OUP user menu

Isolation of a Novel Nanobody Against HER-2/neu Using Phage Displays Technology

Farzaneh Sheikholeslami, Mohammad J. Rasaee, Mohammad A. Shokrgozar, Manijeh Mokhtari Dizaji, Fatemeh Rahbarizadeh, Davoud Ahmadvande
DOI: http://dx.doi.org/10.1309/LM0WXKM0R0DVUZWF 69-76 First published online: 1 February 2010

In 1993, Hamers-Casterman and colleagues discovered a novel type of IgG antibodies.1 Camelid serum contains functional heavy-chain homodimeric antibodies without light chains. The variable domain of these single chain antibodies are named as VHHs (nanobodies). The binding ability of VHH to antigens was found to be similar or even better than classical IgG.2 Single-domain antibody was also identified in special cartilaginous fish by Greenberg and colleagues.3 It was also found that they are suitable for expression as multivalent formats like bispecific, enzyme, or toxin-VHH fusions to increase avidity without using current linkers like those that existed between VL and VH domains, which often result in aggregation and decrease affinity because of mispairing of VL and VH domains.4 Lacking of the mispairing problems allows designing more flexible linkers and successfully achieving functional trivalent-bispecific VHHs.5,6

Sequencing and crystallographic analysis of camel VHH has showed homology to the human VHH. The above facts prepare some advantages for biotechnological applications of VHH, such as facile genetic manipulation, increased functional size of libraries, high physicochemical stability, ease of production of multivalent forms, high stability, recognition of hidden antigenic sites, well expression, rapid tissue penetration, and fast blood clearance.7 Soluble VHHs have been produced in Escherichia coli (E. coli), filamentous fungi, Saccharomyces cerevisiae, and Pichia pastoris.811

VHH single-domains can be used whenever high stability is required, such as use in shampoo for preventing dandruff.12 They were used in immunoaffinity purification as capturing reagents and in biosensor technology.13,14 VHH is also reported to be a good candidate for oral immunotherapy because of its resistance against high pH, and its ability to bind to the target in the presence of high concentrations of many other agents.15 This type of functional VHH antibody can be protected from proteolysis by local VHH production by applying natural elementary channel bacteria. Then diarrhea could be prevented by lactobacilli, which produce rotavirus-neutralizing VHHs fused to a cell surface anchor.16 Because of rapid renal clearance, the usage of VHH in parenteral applications is limited. Therefore, VHH must be attached to large serum proteins such as albumin or PEGylated in order to not only increase half life but also to increase its virus-neutralizing potency.17 VHHs could be used as targeting devices for toxic enzymes or to block a specific molecular interaction. In sleeping sickness, some VHHs bind to trypanosome coat protein and fuse to apolipoprotein L-1 enzyme, resulting in trypanosome lysis.18

Muruganandam and colleagues investigated VHH accumulation in mice brains after intravenous injections in targeting drugs across the blood brain barrier.19 They found this reagent would be useful for neurological disease therapy. Multiple injections of VHHs have not shown any immunogenicity in mice in contrast with specific antibodies. It depends on their high sequence homology to normal VH domains.20

Our laboratory is the first lab which focused on the production of various VHHs against different tumor markers.2123 In this way we produced a nanobody against well-known human epidermal growth factor receptor-2s (HER2) surface tumor markers. EGFRII, also known as HER-2, c-erb B-2, or neu, is a rational target for antitumor strategies. HER-2 gene amplification or protein over expression is found in approximately 30% of human breast carcinomas. HER-2 is a 185 KDa transmembrane tyrosine kinase receptor. Many animal models and in vitro experiments have shown genotypic change of HER-2 amplification plays a pivotal role in oncogenic transformation, tumorgenesis, and metastasis. This obviously found the growth of the HER-2 positive human breast cancer cell lines and tumor xenografts is inhibited by anti-HER-2 monoclonal antibodies or VHH nanobodies. Use of monoclonal antibody (like Herceptin) therapy is an important strategy for HER-2 targeting and receptor inhibition in breast cancer. VHH, however, has many advantages over murine monoclonal antibodies in cancer antigen targeting.24

In this study we produced a recombinant VHH nanobody against ECD-HER2 following phage display technology. Also, we applied a new instrument for targeting ECD-HER2.

