Modeling,tuberous,sclerosis,complex,with,human,induced,pluripotent,stem,cells

Weibo Niu·Benjamin Siciliano·Zhexing Wen

Abstract Background Tuberous sclerosis complex (TSC) is an autosomal dominant genetic disorder with a birth incidence of 1:6000 in the United States that is characterized by the growth of non-cancerous tumors in multiple organ systems including the brain,kidneys,lungs,and skin.Importantly,TSC is also associated with signif icant neurological manifestations including epilepsy,TSC-associated neuropsychiatric disorders,intellectual disabilities,and autism spectrum disorder.Mutations in the TSC1 or TSC2 genes are well-established causes of TSC,which lead to TSC1/TSC2 def iciency in organs and hyper-activation of the mammalian target of rapamycin signaling pathway.Animal models have been widely used to study the eff ect of TSC1/2 genes on the development and function of the brain.Despite considerable progress in understanding the molecular mechanisms underlying TSC in animal models,a human-specif ic model is urgently needed to investigate the eff ects of TSC1/2 mutations that are unique to human neurodevelopment.Data sources Literature reviews and research articles were published in PubMed-indexed journals.Results Human-induced pluripotent stem cells (iPSCs),which capture risk alleles that are identical to their donors and have the capacity to diff erentiate into virtually any cell type in the human body,pave the way for the empirical study of previously inaccessible biological systems such as the developing human brain.Conclusions In this review,we present an overview of the recent progress in modeling TSC with human iPSC models,the existing limitations,and potential directions for future research.

Keywords Induced pluripotent stem cells·Three-dimensional cultures·Tuberous sclerosis complex·Two-dimensional cultures

Tuberous sclerosis complex (TSC) is a rare and multi-system genetic disorder characterized by the growth of numerous benign tumors in all major organ systems [1],including the brain,heart,lungs,kidneys,skin,eyes,and teeth.TSC patients are also predisposed to the development of neurological disorders,such as epilepsy,TSC-associated neuropsychiatric disorders,intellectual disability,and autism spectrum disorder (ASD),which place signif icant burdens on these patients,their families,and society [1,2].The worldwide incidence rate of TSC is 1/6000-10,000 live births.Approximately 50,000 people in the United States suff er from TSC,and 1-2 million patients are affl icted by this disorder worldwide with no apparent gender,race,or ethnicity biases [1].

TSC is an autosomal dominant disease,resulting from inherited or spontaneous mutations inTSC1orTSC2,which respectively encode the hamartin and tuberin proteins.Together,these two proteins form a heterodimer that impedes the hyperactivation of the mammalian target of rapamycin complex 1 (mTORC1),a key regulator of cell growth,proliferation,and diff erentiation [1].A pathogenetic sequence variant inTSC1orTSC2is suffi cient to diagnose or predict TSC independent of clinical f indings[2].TheTSC1gene,which is located on chromosome 9q34.13,has a total length of 53,281 bp (GRCh38) and is comprised of 25 exons.To date,the Leiden Open Variation Database (LOVD) has reported a total of 4464TSC1variants,1356 of which are unique (https:// datab ases.lovd.nl/shared/ genes/ TSC1).The clinical signif icance,molecular consequence,and type of eachTSC1variant cataloged in the Clinvar database are outlined in Table 1.The molecular consequences of theseTSC1variants are predominately missense (53.6%),followed by untranslated region (UTR)(17.6%),then frameshift (16.7%),nonsense (8.1%),and splice site (4.0%).TheTSC2gene,which is located on chromosome 16p13.3,has a total length of 40,699 bp and contains 44 exons.Presently,4046 out of the 12,572 totalTSC2variants reported in the LOVD unique are unique(https:// datab ases.lovd.nl/ shared/ genes/ TSC2).The most common molecular consequence ofTSC2variants are missense (64.8%),followed by frameshift (15.0%),splice site (7.0%),nonsense (6.8%),and UTR (6.3%).These data are consistent with the f inding that approximately 20% of patients clinically diagnosed with TSC possess mutations inTSC1and 70% carry mutations inTSC2[1].Despite the thousands ofTSC1andTSC2pathogenic sequence variants that have been identif ied to date,de novovariants continue to be identif ied and added to these public databases.All manifestations of TSC occur more frequently and with greater severity in cases associated withTSC2rather thanTSC1mutations.However,there is still considerable variability in disease presentation between individuals carrying mutations in the same gene and even among family members with identical mutations [1].Indeed,it has also been reported that phenotypes and severity are not always genotype-dependent [3].To explore howTSC1/2mutations aff ect neural development,researchers have diff erentiated cortical spheroids from human embryonic stem cells (hESCs).After 20 days of diff erentiation,TSC1/2spheroids demonstrated rosette structures resembling cortical ventricular zones and the expression of the neural progenitor markers paired box protein Pax-6 (PAX6) and sex-determining region Y-box 2 (SOX2) were comparable to that of healthy controls,suggesting thatTSC1/2mutations do not strongly aff ect progenitor cell diff erentiation[4].Furthermore,germline inactivation of TSC1 or TSC2 followed by somatic second-hit mutation appears to reliably inactivate the remaining wild-type allele [1].

