Multi-omic modeling of antidepressant response implicates dynamic immune and inflammatory changes in individuals who respond to treatment.

BACKGROUND: Major depressive disorder (MDD) is a leading cause of disability worldwide, and is commonly treated with antidepressant drugs (AD). Although effective, many patients fail to respond to AD treatment, and accordingly identifying factors that can predict AD response would greatly improve treatment outcomes. In this study, we developed a machine learning tool to integrate multi-omic datasets (gene expression, DNA methylation, and genotyping) to identify biomarker profiles associated with AD response in a cohort of individuals with MDD. MATERIALS AND METHODS: Individuals with MDD (N = 111) were treated for 8 weeks with antidepressants and were separated into responders and non-responders based on the Montgomery-Åsberg Depression Rating Scale (MADRS). Using peripheral blood samples, we performed RNA-sequencing, assessed DNA methylation using the Illumina EPIC array, and performed genotyping using the Illumina PsychArray. To address this rich multi-omic dataset with high dimensional features, we developed integrative Geneset-Embedded non-negative Matrix factorization (iGEM), a non-negative matrix factorization (NMF) based model, supplemented with auxiliary information regarding gene sets and gene-methylation relationships. In particular, we factorize the subjects by features (i.e., gene expression or DNA methylation) into subjects-by-factors and factors-by-features. We define the factors as the meta-phenotypes as they represent integrated composite scores of the molecular measurements for each subject. RESULTS: Using our model, we identified a number of meta-phenotypes which were related to AD response. By integrating geneset information into the model, we were able to relate these meta-phenotypes to biological processes, including a meta-phenotype related to immune and inflammatory functions as well as other genes related to depression or AD response. The meta-phenotype identified several genes including immune interleukin 1 receptor like 1 (IL1RL1) and interleukin 5 receptor (IL5) subunit alpha (IL5RA), AKT/PIK3 pathway related phosphoinositide-3-kinase regulatory subunit 6 (PIK3R6), and sphingomyelin phosphodiesterase 3 (SMPD3), which has been identified as a target of AD treatment. CONCLUSIONS: The derived meta-phenotypes and associated biological functions represent both biomarkers to predict response, as well as potential new treatment targets. Our method is applicable to other diseases with multi-omic data, and the software is open source and available on Github (https://github.com/li-lab-mcgill/iGEM).

MicroRNA-210 regulates the metabolic and inflammatory status of primary human astrocytes.

BACKGROUND: Astrocytes are the most numerous glial cell type with important roles in maintaining homeostasis and responding to diseases in the brain. Astrocyte function is subject to modulation by microRNAs (miRs), which are short nucleotide strands that regulate protein expression in a post-transcriptional manner. Understanding the miR expression profile of astrocytes in disease settings provides insight into the cellular stresses present in the microenvironment and may uncover pathways of therapeutic interest. METHODS: Laser-capture microdissection was used to isolate human astrocytes surrounding stroke lesions and those from neurological control tissue. Astrocytic miR expression profiles were examined using quantitative reverse transcription polymerase chain reaction (RT-qPCR). Primary human fetal astrocytes were cultured under in vitro stress conditions and transfection of a miR mimic was used to better understand how altered levels of miR-210 affect astrocyte function. The astrocytic response to stress was studied using qPCR, enzyme-linked immunosorbent assays (ELISAs), measurement of released lactate, and Seahorse. RESULTS: Here, we measured miR expression levels in astrocytes around human ischemic stroke lesions and observed differential expression of miR-210 in chronic stroke astrocytes compared to astrocytes from neurological control tissue. We also identified increased expression of miR-210 in mouse white matter tissue around middle cerebral artery occlusion (MCAO) brain lesions. We aimed to understand the role of miR-210 in primary human fetal astrocytes by developing an in vitro assay of hypoxic, metabolic, and inflammatory stresses. A combination of hypoxic and inflammatory stresses was observed to upregulate miR-210 expression. Transfection with miR-210-mimic (210M) increased glycolysis, enhanced lactate export, and promoted an anti-inflammatory transcriptional and translational signature in astrocytes. Additionally, 210M transfection resulted in decreased expression of complement 3 (C3) and semaphorin 5b (Sema5b). CONCLUSIONS: We conclude that miR-210 expression in human astrocytes is modulated in response to ischemic stroke disease and under in vitro stress conditions, supporting a role for miR-210 in the astrocytic response to disease conditions. Further, the anti-inflammatory and pro-glycolytic impact of miR-210 on astrocytes makes it a potential candidate for further research as a neuroprotective agent.

Recurrent disease progression networks for modelling risk trajectory of heart failure.

