Characterization of novel compounds discovered in neurodegeneration models

Abstract

The prevalence of neurodegenerative diseases is increasing and treatments so far are limited to alleviating symptoms. This creates a severe socioeconomic burden and a need for a better understanding of disease initiation and progression, which could lead to the development of curative or disease-modifying treatments. In vitro neurodegeneration models can offer the opportunity to study disease mechanisms and pharmacological intervention. When working with in vitro models that employ tool compounds to elicit pathological phenotypes, there is always the danger of confusing an experimentally generated artefact for a real effect. Glutathione and its precursors are reportedly capable of protecting neurons from MG-132-induced proteasome inhibition. However, since MG-132 is a peptide-aldehyde known to form hemithioacetals with free thiols, manuscript 1 aimed to determine whether this protection resulted from the thiols’ antioxidative properties (physiological effect) or a chemical inactivation of the inhibitor (artefact). The chemical reaction of MG-132 with glutathione, cysteine and N-acetyl-cysteine was confirmed to occur at very high thiol concentrations. It was irreversible in the case of cysteine, which led to a significant prevention of proteasome inhibition in a cell-free setup. However, transcriptome analysis showed a persistent proteasome inhibition unaffected by thiol co-incubation, which proves that a chemical inactivation does not occur within cells, probably because the necessary thiol concentrations cannot be reached even after supplementation. Instead, downstream damage pathways were attenuated by the thiols’ protective antioxidative properties. Taken together, these results confirm the physiological relevance of thiol-mediated protection of neurons from proteotoxic stress. On a more general level, caution should be exercised when co- or pre-incubating any kind of pharmacological modulator containing an aldehyde moiety with high concentrations of cysteine to prevent artefact generation. Iron chelation by deferiprone is currently being explored as a disease-modifying treatment for Parkinson’s disease. Since clinical trials yielded mixed results, the second manuscript of this thesis aimed to generate improved iron chelators based on deferiprone. As low blood-brain-barrier passage was hypothesized to be responsible for the variable effect of deferiprone in clinical trials, structural elements allowing transport to the brain via the neutral amino acid transporter LAT1 were introduced to the iron chelator. Five candidate molecules were synthesized and screened, revealing SK4 as the most promising iron chelator. Selective iron chelation within the physiological pH range and its uptake by LAT1 were confirmed. Cell death and neurite loss induced by parkinsonian neurotoxicants in LUHMES neurons could be prevented by 10-20 μM of the new iron chelator. Microbial metabolites and their delivery to the brain via the gut-brain-axis are hypothesized to contribute to the increasing prevalence of Parkinson’s disease. The third manuscript of this thesis follows up on the previous discovery of an unidentified metabolite of Streptomyces venezuelae, which caused specific dopaminergic neurodegeneration in Caenorhabditis elegans, and aimed to identify and characterize said metabolite. A screening of sequentially purified bacterial extracts on LUHMES neurons revealed aerugine and aeruginol as active cytotoxic compounds. The further characterization of re-synthesized aerugine (C10H11NO2S; 2-[4-(hydroxymethyl)-4,5-dihydro-1,3-thiazol-2-yl]phenol) revealed its striking cell-type specificity in vitro and in vivo. Half maximal dopaminergic neurotoxicity was triggered by 3-4 μM in vitro, other neurons were less sensitive (10-20 μM) and common human cell lines were unaffected at concentrations of up to 100 μM. In transgenic Caenorhabditis elegans, dopaminergic neurons were also degenerated specifically without compromising general nematode viability or the morphology of γ-aminobutyric acid-neurons. This further led to a functional toxicity as observed by a reduced basal slowing response. Mechanistically, aerugine-induced cell death was shown to be highly dependent on iron and prevented by distinct antioxidants or a caspase inhibitor. In conclusion, this thesis presents several new, neurologically relevant compounds and provides their basic characterization utilizing multiple neurodegeneration models. Aldehyde-thiol reaction products are possible sources for artefacts in in vitro models. SK4 is a novel iron-chelator combining iron selectivity with blood-brain-barrier passage via LAT1. Aerugine is a potential new tool compound for Parkinson’s disease models and could have implications in human pathology.