Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/oxford-university-press/ancient-species-offers-contemporary-therapeutics-an-update-on-shark-MMYfLjqf7p
- Publisher
- Oxford University Press
- Copyright
- © The Author(s) 2020. Published by Oxford University Press on behalf of Antibody Therapeutics.
- eISSN
- 2516-4236
- DOI
- 10.1093/abt/tbaa001
- Publisher site
- See Article on Publisher Site
Abstract
Antibody Therapeutics, 2020, Vol. 3, No. 1 1–9 doi:10.1093/abt/tbaa001 Advance Access Publication on 21 January 2020 Review Article Ancient species offers contemporary therapeutics: an update on shark V single domain antibody NAR sequences, phage libraries and potential clinical applications 1,2 1,2 1,2, Hejiao English , Jessica Hong and Mitchell Ho NCI Antibody Engineering Program, Center for Cancer Research, National Cancer Institute, Bethesda, MD 20892, USA and Laboratory of Molecular Biology, Center for Cancer Research, National Cancer Institute, Bethesda, MD 20892, USA Received: October 6, 2019; Revised: January 6, 2020; Accepted: January 8, 2020 ABSTRACT The antigen binding variable domain (V ) of the shark immunoglobulin new antigen receptor (IgNAR) NAR evolved approximately 500 million years ago and it is one of the smallest antibody fragments in the animal kingdom with sizes of 12–15 kDa. This review discusses the current knowledge of the shark V NAR single domain sequences and ongoing development of shark V s as research tools as well as potential NAR therapeutics, in particular highlighting the recent next-generation sequencing analysis of 1.2 million shark V sequences and construction of a large phage displayed shark V library from six naïve adult nurse NAR NAR sharks (Ginglymostoma cirratum). The large phage-displayed V single domain library covers all the four NAR known V types (Types I–IV) and many previously unknown types. Ongoing preclinical development will NAR help define the utility of shark V single domains as a potentially new family of drug candidates for treating NAR cancer and other human diseases. Statement of Significance: This review discusses recent progress on next-generation sequencing analysis and construction of a large nurse shark V single domain antibody library, the current NAR development of shark V single domain antibodies as research tools, as well as potential therapeutics NAR for treating cancer and other human diseases. KEYWORDS: nurse shark (Ginglymostoma cirratum); phage display library; V single domain; NAR next-generation sequencing; antibody engineering INTRODUCTION complexes (MHCs) that are found in mammals [1, 4, 7]. However, there are certain unique immunological Currently, there are over 400 species of sharks living in features in sharks that are not normally present in humans the marine ecosystem on our planet [1–3]. Their sizes can or other mammals, with the exception of camelids [2, 8]. vary from 7-inch dwarf lanternshark (Etmopterus perryi)to Sharks began to produce a homodimeric heavy chain- 39-foot whale shark (Rhincodon typus)[4]. Sharks initially only antibody known as immunoglobulin new antigen dominated the ocean as top predators about 500 million receptor (IgNAR), a homodimeric protein composed of years ago [1, 4–6]. During that time, their immune system, heavy chains with an antigen-binding region at the end the earliest adaptive immunity to our knowledge, has of each H chain [9] to combat foreign antigens [1, 4, 6, evolved to produce the comparable fundamental factors 10, 11](Fig. 1). The basic Ig fold of modern antibodies is including T cells, B cells and major histocompatibility present in the evolutionary ancient shark IgNAR domains To whom correspondence should be addressed. Dr. Mitchell Ho, the Center for Cancer Research, the National Cancer Institute, Bethesda, MA 20892, USA. Web: https://ccr.cancer.gov/mitchell-ho; https://ostr.ccr.cancer.gov/resources/antibody-engineering-program/; E-mail: homi@mail.nih.gov. © The Author(s) 2020. Published by Oxford University Press on behalf of Antibody Therapeutics. This work is written by US Government employees and is in the public domain in the US. 2 Antibody Therapeutics, 2020 Figure 1. Schematic depiction of human IgG, shark IgNAR and V structures. NAR even with low sequence conservation [6, 10]. The antigen of the total nurse shark V s). Type II and III V shave NAR NAR binding variable domain of these IgNAR (V ) is much at least one extra cysteine at amino acid 28 (21C, 28C, 82C; NAR smaller (12–15 kDa) than classical IgG V + V (Fv 25– 70% of the total V s). Among them, ∼ 57% of the total H L NAR 30 kDa) [7]. The strikingly high homology between V V s are classical Type II, a subtype with a total of four NAR NAR and shark TCR NAR has raised interests in the possible cysteines. Importantly, approximately 30% of V sdo NAR origin of V [6, 10]. It is speculated that V originated not belong to any of the four classical types. The structure NAR NAR from a common NAR domain that was used by both TCR and function of these previously unidentified V types NAR and IgNAR as