7.5 × 10−4 9.0 9.8 × 104 8.4 × 10−4 8.6 
NRC-sdAb029 4.0 × 105 4.0 × 10−3 9.9 4.4 × 105 4.4 × 10−3 10.2 
NRC-sdAb030 2.7 × 105 5.9 × 10−3 21.4 2.9 × 105 6.6 × 10−3 22.7 
NRC-sdAb032 1.0 × 105 1.3 × 10−3 12.91 n.d.2 n.d.2 n.d.2 
NRC-sdAb033 1.5 × 105 1.7 × 10−4 1.1 n.d.2 n.d.2 n.d.2 
1

Determined by amine-coupling NRC-sdAb032 and flowing recombinant EGFR ectodomain.

2

No difference was observed in ELISA binding to human EGFR or EGFRvIII (data not shown); n.d., not determined.

Table 2
Monovalent affinities and kinetics of the interactions between VHHs and human, rhesus and mouse EGFR-Fc (pH 7.4, 25°C)
VHHuman EGFR-Fc Rhesus EGFR-Fc Mouse EGFR-Fc 
kon (M−1 s−1koff (s−1KD (nM) kon (M−1 s−1koff (s−1KD (nM) kon (M−1 s−1koff (s−1KD (nM) 
EG2 8.5 × 105 8.5 × 10−3 11.1 ± 0.2 — — — — — — 
NRC-sdAb022 2.5 × 105 2.2 × 10−3 8.9 ± 0.1 8.5 × 104 1.7 × 10−3 19.6 ± 0.7 — — — 
NRC-sdAb028 9.2 × 104 5.5 × 10−4 6.0 ± 0.1 — — — 6.5 × 105 7.5 × 10−3 11.6 ± 0.3 
NRC-sdAb029 5.3 × 105 4.5 × 10−3 8.5 ± 0.1 5.5 × 105 6.4 × 10−3 11.8 ± 0.3 5.1 × 105 6.3 × 10−3 12.5 ± 1.2 
NRC-sdAb032 3.6 × 105 2.5 × 10−3 6.9 ± 0.02 4.8 × 105 3.9 × 10−3 8.1 ± 0.1 — — — 
NRC-sdAb033 9.0 × 104 1.6 × 10−4 1.8 ± 0.1 1.1 × 105 1.1 × 10−4 1.0 ± 0.1 9.2 × 104 2.8 × 10−4 3.1 ± 0.2 
VHHuman EGFR-Fc Rhesus EGFR-Fc Mouse EGFR-Fc 
kon (M−1 s−1koff (s−1KD (nM) kon (M−1 s−1koff (s−1KD (nM) kon (M−1 s−1koff (s−1KD (nM) 
EG2 8.5 × 105 8.5 × 10−3 11.1 ± 0.2 — — — — — — 
NRC-sdAb022 2.5 × 105 2.2 × 10−3 8.9 ± 0.1 8.5 × 104 1.7 × 10−3 19.6 ± 0.7 — — — 
NRC-sdAb028 9.2 × 104 5.5 × 10−4 6.0 ± 0.1 — — — 6.5 × 105 7.5 × 10−3 11.6 ± 0.3 
NRC-sdAb029 5.3 × 105 4.5 × 10−3 8.5 ± 0.1 5.5 × 105 6.4 × 10−3 11.8 ± 0.3 5.1 × 105 6.3 × 10−3 12.5 ± 1.2 
NRC-sdAb032 3.6 × 105 2.5 × 10−3 6.9 ± 0.02 4.8 × 105 3.9 × 10−3 8.1 ± 0.1 — — — 
NRC-sdAb033 9.0 × 104 1.6 × 10−4 1.8 ± 0.1 1.1 × 105 1.1 × 10−4 1.0 ± 0.1 9.2 × 104 2.8 × 10−4 3.1 ± 0.2 

(—) indicates no binding.

KD values are expressed as the means ± standard deviations of three independent experiments.