Materials and Methods

Preparation of Extracellular Domain of HER-2 (ECD-HER2)

The ECD-HER2 (amino acid residues 1–627) was subcloned in expression vector pPICZαC for soluble expression in yeast Pichia pastoris. Pichia pastoris was grown at 28°C in 250 mL BMGY medium in a culture flask until OD550 reached 6–8. The cells were washed and resuspended in 1,000 mL BMGY, induced by absolute methanol (0.1% total volume), and incubated for 3 days at −28°C. At the end of the incubation time, cells were harvested by centrifugation (4,000g, 4°C for 20 minutes) and the pallets were resuspended in PBS (10mM, pH 7.2) and NaN3 (final concentration 0.05%). Protease inhibitor phenylmethylsulfonyl fluoride (PMSF) was added to the final concentration of 1.0 mM, the supernatant was centrifuged (14,000g for 30 minutes at 4°C), and concentrated 20 fold by ultrafiltration (Millipore cut off 50 kDa). Finally, the protein was purified by loading the centrifuged supernatant onto the Nicle-nitrilo-triacetic acid column (Qiagen, Valencia, CA). The resin was washed with 10 column volume (CV) of adsorption buffer (Na2HPO4 50mM, NaCl 300 mM, Imidazole 10 mM, and pH 8). The binding proteins were eluted using imidazole gradient (100 to 300mM in 10 CV) in adsorption buffer. Fractions were collected in 1.0 mL tubes with a flow rate of 1–2 mL/minute during all chromatographic steps. The product was dialyzed in PBS pH 7.2, measured for total protein concentration according to the Bradford25 method, and analyzed by SDS-PAGE 7.5% as Lammli described.26

Selection of Phage Displayed Antibody

In this research we used the Camelus dromedarius nanobody gene library, which was displayed on phage particles and produced previously in our laboratory.23 The 109 cells of TG1 were grown to mid-logarithmic phase. Serial dilution of M13 KO7 helper phage: 10−3, 10−6, 10−9, 10−12, 10−14 in 1 mL of LB medium was prepared separately. It was mixed and added with 10 μL of each serial dilution of helper phage to 100 μl of TG1 in 1 mL tube, mixed, and incubated at 37°C for 30 minutes. Three mL LB- top agar was mixed with each of the above mentioned serial dilutions in sterile condition, transferred to kanamycin plates and incubated at 37°C overnight. The colonies were counted to evaluate the titer of helper phage. Five μL of TG1 strain of E. coli was grown at 37°C by shaking in 100 mL LB medium in a culture flask until OD550 reached to 0.4–0.5. Then 2 μL of previously prepared library (about 109 cells) was added to these cells and kept for 30 minutes at 37°C. Also, 150 μL of ampicilin was added in 2 parts after 2 hours. Following which M13 KO7 helper phage (Pharmacia) was added to the flask (each unit of OD550 contains 8 × 108 cells/mL and each cell of E. coli needs 20 helper phage), incubated at 37°C for 30 minutes, added with 100 μL of kanamycin and incubated at 37°C overnight. The next day, cells were harvested by centrifugation at 3,500 rpm for 15 minutes at 4°C. Supernatant was collected in sterile centrifuge tubes and was centrifuged at 14,000 rpm for 15 minutes at 4°C. The white revealed pallet on the wall of the tube belonging to virions (about 1012 virions/mL) and ready for panning.

Microtiter wells (Nunc, MaxiSorp) were coated overnight at 4°C with 100 μL of 5μg/mL ECD-HER2 in PBS for the first round of panning. For subsequent pannings 4, 3, 2, and 1 μg/mL of ECD-HER2 in PBS were applied respectively. The wells were blocked with 3% skim milk at 37°C for 1 hour and the virions passed over the BSA coated wells (5, 4, 3, 2, 1 μg/mL). The unbound phages were loaded onto the ECD-HER2 coated wells to improve the specific phage binding particles. After interaction of virions with the immobilized antigen, elutions of binders with 100 mM triethylamine (pH 10.0) were performed according to Barbas and colleagues.27 After each round of panning, eluted phages were neutralized with tris-HCl (1M, pH 7.2) and were subsequently added to growing E. coli TG1 cells (Amersham-Pharmacia-Biotech, London, UK) and plated on LB-ampicillin. The enrichment of phage particles carrying antigen-specific nanobodies were assessed by comparing the results obtained before and after each round of panning following anti-M13/HRP ELISA.