an antigen binding domain over 450 million require further structure and functional investigation. Due years ago and into V or NAR-T RV to allow improved to their high stability, solubility and potential to penetrate NAR C specialization binding ability for both domain types [6]. In tissues and bind to hidden functional sites in target pro- addition, an alternative theory suggests that both NAR- teins, shark V may potentially possess distinct advan- NARs TCR and V may have originated (independently or not) tages over conventional antibodies as novel therapeutics NAR from cell adhesion molecules [5, 12]. Although the origin for human diseases [3, 8, 16]. We previously compared the of IgNAR remains unsolved, the fact that V also exists known structures of a nurse shark V (PDB 1T6V) [17] NAR NAR as an antigen-recognition domain in shark TCRs suggests to a conventional Fv antibody (PDB 2EIZ) [18] in their that V functionally may play a role in both B-cell and complexes with lysozyme, describing the shark V single NAR NAR T-cell immune responses [1]. domain, not the Fv antibody domain, can reach to the Based on the number and position of cysteines, shark buried substrate binding pocket of lysozyme [3]. Based on V s were categorized into four classical types in pre- this observation and other findings targeting virus particles NAR vious studies [8, 9, 13]. A recent next-generation sequenc- [19, 20] and toxins [21], we have postulated that V NAR ing (NGS) on Illumina MiSeq analyzed with custom Perl single domains may have potential to reach buried func- Scripts revealed approximately 1.2 million full-length shark tional cavities or grooves with the extended CDR3 loop. V sequences from a phage-displayed V antibody Currently, there are several promising shark V sinpre- NAR NAR NAR library constructed from six naïve adult nurse sharks (Ging- clinical development aiming to be evaluated in clinical trials lymostoma cirratum)[14]. It is the largest scale shark V as therapeutics in humans [8, 16, 22–26]. NAR sequence analysis reported thus far. The NGS analysis reveals that a number of V sequences do not belong NAR THE EVOLUTIONARY ORIGINS OF SHARK V s NAR to any of the four classical types [15]. As described in Fig. 2, the presence of two canonical cysteines located at The diversification of the vertebrate adaptive immune sys- both amino acid 21 and 82 are used as a key criterion for tem is estimated to have happened over 500 million years us to define Type I–IV V s in our NGS analysis. The ago [10]. Cartilaginous fish such as sharks, skates and rays NAR diverged from the bony fish about 450 million years ago sequences that do not contain one or both of these cysteines and started to produce immunoglobulin-based antibod- are considered as other types (∼5% of the total V s ana- NAR lyzed by NGS). The sequences that have both 21C and 82C ies—IgNARs [6, 10]. Previous studies have demonstrated are further categorized based on their placement of addi- that IgNARs are part of the adaptive immune armory and tional cysteines. Type IV V s contain only two canonical they have the highest potential for antigen-driven affinity NAR cysteines found at positions 21 and 82. We further defined maturation as compared to other Ig types in sharks such as Type I V s as those contain an extra cysteine at position IgW and IgM [6, 11]. In-depth analysis of the evolution of NAR 34 (21C, 34C, 82C; ∼ 24% of the total nurse shark V s). fish and shark immunoglobulin systems has been reviewed NAR This group can be further divided into subtypes based on by others [5, 12]. how many additional cysteines they contain. The classical V phage-displayed libraries and functional valida- NAR Type I V as a subtype has a total of six cysteines (∼ 11% tion of a panel of single domain V binders showed NAR NAR Antibody Therapeutics, 2020 3 Figure 2. Shark V Types I-IV based on the recent NGS analysis of 1.2 million adult nurse shark V sequences. The two canonical cysteines (21C NAR NAR and 82C) are in white circles, whereas the extra cysteines are in black circles. The variable regions are marked as CDR1, HV2, HV4 and CDR3. All the disulfide bridges are shown using connected black lines. Type I (21C, 34C and 82C) and Type II (21C, 28C and 82C) V s contain three canonical NAR cysteines as indicated. Subtypes in Type I and Type II V s contain 3–9 cysteines. NAR that dimerization is not required for high-affinity antigen UPDATES ON SHARK V SEQUENCES NAR binding [8, 14, 21, 27, 28]. These observations may suggest AND TYPES that ancient V s could be a functional single domain NAR The two β sheets of V domains are held together NAR antigen binding domain versus the homodimeric format by two canonical cysteine residues (21C and 82C) in of IgNAR in modern sharks [6]. This differentiates shark framework regions (FR) 1 and 3. The canonical cysteines V from camelid V H (variable domain of camelid NAR H can stabilize the immunoglobulin fold via a disulfide heavy chain antibody) evolutionarily, as camelid V H bond. In addition to these canonical