Comparison of Model Systems for Emulating Human Tissue and Physiology in Psychiatric Research

2D Models of Psychiatric Disorders

2D models utilizing human pluripotent stem cells (hPSCs) offer a human-specific framework for investigating neurodevelopmental and psychiatric disorders. These conditions are difficult to analyze using conventional animal models, due to their subjective symptoms and intricate genetic complex backgrounds. Further, patient-derived hiPSCs retain the individual’s genetic profile, allowing researchers to understand how genetic factors contribute to disease phenotypes. 2D hiPSC models have been used to better understand the underlying cellular and molecular mechanisms that underlie a variety of pyschiatric disorders including ASD, SCZ, and BD [3]. Table 1 summarizes this.

Coculturing, which merges various cell types in a single culture system, can be used to investigate how different cell types influence each other, termed non-cell-autonomous interactions. It also allows for the examination of contact-dependent and independent cell interactions [4]. This enhances 2D models by more accurately mimicking the underlying mechanisms of neuropsychiatric disorder pathogenesis. Cocultures do have certain limitations, in that they lack tissue architecture, cell-type diversity, dynamic growth expansion, and maturation.

The main benefit of 2D models compared to 3D organoid models lies in their simplicity, reproducibility, and scalability, enabling larger experiments and more extensive high-throughput screening. However, the limitations of 2D cell culture models, including their lack of tissue complexity and inability to replicate key in vivo conditions, significantly limit their effectiveness in elucidating the pathophysiology of psychiatric disorders. These challenges underscore the critical need for organoid models that offer a more physiologically relevant and complex platform for understanding the intricacies of psychiatric disease mechanisms.

Idiopathic Autism Spectrum Disorder

iPSCs retain the unique genetic background of the individual which is paramount for studying ASD where roughly 80-85 percent of the population is idiopathic in origin. Research shows that neural cells derived from induced pluripotent stem cells (iPSCs), including both precursor cells and neurons that have undergone maturation, closely mimic the characteristics of human cortical cells during the embryonic stage, specifically between weeks 8 to 24. Therefore, these iPSC-derived neural cells serve as an ideal tool for investigating the initial stages of human neurodevelopment, especially when exploring genetically complex disorders such as autism [5].

Synaptic Dynamics and Connectivity

In ASD, one of the key cellular phenotypes is altered synaptic formation and pruning. This leads to changes in neuronal connectivity, which are thought to underpin many of the behavioral characteristics of ASD [6]. A 2021 study exploring the role of the innate immune complement system in ASD found that complement c4 was reduced in iPSC-derived astrocytes from people with ASD. Since astrocytes participate in synapse elimination, diminished c4 levels are associated with defective synaptic pruning, thus leading to atypically enhanced brain connectivity in ASD [7].

Neuronal Differentiation and Maturation

In general, aberrant neuronal maturation, varied neuronal differentiation, and synaptic formation have been implicated in iPSC models focused on APD [8]. Given the association between ASD and macrocephaly, excessive neural growth was also proposed to be associated with ASD. This has been demonstrated in iPSC models which showed increased neutral progenitor cell growth compared to controls. Additionally, this increased NPC growth in ASD individuals with macrocephaly was associated with an altered DNA replication program and increased DNA damage [9]. Increased NPC growth was hypothesized to be due to the dysregulation of a beta-catenin/BRN2 transcriptional cascade. Neural networks formed from ASD-derived neuronal were found to be dysfunctional, characterized by abnormal neurogenesis and reduced synapse formation. Insulin growth factor 1, a drug in clinical trial for ASD, was able to rescue defects in this neural network, demonstrating the potential utility of iPSC models in modeling cellular mechanisms of therapeutic compounds [10].

Glial Cell Function and Neuron-Glia Interactions

Additionally, there is evidence of increased activation of microglia and astrocytes in the brain, suggesting an inflammatory response. Co culture models investigating the interactions between neurons and astrocytes from individuals with idiopathic ASD using iPSCs have also been illustrative. ASD-derived neurons showed a significant decrease in synaptic gene expression and protein levels, glutamate neurotransmitter release, and reduced spontaneous firing rate. ASD-derived astrocytes interfered with proper neuronal development but this was rescued with control-derived astrocytes. Finally, blocking interleukin-6 secretion from astrocytes resulted in increased synaptogenesis in ASD-derived co cultures [11].

