Assessing the therapeutic potential of glycosaminoglycans in amyloidoses

Although neurodegenerative disorders manifest themselves differently, and target different tissues of the body, there are several common features of each disease, which could represent targets for therapeutic intervention for several diseases. Neurodegenerative disorders are characterised by the aberrant aggregation and subsequent deposition of misfolded proteins into amyloid deposits in their respective target tissues. The proteins associated with each disease adopt a cross-β-sheet conformation, which increases their propensity to aggregate into amyloid fibrils, with the β-strands aligning themselves perpendicular to the axis of the fibril. Another common feature of neurodegenerative diseases is the association of glycosaminoglycans (GAGs) and proteoglycans (PGs) with amyloid deposits. GAGs are long linear polysaccharides, which can be covalently attached to proteinaceous cores to form PGs. PGs have been shown to be associated with deposits of several different diseases, implicating GAGs and PGs in the pathogenesis of these diseases. The central hypothesis that GAGs and PGs may be involved in the pathogenesis of several neurodegenerative diseases raises the possibility that targeting GAGs and PGs could be a therapeutic strategy for treating several diseases. The studies presented here aimed to investigate the role GAGs and PGs play in the pathogenesis of two neurodegenerative diseases, Alzheimer’s disease (AD) and familial amyloidotic polyneuropathy (FAP). AD is characterised by the deposition of a small amyloidogenic protein known as the β-amyloid protein (Aβ). Aβ is produced by sequential proteolytic cleavage of its precursor, the β-amyloid precursor protein (APP). The aspartyl protease β-secretase (BACE1) first cleaves APP to produce the secreted N-terminal fragment sAPPβ, and the C-terminal fragment C99. Following β-secretase cleavage of APP, the γ-secretase complex then cleaves C99 to produce Aβ and the APP intracellular domain (AICD). Alternatively, processing of APP by α-secretase results in cleavage of APP within the Aβ region to produce sAPPα and C83. C83 can then be processed through γ-secretase to yield the fragments p3 and AICD. This pathway thus precludes formation of Aβ. GAGs, in particular heparin, have previously been shown to increase BACE1 activity in vitro, and this stimulation of activity was found to be critically dependent on the presence of the prodomain of BACE1. Therefore, the first aim of these studies was to determine the structural basis of the stimulation of BACE1 activity by GAGs. Using an in vitro assay of BACE1 activity, it was shown that polysaccharide backbone structure contributed to the activation of BACE1 by GAGs, with heparin and heparan sulfate activating BACE1 to the greatest extent compared to other GAGs and polyanions. Negative charge of heparin also contributed to the amount of GAG-stimulated BACE1 activity, as desulfation and decarboxylation of heparin reduced its ability to stimulate BACE1 activity compared to unmodified heparin. Polysaccharide chain length was also important, as the capacity to stimulate BACE1 activity was reduced as the chain length of heparin was reduced. However, small heparin analogues inhibited stimulation of BACE1 by full length heparin, and the degree of sulfation of the heparin analogue was important for inhibition of BACE1 stimulation. Analysis of BACE1 enzyme kinetics showed that the kinetics were non-linear. Furthermore, a theoretical structure of proBACE1 is presented, and a mechanism of proBACE1 activation proposed, whereby conformational changes in a putative inhibitory loop containing residues E46 – D65 of proBACE1 following heparin binding facilitate the stimulation of enzyme activity. There is evidence that self-association of full length BACE1 can regulate enzyme activity in cell culture. Therefore, an aim of this study was to examine self-association of the catalytic subunit of BACE1 in vitro. Using dynamic light scattering, native PAGE and size-exclusion chromatography, self-association of the catalytic subunit of BACE1 was demonstrated, with a small, but measurable percentage of BACE1 detectable as higher molecular weight oligomers. Furthermore, heparin was found to increase self-association of BACE1, and this study provides evidence that promotion of self-association of BACE1 may be a mechanism through which heparin stimulates BACE1 activity. There is evidence from cell culture experiments that GAGs and PGs may play a role in APP processing, however the precise effects of GAGs and PGs on APP processing remain unclear. Therefore, an aim of this study is to examine the role of GAGs and PGs in APP processing in primary cortical neurons (PCNs) isolated from APP transgenic mice (Tg2576). The results obtained from these cell culture experiments suggest complicated effects of GAGs and PGs on multiple processes involved in regulation of APP metabolism. Treatment of Tg2576 PCNs with heparin resulted in a decrease in the secretion of both sAPPα and Aβ, while both C83 and C99 levels were increased, with no change in total APP levels. Blocking endogenous proteoglycan assembly using xyloside resulted in a marked decrease in total APP levels and sAPPα secretion, and an increase in both Aβ secretion and C83 levels. These results suggest GAGs and PGs are involved in several processes, including secretion of APP fragments, regulation of γ-secretase and effects on APP transcription. While this provides potentially important insight into the role of GAGs and PGs in AD, it may be difficult to assess the therapeutic potential of targeting the effect of GAGs or PGs on APP processing. FAP is an inherited disorder that can be caused by the deposition of mutant transthyretin (TTR) in peripheral nerves. This study investigated the effect of GAGs on the aggregation of an amyloidogenic TTR mutant implicated in FAP, L55P TTR, using dynamic light scattering. While heparin and chondroitin sulfates A and B markedly stimulated aggregation of L55P TTR, chondroitin sulfate C inhibited aggregation. Chain length of heparin was also found to have a significant effect on the ability of heparin to stimulate L55P TTR aggregation, as aggregation of L55P TTR in the presence of low molecular weight heparin derivatives was significantly slower than L55P TTR in the presence of full-length heparin. Analysis of the data using discrete exponentials suggests that GAGs may influence the formation of low molecular weight aggregates of L55P TTR, and that aggregation of L55P TTR is nucleation-dependent. The data provided suggest that chondroitin sulfate C, or analogues of chondroitin sulfate C, may be potential therapeutic compounds for FAP. Based on the data presented in these studies, a model of the role of GAGs and PGs in the pathogenesis of AD, and possibly other amyloidoses, is proposed, whereby GAGs and PGs regulate multiple facets of disease progression, notably transcription, processing and secretion of APP and APP fragments, as well as aggregation of amyloidogenic proteins. This work supports the view that GAG-based compounds may be useful for the treatment of neurodegenerative diseases.