MOTIVATION: Recurrent neural networks (RNN) are powerful frameworks to model medical time series records. Recent studies showed improved accuracy of predicting future medical events (e.g., readmission, mortality) by leveraging large amount of high-dimensional data. However, very few studies have explored the ability of RNN in predicting long-term trajectories of recurrent events, which is more informative than predicting one single event in directing medical intervention. METHODS: In this study, we focus on heart failure (HF) which is the leading cause of death among cardiovascular diseases. We present a novel RNN framework named Deep Heart-failure Trajectory Model (DHTM) for modelling the long-term trajectories of recurrent HF. DHTM auto-regressively predicts the future HF onsets of each patient and uses the predicted HF as input to predict the HF event at the next time point. Furthermore, we propose an augmented DHTM named DHTM+C (where "C" stands for co-morbidities), which jointly predicts both the HF and a set of acute co-morbidities diagnoses. To efficiently train the DHTM+C model, we devised a novel RNN architecture to model disease progression implicated in the co-morbidities. RESULTS: Our deep learning models confers higher prediction accuracy for both the next-step HF prediction and the HF trajectory prediction compared to the baseline non-neural network models and the baseline RNN model. Compared to DHTM, DHTM+C is able to output higher probability of HF for high-risk patients, even in cases where it is only given less than 2 years of data to predict over 5 years of trajectory. We illustrated multiple non-trivial real patient examples of complex HF trajectories, indicating a promising path for creating highly accurate and scalable longitudinal deep learning models for modeling the chronic disease.

Astrocytes in the Pathogenesis of Multiple Sclerosis: An In Situ MicroRNA Study.

Astrocytes are increasingly recognized as active contributors to the disease process in multiple sclerosis (MS), rather than being merely reactive. We investigated the expression of a selected microRNA (miRNA) panel that could contribute both to the injury and to the recovery phases of the disease. Individual astrocytes were laser microdissected from brain sections. We then compared the miRNAs' expressions in MS and control brain samples at different lesional stages in white versus grey matter regions. In active MS lesions, we found upregulation of ischemia-related miRNAs in white but not grey matter, often with reversion to the normal state in inactive lesions. In contrast to our previous findings on MS macrophages, expression of 2 classical inflammatory-related miRNAs, miRNA-155 and miRNA-146a, was reduced in astrocytes from active and chronic active MS lesions in white and grey matter, suggesting a lesser direct pathogenetic role for these miRNAs in astrocytes. miRNAs within the categories regulating aquaporin4 (-100, -145, -320) and glutamate transport/apoptosis/neuroprotection (-124a, -181a, and -29a) showed some contrasting responses. The regional and lesion-stage differences of expression of these miRNAs indicate the remarkable ability of astrocytes to show a wide range of selective responses in the face of differing insults and phases of resolution.

Distinct age and differentiation-state dependent metabolic profiles of oligodendrocytes under optimal and stress conditions.

Within the microenvironment of multiple sclerosis lesions, oligodendrocytes are subject to metabolic stress reflecting effects of focal ischemia and inflammation. Previous studies have shown that under optimal conditions in vitro, the respiratory activity of human adult brain-derived oligodendrocytes is lower and more predominantly glycolytic compared to oligodendrocytes differentiated in vitro from post natal rat brain oligodendrocyte progenitor cells. In response to sub-lethal metabolic stress, adult human oligodendrocytes reduce overall energy production rate impacting the capacity to maintain myelination. Here, we directly compare the metabolic profiles of oligodendrocytes derived from adult rat brain with oligodendrocytes newly differentiated in vitro from oligodendrocyte progenitor cells obtained from the post natal rat brain, under both optimal culture and metabolic stress (low/no glucose) conditions. Oxygen consumption and extracellular acidification rates were measured using a Seahorse extracellular flux analyzer. Our findings indicate that under optimal conditions, adult rat oligodendrocytes preferentially use glycolysis whereas newly differentiated post natal rat oligodendrocytes, and the oligodendrocyte progenitor cells from which they are derived, mainly utilize oxidative phosphorylation to produce ATP. Metabolic stress increases the rate of ATP production via oxidative phosphorylation and significantly reduces glycolysis in adult oligodendrocytes. The rate of ATP production was relatively unchanged in newly differentiated post natal oligodendrocytes under these stress conditions, while it was significantly reduced in oligodendrocyte progenitor cells. Our study indicates that both age and maturation influence the metabolic profile under optimal and stressed conditions, emphasizing the need to consider these variables for in vitro studies that aim to model adult human disease.

MicroRNA Expression Patterns in Human Astrocytes in Relation to Anatomical Location and Age.

Anatomic distribution and age are variables linked to functions of astrocytes under physiologic and pathologic conditions. We measured the relative expression of a panel of microRNAs (miRNAs) in astrocytes captured by laser micro-dissection from normal human adult white and grey matter, human fetal white matter and germinal matrix samples. Although expression of most miRNAs was comparable between adult and fetal samples, regional differences were observed. In the adult cerebral cortex, expression of miRNAs in morphologically distinct inter-laminar astrocytes underlying the glial limitans differed from those in deeper cortical layers, suggesting functional specialization possibly related to structural stability and defense from potentially harmful factors in the cerebrospinal fluid. Differences between adult white and grey matter miRNA expression included higher expression of pro-inflammatory miRNAs in the former, potentially contributing to differences in inflammation between grey and white matter plaques in multiple sclerosis. Lower expression of miRNAs in fetal versus adult white matter astrocytes likely reflects the immaturity of these migrating cells. Highly expressed miRNAs in the fetal germinal matrix are probably relevant in development and also recapitulate some responses to injury. Future studies can address regional alterations of miRNA expression in pathological conditions.