Despite their broadly similar monovalent affinities for recombinant EGFR (1–40 nM), the VHHs showed significant variability in their ability to recognize EGFR-positive tumor cell lines. Flow cytometry showed that four of five epitope bins targeted by the VHH monomers were accessible on native cell-surface EGFR; neither binding of NRC-sdAb029 nor, surprisingly, binding of EG2 to MDA-MB-468 cells was detectable at the single concentration (20 µg/ml, equivalent to ∼1.3 µM) used in this assay (Figure 3A). Titration of the VHH-Fc fusions against adherent MDA-MB-468 cells used in mirrorball® microplate cytometry assay revealed very weak binding of EG2-Fc (EC50: 279 nM), moderate binding of NRC-sdAb022-Fc, NRC-sdAb028-Fc and NRC-sdAb033-Fc (EC50: 67, 86 and 49 nM, respectively) and strong binding of NRC-sdAb032 (EC50: 0.5 nM, similar to cetuximab) (Figure 3B). Similar binding patterns were observed for in vivo biotinylated VHH monomers (data not shown), and no binding of either the VHHs or VHH-Fcs was observed to EGFR-low MCF7 cells (Figure 3C). Binding by biparatopic heterodimeric VHH/VHH-Fcs combining NRC-sdAb032 with each of the other possible VHHs was similar to the parental homodimeric NRC-sdAb032-Fc (Figure 3D), suggesting that the monovalent interaction of the NRC-sdAb032 VHH arm with EGFR drove the majority of binding in this assay. Thus, almost all of the VHHs raised by DNA immunization recognized native EGFR on tumor cells better than EG2 VHH, and one VHH-Fc (NRC-sdAb032-Fc) had an EC50 1–2 logs lower than previously reported VHHs (in bivalent form) raised by recombinant protein immunization [4,6].

Binding of VHHs to EGFR-positive MDA-MB-468 cells and EGFR-low MCF7 cells.

Figure 3.
Binding of VHHs to EGFR-positive MDA-MB-468 cells and EGFR-low MCF7 cells.

(A) Binding of VHH monomers (20 µg/ml) to EGFR-positive MDA-MB-468 cells by flow cytometry detected using anti-c-Myc and AF647-labeled anti-mouse antibodies. (B and C) Binding of serially diluted VHH-Fcs to EGFR-positive MDA-MB-468 cells (B) and EGFR-low MCF7 cells (C) by mirrorball® microplate cytometry detected using AF488-labeled anti-human Fc antibody. EC50s were determined by curve fitting in Graphpad Prism using the equation for one-site specific binding with Hill slope. (D) Binding of serially diluted biparatopic VHH-Fcs by mirrorball® microplate cytometry detecting using AF488-labeled anti-human Fc antibody. EC50s were determined by curve fitting in Graphpad Prism using the equation for one-site specific binding with Hill slope.

Figure 3.
Binding of VHHs to EGFR-positive MDA-MB-468 cells and EGFR-low MCF7 cells.

(A) Binding of VHH monomers (20 µg/ml) to EGFR-positive MDA-MB-468 cells by flow cytometry detected using anti-c-Myc and AF647-labeled anti-mouse antibodies. (B and C) Binding of serially diluted VHH-Fcs to EGFR-positive MDA-MB-468 cells (B) and EGFR-low MCF7 cells (C) by mirrorball® microplate cytometry detected using AF488-labeled anti-human Fc antibody. EC50s were determined by curve fitting in Graphpad Prism using the equation for one-site specific binding with Hill slope. (D) Binding of serially diluted biparatopic VHH-Fcs by mirrorball® microplate cytometry detecting using AF488-labeled anti-human Fc antibody. EC50s were determined by curve fitting in Graphpad Prism using the equation for one-site specific binding with Hill slope.

