Sara Diels

• 18 December 2020, 4pm - 6pm
• Online defence
• Promoters: Wim Van Hul, Wim Vanden Berghe

Abstract

Obesity is a highly prevalent condition in which excess body fat has accumulated as a result of the chronic imbalance in energy homeostasis. The disorder is characterized by a heterogeneous phenotype that represents itself in many different isolated forms as well as with varying cardiovascular and metabolic manifestations. Ectopic fat accumulation in the liver as a consequence of weight gain is present in up to 90% of patients and is referred to as non-alcoholic fatty liver disease. The high incidence of fatty liver disease in patients with obesity indicates an interconnection between both conditions. Research into their pathogenesis denoted a multifactorial aetiology with eating behaviour, lifestyle, environment, and genetics as the main contributing factors. Although successful efforts already identified numerous variants, the disease’s onset and the development of metabolic alterations leading to its comorbidities later in life are still incompletely understood. Further unravelling the genetic landscape of obesity is thus essential as it will provide insights into its pathogenesis. The general aim of this thesis was therefore to investigate the genetic architecture of complex obesity phenotypes by examining the role of structural variation in non-syndromic forms of obesity and the combined effect of (epi)genetics in the development of obesity and associated liver pathology. In a first part, fine-mapping and exploration of the 11q11 region in a traditional case-control study indicated an increased prevalence of a ± 80 kb deletion covering OR4C11, OR4P4, and OR4S2. The functional impact of these three olfactory receptor genes was assessed by expression profiling in metabolic relevant tissues and postulates that gene disruption will negatively influence energy metabolism with fat accumulation and obesity as outcome. In a second part, a targeted multi-omics approach investigating the influence of common polymorphisms and DNA methylation variation on PON1 status and the hepatometabolic phenotype exposed a significant relationship for (i) regulatory polymorphism rs705379:C>T with waist-to-hip ratio and indicative features of liver pathology and (ii) coding polymorphism rs854560:A>T with the expression of circadian clock gene ARNTL. These findings demonstrate the promising use of vertical data-integration methods to gain novel mechanistic insights that can improve our understanding of complex disease pathogenesis.

​Julie Hoogmartens​

• 20 October 2020
  
• Supervisors: Christine Van Broeckhoven and Rita Cacace

Abstract

Alzheimer’s disease is a complex neurodegenerative disorder presenting at early (early-onset (EO) AD, ≤ 65 years) and late ages (late-onset (LO) AD, > 65 years) in familial and sporadic patients. Up to 60% of EOAD patients have a positive family history of dementia and 10-15% show clear autosomal dominant transmission. Linkage studies in large informative EOAD families with autosomal dominant inheritance have led to the identification of three causal genes - amyloid precursor protein (APP), presenilin 1 (PSEN1) and presenilin 2 (PSEN2) - and a major risk factor - apolipoprotein E (APOE) e4 - and greatly improved our understanding of the pathomechanisms underlying AD. Causal mutations in APP, PSEN1 and PSEN2, however, explain only 5-10% of EOAD patients, leaving the majority of the patients genetically unexplained. Identification of novel genes will be key in further unravelling the complex molecular basis and pathological processes underlying AD etiology.

The general aim of this PhD project was to contribute to a better insight into the missing genetic etiology of AD using next generation sequencing (NGS). We used two different approaches, on the one hand, we performed a gene-discovery study using whole-genome sequencing (WGS) data of unrelated AD patients and on the other hand, we followed a hypothesis driven approach based on prior data available in literature and performed a gene-based re-sequencing study.

With the first approach, we identified a homozygous missense mutation in the Von Willebrand Factor A Domain Containing 2 (VWA2) gene in WGS data of 17 unrelated EOAD patients from Flanders-Belgium. Mutation screening in extended Flanders-Belgian and European replication cohorts identified additional homozygous and compound heterozygous missense mutations in VWA2 mimicking autosomal recessive inheritance in sporadic AD patients.

