A Machine Learning-Based Holistic Approach to Predict the Clinical Course of Patients within the Alzheimer's Disease Spectrum.

BACKGROUND: Alzheimer's disease (AD) is a neurodegenerative condition driven by multifactorial etiology. Mild cognitive impairment (MCI) is a transitional condition between healthy aging and dementia. No reliable biomarkers are available to predict the conversion from MCI to AD. OBJECTIVE: To evaluate the use of machine learning (ML) on a wealth of data offered by the Alzheimer's Disease Neuroimaging Initiative (ADNI) and Alzheimer's Disease Metabolomics Consortium (ADMC) database in the prediction of the MCI to AD conversion. METHODS: We implemented an ML-based Random Forest (RF) algorithm to predict conversion from MCI to AD. Data related to the study population (587 MCI subjects) were analyzed by RF as separate or combined features and assessed for classification power. Four classes of variables were considered: neuropsychological test scores, AD-related cerebrospinal fluid (CSF) biomarkers, peripheral biomarkers, and structural magnetic resonance imaging (MRI) variables. RESULTS: The ML-based algorithm exhibited 86% accuracy in predicting the AD conversion of MCI subjects. When assessing the features that helped the most, neuropsychological test scores, MRI data, and CSF biomarkers were the most relevant in the MCI to AD prediction. Peripheral parameters were effective when employed in association with neuropsychological test scores. Age and sex differences modulated the prediction accuracy. AD conversion was more effectively predicted in females and younger subjects. CONCLUSION: Our findings support the notion that AD-related neurodegenerative processes result from the concerted activity of multiple pathological mechanisms and factors that act inside and outside the brain and are dynamically affected by age and sex.

Inhibition of de novo ceramide biosynthesis affects aging phenotype in an in vitro model of neuronal senescence.

Although aging is considered to be an unavoidable event, recent experimental evidence suggests that the process can be counteracted. Intracellular calcium (Ca2+i) dyshomeostasis, mitochondrial dysfunction, oxidative stress, and lipid dysregulation are critical factors that contribute to senescence-related processes. Ceramides, a pleiotropic class of sphingolipids, are important mediators of cellular senescence, but their role in neuronal aging is still largely unexplored. In this study, we investigated the effects of L-cycloserine (L-CS), an inhibitor of thede novoceramide biosynthesis, on the aging phenotype of cortical neurons cultured for 22 days, a setting employed as anin vitromodel of senescence. Our findings indicate that, compared to control cultures, 'aged' neurons display dysregulation of [Ca2+]ilevels, mitochondrial dysfunction, increased generation of reactive oxygen species (ROS), altered synaptic activity as well as the activation of neuronal death-related molecules. Treatment with L-CS positively affected the senescent phenotype, a result associated with recovery of neuronal [Ca2+]isignaling and reduction of mitochondrial dysfunction and ROS generation. The results suggest that thede novoceramide biosynthesis represents a critical intermediate in the molecular and functional cascade leading to neuronal senescence and identify ceramide biosynthesis inhibitors as promising pharmacological tools to decrease age-related neuronal dysfunctions.

The pharmacological perturbation of brain zinc impairs BDNF-related signaling and the cognitive performances of young mice.

Zinc (Zn2+) is a pleiotropic modulator of the neuronal and brain activity. The disruption of intraneuronal Zn2+ levels triggers neurotoxic processes and affects neuronal functioning. In this study, we investigated how the pharmacological modulation of brain Zn2+ affects synaptic plasticity and cognition in wild-type mice. To manipulate brain Zn2+ levels, we employed the Zn2+ (and copper) chelator 5-chloro-7-iodo-8-hydroxyquinoline (clioquinol, CQ). CQ was administered for two weeks to 2.5-month-old (m.o.) mice, and effects studied on BDNF-related signaling, metalloproteinase activity as well as learning and memory performances. CQ treatment was found to negatively affect short- and long-term memory performances. The CQ-driven perturbation of brain Zn2+ was found to reduce levels of BDNF, synaptic plasticity-related proteins and dendritic spine density in vivo. Our study highlights the importance of choosing "when", "where", and "how much" in the modulation of brain Zn2+ levels. Our findings confirm the importance of targeting Zn2+ as a therapeutic approach against neurodegenerative conditions but, at the same time, underscore the potential drawbacks of reducing brain Zn2+ availability upon the early stages of development.

Exenatide exerts cognitive effects by modulating the BDNF-TrkB neurotrophic axis in adult mice.

Modulation of insulin-dependent signaling is emerging as a valuable therapeutic tool to target neurodegeneration. In the brain, the activation of insulin receptors promotes cell growth, neuronal repair, and protection. Altered brain insulin signaling participates in the cognitive decline seen in Alzheimer's disease patients and the aging brain. Glucagon-like peptide-1 (GLP-1) regulates insulin secretion and, along with GLP-1 analogues, enhances neurotrophic signaling and counteracts cognitive deficits in preclinical models of neurodegeneration. Moreover, recent evidence indicates that GLP-1 modulates the activity of the brain-derived neurotrophic factor (BDNF). In this study, in adult wild-type mice, here employed as a model of mid-life brain aging, we evaluated the effects of a 2-month treatment with exenatide, a GLP-1 analogue. We found that exenatide promotes the enhancement of long-term memory performances. Biochemical and imaging analyses show that the drug promotes the activation of the BDNF-TrkB neurotrophic axis and inhibits apoptosis by decreasing p75NTR-mediated signaling. The study provides preclinical evidence for the use of exenatide to delay age-dependent cognitive decline.