Degree Name

Doctor of Philosophy


School of Chemistry


Electrospray ionisation mass spectrometry (ESI-MS) was used to investigate noncovalent interactions of E. coli DNA polymerase III (Pol III) protein subunits with metal ions, ligands and other Pol III proteins. Nanospray (nano) ESI-MS was used to examine the interaction between α(polymerase) and a recombinant version of the 16 kDa C-terminal domain of τ (τC16, a clamp loader protein) which retains the α binding site. Firstly, τC16 and various τC16 mutants and truncations were titrated into samples containing a constant amount of α. Based on the nanoESI mass spectra, the relative order of binding affinity towards α was: τC16 > τC16[I618T] ≈ τC16[L635P]>τC16[D636G]≈τC16[F631I]>τC16[S617P]>τ C16[L627P]>τC16△7>τC16△11≈τC14. This is in agreement with the binding order determined in previous work by surface plasmon resonance (SPR; BiacoreTM). The differences in binding affinities were not evident when the complexes were subjected to collision-induced dissociation (CID). The peaks in nanoESI mass spectra from ions corresponding to α and the α-τC16 complex were broad and it was not possible to determine values for dissociation constants (KD) by comparing the relative abundances of ions from α and the α-τC16 complex. One explanation for this is that the response factor for the complex was greater than that for α alone and suggests a conformational change on binding. Hydrogen/deuterium exchange (HDX) experiments were challenging because α was not stable at the low pH required to quench exchange of amide protons. Consequently, exchange of all protons was measured by ‘direct’ HDX. The results of two sets of experiments suggested that a greater number of protons exchanged in the complex than for the sum of that observed for the individual binding partners, consistent with a conformational change of α on binding. This is likely to occur during the normal functioning of the replisome as α participates in many protein-protein complexes that involve interactions with the clamp loader and hand off to the sliding clamp which tethers α as part of the polymerase core to its template DNA.

ESI-MS can detect binding between metal ions and proteins. In some cases, complexes that may not be representative of complexes that are relevant in vivo may be detected since positively charged metal ions may bind non-specifically to acidic groups on the protein. The binding of metal ions to the N-terminal domain of the Pol III exonuclease subunit, ε186, and the effect of the inhibitor, thymidine-5'-monophosphate (TMP), and its binding partner, θ were investigated. The affinity of the metals for ε186 decreased in the order: Fe2+ > Dy3+ > Mn2+ ≈ Zn2+> Mg2+ ≈ Cu2+. In the absence of TMP, the stoichiometry of binding for the metal ion that supports the greatest enzymatic activity, Mn2+, and also for Zn2+ which bound with similar affinity as judged by the ESI mass spectra, was not clear. In contrast, it was clear that two Fe2+ ions and one Dy3+ ion bound. When TMP was added to samples containing ε186 in the presence of metal ions, enzyme-metal complexes in which two Mn2+, Zn2+ or Fe2+ ions were bound in addition to TMP were observed. The observation of a favoured metal ion-binding stoichiometry supports the hypothesis that a biologically relevant active site had been reconstituted; the binding of TMP to ε186 facilitated the binding of two metal ions, which is necessary for the activity of :epsilon;. The above experiments were also performed using the ε186-θ complex. The presence of θ (no TMP) did not potentiate the formation of complexes containing two metal ions and had only a small effect on the hydrolytic activity of the metalloenzyme.

ESI ion mobility mass spectrometry (IM-MS) was used to collect information on the collision cross sections (CCS) of a range of proteins. In IM-MS, the ‘drift time’ (or arrival time distribution, ATD) of an analyte depends not only on its mass and charge but also on its shape, which may be reflected in the CCS. For this, the travelling wave IM-MS instrument known as the Synapt HDMSTM was used. This instrument has only recently become commercially available. Since drift times obtained using this instrument are not directly related to CCS, drift times of proteins of known CCS must be determined under the same conditions as those used to determine the drift time of the protein of unknown CCS. As this is a relatively new technique, the CCS values of Pol III proteins from the clamp loader subassembly were compared with those determined by our collaborators at the University of Cambridge (group of Professor C. V. Robinson). The CCS values determined for δ, δ', γ1, γ2, γ3, γ4, γ2-δ', γ3-δ' and γ3-δ-δ'were 2805, 2694, 2994, 4958, 6443, 8060, 6294, 8071 and 9197 Å2, respectively. These values were in agreement to within 10% of those determined at Cambridge, showing that the method is sufficiently robust to enable comparisons across independent laboratories of CCS of proteins/complexes that are challenging to prepare. In addition, CCS values were determined for some commercially available proteins, α ± τC16 and a very large protein complex, rubisco (ribulose-1,5-bisphosphate carboxylase oxygenase). CCS values of cytochrome c, myoglobin and lysozyme determined here were also within 10% of the literature values, further validating this method. The CCS values of α, τC16, α-τC16 and rubisco were consistent with their molecular weights. Finally, the two literature CCS calibration methods were applied to all the proteins described in this chapter. The values determined by the two methods were in good agreement.