How do repetitive bacterial surface proteins avoid inter-domain aggregation?

Protein aggregation has wide relevance to medicine and biotechnology. Aggregation underlies clinical conditions such as Alzheimer's disease that are placing an increasingly severe burden on health services. Aggregation is also a key problem for production, storage and application of protein therapeutics whether naturally occurring or produced using novel synthetic scaffolds. A range of scaffolds with intrinsic resistance to aggregation would provide a valuable starting point for future biologics.
Aggregation occurs when regions of (normally buried) protein sequence become exposed and interact; aggregation is dependent on high sequence identity between molecules and high concentration. Thus, the observation that adjacent domains in multi-domain proteins usually have <40% sequence identity has been interpreted as an evolutionary response to reduce aggregation between adjacent, covalently-linked domains. In repetitive bacterial surface proteins studied in the JRP group protein sequence identity is a result of highly identical DNA repeats that provide a potential advantage to the organism through recombination events that produce proteins with differing numbers of repeats (a potential immune evasion mechanism). However, the resultant protein sequence repetition should lead to aggregation. Our novel hypothesis is that, in repetitive bacterial proteins immune evasion "wins" and novel solutions to the protein aggregation problem have evolved and will be detectable in the isolated domains. Preliminary work supports this hypothesis.