Improved throughput of affinity reagent production

In order to approach complete coverage of the human proteome, new binders will have to be made in very large numbers. The proteome is estimated to exceed by at least an order of magnitude the number of gene loci (~20,300), due to alternative splicings and post-translational modifications (PTMs). Moreover, for confident detection and quantification of various protein states (folded, denatured, in complexes, ...) in different applications, several binders will be needed for each target. In addition, proteins often have multiple functional and structural subregions of interest, such as active sites, motifs targeted for modification, and the modifications themselves, any of which can be important epitopes. Hence, in order to serve the likely needs of the research community, panels of binders well characterised for affinity and specificity will be sought for many of the targets.

While a true proteomic scale is clearly beyond the scope of a single project, AFFINOMICS will demonstrate feasibility utilising the most efficient methods available for polyclonal (KTH) and monoclonal antibody (EMBL, HMGU) generation, as well as focusing on recombinant binders (BBT, KTH, TUBS, ULUND, UZH, VIB). While purified monospecific polyclonal antibodies can be made quickly and are suitable for certain applications (see below), renewable sources of binders are essential for long-term continuity. Recombinant methods are useful not only for generation of binders against toxic or non-immunogenic targets which cannot be employed in vivo; the pivotal advantage of recombinant selection strategies is that the desired properties of the binders can be enforced by the selection approach used, such as recognition of the native state or of a linear epitope, or recognition of a particular isomer, conformation or modification, or other epitope, as well as affinity. Moreover, as these properties can all be selected for, the characterisation of recombinant binding reagents has usually been done in vastly more detail than that of antibodies obtained by traditional routes. However, while recombinant methods may well fulfil future production needs, and in particular generate binders which greatly extend the range of applications possible with mAbs, a major current hurdle is the speed of selection from molecular libraries. To combine the enormous opportunities inherent in these recombinant technologies, and since high throughput methods operating to maximum efficiency of production and cost will be critical to the success of the effort, significant resources will be deployed in ProtAffin toward developing novel automated recombinant selection procedures, with streamlined robotic protocols for parallel handling of multiple targets, to establish the robustness to deliver on a proteome-wide scale and capable of delivering several thousand binders per year. Engineering processes to require much less target protein and use of miniaturised systems to allow economically feasible production will also be important objectives. Fully established recombinant methods will become an essential part of future proteome-wide efforts.