Camelid and shark single domain antibodies: promising therapeutic tools
The discovery of light-chains lacking functional antibodies produced by camelids (llamas, camels, dromedaries, and alpacas) and cartilaginous fishes (sharks) formed a further breakthrough in therapeutic antibodies (Saerens et al., 2008; Khodabakhsh et al., 2018). The antibodies derived from the single variable domain of these antibodies are called single domain antibodies, abbreviated as sdADs or nanobodies. They are of 2 types: VHH and IgNAR.
VHH single domain antibody
The conventional structure of antibodies is tetramers of two heavy chains and two light chains (Fig 1a). The variable domains of each light and heavy chain together form the antigen-binding site, and there are two antigen-binding sites in an antibody molecule. In 1993, Hamers-Casterman et al. discovered that in addition to the conventional tetrameric antibodies, camelids (llamas, camels, dromedaries, and alpacas) produce functional homodimeric heavy-chain only antibodies, devoid of light chains. Apart from the light chains, these heavy chain antibodies also lack CH1 domains (Fig 1b). Thus, the heavy chain antibodies are composed of two identical heavy chains, each comprising two constant domains (CH2 and CH3), a hinge region, and a variable domain responsible for antigen recognition.
So, in contrast to the paired variable domains of light (VL) and heavy (VH) chains that form an antigen-binding site, the single variable domain of camelid antibodies can bind to the antigen without domain pairing (Hamers-Casterman et al., 1993; Muyldermans, 2013). The variable domains of camelid antibodies are often referred to as VHH domains, meaning VH domain obtained from heavy-chain antibodies (Fig 1c). The single VHH domain size is 15 kDa in contrast to the 150 kDa size of a conventional antibody. Because of the smaller size, single domain antibody fragments can recognize targets and active sites that are not accessible by traditional antibodies. Therefore, the VHH domain can be developed as a single domain antibody with appropriate modification (Hoey et al., 2019).
The antibody fragments Fab or single-chain Fv (scFv) contains VH and VL domains joined either by a disulfide bond (Fab) or by a polypeptide linker (scFv). VH and VL domains together form the antigen-binding site (Fig 2). But the single domain antibody fragment contains the single variable domain VHH of the heavy chain antibodies. The single VHH recognizes and binds to the antigen, thus representing the functional equivalent of a Fab fragment of conventional antibodies.
Properties of VHH domains
Like the V domain of conventional antibodies, the VHH domain of a camelid antibody contains three complementarity determining regions (CDRs) and framework regions (FRs). The CDRs mediate interaction with the antigen. On the other hand, FRs support the binding of CDRs to the antigen.
Antigen binding by VHH domains is mediated by only three CDRs rather than the six CDRs of conventional antibodies (Fig 3).
To make up for the reduced CDRs, the CDRs of VHH domains have certain characteristic features: Firstly, the N-terminal part of CDR1 is more variable. Secondly, VHHs have a long and extended CDR3 that is often stabilized by an additional disulfide bond with a cysteine in CDR1, resulting in the folding of the CDR3 loop to form finger-like projections (Fig 4b). The long and extended CDR3 is responsible for the binding of the VHH to the concave epitopes, such as protein clefts/pockets, including active sites of the enzymes (Iezzi et al., 2018; Soetens et al., 2020).
The framework region (FR) of VHH also exhibits certain differences from the VH domain of conventional antibodies. In the case of conventional antibodies, the amino acid residues in the framework region 2 (FR2) of the VH domain contain hydrophobic residues that are responsible for interaction with the light chains (Fig 4a). But in the case of the VHH domain of heavy-chain antibodies, hydrophobic residues in the FR2 are replaced by more hydrophilic residues (Fig 4b). Thus, it increases the solubility of heavy-chain antibodies.
IgNAR single domain antibody
Two years after the discovery of camelid heavy-chain antibodies, it was reported that sharks and other cartilaginous fish also produce homodimeric heavy chain antibodies called Ig new antigen receptors or IgNARs. IgNARs are composed of two identical heavy chains, each comprising one variable VNAR domain and five constant domains (Fig 5a). Like camelid antibodies, IgNAR targets antigen through a single variable domain VNAR, of size 13 kDa (Fig 5b). The VNAR domain contains 2 CDRs: CDR1 and CDR3, in contrast to 3 CDRs in the VHH domain (Cheong et al., 2020). The CDR3 loop of VNAR is long and extended, which allows it to reach and bind the buried epitopes. The ability to recognize and bind hidden functional sites of a target antigen makes novel single domain antibodies attractive as novel therapeutics for human diseases.