NRC-sdAb032, a VHH elicited by DNA immunization, inhibited EGFR signaling with potency similar to cetuximab

We tested the ability of all of the VHH-Fcs showing significant binding to EGFR-positive tumor cells (NRC-sdAb022-Fc, NRC-sdAb028-Fc, NRC-sdAb032-Fc and NRC-sdAb033-Fc), as well as cetuximab and EG2-Fc as a historical control, to inhibit EGF-induced EGFR phosphorylation. At a single concentration of 500 nM, NRC-sdAb032-Fc was the only VHH-Fc showing significant inhibition of EGFR signaling (Figure 4A,B). Reduction in EGFR phosphorylation in the presence of NRC-sdAb032-Fc was dose-dependent, with an IC50 between 25 and 50 nM, similar to that of cetuximab (Figure 4C,D). No improvement in potency was achieved by biparatopic heterodimeric VHH/VHH-Fcs combining NRC-sdAb032 with each other VHH, although there appeared to be a general advantage of molecules bearing two EGFR-binding arms (other than EG2) over VHH/VHH-Fcs bearing a single EGFR-binding arm (Figure 4E,F).

Inhibition of EGF-induced EGFR phosphorylation in MDA-MB-468 cells by VHH-Fcs.

Figure 4.
Inhibition of EGF-induced EGFR phosphorylation in MDA-MB-468 cells by VHH-Fcs.

(A) Western blotting of phosphorylated EGFR (Tyr1068), total EGFR and β-actin in serum-starved MDA-MB-468 cells treated with the indicated inhibitor (all at 500 nM) in the presence or absence of EGF. (B) Densitometry analysis of bands in (A). (C) Western blotting of phosphorylated EGFR (Tyr1068), total EGFR and β-actin in serum-starved MDA-MB-468 cells treated with the indicated inhibitor in the presence or absence of EGF. (D) Densitometry analysis of bands in (C). The concentrations used are as shown in (C). A20.1-Fc is an irrelevant VHH-Fc against Clostridium difficile toxin A [24] included as a negative control. (E) Western blotting of phosphorylated EGFR (Tyr1068), total EGFR and β-actin in serum-starved MDA-MB-468 cells treated with the indicated inhibitor (all at 5 nM) in the presence or absence of EGF. (F) Densitometry analysis of bands in (E).

Figure 4.
Inhibition of EGF-induced EGFR phosphorylation in MDA-MB-468 cells by VHH-Fcs.

(A) Western blotting of phosphorylated EGFR (Tyr1068), total EGFR and β-actin in serum-starved MDA-MB-468 cells treated with the indicated inhibitor (all at 500 nM) in the presence or absence of EGF. (B) Densitometry analysis of bands in (A). (C) Western blotting of phosphorylated EGFR (Tyr1068), total EGFR and β-actin in serum-starved MDA-MB-468 cells treated with the indicated inhibitor in the presence or absence of EGF. (D) Densitometry analysis of bands in (C). The concentrations used are as shown in (C). A20.1-Fc is an irrelevant VHH-Fc against Clostridium difficile toxin A [24] included as a negative control. (E) Western blotting of phosphorylated EGFR (Tyr1068), total EGFR and β-actin in serum-starved MDA-MB-468 cells treated with the indicated inhibitor (all at 5 nM) in the presence or absence of EGF. (F) Densitometry analysis of bands in (E).

Humanization of VHHs

With the aim of using these VHHs in therapeutic applications, the sequences of the four VHHs showing significant binding to EGFR-positive tumor cells (NRC-sdAb022, NRC-sdAb028, NRC-sdAb032 and NRC-sdAb033) were humanized with reference to human IGHV3-30*01 and IGHJ1*01 germline genes. This process yielded at least one humanized variant with unimpaired solubility and EGFR-binding affinity for three of the four parent VHHs (NRC-sdAb028-H1, NRC-sdAb032-H1 and NRC-sdAb033-H2). The framework regions of these humanized variants bore 89–94% sequence identity with human IGHV3-30*01, with no or minimal impact on the biophysical properties and EGFR-binding affinities of the resulting VHHs (Table 3 and Supplementary Table S2).

