The work described in this Thesis has focused on the application of nuclear magnetic resonance (NMR) spectroscopy to study the Fe(II) and 2-oxoglutarate (2OG) dependent oxygenases, a family of structurally related enzymes that are ubiquitous in plants, micro-organisms and animals. The work included enzyme mechanistic studies, NMR method development, protein NMR and method development for inhibitor discovery. NMR was applied to study human γ-butyrobetaine hydroxylase (BBOX), an enzyme involved in L-carnitine biosynthesis. The studies showed that BBOX exhibits remarkable substrate promiscuity and is able to accept a large number of alternative substrates including the catalytic product L-carnitine and its unnatural enantiomer D-carnitine. Inhibition and mechanistic studies have also been conducted with 3-(2,2,2-trimethylhydrazineyl)propionate (THP), a BBOX inhibitor that is clinically used for the treatments of myocardial infarction and angina. The studies showed that THP is actually a competitive substrate forming a number of products including 3-amino-4-(methylamino)butanoic acid that is observed as a formaldehyde adduct. NMR analyses using 13C-labelled THP revealed that the reaction of THP involves a radical mechanism involving C-N fragmentation and C-C bond formation in a fashion similar to a Stevens rearrangement in organic chemistry. The development and implementation of two competition-based NMR methodologies to study ligand binding to 2OG oxygenases are described. In the first method, 2OG was used as a reporter ligand and its signal was monitored by spin echo-edited 1H experiments upon displacement by competitive ligand. An alternative version involving the use of 13C-labelled 2OG as the reporter and heteronuclear spectra editing is also proposed. This displacement method was applied for ligand screening and for quantifying ligand binding affinities for several 2OG oxygenases including the hypoxia inducible factor (HIF) hydroxylases. It was also used to investigate the catalytic mechanism of deacetoxycephalosporin C synthase (DAOCS), which revealed that 2OG and penicillin (substrate) have distinct binding sites, arguing against the atypical mechanism previously proposed for this enzyme. In the second NMR method, water molecules were used as the reporter ligand. The basis of the method relies on the paramagnetic relaxation enhancement (PRE) effect. By replacing the active site metal of 2OG oxygenases by paramagnetic metal Mn(II), the relaxation rates of bulk water were increased. In the presence of a ligand, the accessibility of water molecules to the paramagnetic metal centre was hindered, leading to a decrease in the bulk water relaxation rates. This method was applied for ligand screening and for ligand binding constant measurements for HIF prolyl hydroxylase (PHD2). Protein NMR was also applied to study substrate binding to PHD2. Isotopically labelled PHD2 was produced and protein resonance assignment was conducted for both the substrate-free and substrate-bound forms of PHD2. Comparison of the two sets of assignment provided solution state evidence for the substrate binding site that was previously identified by X-ray crystallography. The chemical shift perturbation data also suggested that a loop region at the N-terminal side of the core double-stranded β-helix fold of PHD2 is important for substrate binding. Unlike 2OG, the binding of the peptidyl substrate and the prolyl hydroxylated peptidyl product to PHD2 do not require the presence of an active site metal. In fact, the presence of 2OG inside the active site actually weakens the binding of the product whilst the binding of the substrate was not affected, which is likely due to steric clash between 2OG and the hydroxyproline residue. The applicability of reversible boronate ester formation to protein-directed dynamic combinatorial chemistry for the discovery of protein ligands was also investigated. Using 11B NMR, proof-of-principle studies demonstrated selective binding of boronate esters from a dynamic combinatorial library (DCL). The studies also identified factors that should be considered when designing a DCL involving boronate ester formation. The method was then applied to discover inhibitors for PHD2, and NMR was used as a verification and quantification technique. The studies have successfully led to the discovery of three novel potent PHD2 inhibitors that inhibit the enzyme at low nanomolar concentrations.