Molecular and Genetic Signaling Pathways

In 2015, Griesi-Oliveira et al. performed the inaugural iPSC research on non-syndromic ASD, targeting a unique TRPC6 mutation. They discovered that disruption in TRPC6 caused diminished calcium signaling and led to abnormal neuronal development, marked by reduced neurite length and complexity, along with a scarcity of dendritic spines. Correcting these defects was achievable through either activating TRPC6 or enhancing its expression, highlighting TRPC6’s significance in ASD’s neurodevelopmental mechanisms [12].

Neuroinflammatory Responses and Immune System Interactions

Emerging research into ASD points to neuroinflammation and immune system dysregulation as pivotal factors in understanding the disorder’s causes and progression. Research utilizing iPSC-derived neurons highlights the critical role of interferon-$\gamma$ (IFN-$\gamma$) signaling in neurodevelopmental disorders. IFN-$\gamma$ influences neurite outgrowth by up-regulating MHCI genes via mediation by promyelocytic leukemia protein nuclear bodies. Additionally, it alters gene expression associated with autism and schizophrenia, underscoring the interaction between genetic and environmental factors [13].

Neurotransmitter Systems and Imbalances

Imbalances in the excitatory-inhibitory neurotransmitter systems, particularly involving glutamatergic and GABAergic neurons, are also commonly observed in ASD. These imbalances can affect neural network functioning and are linked to the sensory and behavioral symptoms characteristic of the disorder. In a study utilizing co-cultures of iPSC-derived glutamatergic and GABAergic neurons, the focus was on Cadherin-13’s (CDH13) pivotal role in regulating the excitation/inhibition (E/I) balance. Findings indicate that a lack of CDH13 in GABAergic neurons leads to a disrupted E/I balance, primarily through enhanced inhibition [14].

Transcriptomic and Proteomic Profiling

An examination of the transcriptome analysis was performed between normal controls and iPSC-derived neuronal cells from normocephalic ASD. Researchers identified dysregulation in gene modules associated with protein synthesis in neuronal progenitor cells (NPC), synapse/neurotransmission, and translation in neurons. Proteomic analysis of NPCs showed potential molecular links between these dysregulated modules in NPCs and neurons. A synapse-related module was consistently upregulated in iPSC-derived neurons (similar to fetal brain expression) and downregulated in postmortem brain tissues of ASD patients, pointing to its potential as an ASD biomarker and therapeutic target due to its varying dysregulation across developmental stages in ASD individuals. This research demonstrates that traditional 2D models, when integrated with newer technologies such as RNA-seq, transcriptomic analysis, differential gene expression profiling, and the prediction of temporal and regional identity of iPSC-derived neurons through machine learning algorithms, can still provide valuable insights into complex biological processes [15].

3D Models: Organoids

Organoids are 3D structures derived from human stem cells in vitro that mimic the complex biological functions of real organs. This technology represents a significant leap from traditional two-dimensional cell cultures, offering a more accurate and dynamic model for studying patient-specific developments, disease processes, and drug responses. Typical brain organoid models are differentiated from iPSCs in a stepwise manner and grown in Matrigel or other ECM analog using Lancaster’s protocol and the hanging drop method.

In this review, we focus on psychiatric disorders and the advances in integrating cerebral organoids into the disease-modeling research framework. We discuss the typical cellular phenotypes of multiple psychiatric disorders such as ASD, SZ and BD. We aim to identify the advantages of 3D modeling over 2D modeling and discuss the recent studies performed using these models in various neuropsychiatric diseases.

Psychiatric disorders are as such extremely difficult to study and its diagnosis involves components like environment, mood and psychosis. It is very difficult to translate this in an accurate manner to gain complete understanding. The absence of markers like cerebrospinal fluid, brain imaging and EEGs unlike those in neurological disease complicates it further. One way to robustly replicate the disorder is through using 3D models like cerebral organoids as they have the ability to self-organize. These models focus on modelling cellular aspects of the disorder such as cell proliferation, migration, lineage trajectory and differentiation.

