This research program is directed by Dr. Jubao Duan. The main goals are to better understand the molecular and neurobiological basis of major neuropsychiatric (mainly schizophrenia) and neurodegenerative (mainly Alzheimer’s disease) disorders by employing functional genomics approaches in hiPSC-derived neurons or cortical organoids. The main technical platforms include hiPSC reprogramming, 2D and 3D (organoids) neural models, CRISPR-based genome/epigenome editing, multi-omics, and single-cell genomics. Specific research areas include:
1) Functional interpretation of noncoding GWAS risk variants for neurodevelopmental disorders or traits.
Despite the hundreds of disease risk loci identified by GWAS, translating the genetic findings into disease biology and clinical interventions remains a big challenge. One of the major difficulties for the field is that most disease risk variants are in noncoding regions of the genome and lack functional interpretation. Using hiPSC-derived neurons as a neurodevelopmental cellular model, we have mapped variants that show differential allelic chromatin accessibility, i.e., allele-specific open chromatin (ASoC), in Assay for Transposase-Accessible Chromatin by sequencing (ATAC-seq) reads at heterozygous single nucleotide polymorphism (SNP) sites (Zhang, Science 2020). We found that neuronal ASoCs were partially driven by altered transcription-factor-binding, over-represented in brain gene enhancers and expression quantitative-trait-loci (eQTL) in the brain, and frequently associated with distal genes through chromatin contacts (Zhang, Science 2020). Notably, the neuronal ASoC SNPs exhibited strong enrichment for genetic variants associated with schizophrenia and several other brain disorders (Zhang, Science 2020). We continue to map disease risk variants that show ASoC and affect gene expression in various cell types and under different cellular states. We are also integrating cellular ASoC or chromatin accessibility-QTL (caQTL) data with other epigenomic datasets as well as biophysical properties of TF-DNA binding data to build computational models that can better infer functional genetic risk variants for neurodevelopmental and neurodegenerative disorders. Our research provides a powerful framework for identifying the likely disease causal variants/genes.
2) Modeling the molecular and cellular phenotypes of genetic risk variants for neuropsychiatric and neurodegenerative disorders.
Translating the mounting genetic discoveries into disease biology requires disease-relevant experimental models for linking genetic risk variants to molecular and cellular phenotypes. We are using both 2D and 3D hiPSC-derived cellular models to dissect the cellular and molecular phenotypes associated with disease risk variants, either common GWAS risk variants of small effect or rare protein-coding variants of high penetrance, mainly for schizophrenia and Alzheimer’s disease. Despite their small population effect sizes, we have demonstrated that some common GWAS risk variants can have detectable biological effects in hiPSC-derived cellular models (Forrest, Cell Stem Cell 2017; Zhang, Science 2020).
We also showed that the rare loss of function (LoF) of OTUD7A in the schizophrenia-associated 15q13.3 deletion impairs synapse development and function in human neurons (Kozlova, Am J Hum Genet 2022).
Our future research is geared to innovatively scale up the assay (more variants, more genetic backgrounds, and more phenotypes) to better understand the polygenicity and convergence of genetic risk factors of these complex disorders.
3) Build an iPSC bank of neurodegenerative disorders (iBoND) for translational research.
We believe that hiPSC models from a well-powered cohort of donors with clear-cut genetic risk factors and comprehensive clinical characteristics hold promise for translational research toward understanding novel disease biology and developing better clinical diagnosis and treatment approaches. Towards this end, we are building iBoND from more than 60 donors, mostly with selected genotypes at the APOE locus, the strongest Alzheimer’s disease genetic risk locus. With this resource, we generate different neural culture models, including brain organoids that can incorporate specific cell types, such as oligodendrocytes, to understand how APOE and polygenic risk factors can affect disease-relevant cellular phenotypes and the underlying molecular mechanisms. We anticipate these efforts will allow us to better understand some novel disease biology, such as how lipids contribute to disease progression and clinical outcomes. Moreover, studying these cellular models with defined genetic makeup will enable us to develop novel and more tailored therapeutic approaches.
4) Context-dependent gene regulation and gene network of neuropsychiatric disorders at single-cell resolution.
Many risk variants/genes likely act as gene networks in cell type-specific and context-dependent manners. We recently identified TCF4 as a master regulator of a schizophrenia gene network (Torshizi, Science Advances 2019). We envision that it is pivotal to identify disease risk variants and gene networks that function only in certain neuronal cell types and under specific conditions such as neural activation.
Towards this end, we are carrying out large-scale perturbation to model neural activation in hiPSC-derived neural co-culture models. Combined with single-cell multi-omics analyses, we are mapping eQTLs and context-specific caQTLs. We will also identify context-dependent gene networks that are relevant to neuropsychiatric disorders. The research will unravel disease biology from new dimensions, i.e., context-dependent regulation and gene network level.
Behavioral Genetics and Personalized Medicine
Besides ongoing work in schizophrenia genetics, both collaborating with Dr. Duan’s efforts and as part of large consortia such as the Psychiatric Genomics Consortium for Schizophrenia, primarily using our Molecular Genetic of Schizophrenia (MGS) collection (dbGaP phs000021 and phs000167), Dr. Sanders and his group are active in ongoing research on (1) the genetics of sexual orientation and related traits, and (2) personalized medicine.
Our past and recent research was primarily on male sexuality genetics, and our current NIH sexuality grant focuses on female sexual orientation. We also continue to participate in large-scale sexuality genetics meta-analyses.
Our efforts in personalized medicine include being part of a Health Provider Organization (HPO) consortium (Illinois Precision Medicine Consortium, IPMC), helping assemble NIH’s All of Us Research Program cohort of one million participants. In addition, NUH’s institutional research biobank, the Genomics Health Initiative (GHI), conducts research in personalized medicine and enables efforts of other NUH research groups; GHI currently has ~45K NUH enrollees, ~30K with DNA, and ~17K with genome-wide data (array, WES, and/or low-pass sequencing with imputation).