cysteines that are might evolve from a conventional IgG by simultaneous loss highly conserved across animals, the complementarity- of the light chain and the CH1 domain of heavy chain [10]. determining region 3 (CDR3), CDR1 and certain FR can However, in our study and others the selection of phage- have additional cysteines that form disulfide bonds within displayed antibody fragments is an artificial selection the V domain [1, 2, 30]. Based on the numbers and positions system. It is not a biological system to either prove or of these extra cysteines in the V domain, IgNARs NAR disapprove whether a single domain can be automated and are classified into four types including classical [15]and independently folded in animals. non-classical subtypes (Fig. 2). As the sequence diversity Sequence homology analysis revealed that V might NARs of V is primarily in CDR3, the diversity of cysteine NAR have evolved from an ancestral antigen receptor or cell numbers and locations is crucial for V structural NAR adhesion molecule that is incorporated in both IgNAR and diversity [14]. shark NAR-T R as antigen binding domains [6]. Whether Classical Type I V contains two non-canonical cys- NAR modern V might possess more T RV-like properties NAR C teine residues in CDR3, which form two extra disulfide such as the ability to bind with MHC-associated peptide bonds with FR2 (34C) and FR4 to make a relatively rigid complexes with high specificity should be explored and antigen binding surface based on the crystal structure of a validated in future structural and functional studies. Cell Type I shark V in complex with lysozyme [17]. Type I NAR surface proteins and secreted proteins account for only a was initially reported only in the nurse shark (G. cirratum) small portion of tumor antigens. Intracellular proteins are [15, 17]. More recently, Type I V s were also found in NAR degraded by the proteasome to produce short peptides, the wobbegong shark (Orectolobus ornatus)[26]. Our group some of which are ultimately presented at the cell surface has isolated a Type I V binder from a phage displayed NAR as the MHC-associated peptide complexes that can be nurse shark single domain V library, which binds to NAR recognized by TCRs on T cells [29]. The similarity in a MERS spike protein [14] The soluble monomeric single amino acid sequences might enable a phage-displayed domain protein derived from the phage binder, MERS A8, V single domain library to be used as a platform to NAR was produced in Escherichia coli and the binding affinity select TCR-mimic antibodies, which can be used to bind was conserved when produced and folded in a prokaryotic intracellular cancer antigens by targeting MHC-associated system with a yield of 3 mg purified Type I V single NAR peptide complexes. domain protein from 600 mL E. coli culture, while the yield 4 Antibody Therapeutics, 2020 of a Type II V single domain protein was 9 mg from the other libraries. To construct a large antibody phage library, NAR same volume of culture [14]. we developed a library construction method based on Classical Type II V domains have two extra cysteines polymerase chain reaction (PCR)-Extension Assembly NAR located in CDR3 and CDR1 (28C), which form a disulfide and Self-Ligation (named “EASeL”) [14]. Interestingly, bond between CDR1 and CDR3. Both positions of the this library includes V s from shark B cells and shark NAR cysteine in CDR3 and the length of CDR3 have expanded T cells. According to the NGS analysis results of 1.2 the structural diversity of antigen binding surfaces for Type million full-length sequences, most of the V sequences NAR II V domains [28, 31, 32]. Both Type I and II V scan in this library are Type I and Type II [14]. Interestingly, NAR NAR have long fingerlike CDR3s that reach into and bind with over 56 000 of the nurse shark V sequences in our NAR recessive epitopes [7, 11]. While CDR3 in humans generally library (∼5% of the total nurse shark V s) cannot be NAR comprise of 8–12 amino acids, V s can have up to 34 categorized in any of the four known V types due to NAR NAR amino acids as previously reported [1]. Our recent NGS variability in cysteine numbers and locations. A significant analysis of over one million nurse shark V sequences percentage of previously undescribed types of V smerit NAR NAR demonstrates the CDR3 length of both Type I and Type further investigation [14]. Traditionally, phage displayed II V s varies greatly between 0 and 40 amino acids antibody library diversity is analyzed by isolation of dozens NAR with a median CDR3 length of about 20 amino acids or hundreds of clones for Sanger sequencing. Modern [14]. V s lack conventional CDR2 which are present in NGS approaches can be used to assess library diversity NAR another classical IgG and camelid V H. Instead, V s by sequencing over 10 clones [39]. The diversity of the new H NAR have two shorter CDR2-like regions: hypervariable region shark V library was analyzed by NGS, revealing 85% NAR 2 (HV2) and hypervariable region 4 (HV4) separated by a (over 