Study of the Effects of Acylation within the Extracellular Domains of BlaR1

Bacterial antibiotic resistance mechanisms are an emerging problem in the treatment of infectious disease. The overuse of the ̢-lactam class of antibiotics within the last seventy years has caused a rising number of cases where the most pathogenic strain, Methicillin Resistant Staphylococcus Aureus (MRSA), is in the homes and schools of the general population. To defend against the potential disastrous consequences efforts are needed to understand the exact mechanism of drug resistance in strains like MRSA and develop novel treatment strategies.

The drug resistance in MRSA strains is largely due to the BlaR1 protein. BlaR1 is a transmembrane signaling protein that initiates a signal stimulus upon binding and recognition of the ̢-lactam antibiotic by the extracellular sensor domain. This transmitted signal moves across the cell membrane to the intracellular domains. The resulting response is the promotion of BlaZ expression, which is a ̢-lactamase capable of neutralizing the available ̢-lactam antibiotic. The exact mechanism of signal initiation and signal relay from the extracellular to intracellular domains is currently unknown. The work of this thesis aims to understand the effects of antibiotic addition on the sensor domain of BlaR1 (BlaRS) and the modulated interaction with the other extracellular domain (Loop-2).

The BlaRS sensor domain has high structural conservation with other penicillin binding proteins (PBPs) and the Class A and D Ì¢-lactamases, although the proteins have low sequence homology. Previous work has initially suggested that BlaRS utilizes an active site NÌ_-carboxylated lysine (K392) and an interaction with the extracellular Loop-2 domain within the mechanism of signal transduction. The work within this thesis uses Nuclear Magnetic Resonance (NMR) as a non-invasive technique to probe the formation of the covalent antibiotic attachment within BlaRS and the resulting effect to the NÌ_-carboxylated K392 side chain and interaction with the Loop-2 domain peptide mimic (L2short). This work acts to fill in gaps of the previous research and to add a new understanding of the effects of antibiotic binding using the atomic resolution of NMR.

The work presented within this thesis observes the BlaRS protein in its apo form, complexed with L2short, or acylated states to determine the mechanism of signal transduction within the protein. We observe that the NÌ_-carboxylated K392 side chain undergoes fast motions within the active site, and simultaneously the protein is found to have slow timescale (ÌÂs-ms) motions. Upon acylation the lysine undergoes NÌ_-decarboxylation, allowing the Ì¢-lactam antibiotic to remain bound. Throughout the transition from apo to acylated, BlaRS the protein appears to rearrange the physical location of slow motion dynamics. The formation of a hydrogen bond network to the antibiotic or the lack of fast motions in the decarboxylated lysine side chain may be the cause of these changes. The slow motions move within the active site to a region of the protein outside of the active site. This dynamic rearrangement is likely the method of signal initiation in the protein. The synthesized peptide representing the C-terminus of the Loop-2 domain (L2short) circumvented solubility issues found in the full-length peptide. The L2short peptide contains a single γ-helix, flanked by two unstructured termini. The comparison of the 15N backbone dynamics of the L2short peptide, in the presence and absence of the BlaRS, revealed the BlaRS binding epitope through the increased slow timescale dynamics on the peptide. Similarly, the binding epitope on BlaRS was identified using chemical shift perturbations (CSPs) along with the saturation transfer difference (STD) experiment. Together this information served as input for the HADDOCK software in order to determine the likely site of interaction and orientation of the peptide within the binding epitope. Finally, the addition of antibiotic to the BlaRS/L2short complex revealed, contrary to previously published work, that the interaction with the C-terminus of the Loop-2 domain is unbroken by acylation of the sensor domain. This suggests a deeper complexity to the role of the Loop-2 domain to signal transduction within BlaR1 than previously understood.