Please note this is an old version of this entry, which may differ significantly from the current revision.
Rare diseases are conditions that affect a small proportion of the population (fewer than 200,000 persons in the USA or fewer than one in 2000 in Europe). The Orphanet portal for rare diseases and orphan drugs (http://www.orpha.net, accessed on 15 September 2021) currently lists more than 5800 rare diseases. Many are genetically inherited and the genetic causes are clearly identified. From the beginning of the human PSC history, rare genetic disorders have been attractive models for proof-of-concept studies of disease modelling. hESC derived from embryos after pre-implantation genetic diagnosis were a first source of Human pluripotent stem cells (PSCs) with natural, disease-inducing mutations.
1. Fragile-X Syndrome (FXS)
The first advantage is to be able to explore genes and pathways expressed at physiological levels without the need to introduce exogenous gene expression systems. This was exemplified in studies that screened compounds for genetic forms of ASD induced by the loss of function of FRMP 
and duplication of a segment of chromosome 7 
. FXS was the first neurodevelopmental disorder to be modeled with this aim using iPSCs. FXS is a neurodevelopmental disorder characterized by mild-to-severe intellectual disability and abnormal behaviors, such as attention deficit, anxiety and depression 
. FXS is also the most common known monogenic cause of autism, with 43–67% of male patients meeting the criteria of autism spectrum disorders. At the genetic level, FXS is linked to mutations (triplet repeats) in the 5′-untranslated region of the fragile X mental retardation 1 (FMR1) gene, which results in the absence of the FMRP protein. FMRP is a brain-specific RNA-binding protein that regulates the transport and translation of many mRNAs that play an important role in learning and memory 
. Consequently, screenings were conducted in neural progenitors to identify compounds that could increase FRMP levels. Kaufmann and collaborators used FXS-patient iPSCs to develop an image-based HTS assay measuring the levels of FRMP in neural stem cells using immunofluorescence 
. In all, 50,000 compounds were screened, including epigenetic regulators with known mode of action (7%), molecules covering a broad chemical space and biological diversity (46%) and a set of randomly selected compounds from an internal archive (47%). Four hits were identified and further confirmed for efficacy and absence of toxicity in dose-response experiments but were not further investigated for their mode of action or evaluated in an animal model. At the same time, Kumari and collaborators described the screening of 5000 known tool compounds and approved drugs in neural stem cells differentiated from an FXS patient-derived iPSC line using time-resolved fluorescence resonance energy transfer assay for FMRP detection 
. Interestingly, the primary screening was performed in a 1536-well plate format, a format rarely used for cell-based assays and six compounds were identified that modestly increased FMR1 gene expression in FXS patient cells. Although none of these studies resulted in clinically relevant compounds, they provide strong proof of principle of the assays performed on patient-derived neural stem cells in a very high-throughput format to identify new lead compounds for FXS drug development. More recently, Li and collaborator used the newly described CRISPR/Cas9 system to create a reporter line for detecting FMR1 gene reactivation in human neural cells and used it to screen 1262 bioactive compounds 
. This revealed two epigenetic regulators, 5-aza-dC and 5-aza-C, that significantly restored FRMP levels in disease cells. This study demonstrated that CRISPR/Cas9 can successfully be combined with iPSC-derived neural cells to design customized screening lines by knocking the luciferase reporter into endogenous target genes in order to obtain reporter lines and to reduce screening costs while increasing screening performance. This was possible only because iPSC-derived neural stem cells physiologically express endogenous levels of FRMP and proved the value of screening in human cells differentiated from PSCs. Together, these three studies demonstrate the feasibility and relevance of HTS in the neural progeny of PSCs for neurodevelopmental disorders.