Advantages over conventional antibodies
High stability: Contrary to conventional antibodies, which lose their antigen-binding ability upon heat denaturation, sdABs have been shown to remain functional up to 60–80°C (Dumoulin et al., 2002). Also, the sdABs are resistant to high pH. Therefore, they are also suitable for oral immunotherapy.
Recognize inaccessible antigenic targets: Another advantage of sdABs over conventional antibodies is that the VHHs and VNARs are adept at recognizing the antigenic sites that are otherwise inaccessible to conventional antibodies, such as enzyme active sites and conserved cryptic epitopes. The ability to recognize these inaccessible antigenic sites has been attributed to the VHHs or VNARs smaller size and the ability of the extended CDR3 loop to penetrate such sites.
Easy production in the microbial system: The antibody fragments produced from the conventional antibodies are monovalent antibody fragments (Fab) or single-chain Fv (scFv), where a flexible peptide linker joins the VH and VL domains. The main advantage single domain antibodies hold over Fab or scFv is that VHHs and VNARs contain only one domain, making them more convenient for cloning and genetic engineering. Also, the production of Fabs or scFvs in microbial cells is often cumbersome because of the requirement for correct VH and VL domains association. But the sdAbs or nanobodies contain only the VHH domain or VNAR; thus, they do not need domain pairing; therefore, they are easily produced and expressed in microbial systems.
Multivalent antibody fragments: In some cases, like virus neutralization, it is advantageous to engineer monovalent antibody fragments into multivalent formats because they are more effective in virus neutralization. Multivalent formats using conventional recombinant antibodies like scFvs are generated using linkers between the VH and VL domains of a specific length. But the problem of mispairing of VH and VL domains arises, which reduces their affinity for antigen binding (Fig 6). On the other hand, there is no problem with domain mispairing in the case of VHH or VNARs domains; therefore, they are more suitable for producing multivalent formats. Furthermore, VHH or VNARs domains with the same antigen specificity can be fused with flexible linkers to recognize the same repeating antigen to increase the functional affinity of the VHH or VNAR single domain antibodies. For instance, it was observed that a bivalent VHH sdAB targeting H5N1 hemagglutinin was at least 60-fold more effective than the monovalent one in controlling influenza virus replication.
Rapid tissue penetration: In addition, the small size of sdAbs results in their fast tissue penetration. It is advantageous for targeting VHHs coupled to toxins or drugs to tumors and in vivo diagnosis using imaging.
Cross blood-brain barrier: Because of the small size, sdAbs can cross the blood-brain barrier (BBB), thus making them excellent candidates for treating neurological disorders and neurotropic virus infections like rabies virus.
Clearance from the body: Because of their small size of about 15 kDa, sdAbs can rapidly pass the renal filter, which has a cutoff of approximately 60 kDa, resulting in their rapid blood clearance. It is advantageous when using the single domain antibody coupled to a toxic substance.
Limitations
sdAbs have a short serum half-life of about 2 hours, which limits their applications in many therapeutic applications. It can be extended by the chemical addition of polyethylene glycol (PEG), by Fc-fusion, or by binding to long-lived serum components, for example, albumin.
Production of single domain antibodies
To develop a VHH single domain antibody, firstly, it is required to immunize alpaca or llama with an antigen against which single domain antibodies are needed to be generated (Fig 7). Over the course of a few weeks, the animal generates antibodies. Then, the antisera are collected, and peripheral blood mononuclear cells (PBMCs) are isolated. From PBMCs, total RNA is isolated, and cDNA is prepared by the process of reverse transcription. The VHH genes are then amplified from cDNA by the PCR process using the primers specific to the variable regions of the heavy chain antibodies. The PCR products are then cloned into phagemid vectors which are electroporated into the E. coli cells. The E. coli cells are then infected with bacteriophages. After infection, the bacteriophages generate a phage-display library in which the VHH regions are displayed on the phage coat protein.