Table 3
Aggregation propensities and monovalent affinities of humanized VHHs for human EGFR-Fc
VHMonomer (%)1 kon (M−1 s−1koff (s−1KD (nM) 
NRC-sdAb022 >95 2.5 × 105 2.2 × 10−3 8.9 ± 0.1 
NRC-sdAb022-H1 n.d.2    
NRC-sdAb022-H2 <50 n.b.3 n.b.3 n.b.3 
NRC-sdAb022-H3 n.d.2    
NRC-sdAb028 >95 9.2 × 104 5.5 × 10−4 6.0 ± 0.1 
NRC-sdAb028-H1 >95 9.3 × 104 5.8 × 10−4 6.2 ± 0.2 
NRC-sdAb028-H2 n.d.2    
NRC-sdAb028-H3 n.d.2    
NRC-sdAb032 >95 3.6 × 105 2.5 × 10−3 6.9 ± 0.02 
NRC-sdAb032-H1 >95 6.2 × 105 1.0 × 10−2 16.6 ± 0.5 
NRC-sdAb032-H2 <50 n.d.4 n.d.4 254 ± 31 
NRC-sdAb032-H3 n.d.2    
NRC-sdAb033 >95 9.0 × 104 1.6 × 10−4 1.8 ± 0.1 
NRC-sdAb033-H1 >95 1.3 × 105 2.2 × 10−4 1.7 ± 0.03 
NRC-sdAb033-H2 >95 2.0 × 105 2.6 × 10−4 1.4 ± 0.1 
NRC-sdAb033-H3 <50 1.3 × 105 1.1 × 10−2 83.8 ± 3.7 
VHMonomer (%)1 kon (M−1 s−1koff (s−1KD (nM) 
NRC-sdAb022 >95 2.5 × 105 2.2 × 10−3 8.9 ± 0.1 
NRC-sdAb022-H1 n.d.2    
NRC-sdAb022-H2 <50 n.b.3 n.b.3 n.b.3 
NRC-sdAb022-H3 n.d.2    
NRC-sdAb028 >95 9.2 × 104 5.5 × 10−4 6.0 ± 0.1 
NRC-sdAb028-H1 >95 9.3 × 104 5.8 × 10−4 6.2 ± 0.2 
NRC-sdAb028-H2 n.d.2    
NRC-sdAb028-H3 n.d.2    
NRC-sdAb032 >95 3.6 × 105 2.5 × 10−3 6.9 ± 0.02 
NRC-sdAb032-H1 >95 6.2 × 105 1.0 × 10−2 16.6 ± 0.5 
NRC-sdAb032-H2 <50 n.d.4 n.d.4 254 ± 31 
NRC-sdAb032-H3 n.d.2    
NRC-sdAb033 >95 9.0 × 104 1.6 × 10−4 1.8 ± 0.1 
NRC-sdAb033-H1 >95 1.3 × 105 2.2 × 10−4 1.7 ± 0.03 
NRC-sdAb033-H2 >95 2.0 × 105 2.6 × 10−4 1.4 ± 0.1 
NRC-sdAb033-H3 <50 1.3 × 105 1.1 × 10−2 83.8 ± 3.7 
1

Determined by SEC (% monomer peak area).

2

n.d., not determined due to low or no expression.

3

n.b., no binding of the purified monomeric sdAb to human EGFR-Fc could be detected.

4

n.d., not determined because steady-state KD was calculated.

