H. Bea Kuiperij

Inhibition of α-synuclein aggregation by small heat shock proteins

The fibrillization of α‐synuclein (α‐syn) is a key event in the pathogenesis of α‐synucleinopathies. Mutant α‐syn (A53T, A30P, or E46K), each linked to familial Parkinsons disease, has altered aggregation properties, fibril morphologies, and fibrillization kinetics. Besides α‐syn, Lewy bodies also contain several associated proteins including small heat shock proteins (sHsps). Since α‐syn accumulates intracellularly, molecular chaperones like sHsps may regulate α‐syn folding and aggregation. Therefore, we investigated if the sHsps αB‐crystallin, Hsp27, Hsp20, HspB8, and HspB2B3 bind to α‐syn and affect α‐syn aggregation. We demonstrate that all sHsps bind to the various α‐syns, although the binding kinetics suggests a weak and transient interaction only. Despite this transient interaction, the various sHsps inhibited mature α‐syn fibril formation as shown by a Thioflavin T assay and atomic force microscopy. Interestingly, HspB8 was the most potent sHsp in inhibiting mature fibril formation of both wild‐type and mutant α‐syn. In conclusion, sHsps may regulate α‐syn aggregation and, therefore, optimization of the interaction between sHsps and α‐syn may be an interesting target for therapeutic intervention in the pathogenesis of α‐synucleinopathies. Proteins 2011;

Diagnosis of progressive supranuclear palsy: can measurement of tau forms help?

Recently, a new assay for the differential diagnosis of progressive supranuclear palsy (PSP) was proposed. It was shown that the ratio of 33/55 kDa tau forms in cerebrospinal fluid (CSF) was specifically reduced in PSP CSF. We aimed to reproduce these results, but were not able to detect the tau forms in CSF. We demonstrate that i) CSF total tau levels are too low to be detected by the published protocol, and ii) the described 33 and 55 kDa bands are likely the heavy and light chains of IgG used in the assay. We conclude that more sensitive techniques are needed to measure tau forms in CSF.

TDP-43 plasma levels do not differentiate sporadic inclusion body myositis from other inflammatory myopathies.

Sporadic inclusion body myositis (sIBM) is a progressive myopathy, clinically characterized by muscle weakness and atrophy, usually beginning after 50 years of age; sIBM belongs to a group of inflammatory myopathies, also including polymyositis (PM) and dermatomyositis (DM). Differentiating between these inflammatory myopathies can be difficult due to an overlap in clinical symptoms. Recently, the TAR DNA-binding protein-43 (TDP-43) was identified as a component of inclusions in sIBM muscle fibers [7]. We examined if the TDP-43 protein could be identified in the plasma of sIBM patients, and could, as such, differentiate sIBM patients from patients with other inflammatory myopathies, in which TDP-43-positive inclusions are not, or only rarely, identified. The study was performed in a cohort of sIBM (n = 31), PM (n = 48), DM (n = 24) and control (n = 33) patients, previously compiled and characterized for our study on amyloid-b42 plasma levels [1]. We developed an ELISA to quantify TDP-43 protein as previously described [2, 4], with minor modifications. The assay is linear until 20 ng TDP-43/ml, with a lower detection limit of 0.4 ng/ml. TDP-43 plasma levels of the different patient groups were analyzed and compared using one-way ANOVA (Kruskal– Wallis test) with Dunn’s post hoc test for multiple comparisons. Mean TDP-43 plasma levels were significantly increased in all three myopathy groups as compared to the healthy controls (Table 1). There were no significant differences between the sIBM, PM and DM groups. TDP-43 plasma levels in the inflammatory myopathies showed a few outliers with high TDP-43 levels, especially in the DM group (Fig. 1). Still, the majority of samples (58–73%) had levels below the detection limit of the assay (Table 1). But even when these outliers were excluded from analysis, TDP-43 levels were significantly increased in sIBM, PM and DM groups, as compared to controls, indicating a myopathy-specific increase in plasma TDP-43 levels. In 88% of the control cases, TDP-43 plasma levels were below the detection limit. This is in good agreement with previously published plasma TDP-43 levels in a control population [2]. However, a few control samples showed relatively high TDP-43 plasma levels, although well below the highest levels found in the plasma of the inflammatory myopathies (Fig. 1). The high TDP-43 plasma level in some DM, PM, and control cases is unexpected, as in these cases no (DM, controls), or rarely (PM), TDP-43-positive inclusions in muscle fibers have been detected. On the contrary, at least 80% of all sIBM cases are positive for TDP-43 inclusions in muscle fibers [5, 7]. This means that TDP-43 protein is released in the plasma irrespective of the presence of TDP-43 pathology. TDP-43 is an ubiquitously expressed protein, and is therefore also present in (healthy) muscle fibers without TDP-43 pathology. An explanation for high TDP-43 plasma levels in PM and DM cases may H. B. Kuiperij (&) W. F. Abdo B. G. van Engelen H. J. Schelhaas M. M. Verbeek Department of Neurology, Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Medical Centre, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands e-mail: b.kuiperij@neuro.umcn.nl