With the second approach, we studied the genetic contribution of the matrix metalloproteinases (MMPs) to the etiology of AD. MMP13, previously functionally associated with AD in cultured neuronal cells and AD transgenic mice through the translational regulation of β-site APP cleaving enzyme 1 (BACE1) via the phosphoinositide 3-kinases (PI3K) signaling pathway, was selected as most promising AD candidate gene. In our study, we identified a premature stop codon (PTC) mutation in one control individual and ultra-rare missense mutations in AD patients only.

With this PhD work, we identified VWA2 and MMP13 as novel AD candidate genes. By sharing these results, we aim to trigger genetic replication and functional follow-up studies needed to unravel the exact role of the identified genes in the etiology of AD and to assess the effect of the identified mutations on disease relevant processes. This might ultimately provide the opportunity to identify novel therapeutic targets for drug development.

​Stefanie Smolders​

• 26 June 2020
  
• Supervisor: Christine Van Broeckhoven

Abstract

Lewy body diseases (LBD) are a heterogeneous group of neurodegenerative brain diseases characterized by the presence of Lewy bodies and Lewy neurites, mainly composed of aggregated α-synuclein, in neurons. Two disorders with substantial clinical, pathological and etiological overlap are at the two-side extremes of the LBD continuum: dementia with Lewy bodies (DLB) and Parkinson’s disease (PD). Although inherited LBD is rare with an estimated frequency of 5–15%, the identification of causal and risk genes have been instrumental in dissecting disease etiologies. A large part of the genetic etiology of LBD remains unexplained and the identification of novel genes will be instrumental in further unravelling the molecular basis and disease processes underlying LBD.

This PhD project aimed to gain more insights in the molecular basis of LBDs using next generation sequencing based gene identification strategies combined with molecular cell biology approaches. Using whole exome sequencing in unrelated early-onset PD patients, we identified compound heterozygous mutations in ATP10B increasing risk for PD and functionally characterized ATP10B as a late endo-/lysosomal glucosylceramide and phosphatidylcholine lipid flippase, involved in lysosomal functionality and neuroprotection. Both ATP10B and the well-known LBD risk factor GBA play essential roles in the fate of lysosomal glucosylceramide, and dysfunction of both results in intra-lysosomal accumulation of glucosylceramide and lysosomal dysfunction. Additionally, using whole genome sequencing in two siblings affected with autosomal recessive early-onset DLB, we identified compound heterozygous mutations in VPS13C, which encodes a late endo-/lysosomal protein involved in glycerolipid transport. Our genetic and expression data suggest that in addition to previously identified premature termination codon mutations identified in PD patients also missense mutations in VPS13C can contribute to the risk of LBD by loss-of-function. Moreover, genetic profiling of the PD cohort for seven major genes associated with PD, including ATP10B and VPS13C, identified oligogenic variants in 4.76% of patients, suggesting more complex inheritance patterns of PD with oligogenic mutations of interconnected PD pathways. Whole exome sequencing of 172 PD patients implicated a possible role for lysosomal storage disorder gene variants in PD pathogenesis.

Our results support the importance of lipid homeostasis alterations and lysosomal dysfunction in LBD pathogenesis, thereby opening new avenues for therapeutic intervention. The observation of more complex inheritance patterns and oligogenic mutations in patients highlights the crosstalk between interconnected PD pathways. In diagnostics, the presence of a monogenic pathogenic mutation should not serve as exclusion criteria to screen for other genetic variation.

​Vera Eva Alexandra Kühne​

• 11 June 2020
  
• Supervisors: Philippe Büscher and Xaveer Van Ostade

Abstract

Visceral Leishmaniasis (VL) is a fatal disease if it is not correctly diagnosed and treated. It is caused by unicellular parasites of the Leishmania donovani complex.

The endemic areas for VL are mostly low- and middle- income countries. However, the existing diagnostic tests for VL are either not adapted for use in resource poor settings or their diagnostic performance in East Africa is variable.

This thesis aims to contribute to the development of an alternative diagnostic test for VL in East Africa based on antibody detection. The goal is to identify novel antigens, that can be incorporated into an immunochromatographic or rapid diagnostic test. These tests are ideal as point-of-care tests for low- and middle income countries as they are fast, non-invasive and need no equipment and minimum training.