The purified recombinant ECD-HER2 was coated onto the wells of microtiter plates (2 μg/well) and incubated at 4°C overnight (the same concentration of BSA was used as negative control). The wells were emptied, washed, and blocked with a 1% solution of BSA in PBS (10mM, pH 7.2) for 1 hour at 37°C. Then wells were washed and added with virions of each round of panning, incubated for 1 hour at 37°C, washed, added with anti- M13/HRP conjugate (Sigma-Aldrich, Oslo, Norway) and incubated for 1 hour at 37°C. Finally, the wells were washed, added with 50 μL of substrate TMB, and incubated for 8–10 minutes at room temperature. The enzyme reaction was terminated using 50 μL of 2N HCl solution and color development was measured at 450 nm. The resulting colonies from the fourth and fifth round of panning, which revealed higher binding in ELISA, were used in further investigation.

Expression of the Single-Domain Antibody Fragments

The selected positive clones were transformed into the non-suppressor strain of E. coli (Rosetta gami2). The Rosetta gami2 cells harboring the recombinant phagmids were grown at 28°C in 250 mL TB-ampicilin containing 1% glucose in culture flasks until OD550 reached 0.9. The cells were washed and resuspended in 500 mL TB-ampicilin, induced with IPTG (1 mM) and incubated overnight in 28°C. The Rosetta gami2 cells were harvested by centrifugation for 20 minutes in 4,000g at 4°C. The pellets of bacteria were resuspended in 300 μL of 50 mM Tris-HCl buffer (pH 8.0) which consisted of 100 mM NaCl, 1 mM EDTA (adsorption buffer), and 1 mM PMSF. The cytoplasmic protein was extracted by sonication according to Harrison and colleagues28 and was clarified by centrifugation for 30 minutes in 14,000g at 4°C. To confirm the quality of recombinant protein expression, SDS-PAGE of cell lysate was performed on a 12.5% gel as Lammli described.26 The protein known to be VHH was named as SR-87.

Reactivity of Single-Domain Recombinant Monoclonal Antibodies Against ECD-HER2

The purified recombinant ECD-HER2 was coated onto the wells of microtiter plates (2 μg/well) and incubated at 4°C overnight. The same concentration of BSA was used as a negative control. The wells were emptied, washed, and blocked with a 1% solution of BSA in PBS (10mM, pH 7.2) for 1 hour at 37°C, and wells were washed and added with dilutions of supernatant of the cytoplasmic fractions’ of nanobody fragments of the selected clones (C3, C15, and C30). In each assay, 0.1% BSA in PBS (10 mM, pH 7.2) was used as an index of NSB. The wells were incubated for 1 hour at 37°C, washed, added with anti-Hemagglutinin tag (HA tag), conjugated to HRP (Sigma-Aldrich), and incubated for 1 hour at 37°C. Finally, wells were washed, added with 50 μL of substrate TMB, and incubated 8–10 minutes at room temperature. The enzyme reaction was terminated using 50 μL of 2N HCl solution and color development was measured at 450 nm.

Purification and Reactivity of Soluble Nanobody

The protein was purified by loading the centrifuged supernatant obtained from the previous section onto the Nicle-nitrilo-triacetic acid resin as mentioned in the section on purification of ECD-HER2. The product was dialyzed and measured for total protein concentration according to the Bradford method25 and analyzed by SDS-PAGE. The purified ECD-HER2 (0–1,000 ng/well) was coated onto the wells of microtiter plates at 4°C overnight. The same concentration of BSA was used as NSB. The wells were washed and blocked with 1% solution of BSA in PBS for 1 hour at 37°C. Then the wells were washed and added with diluted purified recombinant nanobody tags. In this assay, adsorption buffer was used as a negative control. The microtiter plate was incubated at 37°C for 1 hour, washed, and added with anti-HA conjugated to HRP, and incubated for 1 hour at 37°C. The rest of the experiment was performed as explained in the section on reactivity of single-domain recombinant monoclonal antibodies against ECD-HER2.