Specifically, cerebral organoids have been used by researchers to understand and observe abnormalities in abnormal psychiatric conditions, model brain development, and as frameworks for preclinical drug testing regimens and personalized medicine [42]. Since commonly highlighted psychiatric disorders such as ASD, SZ, anxiety, depression, PTSD and BD involve multiple brain regions, cerebral organoids specific to particular regions can be cultured using human induce pluripotent stem cells (hiPSCs) and fused to form assembloids. This is a powerful tool for scientists to identify multiregional affects and how the connectivity between them contributes towards the disorder, defects in cellular organization and the underlying biological [43,44].

Table 2. Summary of Cellular Phenotypes in 3D Models.

Disease	Cell Type	Region	Observations/Findings	References
BD	Neural Progenitor Cells	Cortex	Store-operated Ca release dysregulated and attenuated in BD NPC lines compared to controls. Thinning of subventricular zone observed in BD organoid lines.	[26]
BD	Neurons	Cortex	Enrichment of genes related to ion storage and homeostasis in BD-derived cortical organoids. Store-operated Ca release dysregulated and attenuated in BD cerebral organoids. BD-derived cortical organoids contain lower proportions of neurons and elevated numbers of radial glial cells. Decreased neuron excitability observed in BD-derived cortex organoids but excitability was rescued by treatment with lithium. DEGs between treated and untreated organoids associated with Na+ homeostasis and regulation of IL-β and TNF-α.	[26,62,63]
BD	Excitatory Cortical Neurons	Cortex	BD Type I-derived organoids display downregulation of cell adhesion associated with abnormal NCAN expression, and downregulation of genes associated with GABA uptake/release. ER-Mitochondria contact sites markedly reduced in BD-derived organoids compared to controls, in both perinuclear region and neurites.	[61]
SCZ	Neurons	Cortex (Dorsal Forebrain), Ventricles	GWAS studies by Kathuria et. al. reveal 23% of genes are differentially expressed in SCZ organoids compared to controls. Significant decreases in mitochondrial activity and cellular respiration are observed in SCZ organoid lines. MEA studies reveal diminished response to stimulation/depolarization in SCZ organoids. BRN2 and PTN are depleted in Scz neurons relative to controls.	[53,55]
SCZ	Non-Neurons	Ventricles	Elevated numbers of cells in ventricular regions of SCZ organoids directed towards myeloid, mural, and endothelial lineages in comparison to controls.	[53]
SCZ	Neurons	Thalamus, Cerebellum	Impaired connection between medial dorsal nucleus of the thalamus and cerebellum	[44]
ASD	Excitatory Neurons	Cortical Plate	Elevated numbers of cortical plate excitatory neurons (ENs) in macrocephalic ASD lines at the expense of preplate ENs, which were diminished. Cortical interneurons are also overproduced in comparison with controls due to DLX6 overexpression.	[49,64]
ASD	Radial Glial Cells	Cortical Plate	Lamination of cortical plate is reduced, and cortical projection neuron maturation is accelerated in SYNGAP1 haploinsufficient organoids. Similar findings observed in vivo in mice.	[49]
ASD	NPCs	General	Excessive growth of NPCs due to mutations in genes such as SUV420H1, CHD8, and PTEN associated with ASD risk and macrocephaly.	[46,47,48]
ASD	GABAergic Neurons	Cortex	Overexpression of GABAergic pathways in ASD organoids results in heightened percentages of GABAergic neurons (as high as 50% of all cells in the organoid) compared to controls.	[46,48]

Table 2. Summary of Cellular Phenotypes in 3D Models.