Table 1 The information of TSC1 and TSC2 genes from the Clinvar database

TSC was f irst described by a French physician named Désiré-Magloire Bourneville in 1880 [5] and the clinical diagnosis was first described by Gomez,Sampson,and Whittemore in 1999 [6].Clinical consensus guidelines on the diagnosis,surveillance and management of TSC were made available in 2013 [7,8] and more recently updated in 2021 [2].Animal models have widely been used to study the roles ofTSC1/2on the development and function of the brain and how variants in these genes contribute to TSC pathogenesis [9-11].However,interspecies diff erences in brain structure,as well as signif icant variance in the cellular and molecular compositions of developing brains [12,13],limit the relevance of these f indings to human patients.Therefore,it is imperative to establish models that precisely recapitulate human-specif ic TSC pathobiology to conf irm the knowledge acquired in animal models and translate these f indings to human patients.Recently,human induced pluripotent stem cells (hiPSCs) have been recognized for their unprecedented potential for studying TSC in vitro with the added advantage of retaining genetic backgrounds that are identical to their donor patients and with fewer ethical challenges than animal models or embryonic stem cells.In this review,we provide an overview of the recent progress in modeling TSC using two-dimensional (2D) and threedimensional (3D) cultures derived from hiPSCs,identify the current challenges,and propose future research directions.

Human neural progenitor cells models of tuberous sclerosis complex

TSC patients experience severe neurological manifestations,such as 40% of those with ASD and 50% with intellectual disabilities [1].Genetic variants associated with a higher risk of ASD risk genes were found to aff ect early phenotypes of neurogenesis including the proliferation of hiPSC-derived neural progenitor cells (NPCs) which mimic neurodevelopmental deficits observed in ASD patients [14,15].Moreover,researchers have demonstrated that NPC proliferation,process outgrowth,and migration prior to neuronal diff erentiation are pivotal in the pathogenesis of ASD as well as schizophrenia [14,16].Patientderived NPCs have been applied to reveal numerous early neurodevelopmental phenotypes in TSC patients (as is shown in Table 2).Clustered regularly interspaced palindromic repeats (CRISPR)/CRISPR-associated nuclease 9(Cas9)-mediated gene editing was used to establish heterozygous,homozygous,and corrected wild type lines of hiPSCs derived from a patient with TSC [17].NPCs diff erentiated fromTSC1heterozygous and homozygous hiPSC lines presented with increased size,enhanced proliferation,and altered neurite outgrowth relative to those differentiated from wild type hiPSCs.Transcriptome analysis inTSC1NPCs has further revealed genotype-dependent changes in the expression of genes associated with ASD,epilepsy,and intellectual disability including annexin A1,contactin-6,human leucocyte antigen-B,protocadherin 19(PCDH19),andPCDH10[17].Furthermore,TSC1heterozygous and homozygous NPCs display basal activation of extracellular signal-regulated kinase 1 and 2 that was further accentuated by treatment with rapamycin.InTSC1-def icient NPCs,MAP kinase-interacting kinase 1/2-eukaryotic translation initiation factor 4E signaling was also found elevated after rapamycin treatment,suggesting that abnormal translation inTSC1-def icient NPCs may play a role in neurodevelopmental def icient [17].Taken together,these f indings indicate that early neurodevelopmental phenotypes are altered inTSC1NPCs.

Additional studies have generatedTSC2patient-specific iPSCs and subsequently differentiated them into NPCs,neurons,and astrocytes to study the cell-specif ic pathogenesis of TSC [19].In doing so,they discovered thatTSC2 NPCs present with enhanced proliferation and greatly increased PAX6 expression in contrast with unaffected control cells.Additionally,somatic enlargement,abnormal process growth,and abnormal intercellular connections were observed in the neurons diff erentiated fromTSC2NPCs [19].These results partially reproduced the development of neural abnormalities known to occur in patients withTSC2mutation,indicating thatTSC2haploinsuffi ciency may play a key role in the neuropathology of TSC.To further validate the role ofTSC2heterozygosity in TSC neuropathology,NPCs were later induced and diff erentiated into neurons in vitro [18].The results of this investigation revealed that neuronal diff erentiation was delayed inTSC2heterozygous NPCs.Further research showed that abnormal phosphoinositide 3-kinase/protein kinase B signaling,rather than direct hyper-activation of mTORC1,likely contributed to delayed neuronal diff erentiation inTSC2heterozygous NPCs [18].These f indings off er novel insight into the mechanistic underpinnings of TSC pathogenesis and illuminate alternative targets for therapeutic intervention.