Determination of Affinity, Specificity, and Sensitivity

Affinity was determined by competitive assay. The bound VHH was detected with a monoclonal antibody specific for the HA-tag (Roche, Basel, Switzerland). We performed the experiments in 3 sets of nanobody concentrations (0, 1.5, 3, 5 μg/mL), antigen concentrations (1, 2 μg/mL), and BSA concentrations (1, 2 μg/mL). Each set of experiments were performed in duplicate. The assay was performed as we had mentioned in our previous report.29

In other sets of experiments, the VHH specificity was determined following ELISA method.30 In this experiment, we applied 1 μg/mL of non-related antigens such as BSA, endoglin, MUC1, and irrelevant peptide (TSAPDTRPAPG-STAPPAHGVTSAPDTR-BSA coupled) along with 1.5 μg/mL of SR-87 or in parallel mouse monoclonal antibody (Alexsis Biochemicals, Lausen, Switzerland) in order to determine the reaction with a standard antibody.

In a competitive assay, in order to characterize the antibody sensitivity of antigen detection limit, SR-87 was incubated with ECD-HER2 protein at increasing concentrations (0–2,000 ng/ml) overnight at 4°C. The assay was performed as explained in a previous section.


Purification of ECD-HER2

Purification of ECD-HER2 was performed by immobilized affinity chromatography (Ni-NTA resin). The collected fractions were measured at 280 nm; a yield in the range of 0.3–0.5 mg of purified protein per liter of yeast culture was calculated. The product was detected by SDS-PAGE (Figure 1).

Phage Selection

The nanobody repertoires were expressed on phage after infection with helper phages. The library was panned using ECD- HER2 and a total of 2 × 1012 phages were used in each round of panning. Finally 6 × 105, 5 × 106, 4 × 107, 4 × 108, and 3 × 109 phages were eluted after the first, second, third, fourth, and fifth rounds of panning. Performing 5 rounds of selection, 120 colonies were chosen for evaluation of their VHH proteins. All of the clones were selected from C. dromedarius nanobody gene library (data not shown). Three colonies were chosen using phage-ELISA (C3, C15, C30); they showed high specificity toward ECD-HER2. The virions from these colonies did not cross-react with BSA. The result of serially diluted virions of the best clone (C15/SR-87) in phage-ELISA confirmed our selection. The DNA phagemid from the positive clone (C15) was extracted, digested, and sequenced (data not shown) (Figure 2). Also, polymerase chain reaction (PCR) was performed for the selected clone C15 using a forward primer: 5′-ATC GGC CCA GGC GGC CAG GTC CAG CTG CTG GAG-3′ and a reverse primer: 5′- ATC GGC CGG CCT GGC CTG AGG AGA CGG AGA CC TG-3′. The resulting PCR fragment produced a band between 350–450 bp (Figure 3).

Figure 1

SDS-PAGE of extracellular domain of human epidermal growth factor receptor-2. M: Protein MW marker (sizes in kDa are indicated), Lane 1: Concentrated and filtered supernatant of yeast product. Lane 2: The yeast product after IMAC. The broad band in lane 2 is due to a various degree of glycosylation of ECD-HER2.

Figure 2

Agarose gel electrophoresis of Phagemid digestion M: DNA MW marker, Lane 1:C15 Phagemid Digestion. The band in 350-450 b.p. is due to VHH gene.

Figure 3

The result of PCR for colony No. 15. Band indicated in lane 1 is at 350 bp. M shows the ladder.

Production and Purification of Soluble Camel’s VHH Fragments

The DNA of the phagmids was transformed into Rosetta gami2. These cells are unable to suppress amber stop codon between the cloned VHH gene and the gene III in pComb3x phagemid vector. In this way, the soluble VHH fragments are produced upon induction with IPTG. The expressed proteins were extracted and visualized by G250 Coomassie brilliant blue staining in SDS-PAGE (Figure 4). A protein band with apparent MW 15kDa was observed and denoted as VHH. Further purification of the nanobody was performed by immobilized affinity chromatography on Ni-NTA resin. The fractions containing the pure antibody were pooled and the absorption measurement was performed at 280 nm. The purified protein was dialyzed in PBS buffer overnight at 4°C resulting in a clear solution. The product was detected by SDS-PAGE (data not shown). The purified nanobody was detected by Western blot analysis using a specific anti-HA/HRP conjugated antibody (Figure 5).

