Search

There are numerous complementary approaches to facilitate the identification of novel inhibitors for biological targets, including high throughput screening and fragment-based drug discovery. Computational tools are often employed to predict binding pose and affinity of the new inhibitors. In this thesis an integrated computational and experimental approach to identify novel inhibitors is described. The approach involves the design of a virtual library of likely synthetically accessible lead-like molecules, followed by virtual high throughput screening (vHTS) against target protein. To exemplify the approach, BACE-1 was selected as an example target protein. BACE-1 is responsible for the formation of amyloidal plaque in brains affected by Alzheimer’s disease and therefore is a potential target for the treatment of the disease. A virtual library of lead-like molecules was generated based on diversity-oriented synthesis methods established in our laboratory. The library underwent virtual high throughput screening (vHTS) against BACE-1 by using eHiTS and two families of putative inhibitors were identified with high predicted ligand efficiency (cLE). The in silico approach employed to identify novel putative BACE-1 inhibitors is schematically represented as follows. [Unable to display image from abstract] A focused library based on the selected putative inhibitors was designed and synthesised, and biological activity was assessed via a fluorimetric assay. Structure-activity relationship (SAR) studies were conducted to rationalise the activity of the inhibitors and to confirm the validity of the integrated approach in identifying new inhibitors for biological targets. A novel series of BACE-1 inhibitors was identified and is herein described.

Alzheimer’s disease (AD) is a progressive form of dementia, which currently has no cure. The metabolism of the amyloid precursor protein (APP) is an important determinant in the development of AD. APP is proteolytically processed either in the amyloidogenic pathway, generating the AD-causing amyloid-beta (Aβ) peptide, or in the non-amyloidogenic pathway which prevents the generation of Aβ. The role of four different proteins, BIN1, contactin 5, neurofascin and Thy-1, in the regulation of APP proteolytic processing was investigated. Contactin 5, neurofascin and Thy-1 have been shown to interact with APP, whereas BIN1 has been genetically related to AD by genome-wide association studies. Over-expression of each protein in cell lines showed that BIN1 and contactin 5, but not neurofascin or Thy-1, regulate the production of Aβ. Contactin 5 over-expression prevented Aβ generation by inhibiting cleavage of APP by the γ-secretase. Western blots of human brain samples showed that contactin 5 levels in the brain are not altered during aging or in AD. BIN1 over-expression decreased APP levels, and decreased Aβ, whereas BIN1 knockdown increased APP, suggesting that BIN1 redirects APP from processing in the amyloidogenic pathway to a compartment of the cell where it is degraded. BIN1 did not alter either Aβ uptake, or the endocytosis or cell surface levels of APP. BIN1 levels were shown to be reduced in the brain during aging and in AD. This has led to the hypothesis that during aging and AD, BIN1 levels decrease, so more APP is trafficked into the amyloidogenic pathway rather than being degraded, resulting in more Aβ generation so increasing the risk of developing AD. This is the first time a mechanism for the role of BIN1 in AD has been suggested.

The function of the integral membrane protein stomatin is as yet poorly understood. Stomatin is deficient from the erythrocyte membrane of patients suffering with Overhydrated Hereditary Stomatocytosis (OHSt). Patient erythrocytes have altered morphology and are known as stomatocytes. It is believed that stomatin is mistrafficked in the developing stomatocyte. These patients suffer grossly abnormal cation fluxes in the stomatocyte membrane which causes increased osmotic fragility of the cell and results in haemolytic anaemia. This study set out to characterise further stomatin and to investigate its role in the cell. The membranes and lipid rafts of stomatocytes were found to have reduced actin levels as compared to erythrocytes, suggesting that stomatin may function as a structural protein linking the cytoskeleton to the membrane. Overexpression of stomatin in nucleated cells caused enhanced actin association with cell membranes and lipid rafts, further confirming the findings from stomatocytes. Calcium-induced vesiculation was found to be significantly enhanced from the stomatocyte as compared to the erythrocyte, with defective partitioning of the flotillin proteins into the vesicles. This suggests that stomatin may function as a negative regulator in this vesiculation, possibly due to its interaction with actin and that the flotillins may substitute for stomatin in this process within the stomatocyte. Mutating the principle cysteine residue for palmitoylation within stomatin caused the protein to show less affinity for the membrane and lipid rafts but an increased affinity for the nucleus. This suggests that palmitoylation of stomatin affects the affinity of stomatin for the membrane and that this modification may be involved in regulating the shuttling of stomatin between the plasma membrane and the nucleus. Prokaryotic stomatin exists in an operon with a serine protease, suggesting a functional link between the two. Using a reporter gene construct approach the potential for mammalian stomatin to be proteolytically processed was investigated. Stomatin was found to be proteolytically processed in the membrane by a serine protease with the subsequent release of a C-terminal fragment.