Predicting progression to Alzheimer's disease with human hippocampal progenitors exposed to serum.

Adult hippocampal neurogenesis is important for learning and memory and is altered early in Alzheimer's disease. As hippocampal neurogenesis is modulated by the circulatory systemic environment, evaluating a proxy of how hippocampal neurogenesis is affected by the systemic milieu could serve as an early biomarker for Alzheimer's disease progression. Here, we used an in vitro assay to model the impact of systemic environment on hippocampal neurogenesis. A human hippocampal progenitor cell line was treated with longitudinal serum samples from individuals with mild cognitive impairment, who either progressed to Alzheimer's disease or remained cognitively stable. Mild cognitive impairment to Alzheimer's disease progression was characterized most prominently with decreased proliferation, increased cell death and increased neurogenesis. A subset of 'baseline' cellular readouts together with education level were able to predict Alzheimer's disease progression. The assay could provide a powerful platform for early prognosis, monitoring disease progression and further mechanistic studies.

Clusterin Is Required for ß-Amyloid Toxicity in Human iPSC-Derived Neurons.

Our understanding of the molecular processes underlying Alzheimer's disease (AD) is still limited, hindering the development of effective treatments, and highlighting the need for human-specific models. Advances in identifying components of the amyloid cascade are progressing, including the role of the protein clusterin in mediating ß-amyloid (Aß) toxicity. Mutations in the clusterin gene (CLU), a major genetic AD risk factor, are known to have important roles in Aß processing. Here we investigate how CLU mediates Aß-driven neurodegeneration in human induced pluripotent stem cell (iPSC)-derived neurons. We generated a novel CLU-knockout iPSC line by CRISPR/Cas9-mediated gene editing to investigate Aß-mediated neurodegeneration in cortical neurons differentiated from wild type and CLU knockout iPSCs. We measured response to Aß using an imaging assay and measured changes in gene expression using qPCR and RNA sequencing. In wild type neurons imaging indicated that neuronal processes degenerate following treatment with Aß25-35 peptides and Aß1-42 oligomers, in a dose dependent manner, and that intracellular levels of clusterin are increased following Aß treatment. However, in CLU knockout neurons Aß exposure did not affect neurite length, suggesting that clusterin is an important component of the amyloid cascade. Transcriptomic data were analyzed to elucidate the pathways responsible for the altered response to Aß in neurons with the CLU deletion. Four of the five genes previously identified as downstream to Aß and Dickkopf-1 (DKK1) proteins in an Aß-driven neurotoxic pathway in rodent cells were also dysregulated in human neurons with the CLU deletion. AD and lysosome pathways were the most significantly dysregulated pathways in the CLU knockout neurons, and pathways relating to cytoskeletal processes were most dysregulated in Aß treated neurons. The absence of neurodegeneration in the CLU knockout neurons in response to Aß compared to the wild type neurons supports the role of clusterin in Aß-mediated AD pathogenesis.

Automatic mining of symptom severity from psychiatric evaluation notes.

OBJECTIVES: As electronic mental health records become more widely available, several approaches have been suggested to automatically extract information from free-text narrative aiming to support epidemiological research and clinical decision-making. In this paper, we explore extraction of explicit mentions of symptom severity from initial psychiatric evaluation records. We use the data provided by the 2016 CEGS N-GRID NLP shared task Track 2, which contains 541 records manually annotated for symptom severity according to the Research Domain Criteria. METHODS: We designed and implemented 3 automatic methods: a knowledge-driven approach relying on local lexicalized rules based on common syntactic patterns in text suggesting positive valence symptoms; a machine learning method using a neural network; and a hybrid approach combining the first 2 methods with a neural network. RESULTS: The results on an unseen evaluation set of 216 psychiatric evaluation records showed a performance of 80.1% for the rule-based method, 73.3% for the machine-learning approach, and 72.0% for the hybrid one. CONCLUSIONS: Although more work is needed to improve the accuracy, the results are encouraging and indicate that automated text mining methods can be used to classify mental health symptom severity from free text psychiatric notes to support epidemiological and clinical research.