Reactivity of Purified Single-Domain Recombinant Monoclonal Antibodies Against ECD-HER2

The immuno-reactivity of chosen clones (C3, C15, and C30) was tested using the ELISA method. Antibodies produced from 1 of these clones (SR-87) showed high immunoreactivity toward ECD-HER2 and no cross-reaction to non-specific protein used in this research (Table 1). The affinity constant measured by ELISA was calculated to be around 1010 M−1. The binding curve is shown in Figure 6A, and the curve was drawn using Beaty’s equations.31 The specificity of the soluble SR-87 was measured by an ELISA experiment. No cross-reactivity could be detected in a direct ELISA with 4 other unrelated immobilized antigens mentioned in the materials and methods section (Figure 6B). Ultimately, Figure 6C shows the rate of sensitivity of the reaction of SR-87 antibody toward ECD-HER2. In this result, it was shown that the minimum detection limit is 2 × 10−3 ng/mL of antigen covering the standard values up to 2,000 ng/mL. Using checkerboard assay (data not shown) it was obtained that the lowest nanobody concentration to give 60%–80% maximum binding was around 2 × 10−3 ng/mL (Figure 6C).


In vitro production of antibodies has been an attractive field for many years. Many efforts failed to make plasmocytes producing the immunoglobin because of their short lifespan. In 1975, Kohler and Milstein32 introduced the hybridoma technique. They produced a fusion hybridoma from a plasmocytes and a cancerous cell line. The resulting cells were able to secrete antibodies, which on colonizing produced the so-called monoclonal antibodies (mAbs) in the laboratory condition. The hybridomas exhibited good proliferation, stability in immunoglobulin production, and a suitable lifespan due to the immortal nature of parent cells.

Use of mAbs with specificity toward tumor associated markers was considered a new approach in cancer diagnosis and therapy. These mAbs were therefore used in many detection methods and almost revolutionized clinical detection. On the other hand, it was found that many of these antibodies have an antitumor activity as a unique agent or they can also be used to deliver conjugated cytotoxic agent such as chemotherapeutic drugs, toxins, and radioactive materials. Some mAbs have been advanced into clinical trials and several mAbs have been approved by the U.S. Food and Drug Administration. However, soon it was found that because of the murine origin of mAbs, serious immunological complications may occur if the new reagent is used in repeated treatments of human beings. In the beginning of a 90 second phage display, technology was introduced by Smith33 and colleagues and McCafferty and Griffiths.34 Following this technology, it was possible to isolate the immunoglobulin genotype by phenotype assessment. Simultaneously, many investigations were carried out in order to decrease the size of conventional immunoglobulin while conserving functional characteristics of antibody using the same phage display technology. These attempts resulted in producing some antibody constructions such as single chain Fv (ScFv), Fvs, and single-domain antibodies (VH).

Figure 4

SDS –PAGE of some selected clones (C3, C15, and C30). M: Protein MW marker (sizes in kDa are indicated). All of the selected clones have broad bands in 14 KDa regions.

Figure 5

Western blot analysis of the purified soluble antibody produced by clone C15. The band in lane 1 at MW near to 14 KD indicates the VHH protein. M shows the protein MW marker.

View this table:
Table 1

Reactivity of Nanobodies With Ecd-Her2 OD450

Cell lysate of SR-870.94 ± 0.100.35 ± 0.10
Cell lysate of Rosetta gami20.38 ± 0.090.32 ± 0.10
Purified nanobody (SR-87)1.42 ± 0.110.37 ± 0.10
PBS0.23 ± 0.080.21 ± 0.10
  • All assays were performed in triplicate.