We started by an attempt to replace an existing antibody detection test for VL: the direct agglutination test (DAT) (Part 2). The DAT has a high sensitivity and specificity in all endemic regions but it is based on entire parasites, which makes it difficult to standardize and to incorporate into an RDT. We analysed the existing literature on the nature of the DAT Ag (Chapter 4). From the retrieved evidence we concluded that the DAT Ag is composed of a mixture of Leishmania specific antigens.

We first aimed to select so-called mimotopes - peptides that mimic the epitopes of the DAT Ag using a phage display approach (Chapter 5). We found peptides that have a diagnostic potential. Unfortunately, their reactivity with VL positive compared to VL negative sera was not sufficient to be incorporated in an RDT.

We then used a more targeted approach and analysed hypotheses found in Chapter 4 experimentally. We reached the conclusion that lipophosphoglycan is part of the DAT Ag while carbohydrates alone are not (Chapter 6).

In Part 3 of this thesis, we analysed the pipeline of serodiagnostic tests for VL. We performed a systematic literature review (Chapter 7), which showed that most tests are not tested on sufficient specimens, especially from East Africa. Moreover, we did not find one non-native antigen (recombinant or synthetic) with a carbohydrate or lipid moiety. Based on this analysis we chose the most prominent glycoprotein of Leishmania – gp63 – and expressed it in a L. tarentolae expression system (Chapter 8). The recombinant glycoprotein has diagnostic potential and thus serves as a proof-of-concept for the use of glycoproteins in the serodiagnosis of VL.

To develop alternative diagnostic tests for VL in East Africa, we propose to focus on the evaluation of mixtures of existing antigens and to investigate the diagnostic potential of a synthetic LPG core-anchor fragment and of differently glycosylated variants of gp63.

​Samuel Van Remoortel​

• 1 June 2020
  
• Supervisors: Jean-Pierre Timmermans and Roeland Buckinx

Abstract

About two decades ago, a novel family of G protein-coupled receptors (GPCRs), termed Mas-related G protein-coupled receptors (Mrgprs), was discovered. Mrgprs are known for their role in the skin neuronal innervation, where they play a central role in the sensation of harmful or painful stimuli, such as pain and itch. The gastrointestinal (GI) tract, just like the skin, is a mucosal surface that is densely innervated by intrinsic and extrinsic neuronal circuitries that control GI functioning. Interestingly, although GI neuronal circuitries operate via distinct neuro-anatomical pathways, they share many receptors, ion channels and signaling pathways with the skin neuronal circuitries. Given these clear parallels, a role for Mrgprs in the GI tract seems plausible, but is currently still poorly characterized. Therefore, the present doctoral work aimed at providing a more profound understanding of the expression and role of Mrgprs in the GI tract and its innervation.

This doctoral work showed that the neuronal pathways responsible for GI pain signaling (i.e. extrinsic spinal afferent dorsal root ganglia (DRG)) are a novel expression site for two Mrgpr family members, mouse Mrgprd and Mrgprc11. Furthermore, in vivo functional studies revealed that mouse Mrgprc11 is a driver of visceral hypersensitivity and abnormally increased pain sensitivity from the GI tract, and the translational potential of these findings to humans was demonstrated. Finally, we explored the expression of Mrgprs in the mucosa of the human GI tract and found that human MRGPRF is specifically expressed in a population of enteroendocrine cells. In addition, in a pilot study in IBS patients, we provided a first indication that MRGPRF might play a role in the IBS-affected mucosa, since its expression tended to be downregulated at the mRNA level in mucosal biopsies of IBS-D patients.

Overall, the results obtained in this doctoral work confirm that the GI tract and its innervation is a novel site for Mrgpr expression. More importantly, our findings put Mrgprs on the map as possible novel therapeutic targets in GI pain research and neuro-immune communication, and suggest that further research is warranted on their role in the pathophysiology of debilitating GI disorders such as Inflammatory Bowel Diseases (IBD) and Irritable Bowel Syndrome (IBS).

​Lore Peeters​

• 6 May 2020
  
• Supervisor: Georgios Keliris

Abstract

The increasing prevalence of neurological disorders poses an ever-increasing healthcare burden on society. A better understanding of the underlying mechanisms mediating these neurological disorders is indispensable for the identification of new potential therapeutic targets.

