24-000803

Although the World Health Organization (WHO) has reported that major depressive disorder (MDD) is the leading cause of suicide and disability worldwide, currently available monoaminergic antidepressants fail to adequately treat an estimated one-third of patients.  In this context, it has been nearly two decades since clinical evidence emerged that a single, sub-anesthetic dose of the glutamatergic modulator ketamine had rapid antidepressant effects, creating a critical paradigm shift in our approach to treating depression.  More recently, a renaissance in psychedelic research has produced compelling clinical evidence for psilocybin-assisted psychotherapy as another rapid-acting treatment for depression.  However, despite their rapid and robust clinical effects, the dissociative and hallucinogenic properties of both compounds limit their broader use because of the need for patient monitoring during treatment.  Moving forward, investigating the pharmacologic profile and mechanism of action of these compounds could help develop targeted treatments with a more favorable side effect profile. 

The prevailing hypothesis suggests that ketamine's inhibition of N-methyl-D-aspartate (NMDA) receptors localized at cortical γ-aminobutyric acid (GABA) interneurons leads to the release of inhibition on cortical pyramidal neurons, acute synaptic glutamate release/cycling, increased α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor activation and downstream neuroplasticity – mammalian target of rapamycin activation (mTOR), release of translational inhibition via eukaryotic elongation factor 2 (eEF2) and local brain-derived neurotrophic factor (BDNF) secretion, ultimately resulting in altered synaptic protein expression.  In contrast, psilocybin is thought to activate the glutamatergic synapse via non-selective agonism at the 5-HT2A receptor, leading to increased α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor activation, brain derived neurotrophic factor (BDNF) secretion, and mechanistic target of rapamycin (mTOR) activation, promoting dendritic spine growth and synapse formation in the cortex. 

We have generated five iPSC lines derived from individuals with treatment-resistant depression (TRD) in addition to two commercially available lines to differentiate neural cultures.  iPSC clones that displayed a normal karyotype were selected, and neural progenitor cells (NPCs) were differentiated into a mixed cortical-like glutamatergic/GABAergic population for eight weeks. 

We then used a combination approach to characterize the effects of ketamine and its metabolite (2R,6R)-hydroxynorketamine (HNK) as well as that of several serotonergic psychedelics (SPs), including lysergic acid diethylamide (LSD), psilocybin, and 2,5-Dimethosy-4-iodoamphetamine (DOI), on synaptic protein expression and dendritic spine morphogenesis. Neuronal cultures treated at multiple doses (1-10uM) and isolated at different times (1-24h) were used for Western blot to quantify specific protein expression levels (p-TrkB/TrkB, p-mTOR/mTOR, p-ERK/ERK, p-eEF2/eEF2, p-elF4E/elF4E, p-4EBP1/4EBP1, GluA1, NR2B, Dab1, PSD-95, Synapsin I, Synaptotagmin).  Another set of differentiated neuronal cultures were stained with various primary antibodies (PSD-95, Synapsin I, MAP2, etc) and fluorescent-secondary antibodies to visualize and quantify neurite synaptic density and spine density.  Immunohistochemistry images acquired via confocal microscopy are currently under quantification and pending analysis. 

In addition, batches of collected neurons and extracted RNA samples were sequenced by scRNAseq for eight and bulk RNAseq for 141 samples.  Despite the technical challenges associated with the iPSC approach to studying downstream pathways for ketamine, HNK and SPs, we have generated a significant amount of transcriptomic data.  We have completed the basic QC pipeline for the bulk and single cell transcriptomic data. 

Project 1: Bulk RNA seq analysis

Part A:

Correlation of known covariates

SVA

Principal Component Analysis

Correlation of known covariates with PCs

Correlation of known covariates with SVs

Find the right covariates to include in the model

Differential Expression Analysis – using mixed models – time series and dosage effects

Functional Enrichment Analysis

Part B:

Weighted Gene Co-expression Network Analysis (WGCNA) for clusters using the module eigengene or an intramodular hub gene, for relating modules to one another and to external sample traits (using eigengene network methodology)

Find modules associated with treatment and dosages

Network-based gene screening to identify candidate biomarkers or therapeutic targets

GO, KEGG, or GSEA pathway analysis for WGCNA modules

Isogenic Treatment Analysis: compare within cell line

Annotation of differentially expressed genes for biological relevance

Project 2: Single cell data analysis

Quality control and filtering

Batch/line effect correction via integration/harmony

Unsupervised clustering

Cell type annotation (based on similar reference datasets or top marker genes)

Trajectory inference (pseudotime)

Differential gene expression analysis for each treatment (while accounting for batch/line as a fixed or random effect)