1 million) of the 1.2 million sequences appeared only FR [28, 32]. once [14]. The unparalleled repertoire of unique locations Classical Type III V s are similar to Type II V s, and numbers of cysteines may produce a large variety of NAR NAR except there is a highly conserved tryptophan residue in loop structures that recognize a wide range of epitopes. CDR1 after the cysteine. The CDR3 is not as diverse in Previous literature showed sequence variability in Type III since Type III V domains are predominantly immunized sharks in CDR1, HV2 and HV4 [28, 32, 40]. NAR found in neonatal sharks [8]. Classical Type IV V s However, the NGS analysis of naïve V sequences NAR NAR domains only have two canonical cysteines (21C and 82C) showed minimal sequence variability in CDR1, HV2 and that hold the V together and have the highest similarity HV4. The diversity of the naïve V s appears to be NAR NAR to V domains in conventional IgG found in other species mostly in CDR3 sequences, while CDR1, HV2 and HV4 including humans. likely acquire more sequence diversity during antigen- Currently, there are several shark V libraries driven affinity maturation [14]. This finding suggests that NAR reported using various shark species: banded wobbegong increasing sequence diversity in CDR1, HV2 and HV4 shark [26], spiny dogfish (Squalus acanthias)[27], smooth through random mutagenesis could potentially mimic in dogfish (Mustelus canis)[21], bamboo shark (Chiloscyllium vivo affinity maturation process in sharks and yield high plagiosum)[33], horn shark (Heterodontus francisci)[24], affinity functional binders [41]. banded houndshark (Triakis scyllium)[34] and nurse shark [14, 35, 36](Table 1). Most of them are either produced from an immunized library or a semi-synthetic library with POTENTIAL CLINICAL APPLICATION FOR SHARK 6 9 a diversity ranging from 10 to 10 . A phage-displayed V S NAR V single domain library was generated from nurse NAR sharks (G. cirratum) by immunization of the model antigen Shark V domains possess unique features and advan- NAR lysozyme [36]. A semi-synthetic phage displayed V tages in potential clinical application compared to con- NAR library derived from wobbegong shark (Orectolobus macu- ventional IgG, including smaller sizes, modifiable half-life, latus) with a diversity of 3 × 10 [37] and expanded to 4 × higher tissue penetration ability [25], higher solubility and 10 [38] was constructed by randomizing the CDR3 region stability [42] and potential to protrude in buried functional [37]. A naïve library previously reported was relatively small sites in antigens [43]. Single domain antibodies, including (1 × 10 ) from naïve adult spiny dogfish (S. acanthias)and camel V Hs [44, 45] and human VH single domains [46– smooth dogfish (M. canis)sharks[21]. The naïve V 48], have the potential to bind the hidden clefts or grooves NAR libraries derived spiny dogfish and smooth dogfish sharks in antigens to block the receptor/ligand interactions [17, 20, have no Type I V [21]. The limitations in the library 49, 50]. On the other hand, the potential immunogenicity NAR of shark antibodies could limit their clinical applications size and types of V s may be the major reasons why NAR including chimeric antigen receptor T cells. To address high affinity V s is generally difficult to isolate from a NAR this potential issue, a study was conducted to humanize naïve shark single domain library. The naïve nurse shark the framework of V by grafting antigen interacting V phage-displayed library that we constructed from six NAR NAR regions to a human framework [51]. The humanized V nurse sharks is currently the largest shark single domain NAR maintained part of the antigen binding; however, more library with a diversity of 1.2 × 10 containing all four structure and computational studies are needed for more known classical types of V , including 11% of classical NAR effective humanization approaches. Furthermore, V s Type I, 57% of classical Type II and about 30% of the NAR have been isolated against vascular endothelial growth V s that do not belong to any of the classical V NAR NAR factor (VEGF) for treating uveitis, diabetic retinopathies types (Table 1), making it a unique tool to potentially and age-related macular degeneration (AMD) [25, 35]. isolate high affinity antibody candidates compared to Antibody Therapeutics, 2020 5 Table 1. Summary of shark V single domain libraries NAR Library Inventors Diversity Shark species Library type Binder types Antigens of binders Reference type isolated from library Phage Commonwealth 3 × 10 O. maculatus Semi-synthetic Type I and II Human periodontal Nuttall et al., 2002; Scientific and (CDR3 disease (Kgp, HRgp), 2003; 2004 [38, 58, 59] Industrial Research randomization) AMA1, Tom70 Organization (CSIRO, Australia) Phage University of 5 × 10 G. cirratum Immunized Types I and II Hen egg-white Dooley et al., 2003 Aberdeen (UK) lysozyme [36] Phage Naval Research 1 × 10 S. acanthias Semi-synthetic Type II and III Ricin, SEB, BoNT/A Liu et al., 2007 [27] Laboratory (US) (CDR3 toxoid randomization) Phage Naval Research 1 × 10 M. canis Naïve Type II and III Cholera toxin Liu et al., 2007 [21] Laboratory (US) Phage Tokyo University of 3.7 × 10 T. scyllium Immunized Unknown Hen egg white Otani et al., 2012 [34] Marine Science and lysozyme Technology (Japan) Phage CICESE (Center for 1.2 × 10 H. francisci Immunized Unknown TNFα Camacho-Villegas et Scientific Research al., 2013 [24] and Higher Education at Ensenada, US) Yeast Technische 2 × 10 C. plagiosum Semi-synthetic Unknown EpCAM, HTRA1, Zielonka et al., 2014 Universität (CDR3 and EphA2 [28] Darmstadt randomization) (Germany) Phage Elasmogen (UK) 10 G. cirratum Immunized Unknown Human and mouse Kovaleva et al., 2017 Induced [35] costimulatory ligand Phage Universiti Sains 1.16 × 10 O. ornatus Immunized Type I and II Malaria Leow et al., 2018 [26] Malaysia (Malaysia) Phage National Cancer 1.2 × 10 G. cirratum Naive Type I-IV GPC3, HER2, PD1, Feng et al., 2019 [14] Institute (NCI) (US) MERS, SARS, P. exotoxin A summary of shark V single domain libraries produced by academic institutes and commercial companies. NAR 6 Antibody Therapeutics, 2020 Table 2. Preclinical development of shark V s NAR Name Target antigen Shark species Library type Application Developmental stage Inventors Reference 14I-1, 14I1-M15 AMA1 O. Maculatus Naïve and Malaria Lead identification CSIRO Health Henderson et synthetic Sciences and al., 2007 [20] Nutrition (Australia) Anti-TNF vNAR TNF H. francisci Repeated Endotoxic shock Pre-clinical Institute Nacional de Bojalil et al., immunized Cardiologia Ignacio 2013 [22] Chavez (Mexico) BFF1 BAFF G. Cirratum Semi-synthetic Multiple sclerosis Lead identification Ossianix (UK) Hasler et al., (CDR3 2016 [60] randomization) vNAR-D01 Aurora-A kinase O. Maculatus Synthetic Solid tumors Lead identification University of Leeds Burgess et al., (UK) 2016 [23] ELN/21, ELN/22 ICOSL G. Cirratum Immunized, naïve, Auto-immune disease, In vivo validation Elasmogen (UK) Kovaleva et and synthetic uveitis al., 2017 [35] D1-BA11-C4, TNF-α G. Cirratum Immunized Polyarthritis Pre-clinical Elasmogen (UK) Ubah et al., D1-Fc-C4 2017; 2019 [52, 53] V13 VEGF H. francisci immunized Vascular eye disease Lead identification CONACYT (Mexico) Camacho- Villegas et al., 2018 [25] F1 GPC3 G. Cirratum Naïve Liver Cancer Lead identification NCI (US) Feng et al., 2019 [14] Her2A6 Her2 G. Cirratum Naïve Her2 positive solid Lead identification NCI (US) Feng et al., tumors 2019 [14] PD1A1 PD1 G. Cirratum Naïve Solid tumors Lead identification NCI (US) Feng et al., 2019 [14] MERS A3, A7, MERS spike G. Cirratum Naïve MERS virus Lead identification NCI (US) Feng et al., A8, B4, B5 protein 2019 [14] SARS binder SARS spike G. Cirratum Naïve SARS virus Lead identification NCI (US) Feng et al., protein 2019 [14] PE38 B6 P. exotoxin G. Cirratum Naïve Pseudomonas Lead identification NCI (US) Feng et al., infection 2019 [14] The V single domains that are being developed by commercial companies and academic institutes for potential clinical applications are listed. NAR Antibody Therapeutics, 2020 7 The designed macular route of delivery differentiated the to penetrate tissues and solid tumors will enable V sto NAR unique features of V s[25, 35]. reach organs that are inaccessible to conventional 150 kDa NAR Currently, several academic institutions and companies IgGs [16]. are developing naïve, immunized, or synthetic V as NAR potential therapeutics for human diseases (Table 2). The FUTURE DIRECTIONS Commonwealth Scientific and Industrial Research Organi- zation (CSIRO) in Australia described semi-synthetic and Structural, computational and functional research is naïve shark libraries and isolated various V s against needed to fully understand the vast sequence and structure NAR apical membrane antigen 1 (AMA1) in malaria [20]. Elas- diversity of shark V s and develop them as a novel NAR mogen Ltd developed several V s from immunized, syn- NAR family of therapeutics for various clinical applications. thetic and naïve libraries targeting inducible T cell costim- With the new report about the white shark genome [57], ulatory ligand (ICOSL) and tumor necrosis factor-alpha a comprehensive comparative genome and V sequence NAR (TNF-α)[52, 53]. Angiogenesis plays an important role in analysis of white shark, whale shark, nurse shark and other many human diseases, including eye diseases, by stimulat- sharks would be intriguing to gain insight on the origin ing the development of therapeutic agents that target the and diversity of shark V s in these ancient fishes that NAR pathological angiogenic process in eyes. Bevacizumab is a are known to have the earliest adaptive immune system humanized mouse monoclonal antibody targeting human during evolution, and with many of them are still living on VEGF for treating cancers. It has been used off-label to the planet today among us. Shark V s may function NAR treat a specific eye disease called AMD, the leading cause as ancient antigen-recognition domains in shark TCRs of age-related