Understanding mis-metalation of native versus heterologously expressed protein

All known life depends on metals, microbial and plant. The special chemical properties of different metals expand the repertoire of what proteins can do unaided. This requires the chemically correct metal(s) to associate with each metalloprotein. It is tempting to assume that such metalation is a sole function of the proteins themselves, but purified proteins typically bind wrong metals many orders of magnitude more tightly than the correct ones. We now appreciate that correct metalation is a combined function of proteins and their biological surroundings. Proteins compete with other molecules for each metal. Over the millennia this competition has evolved such that the correct metals generally win and mis-metalation is largely avoided: But does this mean that proteins are liable to mis-metalation when expressed in different organisms or if the cellular milieu is somehow altered?

DNA-binding metal sensors are tuned to compete for available metal with the intracellular milieu (Nature Chemical Biology 2017 13:409-417). Over the course of nearly a decade, supported by the BBSRC, a series of thermodynamic values were collected for a set of DNA-binding metal sensors (from Salmonella), making it possible to calculate the magnitude of intracellular competition for each metal (Nature Chemical Biology 2019 15:241-0249). By reference to these values, it has become possible to make predictions about the metalation of metalloproteins. We have recently developed a metalation calculator that provides such calculations in Salmonella and closely related E. coli (Nature Communications 2021 12: 1195). It is now necessary to test such calculations in other organisms.

Some metals are embedded into cofactors. Proteins that supply metals to cofactors, such as cobalt to vitamin B12, must also somehow acquire the correct metal. Notably, plants do not make B12 generating interest in the supply of this vitamin for the increasing numbers of individuals turning to low meat diets. E. coli also does not make B12 but the CoI has engineered (by introduction of more than 30 genes) E. coli to generate a B12 hyper-producing strain. We have shown that in the synthetic E. coli system the cobalt delivery protein for B12 is susceptible to mis-metalation with zinc but mismetalation is overcome when cells are supplemented with surplus cobalt. Is such mis-metalation a function of a mis-match between the introduced proteins and the cellular milieu?

We will express, purify and prepare the deduced or known Rhizobium proteins involved in sensing zinc and cobalt, and in cobalt delivery for B12, in a manner suitable for determining metal affinities and DNA-affinities for the sensors. Sensor abundance will also be determined. A hybrid Rhizobium metalation calculator will thus be produced that can determine metalation and mis-metalation of the B12 pathway with cobalt and zinc (or iron) in this native host system. B12 production will be measured and metalation calculated both in free living Rhizobium and in root nodules. One central purpose of this work is to further develop (and exemplify the utility of) the metalation calculator with widespread applications/implications in biology. This work also has implications for the supply of B12 to those on plant-based diets (refer to closing statement on advancing to higher TRLs). Rhizobium makes B12 in nodules of leguminous plant roots and thus cobalt promotes legume growth. An ability to design seed dressings (potentially containing cobalt or with a zinc chelator), to enhance nitrogen fixation also has implications for global sustainability since the manufacture of nitrogen fertiliser represents up to 2% of energy demands. The PI and CoI directed phase I, and now direct phase II, of a BBSRC Network in Industrial Biotechnology and Bioenergy with the express purpose of accelerating the exploitation of such advances in metals in biology to support biotechnology.

We will determine intracellular availabilities of Zn and Co in Rhizobium leguminosarum as free energies for complex formation by establishing the sensitivities of cognate metal-detecting transcriptional regulators ZntR, Zur and DmeR. Parameters will be collected in order to solve the thermodynamic cycles of the sensors (to date this has only been done for the sensors of Salmonella and approximated in E. coli). Fe availability will also be approximated from the Fe-dependent activity of Rhizobium ferrochelatase. A (hybrid) Rhizobium metalation calculator will be created.

Expression of genes regulated by the Rhizobium sensors will be calibrated to read-out internal free energies for metalation. Intracellular metal availabilities will thus be determined in free-living cells and in nodules. The metals acquired by CobW, that supplies Co for vitamin B12 biosynthesis, will be calculated from differences in free energies for metal complex formation having measured the affinities (for Mn, Fe, Co, Ni, Zn and Cu) of Rhizobium MgGTPCobW. This will reveal whether or not, unlike heterologous CobW in engineered E. coli, Zn mis-metalation of Rhizobium CobW is avoided in the natural system.

CobW, DmeR, ZntR and Zur will be expressed, purified and prepared in forms suitable for determining metal-affinities and (for the metal-sensors) DNA-affinities. The number of molecules per cell of DmeR, ZntR and Zur will be measured by MRM-MS in cells grown under metal regimes that give maximum and minimum expression of the respective target genes. The sensitivities of the metal-sensors will provide insight into the extent to which intracellular metal availabilities show different, or broadly similar, ranges in different bacteria. The work will reveal whether the metal-affinities of metalloproteins such as CobW are tuned to a specific intracellular milieu, and then fine-tuned to normal growth conditions, with implications for 'metal-matching' of proteins and cells in biotechnology.