Modeling Psychiatric Disorder Biology with Stem Cells

Abstract

Purpose of review: We review the ways in which stem cells are used in psychiatric disease research, including the related advances in gene editing and directed cell differentiation.

Recent findings: The recent development of induced pluripotent stem cell (iPSC) technologies has created new possibilities for the study of psychiatric disease. iPSCs can be derived from patients or controls and differentiated to an array of neuronal and non-neuronal cell types. Their genomes can be edited as desired, and they can be assessed for a variety of phenotypes. This makes them especially interesting for studying genetic variation, which is particularly useful today now that our knowledge on the genetics of psychiatric disease is quickly expanding. The recent advances in cell engineering have led to powerful new methods for studying psychiatric illness including schizophrenia, bipolar disorder, and autism. There is a wide array of possible applications as illustrated by the many examples from the literature, most of which are cited here.

Keywords: Autism; Bipolar disorder; CRISPR/Cas9; Differentiation; Disease modeling; Embryonic stem cells; Genome editing; Induced pluripotent stem cells; Neurons; Organoids; Schizophrenia.

Fig. 1

Neuropsychiatric disease modeling using iPSCs. Somatic cells, like skin fibroblast, are collected from patients and healthy donors. Yamamaka transcription factors (OCT4, SOX2, KLF4, and c-MYC) are introduced to the cells, reprogramming them to a pluripotent state. These cells can be genetically manipulated using gene-editing tools to generate isogenic lines with specific disease-associated variants. Healthy control, patient, and genetically edited isogenic cells are then differentiated into an array of cell types and subtypes. The iPSC-derived neuronal and non-neuronal cells are then used to study differences between healthy and patient cells, or understand the consequences of disease associate variants using isogenic lines

Fig. 2

CRISPR/Cas9 scarless editing strategies in stem cells (a) mechanism of CRISPR/Cas9 editing. The target site in the genome consists of a 20 bp DNA protospacer sequence (blue) located directly upstream of a protospacer adjacent motif (PAM) site (green). The DNA protospacer is recognized and bound by the spacer sequence of a single-guided RNA (sgRNA, purple). The sgRNA recruits Cas9, which recognizes the PAM site (NGG for the most commonly utilized Cas9 species from S. pyogenes). Cas9 creates a double-stranded cut in the DNA protospacer three base pairs upstream of the PAM site. The resulting dsDNA break can be repaired by non-homologous end joining (NHEJ) resulting in small indels, or alternatively by homology-directed repair (HDR) if a repair template is provided resulting in precise repair which may include the addition of a desired mutation. b–e Scarless editing strategies employed in stem cells. For simplicity, only the spacer region of the sgRNA is depicted. b The simplest strategy is to choose sgRNAs for which the variant of interest itself acts as a blocking mutation for the sgRNA, the PAM site, or both. In this scenario, once the variant is introduced via HDR, further Cas9 cleavage will be prevented, NHEJ will be substantially reduced, and the efficiency of precise base pair editing will increase [205, 224]. Zygosity of edited cells can be modulated by choosing sgRNAs in which M is located in the 5′ end (more distal to the cleavage site), as the further the variant is from the cut site, the less efficient editing will be [205, 224]. This strategy is the most streamlined as it requires only one round of editing and off-target analysis of only one sgRNA. However, this strategy is not always feasible as it is entirely dependent on the existing architecture of the genomic region to be edited. c Gupta and colleagues’ two-step protocol [225] that interferes with Cas9 cleavage. In step 1, a sgRNA1 is chosen that is proximal to, but not overlapping with, M. The first step of HDR removes the PAM site associated with sgRNA1 to prevent further Cas9 cleavage. It also replaces the nearby reference allele of M with a new PAM site. The second step utilizes this new PAM site to create a Cas9 cut right next where to M is then introduced, and the new PAM site destroyed. This not only increases efficiency by preventing Cas9 cleavage but also efficiently utilizes the property of HDR that a Cas9 cleavage close to the variant of interest will have high efficiency [205, 224]. This approach is ideal for genomic regions where PAM sites are sparse. d The CORRECT method [205, 224] consists of two rounds of HDR and can target either the sgRNA binding site (re-Guide) or the PAM site (re-Cas). The first round of editing introduces the mutation of interest (M) as well as a blocking mutation (b, red) that disrupts either the sgRNA binding site (re-Guide) or alters the PAM site from the Cas9-recognized NGG to the VRER-Cas9-recognized NGCG (re-Cas). Edited clones are identified through restriction fragment length polymorphism (RLFP) analysis and confirmed by sequencing. A second round of HDR uses a Re-sgRNA containing the B mutation (re-Guide) or VRER-Cas9 (re-Cas) to recognize the sites altered in the first round, and then corrects the B mutation using the CORRECT HDR template while leaving the M mutation intact. The corrected B site then can no longer be targeted by the modified sgRNA or VRER-Cas9. e Summary of two different methods (referred to as piggyBac, PB, and MACS) involving the genomic insertion, then deletion of a positive/negative selection cassette which incorporates M along the way. In Step 1, a sgRNA targets Cas9 to a genomic region of interest, which creates a DSB and facilitates HDR from a donor plasmid. The donor plasmid contains a positive/negative selection cassette (blue, PB, or MACS) flanked by homology arms of 400–1000 bp (black) matching the genomic region, one of which includes M. For the piggyBac system, the cassette must be immediately flanked by inverted repeats and a “TTAA” motif (not shown) and integrated into an endogenous “TTAA” sequence near the DSB for excision purposes. Once HDR occurs, the cassette and M are stably integrated into the genome. HDR clones are enriched by positive selection with antibiotics (piggyBac), or by magnetic bead-assisted cell sorting (MACS) and manual picking of mCherry positive clones (MACS). Genotyping is performed by PCR and sequencing. In step 2, the selection cassette is excised. In the piggyBac method, the cells are treated with piggyBac transposase, which recognizes the TTAA motifs and inverted repeats flanking the cassette, resulting in scarless excision leaving the original TTAA intact. In the MACS method, a second round of CRISPR/Cas9 is performed using a sgRNA targeting the junction between the genome and the cassette (sgRNA 2). A second step of HDR removes the selection cassette, thus destroying the sgRNA binding site, but leaves M intact. Negative selection is then performed against cells that retained the cassette with drug treatment (piggyBac) or MACS (MACS) and screening performed by PCR. These strategies have the benefit of allowing cells to recover after transfection before selection, and screening is easy with PCR. However, the efficacy of this approach is limited by the requirement of the TTAA motif for the piggyBac system (which is much less frequent in the genome than Cas9’s NGG) and the laborious processes of vector assembly and multiple rounds of selection