When Great is not Good Enough—VHH Antibodies Engineered for 10 Fold Affinity Increase

http://blog.allelebiotech.com/2013/10/1697/

Single Domain antibodies (VHH fragments, nanobodies, or as we call them, nAbs) have been generated by injecting llamas with ligand-bound GPCR for the purpose of obtaining crystals of active-state structures. Such structural information could be critical in understanding drug functions and screening for new drugs. The unique ability of VHH fragments to fit into protein-protein complex crevices and hold proteins together was demonstrated by two Nature publications from Brian Kobilka’s group at Stanford ([1, 2], also see Allele Newsletter of Sep 4th, 2013). The nano antibody used in those studies, Nb80, showed affinity towards only the active state of the target GPCR.

However, even with an antibody as great as Nb80, the authors were only able to co-crystal GPCR beta2-adrenoceptor (b2AR) with high affinity agonists, not its natural agonists such as adrenaline. In yet another Nature paper published just now, the Kobilka lab showed that Nb80 could be further improved by 10 times in affinity, through in vitro evolution [3]. They presented Nb80 on the surface of yeast using an existing yeast display system, then applied standard limited mutagenesis and magnetic separation technologies for screening. After about 5 rounds of selection, a new version of VHH Nb6B9 was isolated that bound to ligand-loaded GPCR with a kD of 6.4 nM. For the first time, a co-crystal of b2AR-adrenoline was made.

Rasmussen et al. Nature, 2011 Structure of a nanobody-stabilized active state of the b2 adrenoceptor
Rasmussen et al. Nature, 2011 Crystal structure of the b2 adrenergic receptor–Gs protein complex
Ring et al. Nature, 2013 Adrenaline-activated structure of b2-adrenoceptor stabilized by an engineered nanobody

Camelid Antibodies--background

When it was discovered that animals in the camel family produce antibodies with no light chains, the idea that a single-domain fragment can bind as well as a full 4-chain antibody formed a breakthrough. So far it has been a relatively less known one.

Smaller antibody fragments have been tested for therapeutic uses because classical IgG antibodies are too bulky to penetrate tissues well, and very expensive to produce. Different combinations of antigen-binding variable regions are used, e.g. scFv, Fab, diabody, all to some degree of success. In comparison, the N-terminal domain of camelid antibodies, termed VHH domain, represents a naturally evolved, only 13-15 kD in size, fully functional target binding fragment with many advantages.

The only other known species outside camelidae family that has heavy chain antibodies is particular cartilaginous fish, nurse shark. Although the arrangement of CDRs is somewhat different between the camel and shark heavy chain variable regions, they share many characteristics such as extremely high stability (maintaining functions after100 C heat and extreme pH treatment).

Accumulating reports have demonstrated the therapeutic potentials of camelid antibody-based fragments in treating cancer, neural diseases, even use in hair dandruff preventing shampoo. For basic research, the tiny antigen binders can be used as tools for quantitative pull down with unmatched efficiency, recognizing previously inaccessible enzyme cleft as antigens, and providing libraries for binding partner selection.

Allele Biotech has been working on display antibody selection from its early days through an NIH grant, and recently carried out an NIH/NCI contract for scFv yeast display in collaboration with AvantGen. By working with Chromotek on camelid antibody products, we hope to combine our superior fluorescent proteins with the best antibody candidates and display technologies to move the capture and signaling fields forward in significant ways.

The product line will be formally announced by AlleleNews shortly.