Journal of Molecular Biology
Identification of a Universal VHH Framework to Graft Non-canonical Antigen-binding Loops of Camel Single-domain Antibodies
Introduction
It is over a decade ago that functional antibodies, composed of heavy chains only, were discovered in the serum of old world and new world camelids.1 Immediately after the discovery of these heavy chain antibodies, a method was designed to clone and select by phage display their antigen-binding domain, referred to as VHH or nanobody™, against any kind of antigen from the heavy chain antibodies of an immunized camelid.2, 3, 4 During the last couple of years, the potential of VHHs as an attractive alternative to antigen-binding fragments from conventional antibodies, such as Fabs and scFvs in biotechnological, diagnostic and therapeutic applications has been demonstrated.5, 6, 7, 8, 9, 10 Especially, the small size and strict monomeric behavior of the VHHs are at the origin of their outstanding performance.2, 3, 4, 5, 6, 7, 8, 9, 10 Moreover, their unique ability to target alternative epitopes compared to conventional antibodies offers additional opportunities.10 However, a number of applications require the use of antibodies or antibody fragments that remain functional after exposure to harsh conditions such as low pH of the stomach or during regeneration of biosensors. There is a growing interest in antibody fragments that preserve their antigen-binding capacity in the reducing environment inside cells. All variable domains, including VHHs, have a conserved disulfide bond between the β-sheets, and therefore their intracellular expression is unpredictable,11 hampering their use as intrabodies. Finally, for other applications like immunofluorescence or micro-array detection, it would be far better if all VHHs directed against a wide variety of antigens possess an identical scaffold, as this would allow the standardization of protocols of probe labeling or chip-immobilization, etc.12
A synthetic library made by the randomization of the codons encoding the loops of a unique, stable scaffold could certainly serve to retrieve binders against a wide variety of possible antigens. Although such libraries based on lipocalins, protein A, scFv, and ankyrins scaffolds have been reported,13, 14, 15, 16, 17, 18 in our hands, synthetic libraries on VHH scaffolds failed to yield consistently good affinity binders after panning. The antigen specificity seemed to be reasonable, but the affinity was most often insufficient to detect low concentrations of analyte, and tedious procedures should be introduced to improve the affinity, mostly involving the creation of a new library by randomizing codons in the periphery of the paratope, and reselection. This approach is time-consuming and it might be faster to first immunize a camelid, whereby the immune system of the host will perform the affinity maturation over a period of two months, and to construct subsequently an immune library of all the VHHs. Several groups, including ours, have reported the successful isolation of high-affinity binders based on camelid single domain antibody fragments from immune libraries.8, 9, 10, 19, 20 In this approach, the antigen-binding loops need to be grafted onto a universal, stable scaffold to obtain a standardized format for labeling, immobilizations, etc. Such a grafting approach has been used successfully for conventional antibody fragments.21, 22, 23 However, from crystallographic data we know that the antigen-binding loops of VHHs do not obey the canonical loop structure rules proposed for human/mouse VHs.24 Moreover, the key elements preserving loop architecture remain to be determined. Some of them might be hidden in the framework regions, and would not be included during the grafting experiment. For example, it is well established that in most VHHs, part of the CDR-H3 loop folds over the framework-2 region and, although all VHHs belong to the same family (equivalent of human VH-III), variations in this region do occur and might influence the CDR-H loop conformation. The loop grafting strategy for VHHs might therefore be far from evident.
Over the last eight years we have generated and characterized more than a hundred VHHs directed against various antigens and possessing a wide range of physico-chemical properties. Such a large VHH collection allows us to generate multiple chimeras by exchanging the CDR-Hs on different frameworks, and to deduce the general principles for successful grafting. In this work, clear evidence is provided that the framework of a VHH, cAbBCII10, is a good candidate to produce chimeras with antigen-binding parameters approaching those of the original CDR-H donor VHH. Moreover, these chimeras are often better expressed and of greater stability than the original VHH. Additionally, they are functional in the absence of the conserved disulfide bond of the immunoglobulin fold or under reducing conditions, a prerequisite for intrabody applications.
Section snippets
Exchange of CDR-Hs between homologous VHH scaffolds
To test the feasibility of loop grafting, we first chose two VHHs having nearly identical frameworks but exhibiting major sequence differences among their CDR-Hs. These VHHs are cAbHuL6 and cAbCA05, which specifically recognize human lysozyme25 and carbonic anhydrase,26 respectively. The amino acid sequences of their framework regions differ by only four amino acid residues, whereas major differences in CDR-H sequence and length occur (Figure 1(a)). In addition, cAbCA05 contains an interloop
Discussion
The technology of CDR grafting has been used extensively for the reshaping (humanization) of murine antibodies.21, 22, 23, 34, 35, 36, 37 VHHs might benefit from this grafting approach to enhance the stability and/or expression level of relevant VHHs. The absence of canonical loop structures,24, 30, 31 and knowledge of their key residues make this strategy less predictable for VHHs.
In this work, we identified the framework of cAbBCII10 as a potential universal VHH framework to graft various
Generation of chimeric VHH constructs
The different CDR-H loops from a donor VHH were transferred to the framework of the recipient VHH by PCR. CDR-H1 sequence of the donor VHH was encoded in a primer containing at the 5′ and 3′ ends the sequences corresponding to the framework residues of the recipient VHH. This primer, together with the universal reverse primer (RP), was used to amplify a fragment of ∼200 bp when the recipient VHH gene cloned in a pUC-based vector was used as template. This fragment, containing framework-1 from
Acknowledgements
We thank Sarah Haesaerts for technical assistance. This work was supported by the Flanders Interuniversity Institute for Biotechnology (VIB), Fonds voor Wetenschappelijk Onderzoek-Vlaanderen (FWO-Vlaanderen) and Vrije Universiteit Brussel (VUB-GOA and GBOU contracts). D.S. and M.P. acknowledge financial support from the Instituut voor de aanmoediging van Innovatie door Wetenschap en Technologie in Vlaanderen (IWT-Flanders). M.D. was supported by a Marie Curie Fellowship from the European
References (43)
- et al.
Functional heavy-chain antibodies in Camelidae
Advan. Immunol.
(2001) - et al.
Comparison of physical chemical properties of llama VHH antibody fragments and mouse monoclonal antibodies
Biochim. Biophys. Acta
(1999) - et al.
Selection and identification of single domain antibody fragments from camel heavy-chain antibodies
FEBS Letters
(1997) - et al.
Efficient targeting of conserved cryptic epitopes of infectious agents by single domain antibodies. African trypanosomes as paradigm
J. Biol. Chem.
(2004) Single domain camel antibodies: current status
J. Biotechnol.
(2001)- et al.
Isolation of endothelial cell-specific human antibodies from a novel fully synthetic scFv library
Biochem. Biophys. Res. Commun.
(2004) - et al.
Phage-displayed antibody libraries of synthetic heavy chain complementarity determining regions
J. Mol. Biol.
(2004) - et al.
High-affinity human antibodies from phage-displayed synthetic Fab libraries with a single framework scaffold
J. Mol. Biol.
(2004) - et al.
A semi-synthetic repertoire of intrinsically stable antibody fragments derived from a single-framework scaffold
J. Mol. Biol.
(2001) - et al.
Fully synthetic human combinatorial antibody libraries (HuCAL) based on modular consensus frameworks and CDRs randomized with trinucleotides
J. Mol. Biol.
(2000)