Oxidized Pin1 and alpha-synuclein in the cerebrospinal fluid as putative biomarkers for Alzheimer's and Parkinson's diseases

Background:Oxidative stress is a general, early feature of several neurodegenerative diseases, including Alzheimer’s disease (AD) and Parkinson’s disease (PD). Importantly, it has been shown that oxidative stress leads to the oxidation of specific proteins including the peptidyl-prolyl isomerase Pin1 in AD, and a-synuclein in PD. These proteins, in their oxidized form, may thus serve as new biomarkers for the early diagnosis of disease. We aim to develop sensitive assays for the quantification of oxidized Pin1 and a -synuclein in cerebrospinal fluid (CSF), and to clinically validate their use as diagnostic biomarkers for AD and PD, respectively. Methods: Sensitive sandwich ELISAs were developed for the quantification of oxidised Pin1 and a-synuclein using a combination of antibodies whereby one antibody is directed against the protein of interest and the other against oxidized groups of the protein, in the form of DNPH-derivatised carbonyl groups. Results: We successfully developed sandwich ELISAs for the specific measurement of oxidized recombinant Pin1 and oxidized recombinant asynuclein. We successfully measured oxPin1 in brain tissue extracts and we will present the results of the CSF analyses of these oxidized proteins in CSF of AD and PD patients compared to controls. Conclusions: We have developed novel ELISAs in which the oxidized from of the proteins Pin1 and a-synuclein can be measured. These ELISAs are used in clinical cohorts to evaluate the potential of these oxidized proteins in CSF as early, diagnostic biomarkers for AD and PD.

MICRO-RNAS AS NOVEL BIOMARKERS IN AD: DIFFERENTIAL EXPRESSION IN HIPPOCAMPUS AND IN CELL-FREE CEREBROSPINAL FLUID

Background: The small, 19-22 nucleotides long micro-RNAs play a prominent role in regulating protein translation in cells. More than 700 miRNAs have been identified in humans, and approximately 70% is expressed in the brain. Therefore, miRNAs have been suggested as indicators for Alzheimer’s Disease (AD). Here, we aim to identify miRNAs that are differentially expressed in AD patients and non-demented controls as possible new biomarkers in both brain tissue and cerebrospinal fluid (CSF).Methods:QPCR was used to study miRNAs let-7f, miR9, miR16, miR34a, miR34c, miR107, miR128a and miR146a. We used hippocampus of 10 nondemented controls, 10 AD patients with Braak III/IV and 10 with Braak VI; and CSF of 20 AD and 20 non-demented controls. For CSF samples, a pre-amplification step was included to increase detection in qPCR. The effects of blood contamination on miRNAs levels was tested by spiking known numbers of blood cells into cell-free CSF. For expression studies, only CSF samples with no or a very low number of cells were selected. In CSF, U6 snRNA served as internal control. In hippocampus, let-7f was used as internal control, because this miRNA was abundantly and equally expressed in all samples. Relative expression levels (REL) were calculated with the formula 2 -DCt. Results: In hippocampus, miR16, miR34c, miR107, miR128a and miR146a were differentially regulated, which was dependent on the Braak stage. In Hela cells, MAP7 was identified as a potential miR16 target. In CSF, out of 8 selected miRNAs only miR16 and miR146a could be reliably detected. Levels of miR-146a were significantly decreased in CSF of AD patients. We identified a strong effect of blood contamination on the CSF levels of miR16, miR24, and miR146a. Conclusions: In conclusion, the abnormal expression of several miRNAs in hippocampus of intermediateand late-stage AD patients suggests their involvement in AD pathogenesis, and low levels of miR146a in CSF were associated with AD.

ABCA-1-mediated beta-amyloid uptake by cells of the blood-brain barrier

out animals. Results: Our data suggests that genetic knockdown of synaptojanin 1 (synj1) decreases Ab levels in cell culture, and reduces amyloid load in AD transgenic animals. The amounts of b CTF are increased. Further, there is a significantly increase of ApoE secretion in cortical neuron/astrocyte culture derived from synj1 -/animals if compared to controls. Conclusions: Together, our results suggest that one phosphoinositol modulator, synaptojanin 1, can regulate amyloid homeostasis, likely through its effects on Ab generation and clearance.

MHPG as an additional marker for the differentiation of dementia with Lewy bodies from Alzheimer's disease, vascular dementia and frontotemporal dementia