irreversible blindness. A smaller Fab frag- and the new large nurse shark phage library that contains ment (48 kDa) of bevacizumab (called ranibizumab) has V s from both shark B and T cells may provide a unique NAR been approved for treating AMD based on clinical trials. source of TCR-mimic single domains targeting MHC- Although Ranibizumab stabilizes and improves vision in peptide complexes for treating cancer and other diseases. over 90% of patients [54], it is administered by intravit- real injection, causing discomfort and increasing the pos- sibility of infection. Recently, a head-to-head compari- ACKNOWLEDGEMENTS son of bevacizumab and ranibizumab showed both drugs This work was supported by the Intramural Research have equivalent effects clinically [55]. Interestingly, a shark Program of the National Institutes of Health, National V (named V13) with a long CDR3 (27 amino acids) NAR Cancer Institute, Center for Cancer Research (NCI CCR). against (VEGF) was isolated from a male H. francisci The construction of large phage displayed nurse shark shark immunized with recombinant human VEGF. The V single domain antibody libraries and the next- group administered V V13 to rabbits with healthy eyes NAR NAR generation sequencing analysis of nurse shark V swas and showed intraocular penetration, indicating the superior NAR funded by the NCI CCR FLEX Intramural Program tissue penetration ability of V s[25]. The finding may NAR Technology Development Award to M.H. We thank Dr. support the use of shark V s as an eye drop for human NAR Martin F. Flajnik (University of Maryland School of eye diseases. Further studies will be needed to compare the Medicine) for expert advice and providing the buffy coat small size V s (e.g. V13) with classical antibodies (e.g. NAR collected from nurse sharks for our construction of a large bevacizumab and ranibizumab) for treating eye diseases phage displayed shark V library at the NCI. We also and other human diseases that require tissue penetration. NAR thank NIH Fellows Editorial Board and NIH Library Our lab has used the new large shark V library to NAR Editing Service for editorial assistance. The content of isolate a panel of shark single domain antibodies that bind this publication does not necessarily reflect the views or tumor antigens (e.g. GPC3, Her2) and pathogens (e.g. Pseu- policies of the Department of Health and Human Services domonas aeruginosa, MERS, SARS) [14]. We have isolated nor does mention of trade names, commercial products, or a Type II shark V single domain (named B6) which NAR organizations imply endorsement by the US Government. binds specifically to Pseudomonas exotoxin with high affin- ity (K = 10 nM). We established a protocol to quickly produce Type I and Type II shark V single domain NAR REFERENCES protein in E. coli for functional analysis [14]. The Type 1. Dooley, H, Flajnik, MF. Antibody repertoire development in II B6 shark V single domain can be expressed and NAR cartilaginous fish. Dev Comp Immunol 2006; 30: 43–56. purified from E. coli culture and has the ability to inhibit P. 2. Feige, MJ, Grawert, MA, Marcinowski, M et al. The structural exotoxin activity. Ongoing studies at the NCI are validating analysis of shark IgNAR antibodies reveals evolutionary principles of immunoglobulins. Proc Natl Acad Sci U S A 2014; 111: 8155–60. the therapeutic potential of these V s and other binders NAR 3. Ho, M. Inaugural editorial: searching for magic bullets. Antibody isolated from our nurse shark V library. NAR therapeutics 2018; 1:1–5. Once immunogenicity issues for therapeutic V sare NAR 4. Frommel, D, Litman, GW, Finstad, J et al. The evolution of the further addressed by improving humanization approaches immune response. XI. The immunoglobulins of the horned shark, for V and identifying an appropriate, fully human Heterodontus francisci: purification, characterization and structural NAR requirement for antibody activity. J Immunol 1971; 106: 1234–43. framework for CDR loop grafting from shark V s, NAR 5. Flajnik, MF. A cold-blooded view of adaptive immunity. Nat Rev V s can complement huge unmet market needs for con- NAR Immunol 2018; 18: 438–53. ventional IgG due to their unique advantages [56]. Inhal- 6. Criscitiello, MF, Saltis, M, Flajnik, MF. An evolutionarily mobile able, macular, or topical administrable therapeutic V s NAR antigen receptor variable region gene: doubly rearranging NAR-TcR genes in sharks. Proc Natl Acad Sci U S A 2006; 103: 5036–41. will expand the utility fields for antibodies, and the ability 8 Antibody Therapeutics, 2020 7. Barelle, C, Gill, DS, Charlton, K. Shark novel antigen receptors–the 29. Xu, Y, Salazar, GT, Zhang, N et al. T-cell receptor mimic (TCRm) next generation of biologic therapeutics? Adv Exp Med Biol 2009; antibody therapeutics against intracellular proteins. Antibody 655: 49–62. therapeutics 2019; 2: 22–32. 