X-ray crystallography is a tool for determining the structures of proteins and protein-ligand complexes. In this thesis the method has been employed to study several proteins of medical relevance.

Cancer is a terrible disease, severely impacting those affected, as well as their family and friends. Current cancer treatments involve a combination of cytostatic drugs, surgery and radiation treatment. Unfortunately many cytostatic drugs also kill healthy cells, which gives rise to serious side-effects. The discovery of treatments which selectively inhibit proteins essential for cancer cell survival but which are non-essential in normal cells, could reduce such side-effects.

MTH1 is a protein that degrades oxidised nucleotides, which when incorporated into DNA cause mutations and subsequent cell death. Cancer cells have higher levels of reactive oxygen species, which create oxidised nucleotides.  In Paper I it was discovered that cancer cells are dependent on MTH1 for their survival. Crystal structures of MTH1 in complex with small molecules guided their development into potent MTH1 inhibitors, capable of killing cancer cells. Cells with increased amounts of oxidised nucleotides, or with induced hypoxia, were more susceptible to MTH1 inhibition, as shown in Paper II. In Paper III several MTH1 orthologues from organisms often used in pre-clinical studies were tested for MTH1 inhibition. Leucine 116 of mouse MTH1 was determined to be important for the lower inhibition of the developed inhibitors towards this enzyme. A virtual fragment screening study using commercial chemicals resulted in several potent MTH1 inhibitors, as shown in Paper IV. The crystal structures with the fragments or optimised inhibitors did in most cases agree with the docking pose determined from the virtual screening. In addition to the known function of MTH1 in the degradation of oxidised nucleotides, Paper V showed that MTH1 also degrades methylated nucleotides.

MTHFD2 is responsible for providing one-carbon units for nucleotide synthesis in cancer cells. As MTHFD2 is present in cancer cells but not in healthy cells, targeting the enzyme would make it possible to selectively kill cancer cells. Paper VI presents the first structure of MTHFD2, along with the first inhibitor of the protein. This information provides a starting point for the development of potent and selective MTHFD2 inhibitors.

The botulinum neurotoxin from the bacterium Clostridium Botulinum is the causative agent of the deadly disease botulism. The action of the botulinum neurotoxin on nerve cells results in paralysis, and is life-threatening if the patient is not helped with breathing support. However, low doses of the neurotoxin are used as a successful treatment for several medical conditions, such as involuntary spasms. In Paper VII the structure of two proteins, P47 and OrfX2, encoded in the gene cluster of a botulinum neurotoxin, were determined. The structures resembled tubular lipid-binding proteins, previously only found in eukaryotes. The proteins were also found to be able to bind lipids. This work gives new insight into the structure and function of this group of proteins, which help the deadly botulinum neurotoxins.

Molecular modeling is increasingly being integrated into the drug discovery process. Although computational methods are already an essential part of the process today, these techniques have vast potential to evolve and refine drug design in the future. The integration of modeling is further catalyzed by the rapidly growing computational power available to both academia and pharmaceutical companies. These resources extend the reach of the computational methods to new time scales in physics-based simulations of biomolecular systems and increase the number of molecules that is possible to dock to a receptor by several orders of magnitude. However, there is also a need for further development of methods that utilize the increased computational power to increase the level of detail in modeling of molecular interactions. The work presented in this thesis has contributed to method development in two different areas and demonstrated how virtual screening can be applied to identify ligands of proteins. The first main project investigated modeling of ordered water molecules in protein binding sites to understand the role of water in molecular recognition and the impact of ligands on hydration networks. In a second project, an approach for discovery and optimization of fragment-sized ligands based on virtual screening was developed and used to identify inhibitors of a cancer drug target. Finally, molecular docking was applied to assess proposed substrates of cytochrome b₅₆₁, a recently crystallized membrane protein.