A human stem cell-derived neuronal model of morphine exposure reflects brain dysregulation in opioid use disorder: Transcriptomic and epigenetic characterization of postmortem-derived iPSC neurons.

Introduction: Human-derived induced pluripotent stem cell (iPSC) models of brain promise to advance our understanding of neurotoxic consequences of drug use. However, how well these models recapitulate the actual genomic landscape and cell function, as well as the drug-induced alterations, remains to be established. New in vitro models of drug exposure are needed to advance our understanding of how to protect or reverse molecular changes related to substance use disorders. Methods: We engineered a novel induced pluripotent stem cell-derived model of neural progenitor cells and neurons from cultured postmortem human skin fibroblasts, and directly compared these to isogenic brain tissue from the donor source. We assessed the maturity of the cell models across differentiation from stem cells to neurons using RNA cell type and maturity deconvolution analyses as well as DNA methylation epigenetic clocks trained on adult and fetal human tissue. As proof-of-concept of this model's utility for substance use disorder studies, we compared morphine- and cocaine-treated neurons to gene expression signatures in postmortem Opioid Use Disorder (OUD) and Cocaine Use Disorder (CUD) brains, respectively. Results: Within each human subject (N = 2, 2 clones each), brain frontal cortex epigenetic age parallels that of skin fibroblasts and closely approximates the donor's chronological age; stem cell induction from fibroblast cells effectively sets the epigenetic clock to an embryonic age; and differentiation of stem cells to neural progenitor cells and then to neurons progressively matures the cells via DNA methylation and RNA gene expression readouts. In neurons derived from an individual who died of opioid overdose, morphine treatment induced alterations in gene expression similar to those previously observed in OUD ex-vivo brain tissue, including differential expression of the immediate early gene EGR1, which is known to be dysregulated by opioid use. Discussion: In summary, we introduce an iPSC model generated from human postmortem fibroblasts that can be directly compared to corresponding isogenic brain tissue and can be used to model perturbagen exposure such as that seen in opioid use disorder. Future studies with this and other postmortem-derived brain cellular models, including cerebral organoids, can be an invaluable tool for understanding mechanisms of drug-induced brain alterations.

Within subject cross-tissue analyzes of epigenetic clocks in substance use disorder postmortem brain and blood.

There is a possible accelerated biological aging in patients with substance use disorders (SUD). The evaluation of epigenetic clocks, which are accurate estimators of biological aging based on DNA methylation changes, has been limited to blood tissue in patients with SUD. Consequently, the impact of biological aging in the brain of individuals with SUD remains unknown. In this study, we evaluated multiple epigenetic clocks (DNAmAge, DNAmAgeHannum, DNAmAgeSkinBlood, DNAmPhenoAge, DNAmGrimAge, and DNAmTL) in individuals with SUD (n = 42), including alcohol (n = 10), opioid (n = 19), and stimulant use disorder (n = 13), and controls (n = 10) in postmortem brain (prefrontal cortex) and blood tissue obtained from the same individuals. We found a higher DNAmPhenoAge (ß = 0.191, p-value = 0.0104) and a nominally lower DNAmTL (ß = -0.149, p-value = 0.0603) in blood from individuals with SUD compared to controls. SUD subgroup analysis showed a nominally lower brain DNAmTL in subjects with alcohol use disorder, compared to stimulant use disorder and controls (ß = 0.0150, p-value = 0.087). Cross-tissue analyzes indicated a lower blood DNAmTL and a higher blood DNAmAge compared to their respective brain values in the SUD group. This study highlights the relevance of tissue specificity in biological aging studies and suggests that peripheral measures of epigenetic clocks in SUD may depend on the specific type of drug used.

Angiogenic gene networks are dysregulated in opioid use disorder: evidence from multi-omics and imaging of postmortem human brain.

