The role of Hsp20 in Alzheimer's disease

Cameron, Ryan T. (2014) The role of Hsp20 in Alzheimer's disease. PhD thesis, University of Glasgow.

Abstract

Alzheimer’s disease is the most common of the degenerative brain diseases and is characterised by impairment of cognitive function. Patients with this disorder lose the ability to encode new memories. Eventually, both declarative and non-declarative memory is significantly impaired, resulting in the capacity for reasoning, abstraction and language becoming progressively reduced. Alzheimer’s disease and other dementias have devastating effects on families and caregivers, and is an increasing burden in an ageing society. It is estimated that 36 million people worldwide are living with dementia and this figure is expected to double every 20 years. The worldwide costs of dementia in 2010 were estimated to be $604 billion, an exorbitant figure that represents 1% of global GDP (World Alzheimer Report 2011).
Alzheimer’s disease is the fourth leading cause of death in industrialised nations, preceded by cardiovascular disease, cancer and stroke. As yet there are currently no disease-modifying drugs approved to treat Alzheimer’s disease. The therapeutics that are available only temporarily alleviate symptoms of cognitive impairment, however, they do not halt the inevitable progression of the disease. As such, major scientific efforts are underway in order to develop drugs which can help stabilise the disease. The publication of the “Amyloid Hypothesis” by Dennis Selkoe in 1991 helped to focus research efforts towards a causative protein involved in the disease, the amyloid β protein (Aβ).
Aggregation and deposition of the Aβ protein is fundamental in the aetiology of Alzheimer’s disease and its importance has been demonstrated by a number familial heterogeneous mutations in the amyloid precursor protein that promote increased Aβ deposition, resulting in early onset phenotypes. There are several other aspects involved in disease progression such as neuroinflammation and aberrant neuronal signalling, however, therapies targeting amyloid β aggregation have the potential to slow or even halt further neurodegeneration and anti-Aβ therapies are regarded as a logical approach to treating Alzheimer’s disease.
Several endogenous pathways exist to prevent protein misfolding and subsequent aggregation following stressful cellular conditions. One pathway includes the amateur chaperones of the small heat-shock protein family, which have recently garnered interest due to their ability to inhibit the aggregation of amyloid-like proteins. In particular, Hsp20 was previously identified as having the ability to inhibit the aggregation of Aβ and could
Abstract
ii
attenuate subsequent toxicity associated with Aβ peptides. Hsp20 was of particular interest to the Baillie group as it has well established cardio-protective functions, which are triggered by the phosphorylation of a serine residue (S16) at a consensus protein kinase A/G site. Hsp20 “activity” can therefore be readily modulated via inhibition of second messenger signal degradation by phosphodiesterases.
The first part of this thesis investigated the interaction between Hsp20 and Aβ using Peptide Array technology. This technique allowed rapid characterisation of interacting domains and pinpointed key residues that mediated the protein-protein interaction. Using this approach, I demonstrated that the domain within Hsp20 that interacted with Aβ included the consensus PKA phosphorylation site (R-R-X-S). Upon introduction of a phospho-serine residue or a phospho-mimetic substitution, I was able to show that the binding of Aβ was enhanced. Reciprocal peptide array experiments highlighted that Hsp20 bound to a domain within Aβ, which is key to the aggregation of the Aβ peptide and is required to produce the higher order toxic Aβ species. The Peptide Array data was then verified using full-length recombinant proteins and several Hsp20 mutants were developed including a phospho-mimetic. The phospho-mimetic Hsp20 was shown to outperform the wild-type variant in several assays such as, in vitro pull-down assays, Aβ aggregation measured using nuclear magnetic resonance spectroscopy, and also a novel Aβ aggregation assay which can differentiate between two distinct aggregation pathways, namely fibrillisation and oligomerisation. These data demonstrated for the first time how the interaction between Hsp20 and Aβ may be modulated by cell signalling cascades.