Figure 6

(A) Binding affinity for different concentrations of SR-87(0, 1.5, 3, 5 μg/mL). Ag and Ag’ are 2 different concentrations of antigen (1 and 2 μg/ml). (B) Specifity of SR-87 and anti-HER2 murine monoclonal antibody against different antigens (BSA, Peptide, MUC1, Endoglin, and ECD-HER2). (C) Sensitivity rate of SR-87. A competitive assay in which antigen is coated and various concentrations of soluble antigens displaces added antibody. Each point is the result of duplicate experiments.

However, it was found that by the end of 20th century, producing functional ScFv would be much more difficult than it was thought to be before and may not be a good strategy in size reduction. Some drawbacks reported in the literature were that the purification of ScFv caused aggregation, misfolding, proteolysis, and instability of these small antibodies.35 In 1993, scientists at Vrije University of Brussels reported natural antibodies devoid light chains in camels.1 It was then demonstrated the hyper variable region of this heavy chain antibodies (VHH) exhibits the whole antibody’s properties. The recombinant VHH is a minimal sized nanobody (around 13.5 KD) derived from matured camel heavy chain antibody. Moreover, VHH antibodies are found to be very similar to human immunoglobulin. The human VH and VHH amino acid sequences share a high degree of identity and are most similar (80%) to the human VH of family III. These nanobodies provide the advantage of interacting with novel epitopes that are inaccessible to intact antibodies. Since murine and engineered mAbs have been discovered, their applications have grown in biotechnology and in the year 2010, the engineered mAbs are predicted to account for more than 30% of all revenues in the biotechnology market.36 This fact alone indicates the great potential of antibodies in the future of clinical treatment. In this report, we prepared a novel VHH antibody against HER-2 receptor in order to investigate the possibility of using this product in immunotherapy of breast cancer. Approximately 20%–25% of breast cancers are HER-2 positive and have fast metastasis with poor prognosis, different response to treatment, and rapid development. In breast cancer therapy, 1 important strategy is to target the HER-2 by mAbs such as Herceptin (a 95% humanized anti-HER-2 antibody targeting HER-2 oncoprotein and decreases the potential for immunogenicity). Herceptin has an excellent affinity and specificity toward HER-2(Kd =0.1 nM). Derbin and colleagues37 had described a monoclonal antibody reactive toward an important cell surface marker, P185HER2. In vivo experiments with mAb were able to significantly inhibit the tumorgenic growth of neu-transformed NIH 3T3 cells implanted into nude mice. In another investigation, Shalaby and colleagues38 developed a bi-specific F (ab')2 antibody molecule containing a humanized arm with a specificity toward P185HER2 linked to another arm derived from a murine anti-CD3 monoclonal antibody, which was cloned from UCHT1 hybridoma. In 1992, Carter and colleagues39 found that the murine mAb 4D5 directed against HER2 specially inhibits proliferation of human tumor cell expressing P185HER2 but the efficacy of this mAb was found to be limited by a human anti-mouse antibody response (HAMA) and lack of effector functions. Later, this antibody was humanized rh4D5 and was named as Herceptin/Trastuzumab, and today is distributed by Genetech.

In recent years, we produced some promising single-domain antibodies for some tumor markers in our laboratory. In 2004, we developed a new nanobody which could recognize the tandem repeat region of MUC1.21 That was the first example of the isolation of camel anti-peptide VHH domains and the first nanobody against a tumor marker. EGFRvIII is the type III deletion mutant form of the epidermal growth factor receptor (EGFR) with transforming activity. This tumor-specific antigen is ligand independent, contains a constitutively active tyrosine kinase domain, and has been shown to be present in a number of human malignancies. We reported the production and characterization of camel antibodies that were directed against the external domain of the EGFRvIII,22 which may be used in tumor imaging and cancer therapy. In 2008, the anti-CD105 nanobody (AR-86a) was isolated in our lab from the same camel immune library by selections on purified antigens and target cells (so-called cell selection). The selected nanobody potently inhibited proliferation of human endothelial cells and formed capillary-like structures.23

