Abstract
Peroxiredoxins (Prx) are important antioxidant enzymes, acting to reduce peroxides, including hydrogen peroxide (H2O2) in cells. High concentrations of these peroxides can be toxic to cells and are thought to lead to various diseases including cancer and cardiovascular disease. Despite its detrimental effects, cells still require a certain level of H2O2 to carry out redox based signalling for regular cell function. Alongside their peroxidase activity, Prx are proposed to facilitate this redox signalling. Prx have been observed in vitro to form dimers, made up of two Prx monomers, and decameric/dodecameric rings made up of five or six dimers. The dimer oligomeric state is acknowledged as the minimal catalytic unit of most Prx. It is proposed that Prx dynamically switch between the dimer and decamer/dodecamer oligomeric state in cells depending on a range of factors including the protein redox state, pH, concentration or post-translational modifications. These proposed oligomerisation dynamics have not yet been observed in cells. It is also not known whether Prx dimers and decamer/dodecamers have different functions, localisations, or binding partners within cells. As the two oligomeric states of Prx expose different surfaces for interactions with other proteins, it could be reasonably hypothesised that Prx could interact with different proteins depending on its oligomeric state. There are currently very few tools that can specifically detect dimeric or decameric Prx in cells. The aim of this project was to develop a tool that could be used to study the two oligomeric states of human Prx2 in cells. Prx2 is one of the six human peroxiredoxins and localises to the cytosol.
Two methods were employed to develop a functional tool for probing Prx2 oligomerisation dynamics. Firstly, nanobodies were selected from a yeast surface display library (NbLib) using Prx2 variants stable as a dimer or a decamer. The nanobodies expressed in the yeast selected with either the Prx2 dimer or decamer were sequenced and the most enriched nanobody sequences were identified. The 6 most enriched nanobodies from each library screen were expressed and purified. They were then analysed for binding to purified Prx2 dimer or decamer using co-elution size exclusion chromatography. From this analysis, 3 nanobodies co-eluted specifically with the Prx2 dimer. Furthermore, these 3 nanobodies were found to be Prx2 specific as they did not co-elute with Prx1, another human Prx that localises to the cytosol. Two of the 3 nanobodies successfully isolated wildtype Prx2 from HEK293FT cell lysates, and one of these two nanobodies also isolated Prx1 from HEK293FT cell lysates. This could potentially be a Prx2:Prx1 heterooligomer as in vitro analysis showed this nanobody to be Prx2 specific. There were no nanobodies identified to bind specifically to the Prx2 decamer.
The second method employed was designing novel protein binders targeting the Prx2 decamer and the Prx2 dimer using the protein design pipeline, BindCraft. BindCraft is an open-access pipeline that leverages AlphaFold 2 and ProteinMPNN neural networks to design novel protein binders. A structure of Prx2 in the reduced state was used to identify suitable targets that could be used to design Prx2 oligomeric state specific novel protein binders. These target hotspots were then used to design novel protein binders using the BindCraft pipeline. There were over 200 binders designed targeting the Prx2 decamer, and 144 binders designed to target the Prx2 dimer. In silico screening using AlphaFold 3 and visual assessment identified 52 novel protein binders for the Prx2 decamer, and 26 novel protein binders for the Prx2 dimer that were taken forward for in vitro binding analysis. A S. cerevisiae surface display strain was transformed with the novel protein binders, and on-yeast binding was analysed by flow cytometry. No in vitro binding to either the Prx2 decamer or dimer was observed.
Although the tools developed here still require some optimisation, a tool that is eventually operable and can differentiate between the two oligomeric states of Prx2 would enable continued investigation into the functions of the Prx2 decamer and dimer.