Design of cyborg proteins by loop region replacement with oligo(ethylene glycol): exploring suitable mutations for cyborg protein construction using machine learning

We synthesized a “cyborg protein,” wherein a synthetic molecule partially substitutes the main peptide chain by linking 2 protein domains with a synthetic oligomer. Green fluorescent protein (GFP) served as the model for constructing the cyborg proteins. We prepared circularly permuted GFP (cpGFP) with new termini between β10 and β11, where the original N- and C-termini were linked by a cleavable peptide loop. The cyborg GFP was constructed from cpGFP by linking the β10 and β11 with oligo(ethylene glycol) (OEG) using maleimide–cysteine couplings, followed by the enzymatic cleavage of the N- and C-termini linking loop by thrombin. With the help of machine learning, we were able to obtain the cpGFP mutants that significantly alter the fluorescence intensity (53% increase) by thrombin treatment, which splits cpGFP into 2 fragments (fragmented GFP), and by heat shock. When the cyborg GFP was constructed using this mutant, the fluorescence intensity increased by 13% after heat treatment, similar to cpGFP (33% increase), and the behavior was significantly different from that of the fragmented GFP. This result suggests the possibility that the OEG chain in the cyborg protein plays a similar role to the peptide in the main chain of the protein.