Various neurological disorders have been associated with disturbances in the interplay of functional brain networks. This thesis focuses on unraveling the relationships between functional brain networks, with particular interest for the default mode-like network and task-positive networks in rodents. We utilized a relatively new multi-modal approach combining chemogenetics and non-invasive functional magnetic resonance imaging (fMRI). This combined approach allows collecting valuable information about the role of specific subsets of cells in brain network function and dysfunction.

In the first part of this thesis, we found that chemogenetic manipulation of a specific area, i.e. the anterior cingulate area (ACC), is able to modulate the neural activity and functional connectivity of brain networks. More specifically, silencing of the ACC resulted in altered neural activity and functional connectivity in brain networks including connections with sensory cortex, thalamus, basolateral amygdala and ventral pallidum, areas involved in attention processes, working memory, fear behavior and reward, respectively. In addition, we found that damage to the ACC is associated with decreased functional connectivity in the default mode-like network, the task-positive network and visual system. Interestingly, these affected functional networks demonstrated distinguishable recovery patterns over time.

In the second part of this thesis, we focused on the interactions between the default mode-like network and attentional processes. We found that stimulating bottom-up sensory processes (by presenting continuous random flickering light to sedated rats) could decrease the activity and functional connectivity in the default mode-like network, similarly as task-related deactivations. In a next study, we aimed to shed light on the mechanisms involved in the deactivation of the default mode-like network during attentional processes. Our results indicated functional connectivity decreases in the default mode-like network upon chemogenetic stimulation of basal forebrain cholinergic neurons, suggesting a role for the cholinergic system in default mode-like network regulation.

Taken together, this thesis provides unique insights on functional and dysfunctional brain networks. Furthermore, this thesis underscores the significance of combining chemogenetics with fMRI for future studies as well as its promise for future neurotherapeutics as it can eliminate the off-target effects of current therapeutics and allows the on-demand control of specified neurons.

​Rukun Hinz​

• 5 May 2020
  
• Supervisors: Georgios Keliris and Annemie Van der Linden

Abstract

The default mode network (DMN) is a large-scale brain network that is thought to play a fundamental role in internally oriented cognition, self-referential thought, mind-wandering and autobiographical memory. This brain network has consistently been shown in humans to be highly active and synchronized during rest. However, during an attentive brain state, e.g. during the performance of an external attention demanding task, DMN deactivates and desynchronizes. This functionality of the DMN is thought to facilitate task performance by suppressing internally based processing and task-independent brain activity during a task. Furthermore, it allows to shift additional cognitive processing resources towards the execution of the task. In recent years, preclinical imaging studies have discovered the existence of a default mode-like network (DMLN) in rodents. This DMLN consist out of anatomical homologous regions to the DMN in humans, however the resemblance in functionality remains to be proven. In this thesis, we investigated the functionality of the DMLN in rodents by using functional magnetic resonance imaging (fMRI) techniques during a visual attentive brain state. Here we could demonstrate that similarly as the human DMN, DMLN in rats was shown to deactivate and desynchronize during a visual attentive brain state. This result provides evidence that DMLN in rodents is functionally alike to the DMN in humans. As both networks show high correspondence in anatomy and functionality, results coming from DMLN animal studies could aid to some extent in the understanding of the DMN in humans. Next, we investigated a potential mechanism through which attention might modulate the DMLN i.e. the cholinergic system. As such, we used chemogenetics to stimulate cholinergic neurons in the basal forebrain inducing a cholinergic stimulated brain state and its effect was investigated using fMRI techniques. Results coming from this experiment demonstrated a clear roll of the cholinergic system in the control of synchronization and activity of the DMLN. Due to the strong link between attention and the cholinergic system, we hypothesize that the cholinergic system is the neuromodulatory system through which the attention system exert its modulation on the DMLN. Lastly, in the final part of the thesis, we focused on improving analysis methods for fMRI techniques. To do so, we developed a probabilistic vascular mouse brain atlas to objectively detect vascular influences within fMRI data. The vascular atlas was applied to fMRI data and was demonstrated to be an effective method to objectively identify vascular influences.