Conclusions and implications

The VHHs isolated and characterized here were generated by immunization of a llama with DNA alone (no protein or cell boost), using biolistic transfection (gene gun) followed by intradermal injection (DERMOJET). Many of the VHHs had high affinities for recombinant human EGFR, cross-reacted with rhesus and/or mouse EGFR, and recognized native cell-surface EGFR on tumor cell lines; in all of these respects, the VHHs described here are dramatically superior to those previously isolated by our group using recombinant protein immunization [4]. One VHH, NRC-sdAb032, showed clear functional activity as an inhibitor of EGFR signaling, and any of the humanized VHHs may have other therapeutic applications. NRC-sdAb032 was isolated by panning on streptavidin-captured EGFR but not on passively adsorbed EGFR, and its epitope appears to be present on native EGFR and recombinant EGFR ectodomain in solution but not in the same ectodomain when it is passively adsorbed to microtiter plates or amine-coupled to sensor chips. This epitope, which overlaps the cetuximab epitope and the EGF-binding site in EGFR domain III, is apparently poorly available for binding by NRC-sdAb032 or cetuximab in ELISAs against directly adsorbed EGFR or in SPR experiments in which these antibodies are flowed over amine-coupled EGFR surfaces; in contrast, both NRC-sdAb032 and cetuximab bound EGFR+ tumor cells with lower EC50s than other VHHs with similar monovalent binding affinities for recombinant EGFR. We speculate that the NRC-sdAb032 epitope, though easily destroyed, contributes to this VHH's superior recognition of native EGFR displayed on the tumor cell surface. Other groups have had similar experiences using streptavidin-captured antigens as ‘bait’ during panning of phage-displayed VHH libraries [23].

We are currently investigating whether the DNA immunization schedule presented here is consistent and generalizable to other targets. Our preliminary findings have indicated that up to five intradermal injections alone were unable to elicit a polyclonal serum Ab response against EGFR in several other llamas, and thus one possibility is that either a prolonged immunization schedule or a combination of two immunization routes (particle bombardment; intradermal injection) may be necessary to trigger seroconversion. These hypotheses are under investigation (Mehdi Arbabi-Ghahroudi, in preparation), although evaluating the impact of immunization strategies in the context of the significant heterogeneity expected in large outbred animals is a major challenge.

Abbreviations

     
  • Abs

    antibodies

  •  
  • AF488

    Alexa Fluor® 488

  •  
  • APC

    allophycocyanin

  •  
  • BSA

    bovine serum albumin

  •  
  • EGF

    epidermal growth factor

  •  
  • EGFR

    epidermal growth factor receptor

  •  
  • GPCR

    G protein-coupled receptor

  •  
  • HBSS

    Hank's Balanced Salt Solution

  •  
  • HRP

    horseradish peroxidase

  •  
  • IMAC

    immobilized metal affinity chromatography

  •  
  • LCIB

    live cell imaging buffer

  •  
  • PBS

    phosphate-buffered saline

  •  
  • PE

    R-phycoerythrin

  •  
  • RU

    resonance unit

  •  
  • sdAb

    single-domain antibody

  •  
  • SPR

    surface plasmon resonance

  •  
  • TMB

    3,3′,5,5′-tetramethylbenzidine

  •  
  • VHH

    variable domain of camelid heavy chain-only antibody

Author Contribution

K.A.H., C.R.M. and M.A.-G. designed the immunization study. K.A.H. performed serology and built the phage-displayed VHH library. K.A.H. and M.A.R. isolated VHHs by panning. H.F. and G.H. conducted SPR experiments. D.C. performed mirrorball® experiments. M.A.R. produced VHH-Fc fusions and performed western blotting experiments to assess EGFR phosphorylation status. M.A.R. and J.T. designed and produced biparatopic VHH-Fc fusions. K.A.H. and G.H. made the figures and K.A.H. wrote the paper. J.T. and M.A.-G. proofread the text. All authors approved the final manuscript.

Funding

This work was supported by funding from the National Research Council Canada.