Disease	Cell Type	Region	Observations/Findings	References
BD	Neural Progenitor Cells	Cortex	Store-operated Ca release dysregulated and attenuated in BD NPC lines compared to controls. Thinning of subventricular zone observed in BD organoid lines.	[26]
BD	Neurons	Cortex	Enrichment of genes related to ion storage and homeostasis in BD-derived cortical organoids. Store-operated Ca release dysregulated and attenuated in BD cerebral organoids. BD-derived cortical organoids contain lower proportions of neurons and elevated numbers of radial glial cells. Decreased neuron excitability observed in BD-derived cortex organoids but excitability was rescued by treatment with lithium. DEGs between treated and untreated organoids associated with Na+ homeostasis and regulation of IL-β and TNF-α.	[26,62,63]
BD	Excitatory Cortical Neurons	Cortex	BD Type I-derived organoids display downregulation of cell adhesion associated with abnormal NCAN expression, and downregulation of genes associated with GABA uptake/release. ER-Mitochondria contact sites markedly reduced in BD-derived organoids compared to controls, in both perinuclear region and neurites.	[61]
SCZ	Neurons	Cortex (Dorsal Forebrain), Ventricles	GWAS studies by Kathuria et. al. reveal 23% of genes are differentially expressed in SCZ organoids compared to controls. Significant decreases in mitochondrial activity and cellular respiration are observed in SCZ organoid lines. MEA studies reveal diminished response to stimulation/depolarization in SCZ organoids. BRN2 and PTN are depleted in Scz neurons relative to controls.	[53,55]
SCZ	Non-Neurons	Ventricles	Elevated numbers of cells in ventricular regions of SCZ organoids directed towards myeloid, mural, and endothelial lineages in comparison to controls.	[53]
SCZ	Neurons	Thalamus, Cerebellum	Impaired connection between medial dorsal nucleus of the thalamus and cerebellum	[44]
ASD	Excitatory Neurons	Cortical Plate	Elevated numbers of cortical plate excitatory neurons (ENs) in macrocephalic ASD lines at the expense of preplate ENs, which were diminished. Cortical interneurons are also overproduced in comparison with controls due to DLX6 overexpression.	[49,64]
ASD	Radial Glial Cells	Cortical Plate	Lamination of cortical plate is reduced, and cortical projection neuron maturation is accelerated in SYNGAP1 haploinsufficient organoids. Similar findings observed in vivo in mice.	[49]
ASD	NPCs	General	Excessive growth of NPCs due to mutations in genes such as SUV420H1, CHD8, and PTEN associated with ASD risk and macrocephaly.	[46,47,48]
ASD	GABAergic Neurons	Cortex	Overexpression of GABAergic pathways in ASD organoids results in heightened percentages of GABAergic neurons (as high as 50% of all cells in the organoid) compared to controls.	[46,48]

Idiopathic Autism Spectrum Disorder

Autism spectrum disorders are characterized by an array of symptoms that can range from mild to severe, including repetitive behaviors, social interaction difficulties, and impaired neurodevelopment. This review focuses on idiopathic autism spectrum disorders, which occur spontaneously and have unclear and polygenic causal factors, intriguing scientists. Studying these complexities requires a comprehensive understanding of brain development and the relationship between cellular and molecular results associated with clinical symptoms.

Previous sections have highlighted extensive studies using 2D and animal models, but the limitations of both hinder the clinical translation of findings. Studies have shown that culturing cells in monolayers alters their genetic expressions. In rodent models of ASD, neurons originate from the subventricular zone, whereas in humans, they originate from proliferative regions outside this zone [45]. To bridge species-specific differences, human brain organoids derived from induced pluripotent stem cells (iPSCs) are now being modeled to gain deeper insights into the etiology of autism spectrum disorder.

Macrocephaly is one of the most well-studied phenotypes associated with ASD, leading researchers to use organoids in studying its etiology. Brain organoids derived from individuals with macrocephalic ASD show larger sizes compared to control organoids due to the involvement of tumor suppressors such as PTEN, CHD8, and RAB39B. This increase in organoid size is also attributed to the overexpression of neural progenitor cells (NPCs) caused by impaired maturation during development [46]. In particular, organoids derived from a cells with shortened form of CHD8 contribute to the increase in organoid size and are implicated in macrocephalic autism, along with abnormal PTEN and SUV420H1 expression, an imbalance of neuronal types and increased differentiation of GABAergic neurons [46,47,48].

