Design Considerations For New Scaffolds

To date, all scaffolds have been derived from natural proteins. In principle, any protein that can tolerate extensive surface mutations can potentially be used as a molecular scaffold. However, as structural classifications of proteins have shown that a small number of protein folds are used in many proteins with diverse functions, not all protein architecture classes are equally suited as molecular scaffolds.

What are the requirements for a highly functional molecular scaffold? As outlined in the previous section, one of the main motivations in the development of new scaffolds is to avoid the issues associated with undesirable physical properties of the immunoglobulin scaffolds. These include the presence of multiple disulfide bonds and a heterodimeric architecture. Clearly, a scaffold must also have multiple surface positions that can be mutated to construct a contiguous interface. Extensive mutations frequently reduce the conformational stability of a protein. Decreased stability due to introduced mutations can also promote transient exposure of secondary structure elements (e.g., helix or strand) and the hydrophobic interior, which can lead to protein aggregation and domain swapping (Liu and Eisenberg 2002). For these reasons, it is advantageous to start with a highly stable protein as a molecular scaffold. Depending on the intended application, different types of stability, such as thermal stability and resistance to extreme pH or organic solvent, might be important. It is often possible to improve the conformational stability of a protein using rational design, library selection, or computational design (Malakauskas and Mayo 1998; Sieber et al. 1998; Perl et al. 2000; Koide et al. 2001). Therefore, a high degree of stability in a starting scaffold is desirable but not absolutely required.

A scaffold should be small (less than ~100 residues) for the ease of manipulation, production, and potential access by chemical synthesis, and simple in terms of its functional structure, meaning it should not contain prosthetic groups or dis-ulfide bonds. The absence of disulfide bonds is particularly important if intended uses include intracellular expression. The scaffold should also be easily produced in bacteria and highly soluble. Solubility in particular is an important point to consider for downstream applications. Not unexpectedly, introducing amino-acid residues that promote new molecular recognition often reduces the solubility of the resulting binders (Bianchi et al. 1994). These issues often need to be dealt with empirically because it is usually difficult to pinpoint the cause of protein precipitation and aggregation. Furthermore, a scaffold should be compatible with common molecular display techniques, and it should have minimal spectroscopic signatures because strong inherent signals limit flexibility in derivatization (e.g., fluorescence labels). Several systems that meet these stringent requirements have been successfully developed and are reviewed in the next section.

0 0

Post a comment