8. Matz, H, Dooley, H. Shark IgNAR-derived binding domains as 30. Kovalenko, OV, Olland, A, Piche-Nicholas, N et al. Atypical antigen potential diagnostic and therapeutic agents. Dev Comp Immunol recognition mode of a shark immunoglobulin new antigen receptor 2019; 90: 100–7. (IgNAR) variable domain characterized by humanization and 9. Roux, KH, Greenberg, AS, Greene, L et al. Structural analysis of the structural analysis. J Biol Chem 2013; 288: 17408–19. nurse shark (new) antigen receptor (NAR): molecular convergence of 31. Wesolowski, J, Alzogaray, V, Reyelt, J et al. Single domain NAR and unusual mammalian immunoglobulins. Proc Natl Acad antibodies: promising experimental and therapeutic tools in infection Sci U S A 1998; 95: 11804–9. and immunity. Med Microbiol Immunol 2009; 198: 157–74. 10. Clem, LW, Leslie, GA. Phylogeny of immunoglobulin structure 32. Zielonka, S, Empting, M, Konning, D et al. The shark strikes twice: and function. XIV. Peptide map and amino acid composition Hypervariable loop 2 of shark IgNAR antibody variable domains studies of shark antibody light chains. Dev Comp Immunol 1982; 6: and its potential to function as an autonomous Paratope. Mar 263–9. Biotechnol (NY) 2015; 17: 386–92. 11. Diaz, M, Stanfield, RL, Greenberg, AS et al. Structural analysis, 33. Cabanillas-Bernal, O, Dueñas, S, Ayala-Avila, M et al. Synthetic selection, and ontogeny of the shark new antigen receptor (IgNAR): libraries of shark vNAR domains with different cysteine numbers identification of a new locus preferentially expressed in early within the CDR3. PLoS One 2019; 14: e0213394. development. Immunogenetics 2002; 54: 501–12. 34. Otani, M, Hikima, J-i, Jung, TS et al. Construction of an artificially 12. Mashoof, S, Criscitiello, MF. Fish Immunoglobulins. Biology randomized IgNAR phage display library: screening of variable (Basel) 2016; 5:45. regions that bind to hen egg white lysozyme. Mar Biotechnol (NY) 13. Rumfelt, LL, Avila, D, Diaz, M et al. A shark antibody heavy chain 2012; 15: 56–62. encoded by a nonsomatically rearranged VDJ is preferentially 35. Kovaleva, M, Johnson, K, Steven, J et al. Therapeutic potential of expressed in early development and is convergent with mammalian shark anti-ICOSL VNAR domains is exemplified in a murine model IgG. Proc Natl Acad Sci U S A 2001; 98: 1775–80. of autoimmune non-infectious uveitis. Front Immunol 2017; 8: 1121. 14. Feng, M, Bian, H, Wu, X et al. Construction and next-generation 36. Dooley, H, Flajnik, MF, Porter, AJ. Selection and characterization sequencing analysis of a large phage-displayed VNAR single-domain of naturally occurring single-domain (IgNAR) antibody fragments antibody library from six naive nurse sharks. Antibody therapeutics from immunized sharks by phage display. Mol Immunol 2003; 40: 2019; 2:1–11. 25–33. 15. Zielonka, S, Empting, M, Grzeschik, J et al. Structural insights and 37. Nuttall, SD, Krishnan, UV, Hattarki, M et al. Isolation of the new biomedical potential of IgNAR scaffolds from sharks. MAbs 2015; antigen receptor from wobbegong sharks, and use as a scaffold for 7: 15–25. the display of protein loop libraries. Mol Immunol 2001; 38: 313–26. 16. Ubah, OC, Buschhaus, MJ, Ferguson, L et al. Next-generation 38. Nuttall, SD, Krishnan, UV, Doughty, L et al. A naturally occurring flexible formats of VNAR domains expand the drug platform’s NAR variable domain binds the Kgp protease from Porphyromonas utility and developability. Biochem Soc Trans 2018; 46: 1559–65. gingivalis. FEBS Lett 2002; 516: 80–6. 17. Stanfield, RL, Dooley, H, Flajnik, MF et al. Crystal structure of a 39. Rouet, R, Jackson, KJL, Langley, DB et al. Next-generation seque- shark single-domain antibody V region in complex with lysozyme. ncing of antibody display repertoires. Front Immunol 2018; 9:118. Science 2004; 305: 1770–3. 40. Anderson, GP, Teichler, DD, Zabetakis, D et al. Importance of 18. Nakanishi, T, Tsumoto, K, Yokota, A et al. Critical contribution of Hypervariable region 2 for stability and affinity of a shark VH-VL interaction to reshaping of an antibody: the case of single-domain antibody specific for Ebola virus nucleoprotein. PLoS humanization of anti-lysozyme antibody, HyHEL-10. Protein One 2016; 11: e0160534. Science: A Publication of the Protein Society 2008; 17: 261–70. 41. Kopsidas, G, Roberts, AS, Coia, G et al. In vitro improvement of a 19. Walsh, R, Nuttall, S, Revill, P et al. Targeting the hepatitis B virus shark IgNAR antibody by Qbeta replicase mutation and ribosome precore antigen with a novel IgNAR single variable domain display mimics in vivo affinity maturation. Immunol Lett 2006; 107: intrabody. Virology 2011; 411: 132–41. 163–8. 20. Henderson, KA, Streltsov, VA, Coley, AM et al. Structure of an 42. Liu, JL, Zabetakis, D, Brown, JC et al. Thermal stability and IgNAR-AMA1 complex: targeting a conserved hydrophobic cleft refolding capability of shark derived single domain antibodies. Mol broadens malarial strain recognition. Structure 2007; 15: Immunol 2014; 59: 194–9. 1452–66. 43. Irving, RA, Coia, G, Roberts, A et al. Ribosome display and affinity 21. Liu, JL, Anderson, GP, Delehanty, JB et al. Selection of cholera maturation: from antibodies to single V-domains and steps towards toxin specific IgNAR single-domain antibodies from a naive shark cancer therapeutics. J Immunol Methods 2001; 248: 31–45. library. Mol Immunol 2007; 44: 1775–83. 