​Paulius Palaima​

• 29 April 2020
  
• Supervisors: Albena Jordanova and Kristien Peeters

Abstract

Charcot-Marie-Tooth disease (CMT) is the most common inherited peripheral neuropathy and it is well-known for its incredible genetic and phenotypic heterogeneity. Recent advances in sequencing technologies and genome annotation have led to an explosion of known CMT- causes with mutations in over 80 genes so far described. Despite this, the overall rate of underlying genetic defect identification of in patients ranges from 18.5%-63.2% suggesting that we are still missing a large part of the disease aetiology.

In this thesis we illuminated part of this missing heritability of dominantly inherited CMT and explored where the unknown part could lie. We show that an unbiased approach combining next generation sequencing technologies with positional cloning and filtering based on the online genomic databases is very potent method in CMT-causing gene discovery. As a result, we were one of the first to report mutations in PMP2 as a novel cause of CMT and later expanded on this by adding novel mutations and associated phenotypes. Also, through close collaboration with clinical scientists we were able to redefine a known genetic entity CMT2G to CMT2P by identifying a missense mutation in LRSAM1. A broader analysis of the current literature revealed LRSAM1 involvement in other neurodegenerative diseases blurring the divide between them and CMT.

Finally, we explored the contribution of non-protein coding parts of the genome by analysing the long-standing locus on 10q24.1-q25.1 attributed to dominant intermediate CMT type A. We interrogated this region for structural variants using third generation sequencing as well as studied the possible functional implications of identified non-coding variants. In particular, one variant in the PAX2 3’UTR sequence was predicted to have an effect on a possible regulatory element. Unfortunately, we were not able to verify the functional relevance of that sequence exemplifying the difficulty in studying the contribution of the non- coding genome to human pathology. Still, the DI-CMTA family remains the prime candidate for a discovery of novel CMT-causing variant in the non-coding part of the genome.

​Sebastiaan Van De Vijver​

• 23 April 2020
  
• Supervisors: Michele Giugliano and Erik De Schutter

Abstract

The human brain is capable of absorbing large amounts of information due to its plastic character. The synapses of neuronal networks are presumed to be the location of the plasticity required for this learning process. However, the process of synaptic plasticity and how it culminates in learning and memory requires further investigation. In this thesis, plasticity is studied using parallel electrophysiological measurements coupled with optogenetic interrogation of in vitro dissociated rat cortical cell cultures. Using this experimental approach, four aspects related to the plasticity process were investigated.

Firstly, it was discovered that channelrhodopsin-2 (ChR2), a commonly used opsin, is a facilitator of short-term plasticity. Activation of ChR2 located in synapses leads to synaptic Ca²⁺ build-up that transiently facilitates neurotransmission. Since ChR2 was restricted to the excitatory neuronal population, this synaptic potentiation led to an interesting interplay between the excitatory and inhibitory population, presenting itself through network-wide reverberations of electrical activity.

Secondly, it was discovered that serine proteases, located in the extracellular matrix, are capable of influencing the functional connectivity of neuronal networks. Inhibition of serine proteases leads to altered electrophysiological activity patterns and a shift in the balance between excitation and inhibition without affecting intrinsic cellular properties. A change in functional connectivity combined with the aforementioned lack of change in single-cell properties indicates synapses are being regulated by serine proteases.

Thirdly, the requirements for neuronal ensemble creation were investigated. The ensemble is hypothesised to be a fundamental building block of memory and consists of a group of tightly interconnected neurons. With targeted repetitive laser stimulation, it was attempted to create novel ensembles through the principle of spike-timing dependent plasticity. Despite effective stimulation, the functional connectivity of the network was not altered permanently, indicating that sequential and repetitive activation of neurons is not sufficient to induce plasticity. Finally, the relationship between Fragile-X syndrome and plasticity was investigated. Plasticity is a critical ability of the brain and involved in various brain disorders. FMR1 protein is known to regulate development of synapses and networks. Loss of the protein is shown here to lead to abnormal network activity, an imbalance between excitation and inhibition and structural alterations of synaptic terminals. The results indicate a faulty configuration of connectivity during development. Therapeutic avenues focusing on restoring normal connectivity and synaptic development would thus be highly valuable. Together, these findings shed new light on multiple facets of the plasticity process in neuronal networks.

