Towards patient-relevant structures: reviewing body-temperature biological macromolecules and their ligands for pharmaceutical applications.

The aim of our structures for eventual clinical application is to be relevant. Regulation of pharmaceutical lead compounds, however, does not yet involve the need for patient-relevant macromolecular structures determined at 37°C, as it is not yet known whether crystal growth and diffraction at 37°C versus standard cryo-condition practices will reveal significant binding variations applicable for drug development or, in the case of extremophiles, provide insight into their function. However, for select examples in the literature interesting changes occurred, and support the initiative that data collection at high temperatures should be considered. This topical review considers a Protein Data Bank (PDB) and Cambridge Structural Database (CSD) data survey of crystal structures that have been determined at elevated temperatures, i.e. neither under cryogenic conditions nor at typical room-temperature conditions of 20-25°C, and reveals a few hurdles as well as many successes in reaching such patient-relevant structures. It highlights key methodology that appears in the literature which could benefit those considering related research. Since it is possible for crystallographic structure-determination methods to be adapted to 37°C, amid some challenges, we encourage the initiative that many more could be determined at 37°C. Included in the studies deposited in the PDB are some that have been performed at temperatures in excess of >37°C, and surprisingly several at even higher temperatures (i.e. 50-90°C). The overall aim of determining the 3D structure of a biological macromolecule at its natural body temperature has in principle to include crystallization and diffraction data collection. In the survey we find very few crystallizations performed at 37°C followed by data collection at the same temperature, and few have conducted a systematic study of comparing the changes occurring at 100 K versus 37°C. It is of course assumed that some key drug binding in proteins may occur over a narrow temperature range appropriate for mesophilic organisms, whereas for thermophilic organisms the protein may well exist over a wide temperature range reflecting that in which the organism is able to thrive. For the higher temperature structure solutions, those in the range which is more appropriate for thermophiles or hyperthermophiles, no crystallizations at these extreme temperatures have yet been conducted. The ability to conduct crystallization at 37°C and obtain acceptable high-resolution data at the same temperature is surely encouraging to the crystallographic community to build on these achievements for this and the full temperature range. We describe aspects of crystallization, mounting and transfer of crystals, data collection, reporting of metadata within databases etc. that have been notable during the survey of the data and highlight them here for the benefit of the community which may be considering 37°C data analysis from pre-crystal growth to re-refinement of data. For comparable data and to avoid any experimental bias, we also encourage the community to complete the analysis sequentially, as few have considered this holistic analysis of solid-state variations which may occur over the low-to-high temperature range.