Human neuronal models of tuberous sclerosis complex

Loss ofTSC1/2function has been shown to result in neuronal hyperexcitability and synaptic dysfunction in human and rodent models,but the eff ects of these changes on central nervous system development remain unknown.Several iPSC-derived neuronal studies have been used to explore TSC pathogenesis (as is shown in Table 2).Induced expression of neurogenin 2 (NGN2) has been utilized to generate excitatory cortical neurons fromTSC2heterozygous and homozygous iPSCs [24].InTSC2+/-neurons,mTORC1 hyperactivation was found to be associated with enlarged soma and enhanced process outgrowth.These abnormalities were even more pronounced inTSC2-/-neurons,which also showed hyperactivity and transcriptional dysregulation in cortical tubers.These results suggest that the loss of one allele ofTSC2is suffi cient to contribute to changes in morphology and physiology of cortical neurons.Importantly,the abnormalities in theseTSC2def icient neurons could be reversed by rapamycin administration [24].It is worth noting that no signif icant diff erences in neuronal diff erentiation were observed between theTSC2heterozygous and homozygous NGN2 induced neurons [24].One possible explanation for this f inding is that the direct iPSC-to-neuron conversion achieved through NGN2 overexpression bypasses the intermediary NPC state and TSC-associated NPC irregularities mentioned in the previous section.

Neurological manifestations,including epileptic seizures,are the main clinical indicator of TSC.Epilepsy is the most common neurological symptom,aff ecting 90% of individuals with TSC [1].However,the molecular and cellular mechanisms underlying epilepsy in TSC patients remain unclear.Most recently,abnormally extended neurites were found in neurons derived fromTSC2-/-iPSCs [20].TheseTSC2-/-neurons also showed increased highly synchronized neuronal activity in the form of calcium ion spikes.Further investigation into the physiological properties ofTSC2-/-neurons,revealed that the abnormal neurite extension in these neurons was likely the result of elevated Ca2+inf lux by L-type Ca2+channels,thus providing a link between the TSC-mTOR pathway and Ca2+signaling [20].These f indings vitally elaborate our understanding of the molecular mechanisms underlying the abnormal neurodevelopment and psychiatric symptoms associated with TSC.

TSC is also associated with significant neurological manifestations including ASD.Recently,TSC2deficient neurons were generated from ASD patient-derived iPSCs[23].Using multielectrode array to explore the changes ofTSC2loss on neuronal network activity,they detected differences in the temporal synchronization of neuronal bursting and spatial connectivity between electrodes across networks of these ASD patient-derived neurons relative to controls [23].The results showed that ASD patient-derived neurons withTSC2gene mutations exhibit neuronal hyperactivity,resulting in dysfunctional neuronal networks as characterized by decreased synchronicity of neuronal bursting and lower spatial connectivity.Additionally,transcription analysis revealed that the expression of genes encoding enzymes and receptor proteins associated with glutamate and gamma-aminobutyric acid (GABA) signaling was significantly altered inTSC2mutant neurons.For instance,the expression of the glutamate receptor genesGRIN1,GRIN3A,andGRIA1,as well as the postsynaptic GABA receptor genesGABAα2,GABAβ1,andGABAγ1were all elevated in ASD patient-derived neuronsTSC2gene mutations [23].Taken together,these observations indicate that mutations inTSC2are associated with disruptions in synaptic excitatory-inhibitory (E/I) balance.

Previous studies have observed impaired autophagic f lux and an accumulation of organelles,including mitochondria,in neuropathological studies of TSC patients and rodent models.To thoroughly investigate the eff ects ofTSC2on the turnover of neuronal mitochondria,researchers generatedTSC2def icient neurons wherein they observed a gradual accumulation of mitochondria in the neuronal cell body[25].However,functional mitochondria were exhausted in axons of these neurons,including those supporting presynaptic sites.In addition,the dynamics of mitochondrial autophagy were reported to be impaired both locally in axons and throughoutTSC2def icient neurons [25],which indicates that reduced turnover could impair mitochondrial metabolism.Additionally,mitochondrial homeostasis was shown to be rescuable by mTORC1 blockade or by inducing mTOR-independent autophagy.These data,therefore,provide evidence for the role of impaired neuronal mitophagy in the pathogenesis of TSC and present a possible means of novel therapeutic intervention.