​Jef Hens​

• 2 April 2020
  
• Supervisor: Luc Kestens

Abstract

The human immune-deficiency virus 1 (HIV-1) is spread worldwide and causes an infection leading to AIDS which, to date, cannot be cured. In search for a better understanding of the virus, scientists investigate the virus by inspecting infected individuals. Especially people that are exposed to HIV-1 but remain seemingly uninfected are crucial for understanding how the virus can be controlled naturally. In general, it is believed that the earlier the virus is controlled, the slower the infection will progress. Recent research even indicates that events during the transmission of HIV-1 could make these individuals resistant to infection. More specifically, immune responses were suggested to obstruct the viral threat by eliminating the cells derived from the HIV-1 infected sexual partner. These immune responses were believed to be exerted by Natural killer (NK) cells, triggered by a mechanism called “missing-self”. To further explore this hypothesis, we assessed the elimination of (non-self) cells derived from the sexual partner when encountered by (self) NK cells, and whether the missing-self mechanism could trigger this or not.

In my PhD, it became clear that the missing-self mechanism enabled the elimination of non-self cells by NK cells. However, NK cells were unable to eliminate the non-self cells when they were infected with HIV-1. Other factors such as MIC-A/B expressed on the HIV-1 infected cells did seem to have an impact on elimination by missing-self NK cell responses. These results show a rather limited impact of the missing-self mechanism on the specific elimination of HIV-1 infected cells. Nevertheless, the results also show that MIC-A/B might be an interesting target when looking at NK cell responses against HIV-1 infected cells. Interestingly, the strength of the missing-self response varied between the different subtypes of NK cells, demonstrating the need for more extensive investigation of the missing-self principle. Especially missing-self responses triggered by KIR2DL1 on NK cells were seen to elicit a strong response. In general, the missing-self mechanism activates the NK cells against non-self cells, whereas NK cell activation against HIV-1 infected cells was rather limited in missing-self conditions.

​Jaana van Gastel​

• 21 February 2020
  
• Supervisors: Stuart Maudsley and Jonathan Janssens

Abstract

Most treatments for age-related disorders have been symptomatic, or target one aspect at a time, we hypothesize however that to combat these disorders at a global early level, we need to prevent these pathologies at a network level. This could be facilitated via the identification of a network-regulating master controller proteins whose modulation would thus possess multidimensional effects, termed ‘keystone’. One such aging keystone has recently been discovered, G protein-coupled receptor (GPCR) kinase interacting protein 2 (GIT2), a scaffolding protein and thus unfortunately not a canonical drug target. Our recent work however has demonstrated that G protein-coupled receptors (GPCRs) can have transcriptional control of protein expression.

As such, a GPCR-based control of GIT2 expression/functionality may be an important mechanism to therapeutically control the aging process. In this dissertation, we will discuss the discovery of one such a receptor, Relaxin family peptide 3 receptor (RXFP3). We have discovered that these two proteins activate different aspects of DNA damage response and repair. In addition, they have the ability to function together and facilitate these processes. DNA damage has been classified as one of the most important hallmarks of aging, as such suggesting that if we can compel RXFP3 to engage a GIT2-dependent signaling pathway, we could potentially ameliorate multiple pathologies associated with diverse age-related disorders. Through the combined use of proteomics, interactomics and bioinformatic analysis, we were able to establish the role of the GIT2-RXFP3 system and already identify a potential GIT2-biased ligand.

​Jana Janssens​

• 14 February 2020
  
• Supervisors: Peter Paul De Deyn and Debby Van Dam

Abstract

Neurodegenerative disorders (NDDs) pose an ever-increasing healthcare burden on society. Studying neurochemical disturbances in these diseases could help identify novel targets of interest and pinpoint disease biomarkers. This thesis focused on the neurochemical characterization of Alzheimer’s disease (AD), frontotemporal lobar degeneration (FTLD), dementia with Lewy bodies/Parkinson’s disease dementia (DLB/PDD) and amyotrophic lateral sclerosis (ALS), primarily with respect to monoaminergic neurotransmitter systems. In addition, we addressed several methodological issues, including the influence of circadian rhythm on monoamines in cerebrospinal fluid (CSF) and plasma and the rostrocaudal concentration gradient (RCG) in CSF. We also verified whether B elements, a type of short interspersed nuclear elements, could be applied as a normalization strategy in reverse transcription quantitative PCR (RT-qPCR) experiments.