After this, affinity screening of the phage-display antibody library is done by a process called biopanning. In this, the phages expressing the VHH region of antibodies on the surface are added to the antigens immobilized on a solid surface, for example, on ELISA plates (Fig 8). The antigen can be a protein on cancer cells or other cells against which single domain antibodies are required to be generated. The phages that have VHH regions specific for the antigens bind to the coated antigen. Non-specific phages are removed by stringent washing. Antigen-bound phages are then eluted and are re-infected into E. coli to produce a subset of phages for the next cycle of panning. After the multiple rounds of affinity panning (phage display method), we can retrieve those phage particles that express single domain antibodies of interest, specific for the desired antigen. Finally, the positive clones are then expressed in bacterial cells to produce a high yield of llama monoclonal single domain antibodies.
Similarly, the shark monoclonal single domain antibodies VNAR can be generated.
The conventional functional antibodies can only be efficiently produced using mammalian cells, especially when appropriate glycosylation and other post-translational modifications are required for therapeutic applications. However, antibody fragments like VHH and VNAR lack the Fc region with N-linked oligosaccharides. Thus they can be efficiently produced in microbial systems like E. coli, yeasts, or filamentous fungi.
Immunogenicity and humanization of sdAbs
Immunogenicity of sdAbs: Certain features of sdAbs, such as small size which decreases the number of potentially immunogenic epitopes, high solubility which reduces the formation of immunogenic aggregates, and rapid clearance of the sdAb favors low immunogenicity in humans. But still, the nonhuman origin of VHHs and VNARs may increase the chance of triggering unwanted immune responses when administered as therapeutic agents. For example, we have already discussed that the transition from fully murine to chimeric, humanized, or fully antibody therapy significantly reduced the frequency of anti-drug antibody (ADA) development and thus immunogenicity. Similar is the case with single domain antibodies. Since the interface between the VH and VL domains of conventional antibodies is exposed in sdAbs, which have just one domain and no pairing partner; therefore, ADAs may be generated against them. In addition, while the CDR1 and CDR2 loops of VHHs adopt similar canonical structures as human antibodies, their long and extended CDR3 loops may adopt conformations absent in the human repertoire, thus raising the possibility of generating ADAs against the CDR3 loop. Therefore, it becomes imperative to humanize sdAds as much as possible.
Humanization of sdAbs: Humanization is defined as replacing xenogeneic sequences with human sequences in the antibody variable domain.
The first approach of humanization of sdAbs is CDR grafting (Fig 9a). It is a process by which CDRs from xenogeneic antibodies (e.g., CDRs from mouse Abs) are grafted into the human framework regions (FRs) of the VH domains of human antibodies. Also, the mutations in the FR residues of the VH domain must be done to restore the properties of non-human sdAbs (Ju et al., 2017; Rossotti et al., 2021). A conventional antibody consists of two domains (VH and VL), which tend to dimerize or aggregate because of their lipophilicity. Therefore, hydrophobic amino acid residues of the FR region are replaced by hydrophilic amino acids to mimic the VHH domain and inhibit its interaction with the VL domain (Fig 9b). Thus, the single domain VH antibodies generated are similar to the sdAbs or nanobodies.
The second approach of humanization of sdAbs is resurfacing of xenogenic sdAb (camelid sdAbs) sequence to mimic VH domains of human antibodies (Fig 10). For this, amino acids in framework regions (FRs) of sdAbs are modified at fixed positions to mimic the VH of human Abs. Also, the mutations should not affect the properties of sdAbs. For this, two of the FR2 hydrophilic amino acids at positions 49 and 50 responsible for the solubility of VHHs are substituted without compromising the solubility, monomeric behavior, and function of the nonhuman sdAbs. The remaining two FR2 amino acids at positions 42 and 52 cannot be humanized as they are crucial in maintaining the structural integrity of the VHH domain and are essential for the proper conformation of the CDR3 loop.
