performing a Sensory Motor Control Task (control) and another while performing the Montreal Imaging Stress Task (stress). Dopamine release was defined as percent change in D2 receptor binding potential between control and stress scans using a novel correction for injected mass of [11C] FLB457. Results: Stress-induced dopamine release (DBPND) was significantly different between groups in mPFC (F(2,30)¼5.40, p¼0.010) with CHR-CU exhibiting lower DBPND compared to CHR (p¼0.008). Similarly, salivary cortisol response (DAUCI) was significantly lower in CHR-CU as compared to CHR (F(2,29)¼5.08, p¼0.013; posthoc p¼0.018). There were no differences among groups in dlPFC (p>0.05). Conclusions: Given the global trend to legalize cannabis, this study is important as it highlights the effects of chronic cannabis use on cortical dopamine function in high-risk youth. Supported By: CIHR Keywords: Adolescent Stress, Cannabis, Dopamine, Cortex, PET 139. Methlyome-Wide Changes Associated With Childhood Trauma and its Long-Term Outcomes William Copeland 1 , Karolina Aberg 2 , Robin Chan 2 , Min Zhao 2 , Lin Ying Xie 2 , E. Jane Costello 3 , and Edwin van den Oord 2 1 University of Vermont, 2 Center for Biomarker Research and Precision Medicine, School of Pharmacy, Virginia Commonwealth University, 3 Duke University Medical Center Background: It is not well understood how childhood trau- matic experiences are biologically embedded. This study uses a prospective longitudinal study that began in childhood, continued into adulthood, and collected bloodspots at each timepoint to test whether childhood trauma was linked to DNA methylation status. Methods: DNA was extracted from 1,233 bloodspots from subjects (median age 15.23) in the Great Smoky Mountains Study before trauma exposure, after trauma exposure and a decade later in adulthood. A sequencing-based approach was used that provided almost complete coverage of all 28 million CpGs in the genome. We analyzed whole blood methylation as well as cell type specific analyses within constituent populations of granulocytes, T cells, B cell and monocytes. Results: Childhood trauma was associated with 200+ methylation sites across cell types (top MWAS p vales < 1.0 x 10-9). Pathways identified overlapped with genes associated with depression, inflammation, cannabis, and schizophrenia in GWAS. Machine learning algorithms combined with 10-fold cross validation were used to obtain an unbiased estimate of the combines effect of all associated sites in whole blood and summarize results into a single methylation risk score (MRS). Results suggest that the MRS shared over 10% of the variation in cumulative trauma exposure. The MRS made a unique contribution to the prediction of health outcomes later in life that could not be captured by clinical data or the number of adverse events. Conclusions: Childhood trauma has a significant impact on the methylome, particularly T-cells and granulocytes, and these changes are associated with adult health outcomes. Supported By: R01 NIMH Keywords: DNA Methylation, Childhood Trauma, Depression, Childhood Maltreatment SYMPOSIUM From Inflammation to Circuits - Cells, Synapses and Signaling Pathways in Depression 3:00 p.m. - 5:00 p.m. Chair: Ebrahim Haroon Co-Chair: Robert McCullumsmith 140. Inflammatory Cytokines and Stem Cell Progeny in Major Depression Maura Boldrini 1 , Yan Liu 2 , Alexandria Tartt 3 , Camille Fulmore 3 , Gorazd Rosoklija 4 , Andrew Dwork 4 , Mark Underwood 4 , Victoria Arango 4 , and J. John Mann 2 1 College of Physicians & Surgeons, Columbia University, 2 New York State Psychiatric Institute, 3 Columbia Univer- sity/NYSPI, 4 Columbia University Background: Inflammation and microglia activation may contribute to the neuroplasticity deficits we observed in sub- jects with major depression (MDD), including smaller anterior dentate gyrus (DG). MDD increases pro-inflammatory cyto- kines interleukin (IL)-1 and IL-6 in the peripheral blood and cerebral spinal fluid, and a depressive-like phenotype is induced by increasing inflammation. Interleukins have anti- neurogenic and pro-gliogenic effect in the hippocampus. Methods: In the whole DG of matched untreated subjects with (n¼36), selective serotonin reuptake inhibitor (SSRI)-treated MDD (n¼12), and non-psychiatric non-suicide controls (n¼36), we performed immunocytochemistry and stereology to quan- tify neural progenitor cells (NPC), neuroblasts and mature granule neurons (GNs), as well as DG volume, using our established methods. We assayed by Luminex 51-Plex (Affy- metrix, Inc.) 51 cytokines and chemokines, and performed Western Blots to quantify interleukin-6, IL-1beta, IL-1R, and TNF-a. Results: MDD had smaller anterior (p¼.005), mid (p¼.000010) and posterior (p¼.013) DG volume than controls, as well as fewer NPCs (p ¼.001), neuroblasts (p¼.027) and GNs (p ¼.032) in anterior DG. SSRI-treated MDD did not differ from non- psychiatric non-suicide controls in any of these measures. Among 51 cytokines and chemokines tested, we found greater levels of inflammatory cytokine TNFa (p¼.019) in MDD brain tissue compared with controls. We are further testing levels of pro- and anti-inflammatory cytokines and correlating them with a number of NPCs, neuroblasts, and GNs in MDD-Suicide and controls. Conclusions: If alteration of TNFa, IL-1b or IL-6 signaling is confirmed, together with a relationship with neurogenesis deficits in MDD, future treatments targeting neuroinflammation may restore MDD symptomatology and DG cell viability MDD. Plenary and Symposium Abstracts S58 Biological Psychiatry May 15, 2019; 85:S1eS104 www.sobp.org/journal Biological Psychiatry: Celebrating 50 Years