neous disease, characterized by atrophy of the frontal and temporal lobes. It is the second most common form of dementia after Alzheimer’s disease (AD), and clinically characterized by behavioural and/or language dysfunction. At the pathological level, most cases are either characterized by the presence of tau inclusions (FTLD-tau), or the presence of TDP-43 inclusions (FTLD-TDP). Although neuropathology seems to be linked to the clinical picture, there are no CSF biomarkers available to diagnose these FTLD subtypes. We aim to quantify both tau and TDP-43 proteins in the cerebrospinal fluid (CSF) and to analyse if a combination of these can function as biomarker to improve discrimination of FTLD subtypes. Methods:We optimized our previously developed TDP-43 ELISA (1) for quantification in CSF. We used post-mortem obtained, pathology-confirmed ventricular CSF to quantify TDP-43, and total and phosphorylated tau (t-tau, p-tau) levels by ELISA. Statistical analyses were performed to determine the optimal combination of markers for discriminating FTLDtau and FTLD-TDP cases. Results: Our analysis showed that from the tested markers, quantification of a combination of TDP-43 and t-tau in ventricular CSF results in the most optimal differentiation of FTLD-tau and FTLD-TDP cases. These combined markers could significantly differentiate the subtypes with a sensitivity of 64% and specificity of 93%. Conclusions: This is the first study to combine the quantification of CSF tau and TDP-43 levels of FTLD cases. Although either biomarker alone did not discriminate FTLD subtypes, combined analysis of TDP-43 and t-tau significantly improved this differentiation. It remains to be shown if TDP-43 in its pathological, phosphorylated form may even further improve this differentiation. We conclude that ante-mortem CSF analysis based on t-tau and TDP-43 levels might aid in the diagnosis of FTLD subtypes, and possibly, the differentiation from AD.

Development of a gold nanoparticle ELISA to detect low-abundant proteins in cerebrospinal fluid for diagnosis of Alzheimer's disease

Background: Cerebrospinal fluid (CSF) analysis is currently becoming integrated into the diagnostic work-up of Alzheimer’s disease (AD). A combination of amyloid-b42, total and phosphorylated tau proteins can identify AD with a sensitivity and specificity above 90%. Nonetheless, their use is limited in differentiating between AD and other types of dementia, which leads to the requirement of new biomarkers with more discriminative capacity. Yet, many protein biomarkers are present at low concentrations in CSF. We hypothesize that the complex of amyloid-b with apolipoprotein E (ApoE) may serve as such low-abundant, specific biomarker for AD. We therefore aim to measure this protein complex in CSF by sandwich ELISA using gold nanoparticles as detection. Because these nanoparticles are able to bind large amounts of detection enzyme, the signal strongly increases and allows measurements of amyloid-b/ApoE complex at low concentrations. Methods: Magnetic particles, coated with an antibody against ApoE, capture the amyloid-b/ApoE complex. Gold nanoparticles are coated simultaneously with an antibody against amyloid-b, and a detection enzyme. After adding the gold nanoparticles to the magnetic-particle-antigen solution, the sandwich is formed and can be detected. The new sandwich ELISA set-up is compared to the conventional ELISA in terms of limit of detection, and amyloid-b/ApoE complex will be quantified in CSF samples. Results: Our current results demonstrate that antibodies can be coupled to both magnetic particles and gold nanoparticles. Moreover, enzymes can be successfully coupled to gold nanoparticles. Subsequently, both particles containing an antibody are also able to bind the antigen of interest, and can be implemented in a sandwich ELISA. Using gold nanoparticles, we can detect lower concentrations of the antigen than the conventional sandwich ELISA does. Conclusions: It is feasible to construct the different elements of the gold nanoparticle ELISA and to measure the antigen at low levels. Future studies will indicate whether the assay is able to measure amyloid-b/ApoE complex in CSF, and whether this complex can serve as a novel CSF biomarker for discrimination of AD and other dementias.

Low glycine levels in brain homogenates of TgSwDI mice compared to wild-type mice

Background: In Alzheimer’s disease (AD), accumulation of amyloid b (Ab) protein can disturb normal glutamatergic neurotransmission. Glutamate is an important neurotransmitter in learning and memory but, when present in high amounts, can also be excitotoxic, resulting in neuronal loss. Indeed, the loss of glutamatergic neurons can be linked to the cognitive impairments seen in AD patients. As disturbances in the glutamatergic system are closely linked to AD, it is our aim to search for potential biomarkers of the glutamatergic system. To this end, we characterized levels of glutamate-related metabolites in the TgSwDI mouse model for AD. Methods: Male and female TgSwDI mice with mild (3 months old) and severe (9 months old) Ab pathology and non-transgenic (C57Bl/6) age-matched controls, were sacrificed by cervical dislocation. Afterwards, their brains were dissected into different brain regions (i.e. hippocampus, cortex) and snapfrozen until analysis. Distilled water was used to dissolve metabolites from tissue. Using an amino acid analyzer, concentrations of various glutamate-related amino acids (i.e. glutamate, glutamine, GABA, glycine, aspartate, asparagine, alanine) were determined. Results: Glutamate levels were hardly changed in TgSwDI mice. However, a consistent significant decrease in glycine concentration was observed in TgSwDI mice compared to agematched non-transgenic mice, independent of age, sex or the brain area tested. Furthermore, in most of the brain samples, aspartate levels were also (significantly) decreased. Levels of other amino acids were unaltered or inconsistently changed in the various brain areas. Conclusions: In TgSwDI mice, glutamate levels remain unchanged, but brain glycine levels, and to a lesser extent aspartate levels, are decreased. This result suggests that these metabolites may be candidate biomarkers reflecting disturbances in glutamatergic neurotransmission.