Opioid use disorder (OUD) is a public health crisis in the U.S. that causes over 50 thousand deaths annually due to overdose. Using next-generation RNA sequencing and proteomics techniques, we identified 394 differentially expressed (DE) coding and long noncoding (lnc) RNAs as well as 213 DE proteins in Brodmann Area 9 of OUD subjects. The RNA and protein changes converged on pro-angiogenic gene networks and cytokine signaling pathways. Four genes (LGALS3, SLC2A1, PCLD1, and VAMP1) were dysregulated in both RNA and protein. Dissecting these DE genes and networks, we found cell type-specific effects with enrichment in astrocyte, endothelial, and microglia correlated genes. Weighted-genome correlation network analysis (WGCNA) revealed cell-type correlated networks including an astrocytic/endothelial/microglia network involved in angiogenic cytokine signaling as well as a neuronal network involved in synaptic vesicle formation. In addition, using ex vivo magnetic resonance imaging, we identified increased vascularization in postmortem brains from a subset of subjects with OUD. This is the first study integrating dysregulation of angiogenic gene networks in OUD with qualitative imaging evidence of hypervascularization in postmortem brain. Understanding the neurovascular effects of OUD is critical in this time of widespread opioid use.

Genome-Wide Correlation of DNA Methylation and Gene Expression in Postmortem Brain Tissues of Opioid Use Disorder Patients.

BACKGROUND: Opioid use disorder (OUD) affects millions of people, causing nearly 50 000 deaths annually in the United States. While opioid exposure and OUD are known to cause widespread transcriptomic and epigenetic changes, few studies in human samples have been conducted. Understanding how OUD affects the brain at the molecular level could help decipher disease pathogenesis and shed light on OUD treatment. METHODS: We generated genome-wide transcriptomic and DNA methylation profiles of 22 OUD subjects and 19 non-psychiatric controls. We applied weighted gene co-expression network analysis to identify genetic markers consistently associated with OUD at both transcriptomic and methylomic levels. We then performed functional enrichment for biological interpretation. We employed cross-omics analysis to uncover OUD-specific regulatory networks. RESULTS: We found 6 OUD-associated co-expression gene modules and 6 co-methylation modules (false discovery rate <0.1). Genes in these modules are involved in astrocyte and glial cell differentiation, gliogenesis, response to organic substance, and response to cytokine (false discovery rate <0.05). Cross-omics analysis revealed immune-related transcription regulators, suggesting the role of transcription factor-targeted regulatory networks in OUD pathogenesis. CONCLUSIONS: Our integrative analysis of multi-omics data in OUD postmortem brain samples suggested complex gene regulatory mechanisms involved in OUD-associated expression patterns. Candidate genes and their upstream regulators revealed in astrocyte, and glial cells could provide new insights into OUD treatment development.

Altered levels of interleukins and neurotrophic growth factors in mood disorders and suicidality: an analysis from periphery to central nervous system.

Interleukins and neurotrophins levels are altered in the periphery of patients with major depression and suicidal behavior, however it is not clear if similar abnormalities occur in the central nervous system. Our objective was to examine the association of IL-6, IL-1ß, BDNF, and GDNF levels between postmortem plasma, cerebrospinal fluid (CSF), and brain tissue in a heterogeneous diagnostic subject groups including normal controls, mood disorders only, mood disorders with AUD/SUD (alcohol abuse disorder, substance abuse disorder), and AUD/SUD without mood disorders. To address these questions we collected postmortem plasma (n = 29), CSF (n = 28), and brain (BA10) (n = 57) samples from individuals with mood disorder, mood disorder with AUD/SUD, AUD/SUD and normal controls. These samples were analyzed using a multiplex based luminex assay with a customized 4-plex cytokine/interleukins- IL-6, IL-1ß, BDNF, and GDNF human acute phase based on xMAP technology platform. Protein levels were determined using a Luminex 200 instrument equipped with Xponent-analyzing software. We observed IL-6 (p = 2.1e-07), and GDNF (p = 0.046) were significantly correlated between brain and CSF. In addition, IL-6 (p = 0.031), were significantly correlated between brain and plasma. Overall diagnostic group analysis showed a significant difference with brain GDNF, p = 0.0106. Pairwise comparisons showed that GDNF level is-39.9 ± 12 pg/ml, p = 0.0106, was significantly higher than in the brains derived from mood disorders compared to normal controls, -23.8 ± 5.5 pg/ml, p = 0.034. Brain BDNF was higher in suicide (p = 0.0023), males compared to females (p = 0.017), and psychiatric medication treated vs. non-treated (p = 0.005) individuals. Overall, we demonstrate that blood IL-6, GDNF and BDNF could be informative peripheral biomarkers of brain biology associated with mood disorders, substance disorders, and suicide.