2. Duplication of a Segment of Chromosome 7 (7Dup)
Validation that screenings can successfully be conducted for larger genetic aneuploidies was then reported. Duplication of a segment of chromosome 7 at 7q11 comprising 26–28 genes is one of the best-characterized copy number variations (CNVs) underlying autism. 7Dup patients show a range of autism spectrum disorder traits, especially varying degrees of language impairments and social restrictions 
. Among the genes of the 7q11.23 region, general transcription factor II-I (GTF2I) has key relevance. This gene mediates signal-dependent transcription and plays a prominent role in various signaling pathways 
. Most importantly, convergent evidences have implicated GTF2I as a major mediator of the cognitive–behavioral alterations in 7Dup 
. Interestingly, deletion of this gene is also related to another rare disease, the Williams–Beuren syndrome. Cavallo and collaborators screened, using RT-PCR, 1478 compounds for their potential to increase GTF2I mRNA levels in 7Dup iPSC-derived cortical glutamatergic neurons. Some HDAC inhibitors were selected and further validated by quantifying the modulation of genes included in the segment duplication and involved in the Williams–Beuren syndrome 
3. Metabolic Disorders
Neurodegeneration is another aspect that requires authentic neurons to develop predictive models for drug screening, since these post-mitotic cells are more sensitive to metabolic stressors than peripheral cells. This was illustrated by studies of GM1 gangliosidosis and Lesch–Nyhan disease 
. GM1 gangliosidosis is a lysosomal storage disorder characterized by abnormal accumulation of GM1 ganglioside. The main clinical feature of the disease is neural dysfunction due to massive GM1 ganglioside deposition in the central nervous system 
. This abnormal deposition is caused by a deficiency in lysosomal β-galactosidase (β-GAL) activity which limits the body’s ability to degrade GM1 ganglioside in lysosomes leading to excessive GM1 ganglioside accumulation and eventual impairment of several pathways, including the unfolded protein response (UPR), endoplasmic reticulum calcium signaling and autophagy. Altogether, this induces progressive neurodegeneration. Kajihara and collaborators generated induced pluripotent stem cells (iPSCs) derived from patients with GM1 gangliosidosis, differentiated neurons and developed an image-based HTS assay to detect GM1 ganglioside accumulation. A collection of 2217 compounds containing already approved drugs and major chemicals used in pathway analyses was screened. The two best compounds, amodiaquine and thiethylperazine, were then shown to restore the presynaptic deficit in disease-derived neurons, upregulate the enzymes responsible for lysosomal glycosphingolipid degradation and activate autophagy. Interestingly, the authors also validated the hit compounds in a mouse model of GM1 gangliosidosis, demonstrating efficacy in reducing ganglioside accumulation in the brain and protecting it from degeneration 
LND is caused by deficiency of the purine salvage pathway enzyme hypoxanthine-guanine phosphoribosyltransferase (HGPRT), an X chromosome-encoded protein 
. LND is characterized by severe neuropsychiatric disorders, which present with choreoathetosis, dystonia, aggression and self-injurious behavior 
. Mutations in the HPRT1 gene, which code for HGPRT, are different for each individual but patients exhibiting the most severe neurological symptoms consistently have mutations that totally block protein synthesis 
. To optimize the chance that a compound is efficient for most children with LND, independent of the type of mutation affecting HPRT1 gene, one strategy is to identify the compounds that activate alternative metabolic pathways that compensate for the deficiency of purine salvage in a target-agnostic manner. From a metabolic point of view, neural stem cells and neurons mainly rely upon recycling as a source of purine, while most other somatic cells rely, instead, upon de novo synthesis, a specificity that renders the brain more vulnerable to HGPRT deficiency than other organs 
. In this context, the use of authentic human neural stem cells and neurons rather than peripheral cells such as fibroblast or blood cells was instrumental. Ruillier and collaborator decided to conduct a functional screening in neural stem cells and neurons derived from iPSCs of children affected with LND treated with azaserine, an inhibitor of the synthesis of purine de novo, in order to selectively induce cell death in HGPRT-deficient cells 
. More than 3000 molecules were screened for their ability to rescue HGPRT-deficient cells from azaserine toxicity. Six pharmacological compounds were identified, all possessing an adenosine moiety, that corrected HGPRT deficiency-associated neuronal phenotypes by promoting metabolic compensations in an HGPRT-independent manner. Among these compounds, S-adenosylmethionine was reported in several case studies to ease the neuropsychiatric symptoms in LND 
, demonstrating the relevance of the screening strategy.