44. Akiba, H, Tamura, H, Kiyoshi, M et al. Structural and 22. Bojalil, R, Mata-Gonzalez, MT, Sanchez-Munoz, F et al. thermodynamic basis for the recognition of the substrate-binding Anti-tumor necrosis factor VNAR single domains reduce lethality cleft on hen egg lysozyme by a single-domain antibody. Sci Rep and regulate underlying inflammatory response in a murine model of 2019; 9: 15481. endotoxic shock. BMC Immunol 2013; 14: 17. 45. Rouet, R, Dudgeon, K, Christie, M et al. Fully human VH single 23. Burgess, SG, Oleksy, A, Cavazza, T et al. Allosteric inhibition of domains that rival the stability and cleft recognition of Camelid aurora-a kinase by a synthetic vNAR domain. Open Biol 2016; 6: antibodies. J Biol Chem 2015; 290: 11905–17. 160089. 46. Li,N,Wei,L,Liu,X et al. A frizzled-like cysteine rich domain in 24. Camacho-Villegas, T, Mata-Gonzalez, T, Paniagua-Solis, J et al. glypican-3 mediates Wnt binding and regulates hepatocellular Human TNF cytokine neutralization with a vNAR from carcinoma tumor growth in mice. Hepatology 2019; 70: 1231–45. Heterodontus francisci shark: a potential therapeutic use. MAbs 47. Feng, M, Gao, W, Wang, R et al. Therapeutically targeting 2013; 5: 80–5. glypican-3 via a conformation-specific single-domain antibody in 25. Camacho-Villegas, TA, Mata-Gonzalez, MT, Garcia-Ubbelohd, W hepatocellular carcinoma. Proc Natl Acad Sci U S A 2013; 110: et al. Intraocular penetration of a vNAR: in vivo and in vitro E1083–91. VEGF165 neutralization. Mar Drugs 2018; 16: 113. 48. De Genst, E, Silence, K, Decanniere, K et al. Molecular basis for the 26. Leow, CH, Fischer, K, Leow, CY et al. Isolation and characteriz- preferential cleft recognition by dromedary heavy-chain antibodies. ation of malaria PfHRP2 specific VNAR antibody fragments from Proc Natl Acad Sci U S A 2006; 103: 4586–91. immunized shark phage display library. Malar J 2018; 17: 383. 49. Stanfield, RL, Dooley, H, Verdino, P et al. Maturation of shark 27. Liu, JL, Anderson, GP, Goldman, ER. Isolation of anti-toxin single single-domain (IgNAR) antibodies: evidence for induced-fit binding. domain antibodies from a semi-synthetic spiny dogfish shark display J Mol Biol 2007; 367: 358–72. library. BMC Biotechnol 2007; 7: 78. 50. Li, N, Fu, H, Hewitt, SM et al. Therapeutically targeting glypican-2 28. Zielonka, S, Weber, N, Becker, S et al. Shark attack: high affinity via single-domain antibody-based chimeric antigen receptors and binding proteins derived from shark vNAR domains by stepwise in immunotoxins in neuroblastoma. Proc Natl Acad Sci U S A 2017; vitro affinity maturation. J Biotechnol 2014; 191: 236–45. 114: E6623–31. Antibody Therapeutics, 2020 9 51. Steven, J, Muller, MR, Carvalho, MF et al. In vitro maturation of a 56. Van Bockstaele, F, Holz, JB, Revets, H. The development of humanized shark VNAR domain to improve its biophysical nanobodies for therapeutic applications. Current Opinion in properties to facilitate clinical development. Front Immunol 2017; 8: Investigational Drugs (London, England : 2000) 2009; 10: 1361. 1212–24. 52. Ubah, OC, Steven, J, Porter, AJ et al. An anti-hTNF-alpha variable 57. Marra, NJ, Stanhope, MJ, Jue, NK et al. White shark genome new antigen receptor format demonstrates superior in vivo reveals ancient elasmobranch adaptations associated with wound preclinical efficacy to Humira(R) in a transgenic mouse autoimmune healing and the maintenance of genome stability. Proc Natl Acad Sci polyarthritis disease model. Front Immunol 2019; 10: 526. USA 2019; 116: 4446–55. 53. Ubah, OC, Steven, J, Kovaleva, M et al. Novel, anti-hTNF-alpha 58. Nuttall, SD, Humberstone, KS, Krishnan, UV et al. Selection and variable new antigen receptor formats with enhanced neutralizing affinity maturation of IgNAR variable domains targeting potency and multifunctionality, generated for therapeutic plasmodium falciparum AMA1. Proteins 2004; 55: 187–97. development. Front Immunol 2017; 8: 1780. 59. Nuttall, SD, Krishnan, UV, Doughty, L et al. Isolation and 54. Ng, EW, Adamis, AP. Targeting angiogenesis, the underlying characterization of an IgNAR variable domain specific for the disorder in neovascular age-related macular degeneration. Canadian human mitochondrial translocase receptor Tom70. Eur J Biochem J ophthalmol Journal canadien d’ophtalmologie 2005; 40: 352–68. 2003; 270: 3543–54. 55. Martin, DF, Maguire, MG, Ying, GS et al. Ranibizumab and 60. Hasler, J, Flajnik, MF, Williams, G et al. VNAR single-domain bevacizumab for neovascular age-related macular degeneration. N antibodies specific for BAFF inhibit B cell development by Engl J Med 2011; 364: 1897–908. molecular mimicry. Mol Immunol 2016; 75: 28–37.
Journal
Antibody Therapeutics – Oxford University Press
Published: Jan 21, 2020
Keywords: nurse shark ( Ginglymostoma cirratum ); phage display library; V NAR single domain; next-generation sequencing; antibody engineering