Therapeutic applications of sdAbs
VHHs have a wide range of therapeutic applications. For instance,
Therapeutic potential for cancer treatment: VHHs may outperform mAbs in treating solid tumors as their small size allows tumor penetration. The epidermal growth factor receptor (EGFR) is overexpressed in many cancer cell types, leading to increased cell proliferation, migration, and angiogenesis. VHHs binding to EGFR can block epidermal growth factor (EGF) binding to its receptor, inhibiting the EGFR signaling pathway that ultimately induces apoptosis in tumor cells by blocking tumor cell proliferation and survival invasion. Thus, VHHs can be used to treat solid tumors.
Even the sdAbs can be conjugated to the tumor penetrating peptides (TPPs) to enhance their antitumor activity by penetrating the tumor cells.
Additionally, sdAbs can be used to generate immunotoxins. For instance, anti-CEA VHH sdAb directed against tumor antigen CEA is fused with the β-lactamase enzyme. When sdAb binds to the CEA antigen on the tumor cell, the fused enzyme converts an injected non-toxic prodrug into a toxic drug in the vicinity of the targeted tumor cells, leading to their killing.
Therapeutics against human viral diseases: Currently, sdAbs are being developed against several viruses, including human immunodeficiency virus-1 (HIV-1), respiratory syncytial virus (RSV), influenza viruses, hepatitis C virus (HCV), etc.
Therapeutics against arthritis and sepsis: VHHs binding to TNF-α can be used to treat rheumatoid arthritis. Studies have shown that the potency of bivalent formats was 500 times more as compared to monovalent VHHs and even exceeded the potency of clinically used conventional antibodies both in vitro and in a murine arthritis model.
Similarly, lipopolysaccharide (LPS)-binding VHHs work by blocking LPS binding and signaling to host cells to treat sepsis.
sdAbs as intrabodies: sdAbs can also be used as “intrabodies.” The term intrabodies is derived from intracellular antibodies and refers to antibodies that are expressed intracellularly. In contrast to naturally expressed antibodies or administered antibodies, which target extracellular proteins, intrabodies target proteins inside the cell. Originally, intrabodies mainly existed as single-chain variable fragments (scFvs) that could be used to treat diseases. The inherent stability of single domain antibodies makes them much more suitable for intracellular expression compared to the scFvs. But the reducing environment of the cytoplasm, in which the intra-domain cysteine bridges fail to form, often inhibits their proper folding. For designing intrabodies, the cysteine residues VHHs are replaced by serine residues, without affecting their ability to bind their cytosolic targets. Additionally, the sdABs are easier to engineer and allow targeting of epitopes that would not be available to scFvs. The propensity of VHHs to bind cleft-like structures makes them ideal for binding to the active sites of enzymes inside the cells. Intracellular expression of these VHHs opens up a whole new range of potential therapeutic targets, which were previously inaccessible using traditional monoclonal antibody (mAb) therapy. For instance, intrabodies have the potential to treat diseases caused by protein misfolding.
Misfolding and aggregation of intracellular proteins like α-synuclein in the brain is a hallmark of many neurodegenerative diseases like Parkinson’s disease. In addition, aggregation of proteins leads to neuron toxicity. The two VHH intrabodies NbSyn2 and NbSyn87 can bind distinct epitopes in the C-terminus of α-synuclein. VHH binding leads to a destabilization of α-synuclein fibrils and reduced cellular toxicity (Soetens et al., 2020). To increase the efficacy of the α-synuclein-specific VHHs, a proteasomal targeting PEST motif can be added to the VHH C-terminus, which allows the VHHs to target monomeric as well as oligomeric α-synuclein for proteasomal degradation and effectively preventing fibril formation; thus leading to the treatment of neurodegenerative disease.