4. Cyclin-Dependent Kinase-Like 5 (CDKL5) Deficiency
Working with authentic and neuronal networks offers the opportunity for phenotypic and functional screening. This is of particular interest for diseases that involve abnormal excitability such as epilepsy. This was exemplified in a study by Negraes and collaborators, who conducted a phenotypic and target agnostic assay monitoring spontaneous calcium activity in 3D neuronal cultures as a read-out for network electric activity, which is abnormally increased in CDK5L-deficient neurons 
. CDKL5 gene encodes for a serine/threonine kinase highly expressed in the central nervous system. Mutations in this gene cause CDKL5 deficiency disorder (CDD), characterized by neurodevelopmental delay, motor dysfunction, autistic features and early-onset intractable seizures, a defining trait that led to the standalone classification of this pathology 
. Patients iPSC-derived 3D cortical spheroids exhibited hyperexcitability as measured as spontaneous calcium oscillations allowing a collection of 1112 compounds modulating different neuronal signaling pathways to be screened. Ivabradine, solifenacin, AZD1080 and crenigacestat were shown to reverse the phenotypic abnormality and were further investigated for their ability to ameliorate other CDD cellular phenotypes, including outward radial cellular migration defects. Due to their ability to regulate abnormal epileptic electrical activity in human neurons, these compounds open new therapeutic opportunities for other types of pathologies that include intractable seizures regardless of the initial trigger.
5. Phelan–McDermid Syndrome (PMS)
One final advantage of modelling neurological disorders with authentic neurons obtained from patient-derived iPSC is the development of personalized medicine strategies. This was achieved in PMS 
. This is the first study in which a compound identified by HTS in a patient-derived cell line was actually evaluated directly and repurposed in the same patient. PMS is a neurodevelopmental disorder characterized by global developmental delay, intellectual disability, severe speech delays, poor motor tone and function, and ASD 
. Genetic screening of the genome region identifies SHANK3 as the main gene involved in the ASD features associated with PMS. De novo truncating mutations inducing haploinsufficiency of the SHANK3 gene were estimated to be present in 0.69–2.12% of individuals with ASD 
. SHANK3 is an abundant component of the postsynaptic density, where it acts as a scaffolding protein recruiting key post-synaptic elements, such as glutamate receptors, and linking them to the actin cytoskeleton 
. Neurons differentiated from iPSCs of individuals with SHANK3 haploinsufficiency exhibited impaired electrophysiological responses to glutamatergic synapses’ stimulations, which could be corrected by re-introducing SHANK3 cDNA expression, validating their value for drug discovery 
. In this study, iPSCs were derived from two children with PMS in order to constitute neuronal networks in a screening format (384-well plates). Patient-derived neurons exhibit reduced SHANK3 mRNA and protein expression, reduced neurite size, decreased glutamatergic synapses and decreased spontaneous network activity. In all, 202 marketed drugs were tested on these neurons and two of them, lithium and valproic acid, were demonstrated to increase SHANK3 levels (mRNA and synaptic protein), rescue neurite length and synapse numbers and, at least partially, restore network activity. Lithium was consecutively administrated during one year to one of the two patients and clinical examination showed significant improvement in the child’s social performance. This study demonstrated the feasibility of using patient derived-iPSC to select patient specific treatment, an approach described as personalized medicine.
PSC-derived neural stem cell and neurons can be suitable biological materials to conduct compound screening at high throughput, can allow compounds that modulate endogenous targets that are not physiologically expressed in other cell models to be identified, are suitable for neurodevelopmental and neuropsychiatric diseases as well as neurodegenerative diseases and offer unique opportunities for phenotypic screening. This opens a path for research in prevalent multifactorial diseases and the promotion of precision medicine.
This entry is adapted from the peer-reviewed paper 10.3390/cells10123290