In search of novel folds : protein evolution via non-homologous recombination : a dissertation presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biochemistry at Massey University, Albany, New Zealand

Thumbnail Image
Open Access Location
Journal Title
Journal ISSN
Volume Title
Massey University
The Author
The emergence of proteins from short peptides or subdomains, facilitated by the duplication and fusion of the minigenes encoding them, is believed to have played a role in the origin of life. In this study it was hypothesised that new domains or basic elements of protein structure, may result from nonhomologous recombination of the genes coding for smaller subdomains. The hypothesis was tested by randomly recombining two distantly related (βα)8-barrel proteins: Escherichia coli phosphoribosylanthranilate isomerase (PRAI), and β subunit of voltage dependent K+ channels (Kvβ2) from Rattus norvegicus. The aim was to identify new, folded structures, which may or may not be (βα)8-barrels. Incremental truncation (ITCHY), a method for fragmenting and randomly recombining genes, was used to mimic in vivo non-homologous recombination and to create a library of chimeric variants. Clones from the library were selected for right reading frame and solubility (foldability) of the recombined chimeras, using the pSALect selection system. Out of the six clones identified as soluble by pSALect, only one (P25K86) was found to be actually soluble. The protein, P25K86, was found to form oligomers and on treatment with a reducing agent, β-mercaptoethanol the multimeric state disappeared. The protein has three cysteines and one of the cysteines (Cys56) was found to mediate in the bond formation, thus giving a dimeric state. An engineered version of P25K86 that has the Cys56 replaced by serine was expressed as a monomer and additionally it was found to be ! iv! more stable. As the pSALect folding selection system reported false positives, i.e. only one of the six chimeras was actually soluble, it was concluded that the in vivo solubility selection system was leaky. A series of experiments were conducted so as to improve pSALect that led to the creation of pFoldM – a more stringent selection system, discussed in chapter 4. Comparing the newer improved version with the old, two more interesting chimeras were discovered. A total of 240,000 non-homologous recombination events were created in vitro and three soluble chimeras (evolutionary solutions) were found. Data from circular dichroism spectroscopy (CD) combined with heteronuclear single quantum coherence (HSQC) spectra suggest that the proteins, P24K89 and P25K86, are present in a molten globule state. ITCHY, as a means of mimicking the subdomain assembly model, was applied in vitro. The discovery of two interesting chimeras (P25K86 and P24K89) using highthroughput engineering experiments widens the possibilities of exploring the protein structure space, and perhaps offers close encounters with these never born proteins that may be trapped in an ensemble of fluctuating (structured and unstructured) states.
Protein evolution, Protein structure, Genetic recombination