I then moved to investigate the cytotoxicity of Aβ in order to investigate whether increasing Hsp20 expression in neuronal-like cells would confer protection against Aβ-mediated toxicity. This was initially carried out using a standard MTT-based cell viability assay, before utilising a real-time cell monitoring device to develop a novel Aβ toxicity assay. In both assays, increasing Hsp20 expression was shown to be cytoprotective. The wild-type variant of Hsp20 was found to be more effective in cell-based assays due to increased levels of phosphorylated Hsp20. The real-time Aβ toxicity monitoring assay also gave me a platform for testing agents with potential neuroprotective properties.
Given that increasing levels of phosphorylated Hsp20 could attenuate Aβ-mediated cytotoxicity, I logically was drawn to study ways that this event could be targeted therapeutically. Several drugs that target cAMP- and cGMP-dependent phosphodiesterases have been shown to be effective in alleviating symptoms of Alzheimer’s disease in rodent
iii
models have also been studied here in cellular systems. These included the “blockbuster” PDE5 inhibitor Viagra® (sildenafil), two novel compounds which selectively inhibit PDE9, which were developed by the pharmaceutical company Lundbeck specifically as Alzheimer’s treatments, and rolipram, a well established cognitive enhancer that was developed originally as an anti-depressant. All of these compounds were shown to “activate” endogenous Hsp20 to varying degrees in neuronal-like cells and the levels of Hsp20 activation was found to correlate with both the level of induced Hsp20/Aβ co-localisation, and subsequent attenuation of Aβ-mediated cytotoxicity. This suggests that this endogenous protection pathway can be targeted by currently available therapeutics in order to reduce the neurotoxic effects of Aβ.
Finally, we wanted to develop novel agents of our own that could promote Hsp20 phosphorylation. To do this, in silico docking of all FDA approved drugs against the catalytic domain of PDE4 was undertaken in an attempt to find a novel compound with the potential to reposition as an Alzheimer’s treatment. Using this methodology we discovered an angiotensin converting enzyme inhibitor, moexipril to be a PDE4 inhibitor in the low micro molar range. Unfortunately, moexipril works as an ACE inhibitor in the low nano molar range making repositioning unviable. However, moexipril treatment was more effective than rolipram in reversing Aβ toxicity and I speculate that this may be due to the sub-family selective nature of its (moexipril) PDE4 inhibition. Furthermore, the lack of emetic side effects associated with moexipril makes this compound an ideal starting point for the development of isoform selective and/or non-emetic PDE4 inhibitor.
In summary, these studies describe a novel endogenous mechanism for combating the toxic effects of the Aβ protein, which underpins the development and progression of Alzheimer’s disease. Given that the interaction between Hsp20 and Aβ can be manipulated via cAMP/cGMP signalling, the interaction could be targeted therapeutically. As there are currently no effective drugs on the market for stabilising Alzheimer’s disease, I believe that the data presented here opens up a potential new avenue that could lead to the development of a new class of AD drugs.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: Alzheimer's Disease; Heatshock proteins; Hsp20 chaperone; Aβ oligomerisation; Hsp20; Peptide array; Aβ cytotoxicity, second messenger signalling, phosphodiesterases, cognition
• Subjects: Q Science > QH Natural history > QH301 Biology; Q Science > QH Natural history > QH345 Biochemistry; R Medicine > RC Internal medicine > RC0321 Neuroscience. Biological psychiatry. Neuropsychiatry; R Medicine > RM Therapeutics. Pharmacology
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Cardiovascular & Metabolic Health
• Supervisor's Name: Baillie, Prof George S.
• Date of Award: 2014
• Depositing User: Dr Ryan T Cameron
• Unique ID: glathesis:2014-5128
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 10 Jul 2014 13:30
• Last Modified: 27 Jun 2017 12:39
• URI: https://theses.gla.ac.uk/id/eprint/5128