FTY720 Improves Behavior, Increases Brain Derived Neurotrophic Factor Levels and Reduces α-Synuclein Pathology in Parkinsonian GM2+/- Mice.

Parkinson's disease (PD) is a progressive aging disorder that affects millions worldwide, thus, disease-modifying-therapies are urgently needed. PD pathology includes α-synuclein (aSyn) accumulation as synucleinopathy. Loss of GM1 gangliosides occurs in PD brain, which is modeled in GM2 synthase transgenic mice. GM2+/- mice have low, not absent GM1 and develop age-onset motor deficits, making them an excellent PD drug testing model. FTY720 (fingolimod) reduces synucleinopathy in A53T aSyn mice and motor dysfunction in 6-OHDA and rotenone PD models, but no one has tested FTY720 in mice that develop age-onset PD-like motor problems. We confirmed that GM2+/-mice had equivalent rotarod, hindlimb reflexes, and adhesive removal functions at 9 mo. From 11 mo, GM2+/- mice received oral FTY720 or vehicle 3x/week to 16 mo. As bladder problems occur in PD, we also assessed GM2+/- bladder function. This allowed us to demonstrate improved motor and bladder function in GM2+/- mice treated with FTY720. By immunoblot, FTY720 reduced levels of proNGF, a biomarker of bladder dysfunction. In humans with PD, arm swing becomes abnormal, and brachial plexus modulates arm swing. Ultrastructure of brachial plexus in wild type and GM2 transgenic mice confirmed abnormal myelination and axons in GM2 transgenics. FTY720 treated GM2+/- brachial plexus sustained myelin associated protein levels and reduced aggregated aSyn and PSer129 aSyn levels. FTY720 increases brain derived neurotrophic factor (BDNF) and we noted increased BDNF in GM2+/- brachial plexus and cerebellum, which contribute to rotarod performance. These findings provide further support for testing low dose FTY720 in patients with PD.

SNAP-25 in Major Psychiatric Disorders: A Review.

Synaptosomal Associated Protein-25 kilodaltons (SNAP-25) is an integral member of the SNARE complex. This complex is essential for calcium-triggered synaptic vesicular fusion and release of neurotransmitters into the synaptic cleft. In addition to neurotransmission, SNAP-25 is associated with insulin release, the regulation of intracellular calcium, and neuroplasticity. Because of SNAP-25's varied and crucial biological roles, the consequences of changes in this protein can be seen in both the central nervous system and the periphery. In this review, we will look at the published literature from human genetic, postmortem, and animal studies involving SNAP-25. The accumulated data indicate that SNAP-25 may be linked with some symptoms associated with a variety of psychiatric disorders. These disorders include bipolar disorder, schizophrenia, major depressive disorder, attention deficit hyperactivity disorder, autism, alcohol use disorder, and dementia. There are also data suggesting SNAP-25 may be involved with non-psychiatric seizures and metabolic disorders. We believe investigation of SNAP-25 is important for understanding both normal behavior and some aspects of the pathophysiology of behavior seen with psychiatric disorders. The wealth of information from both animal and human studies on SNAP-25 offers an excellent opportunity to use a bi-directional research approach. Hypotheses generated from genetically manipulated mice can be directly tested in human postmortem tissue, and, conversely, human genetic and postmortem findings can improve and validate animal models for psychiatric disorders.