Additionally, organoid models have implicated imbalances in neuronal subtypes as possible causes for behavioral symptoms of autism. The overexpression of upper-layer colossal projection neurons, which are responsible for human social behavior, is observed in autism and can be further explored using organoid models. Cortical organoid models were created to illustrate that the expression of SYNGAP1 in human radial glia is reduced in individuals with ASD. This reduction impacts the lamination of the cortical plate and influences the maturation of cortical projection neurons [49]. Additionally, increased expression of DLX6 leads to the overproduction of cortical interneurons compared to control organoids. Cortical organoids derived from iPSCs affected by idiopathic ASD show upregulated expression of the transcription factor forkhead box G1, leading to shortened cell cycle length, abnormal cell proliferation, and unbalanced neuron differentiation. The imbalance between inhibitory and excitatory neurons during early development suggests accelerated brain growth in the second trimester [50]. This imbalance is also studied in forebrain organoids, comparing the cellular and molecular profiles of macrocephalic and non-macrocephalic ASD. The analysis concludes that there is an overlap between the two in terms of upregulated genes in transcription, with patient-derived organoids showing decreased cortical area-specific transcripts [51]. The size of organoids is also assessed using proteomic techniques and time-point assessments [52].

Schizophrenia

Schizophrenia, a neurodevelopmental disorder with an varied molecular origin, was initially modeled in 2011 using patient-derived hiPSCs carrying mutated DISC1 genes. Subsequent research delved into investigating dysregulation in synapse-associated gene transcription, defective synaptic vesicle release, cell cycle regulation, and its impact on cortical size. Progress in this area led to the utilization of 3D modeling techniques, specifically cerebral organoids, to dissect the cellular phenotype of schizophrenia. In 2011, the Gage lab demonstrated neuronal phenotypic changes derived from patient cells, revealing reduced neuronal connectivity, decreased levels of PSD-95 protein, alterations in glutamate expression, and dysregulation in cAMP and Wnt signaling pathways. These advancements transitioned into using organoid models and conducting cell patterning assays, which identified decreased connectivity between the medial dorsal nucleus of the thalamus and cerebellum in schizophrenic patients [44]. Presently, organoid models mimicking neuropathology of schizophrenia are employed to assess risk factors such as amalgamate, PTN, and BRN2 [53]. Other research has focused on patterning, time-dependent organoid generation, abnormal distribution of proliferating (Ki67+)neural progenitor cells, and the role of genes like FGFR1 in cortical development [54]. In addition, studies have identified DEGs associated with significant decreases in mitochondrial activity and cellular respiration in SCZ organoid lines [55,56].

Bipolar Disorder

Bipolar disorder (BD), a neuropsychiatric condition, remains poorly understood in terms of its structural, cellular, and circuit-level characteristics. The absence of clear diagnostic biomarkers and the complexity of its behavioral traits pose significant challenges for treatment [57]. Given its involvement in higher-order brain functions and the presence of multiple cell types across various brain regions like the thalamus, cortex, and basal ganglia, utilizing 3D models such as cerebral organoids offers a potent strategy for analysis. Studies on patient-derived samples have revealed dysregulated genes and signaling pathways such as the Wnt pathway, which has also been observed in other neuropsychiatric disorders. BD-derived organoids exhibit reduced neuron count, resulting in decreased excitability and network activity, along with smaller size. Additionally, there is evidence of overexpression of genes related to membrane receptors and calcium channels [58]. Lithium has been used as a treatment in bipolar disorder for its mood-stabilizing properties; however, its mechanism was not completely understood until studies were performed using brain organoids. In patient derived organoids, the administration of lithium increases excitability, neuroprotection, mitochondrial reverse capacity and regulated pro-inflammatory cytokine secretions. This helped to identify Li-associated DEGs (differentially expressed genes) which can serve as targets for drug delivery [59]. iPSC-derived cerebral organoids from patients serve as a valuable model to understand aberrations in ER biology and electrophysiological responses, as well as the dysregulation of cell adhesion, upregulation of immune signaling genes, and neurodevelopmental aspects in bipolar disorder [60,61].