We found an influence of circadian rhythm on homovanillic acid (HVA), and identified the presence of a RCG for HVA and 5-hydroxyindoleacetic acid. These findings underscored the importance of standardization of pre-analytical procedures. Furthermore, we could conclude that B elements could be adopted as normalization factors in hippocampus, but less so in cortex.

We further investigated monoaminergic systems in AD and behavioral variant FTLD and found that these two NDDs are characterized by distinct serotonergic and noradrenergic systems. Next, we analyzed CSF and serum samples from patients with AD, FTLD, DLB/PDD and CONTR and noted altered 3-methoxy-4-hydroxyphenylglycol levels in DLB/PDD subjects. In addition, we confirmed that CSF and serum levels of this noradrenergic metabolite improved the AD versus DLB/PDD differential diagnosis.

Because of considerable clinical, genetic and neuropathological overlap between FTLD and ALS, we analyzed CSF and serum samples for monoaminergic and kynurenergic content. A general dopaminergic disturbance was observed in FTLD and ALS subjects, while the KP did not appear to be altered in the ALS-FTLD continuum.

The expression of G protein-coupled receptors (GPCRs) in brain tissue of AD and CONTR subjects was assessed using RT-qPCR. Our results indicated alterations in serotonergic and noradrenergic GPCR expression levels in limbic and frontal brain areas of AD subjects. Glutamatergic and cholinergic GPCR expression levels in the hippocampus of APP23 mice were influenced by age, but not by genotype. Protein-level analyses of the 5-HT₆ receptor in the APP23 model did not show any age- or genotype effects.

Taken together, this dissertation offers more insight into the neurochemical imbalance in several NDDs. Our hypotheses regarding neurochemical underpinnings of NDDs remain to be strengthened, and the biomarker potential of the investigated neurochemicals should be further confirmed.

​Philipp Adams​

• 12 February 2020
  
• Supervisors: Guido Vanham and Carole Devaux

Abstract

​Dietmar Hammerschmid​

• 17 January 2020
  
• Supervisors: Sylvia Dewilde and Frank Sobott

Abstract

For both GsGCS and RAGE it became possible for the first time to obtain structural data on the full-length protein. In comparison with the globin domain (GsGCS¹⁶²), GsGCS highlighted a membrane-domain driven oligomerization into an equilibrium between trimers and tetramers, and, more intriguing, a much higher kinetic stability of the ferrous (Fe²⁺) compared to the ferric (Fe³⁺) form of GsGCS, which was not observed for GsGCS¹⁶². The fact that the higher stability was only observed for GsGCS indicates an oxidation-state dependent helix rearrangement, i.e. (in)activation of a proton transport, for example, of the membrane domain. RAGE has a relatively complex domain architecture with one V- and two C-type domains (akin to Ig domains) on the extracellular side, a transmembrane helix, and a cytoplasmic tail. Therefore, unsurprisingly, the characterization of RAGE showed different oligomerization states depending on which protein construct was analyzed. The full-length and the C2_TM_CT construct exhibited an oligomerization of up to tetramers and hexamers, respectively. These higher oligomerization tendencies were only observed in the presence of the membrane domain. Moreover, collision-induced unfolding experiments, i.e. stability assays, performed on V_C1, V_C1_C2, and FL_RAGE showed a significantly higher stability of V_C1_C2 and FL_RAGE as compared to V_C1, suggesting the C2, which can form a disulfide-mediated dimer, and the transmembrane domain as crucial for the protein’s stability.

Upon oxidative plasma treatment of CYGB, I could identify three major chemical modifications in the protein. Beside the induction of a number of various oxidations, the covalent dimerization via disulfide bridge formation and the binding of NO₂ to the heme group could be detected as well.

Overall, native and structural MS have proven as reliable biophysical tools particularly valuable when high-resolution techniques such as X-ray crystallography and/or cryo-electron microscopy lead to unsatisfying results.