Acknowledgements

We gratefully acknowledge the excellent technical help of Yonghong Guan, Hong Tong-Sevinc, Qingling Yang and Shalini Raphael.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Herbst
,
R.S.
and
Shin
,
D.M.
(
2002
)
Monoclonal antibodies to target epidermal growth factor receptor-positive tumors: a new paradigm for cancer therapy
.
Cancer
94
,
1593
1611
2
Jonker
,
D.J.
,
O'Callaghan
,
C.J.
,
Karapetis
,
C.S.
,
Zalcberg
,
J.R.
,
Tu
,
D.
,
Au
,
H.-J.
et al.  (
2007
)
Cetuximab for the treatment of colorectal cancer
.
N. Engl. J. Med.
357
,
2040
2048
3
Bonner
,
J.A.
,
Harari
,
P.M.
,
Giralt
,
J.
,
Azarnia
,
N.
,
Shin
,
D.M.
,
Cohen
,
R.B.
et al.  (
2006
)
Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck
.
N. Engl. J. Med.
354
,
567
578
4
Bell
,
A.
,
Wang
,
Z.J.
,
Arbabi-Ghahroudi
,
M.
,
Chang
,
T.A.
,
Durocher
,
Y.
,
Trojahn
,
U.
et al.  (
2010
)
Differential tumor-targeting abilities of three single-domain antibody formats
.
Cancer Lett.
289
,
81
90
5
Gottlin
,
E.B.
,
Xiangrong
,
G.
,
Pegram
,
C.
,
Cannedy
,
A.
,
Campa
,
M.J.
and
Patz
, Jr,
E.F.
(
2009
)
Isolation of novel EGFR-specific VHH domains
.
J. Biomol. Screen.
14
,
77
85
6
Roovers
,
R.C.
,
Laeremans
,
T.
,
Huang
,
L.
,
De Taeye
,
S.
,
Verkleij
,
A.J.
,
Revets
,
H.
et al.  (
2007
)
Efficient inhibition of EGFR signaling and of tumour growth by antagonistic anti-EFGR nanobodies
.
Cancer Immunol. Immunother.
56
,
303
317
7
Salema
,
V.
,
Mañas
,
C.
,
Cerdán
,
L.
,
Piñero-Lambea
,
C.
,
Marín
,
E.
,
Roovers
,
R.C.
et al.  (
2016
)
High affinity nanobodies against human epidermal growth factor receptor selected on cells by E. coli display
.
MAbs
8
,
1286
1301
8
Roovers
,
R.C.
,
Vosjan
,
M.J.
,
Laeremans
,
T.
,
el Khoulati
,
R.
,
de Bruin
,
R.C.
,
Ferguson
,
K.M.
et al.  (
2011
)
A biparatopic anti-EGFR nanobody efficiently inhibits solid tumour growth
.
Int. J. Cancer
129
,
2013
2024
9
Niu
,
G.
,
Murad
,
Y.M.
,
Gao
,
H.
,
Hu
,
S.
,
Guo
,
N.
,
Jacobson
,
O.
et al.  (
2012
)
Molecular targeting of CEACAM6 using antibody probes of different sizes
.
J. Control Release
161
,
18
24
10
Babiuk
,
L.A.
,
Pontarollo
,
R.
,
Babiuk
,
S.
,
Loehr
,
B.
and
van Drunen Littel-van den Hurk
,
S.
(
2003
)
Induction of immune responses by DNA vaccines in large animals
.
Vaccine
21
,
649
658
11
Peyrassol
,
X.
,
Laeremans
,
T.
,
Gouwy
,
M.
,
Lahura
,
V.
,
Debulpaep
,
M.
,
Van Damme
,
J.
et al.  (
2016
)
Development by genetic immunization of monovalent antibodies (nanobodies) behaving as antagonists of the human ChemR23 receptor
.
J. Immunol.
196
,
2893
2901
12
Peyrassol
,
X.
,
Laeremans
,
T.
,
Lahura
,
V.
,
Debulpaep
,
M.
,
El Hassan
,
H.
,
Steyaert
,
J.
et al.  (
2018
)
Development by genetic immunization of monovalent antibodies against human vasoactive intestinal peptide receptor 1 (VPAC1), new innovative, and versatile tools to study VPAC1 receptor function
.
Front. Endocrinol.
9
,
153
13
van der Woning