FDA approved single domain antibody
Currently, more than 30 sdAbs are being in preclinical to clinical development. In February 2019, the FDA approved the first VHH-based therapy, Cablivi, to treat acquired thrombotic thrombocytopenic purpura, abbreviated as aTTP (Duggan, 2018). It was developed by Belgian biotech company Ablynx (now acquired by Sanofi).
aTTP is a rare life-threatening autoimmune blood-clotting disorder. During the normal blood clotting process, the plasma protein von Willibrand factor (vWF) is essential in recruiting blood-clotting cells called platelets to the damaged vessels and stopping blood loss at the sites of blood vessel injury. In healthy individuals, the von Willibrand factor is immediately cut up into smaller pieces by an enzyme called ADAMTS13 to prevent the spontaneous recruitment of platelets when blood vessels are not damaged. However, in aTTP patients, anti-ADAMTS13 autoantibodies are present that block ADAMTS13 from working correctly (Fig 11). It results in an accumulation of ultra-large von Willibrand factor molecules in the blood, which bind to the platelets and cause the excessive formation of blood clots. The formation of blood clots in small blood vessels throughout the body leads to a severe reduction in platelets that circulate in the blood (thrombocytopenia), breakdown of red blood cells (hemolytic anemia), and an insufficient blood supply to tissue (tissue ischemia). In addition, it results in organ dysfunction, especially brain, heart, and kidneys. There is currently no authorized treatment for aTTP. Patients often receive immunosuppressants and plasma exchange but remain at risk of significant morbidity and mortality.
Cablivi (Caplacizumab) is a bivalent single domain humanized anti-vWF antibody in which two anti-vWF sdAbs are linked by a tri-alanine (AAA) linker (Fig 12a).
The drug binds to the von Willibrand factor and blocks its interactions with the GP1b receptors on platelets (Fig 12b). Thus, it results in the reduction of platelet aggregation, and ultimately formation and accumulation of blood clots in the blood vessels are reduced (Elliott and Chan, 2019). The drug is used in combination with plasma exchange and immunosuppressive therapy.
References
Cheong, W. S., Leow, C. Y., Majeed, A. B. A., & Leow, C. H. (2020). Diagnostic and therapeutic potential of shark variable new antigen receptor (VNAR) single domain antibody. International journal of biological macromolecules, 147, 369–375.
Duggan, S. (2018). Caplacizumab: first global approval. Drugs, 78(15), 1639–1642.
Dumoulin, M., Conrath, K., Van Meirhaeghe, A., Meersman, F., Heremans, K., Frenken, L. G., … & Matagne, A. (2002). Single‐domain antibody fragments with high conformational stability. Protein Science, 11(3), 500–515.
Elliott, W., & Chan, J. (2019). Caplacizumab-yhdp for Injection (Cablivi). Internal Medicine Alert, 41(5).
Hamers-Casterman, C. T. S. G., Atarhouch, T., Muyldermans, S. A., Robinson, G., Hammers, C., Songa, E. B., … & Hammers, R. (1993). Naturally occurring antibodies devoid of light chains. Nature, 363(6428), 446–448.
Hoey, R. J., Eom, H., & Horn, J. R. (2019). Structure and development of single domain antibodies as modules for therapeutics and diagnostics. Experimental Biology and Medicine, 244(17), 1568–1576.
Iezzi, M. E., Policastro, L., Werbajh, S., Podhajcer, O., & Canziani, G. A. (2018). Single-domain antibodies and the promise of modular targeting in cancer imaging and treatment. Frontiers in immunology, 9, 273.
Ju, M. S., Min, S. W., Lee, S. M., Kwon, H. S., Park, J. C., Lee, J. C., & Jung, S. T. (2017). A synthetic library for rapid isolation of humanized single-domain antibodies. Biotechnology and Bioprocess Engineering, 22(3), 239–247.
Khodabakhsh, F., Behdani, M., Rami, A., & Kazemi-Lomedasht, F. (2018). Single-domain antibodies or nanobodies: a class of next-generation antibodies. International reviews of immunology, 37(6), 316–322.
Muyldermans, S. (2013). Nanobodies: natural single-domain antibodies. Annual review of biochemistry, 82, 775–797.
Rossotti, M. A., Bélanger, K., Henry, K. A., & Tanha, J. (2021). Immunogenicity and humanization of single‐domain antibodies. The FEBS Journal.
Saerens, D., Ghassabeh, G. H., & Muyldermans, S. (2008). Single-domain antibodies as building blocks for novel therapeutics. Current opinion in pharmacology, 8(5), 600–608.
Soetens, E., Ballegeer, M., & Saelens, X. (2020). An Inside Job: Applications of Intracellular Single Domain Antibodies. Biomolecules, 10(12), 1663.