In the present work, we studied the possibility of identification of single domain antibody fragment (VHH) against recombinant extracellular domain of epidermal growth factor receptor-2 (ECD-HER2). The chosen antigen is routinely measured as an over-expressed tumor marker in breast cancer.40 Working with living cells, which present native antigen for the biopanning process, requires many complicated and sterile conditions. However, there is no need to apply them when the purified antigen is available in large quantities. In this work, we produced a large amount of recombinant ECD-HER2 by microbial expression systems. The most favored yeast expression host for soluble expression of ECD-HER2 is Pichia pastoris methylotrofic strains. Yeast produced ECD-HER2 with an engineered N-glycosilation machinery able to produce proteins with specific native glycoform. The purified ECD-HER2 was applied for specific biopanning and isolation processes of a nanobody. Then, purified nanobody successfully reacted with native antigen presented on the surface of the SK-Br-3, which is HER-2 positive cell lines (data not shown). The original library used in our experiment consisted of the variable region of heavy chains from immunized Camelus dromedarius blood lymphocytes, against ECD-HER2 produced in our laboratory. After 5 rounds of panning, we isolated the clone in which the amino acid sequence was found to be raised from camel VHHs (gene bank accession no GQ281127).

Suitable host cells for expression of recombinant proteins include prokaryotes, yeast, or higher eukaryotic cells under the control of appropriate promoters. Most functional antibodies can only be efficiently produced using mammalian cells, especially when their appropriate glycosilation is required for therapeutic applications. Also, antibody fragments (eg, ScFv and Fab), which do not have Fc region with its N-linked oligosaccharide, are preferably produced in a microbial system.

However, it was found that many of such antibodies exhibit solubility and activity problems. Aggregation of bulk antibody fragments, which were raised from linker fragments, caused antibody insolubility and decreased activity. Meanwhile, VHH could be produced in all microbial systems such as E. coli, yeast, and fungi without solubility and glycosilation problems that normally decrease nanobody activity. Lacking of the VL domain, containing hydrophobic residues, and having single domain nature give VHHs the ability to be a soluble product without any mispairing or misfolding, resulting in aggregation. Also, efficient folding due to increased hydrophilicity give VHHs high physicochemical stability, high solubility, and well expression during the production processes. In this report, we produced SR-87 nanobody in E. coli in high yield and soluble form.

Spirodon and colleagues41 characterized 3 murine mAbs (HER50, HER66, and HER70) and compared them with Herceptin. All of them had shown the ability to bind with various affinities toward different epitopes on the surface of ECD-HER2. The relative-binding affinities of these mAbs and Herceptin were determined, and it was found that HER-50 showed the highest relative-binding affinities (respectively 0.4 and 0.8 nM) on BT474 cell line over expressing HER2 tumor marker. HER-70 had the lowest relative-binding affinity, about 13.5nM, and HER-66 was intermediary (3.8 nM). All mAbs were reactive and inhibited the growth of the cells expressing medium HER2, such as T47D, and expressed low HER2 such as MDA-MB-231.

The affinity of produced nanobody (SR-87) was found to be about 1010 M−1 for 1 of the clones (C15). Our experiments indicated high specificity toward HER 2 protein, since there was minimal cross reaction with related proteins such as BSA, MUC1 protein, MUC1, related peptide, and endoglin. The high affinity and specificity of SR-87 for antigens comes from the long CDR3 loop, providing a sufficiently large antigen binding site, which is able to react with purified recombinant ECD-HER2. Thus, this single domain antibody may be useful for targeting a HER-2 marker on the surface of tumor cells. In comparison to Herceptin, our VHH was a tenth of the size and is the smallest antibody fragment retaining specificity reported thus far. The VHH size prepares good capacity to faster penetration in tissues and tumors with a longer lifespan. Also, they can break through the blood brain barrier, an extraordinary action Herceptin is not capable of, which may be a good basis for novel central nervous system drugs when HER-2 expressing cells in the brain are to be targeted. Furthermore, appropriate resistance to pH levels and temperatures as high as 90°C,42 which has never been observed in any murine mAbs, makes VHH an excellent candidate for oral drug delivery in elementary channel diseases.43


The authors thank Dr. Mamalaki (Helenic Institute Pasteur, Athens) for her donation of extracellular domain of human epidermal growth factor receptor 2, which was subcloned in pPICZαC (X33). This work was supported by grants from the Iran Nanotechnology Initiative Council.


View Abstract