,
B.
,
De Boeck
,
G.
,
Blanchetot
,
C.
,
Bobkov
,
V.
,
Klarenbeek
,
A.
,
Saunders
,
M.
et al.  (
2016
)
DNA immunization combined with scFv phage display identifies antagonistic GCGR specific antibodies and reveals new epitopes on the small extracellular loops
.
MAbs
8
,
1126
1135
14
Koch-Nolte
,
F.
,
Reyelt
,
J.
,
Schössow
,
B.
,
Schwarz
,
N.
,
Scheuplein
,
F.
,
Rothenburg
,
S.
et al.  (
2007
)
Single domain antibodies from llama effectively and specifically block T cell ecto-ADP-ribosyltransferase ART2.2 in vivo
.
FASEB J.
21
,
3490
3498
15
Danquah
,
W.
,
Meyer-Schwesinger
,
C.
,
Rissiek
,
B.
,
Pinto
,
C.
,
Serracant-Prat
,
A.
,
Amadi
,
M.
et al.  (
2016
)
Nanobodies that block gating of the P2X7 ion channel ameliorate inflammation
.
Sci. Transl. Med.
8
,
366ra162
16
Fumey
,
W.
,
Koenigsdorf
,
J.
,
Kunick
,
V.
,
Menzel
,
S.
,
Schütze
,
K.
,
Unger
,
M.
et al.  (
2017
)
Nanobodies effectively modulate the enzymatic activity of CD38 and allow specific imaging of CD38(+) tumors in mouse models in vivo
.
Sci. Rep.
7
,
14289
17
Durocher
,
Y.
,
Perret
,
S.
and
Kamen
,
A.
(
2002
)
High-level and high-throughput recombinant protein production by transient transfection of suspension-growing human 293-EBNA1 cells
.
Nucleic Acids Res.
30
,
E9
18
Henry
,
K.A.
,
Hussack
,
G.
,
Collins
,
C.
,
Zwaagstra
,
J.C.
,
Tanha
,
J.
and
MacKenzie
,
C.R.
(
2016
)
Isolation of TGF-beta-neutralizing single-domain antibodies of predetermined epitope specificity using next-generation DNA sequencing
.
Protein Eng. Des. Sel.
29
,
439
443
19
Henry
,
K.A.
,
Tanha
,
J.
and
Hussack
,
G.
(
2015
)
Identification of cross-reactive single-domain antibodies against serum albumin using next-generation DNA sequencing
.
Protein Eng. Des. Sel.
28
,
379
383
20
Baral
,
T.N.
,
MacKenzie
,
R.
and
Arbabi Ghahroudi
,
M.
(
2013
)
Single-domain antibodies and their utility
.
Curr. Protoc. Immunol.
103
,
Unit 2 17
21
Rossotti
,
M.A.
,
Pirez
,
M.
,
González-Techera
,
A.
,
Cui
,
Y.
,
Bever
,
C.S.
,
Lee
,
K.S.
et al.  (
2015
)
Method for sorting and pairwise selection of nanobodies for the development of highly sensitive sandwich immunoassays
.
Anal. Chem.
87
,
11907
11914
22
Zhang
,
J.
,
Liu
,
X.
,
Bell
,
A.
,
To
,
R.
,
Baral
,
T.N.
,
Azizi
,
A.
et al.  (
2009
)
Transient expression and purification of chimeric heavy chain antibodies
.
Protein Expr. Purif.
65
,
77
82
23
Pardon
,
E.
,
Laeremans
,
T.
,
Triest
,
S.
,
Rasmussen
,
S.G.
,
Wohlkönig
,
A.
,
Ruf
,
A.
et al.  (
2014
)
A general protocol for the generation of nanobodies for structural biology
.
Nat. Protoc.
9
,
674
693
24
Hussack
,
G.
,
Arbabi-Ghahroudi
,
M.
,
van Faassen
,
H.
,
Songer
,
J.G.
,
Ng
,
K.K.
,
MacKenzie
,
R.
et al.  (
2011
)
Neutralization of Clostridium difficile toxin A with single-domain antibodies targeting the cell receptor binding domain
.
J. Biol. Chem.
286
,
8961
8976

Author notes

*

These authors contributed equally to this work.