Another recent study found thatTSC2+/-neurons are less responsive to repulsive cues and defective axon guidance[21].They also reported that chronic mTOR inhibition signif icantly aff ects the ability of normal neurons migration in vitro,thus providing a guideline for the use of mTORC1 inhibitors.More importantly,mTOR-independent TSC2 signaling through Ras homolog family member A (RhoA),a small GTPase,was observed in growth cones [21].These results showed that RhoA-mediated cytoskeletal dynamics during neuronal development may lead to aberrant neural network connectivity in TSC patients.Neurons diff erentiated fromTSC2+/-NPCs have also been shown to exhibit enhanced soma size,abnormal neurite outgrowth,and aberrant intercellular connections [19].Taken together,the literature reviewed above indicates thatTSC1orTSC2mutations cause changes in neuronal morphology,physiology,hyperexcitability,and mitophagy inTSC1/2def icient hiPSC-derived neurons that likely contribute to TSC pathogenesis in human patients.

Human astrocytes models of tuberous sclerosis complex

Disruption of synaptic E/I balance has been observed in TSC def icient neurons [23],and studies in rodent models of TSC have shown that synaptic E/I balance is disrupted as the consequence of hyperactivated mTOR [32].This disruption in synaptic E/I balance likely manifests as the neurological symptoms of TSC patients,such as ASD.Because it is well-established that astrocytes play a central role in neuronal development [33] and postmortem tissue analyses have revealed large numbers of reactive astrocytes in both the cortical and subcortical white matter of TSC patient brains [34],synaptic E/I balance and the associated neurological def icits observed in TSC patients could be the result of astrocytic dysfunction.TSC2-def icient astrocytes have been reported to display elevated saturation density and higher proliferative activity than TSC2-suffi cient controls [19].This result is consistent with previous reports of increased proliferative capacity in astrocytes derived from TSC patients relative to those derived from healthy controls [22].Additional RNA sequencing in these astrocytes uncovered changes in the expression of genes that are associated with epidermal growth factor signaling and those encoding secreted and transmembrane proteins.Moreover,control neurons cultured in TSC astrocyte conditioned medium presented an altered synaptic balance that was accompanied by an enhanced percentage of vesicular GABA transporter synapses [22],suggesting thatTSC1/2def icient astrocytes could secrete factors that aff ect the balance of neuronal synapses.Given the numerous neurological symptoms of TSC that are likely caused by a disrupted E/I balance,astrocytes may off er a new therapeutic target for the restoration of normal neuronal signaling in TSC patients.Along these lines,a combined microformulator and an electrochemical sensor,known as a micro-clinical analyzer,has been designed,that is capable of monitoring the glutamate metabolism of human astrocytes in real-time [26].Studies with this innovative device have revealed that,when challenged,healthy control astrocytes metabolize glutamate at a higher rate thanTSC2-def icient astrocytes [26].This altered glutamate homeostasis may contribute to glutamate-induced excitotoxicity,epilepsy,and intellectual disabilities in TSC patients.

Human oligodendrocytes models of tuberous sclerosis complex

Glial dysfunction,such as enhanced oligodendrocyte (OL)progenitor proliferation and elevated cell death,were found in rodent models of TSC [35,36].Pure TSC patient-derived neuronal cultures and co-cultures with OLs were used to investigate neuron-glia interactions in TSC phenotypes (as is shown in Table 2) [27,28].In TSC def icient neuronal monocultures,elevated neuronal activity was observed.When neurons were co-cultured with OLs,neuronal defects,such as elevated axonal density and hypertrophy,became more apparent.Enhanced OL proliferation and reduced OL maturation were observed in co-cultures of TSC neurons and OLs [28],indicating that a complex cellular phenotype of TSC arises from neuronal and glial cell interactions.Additionally,a protocol for the co-culturing of neurons and oligodendrocyte progenitor cells (OPCs) was also established[27].Using this protocol,they found that when TSC neurons were co-cultured with TSC OPCs,the TSC neuronal abnormalities became more apparent.These two studies demonstrate that neuron and OL co-cultures are a promising tool for exploring the interactive mechanisms involved in TSC.

Human Purkinje cells models of tuberous sclerosis complex

TSC is a neurodevelopmental disorder in which 40% of patients are diagnosed with ASD [1],which is associated with early cerebellar dysfunction [29].To study the molecular mechanism of cerebellar abnormalities in ASD and TSC,a cerebellar Purkinje cell (PC) model was established using TSC patient-derived iPSCs (as is shown in Table 2) [29].In this model,diff erentiation def icits,increased soma size,autophagy activation,oxidative stress,and cell death were observed inTSC2-def icient PCs compared to controls.In addition,RNA sequencing revealed that several components of fragile X mental retardation protein targets,including FMR1,FXR1,TIA1,TIA1L,were reduced inTSC2-def icient PCs [29].The results indicate that these cellular abnormalities may cause PCs dysfunction in TSC patients.

The progress of 3D culture techniques,such as organoids or spheroids,has furthered the investigation of neurodevelopment disorders,including Fragile X syndrome[37-39],ASD [40-42],and microcephaly [43-45].These advanced 3D cultures enable more interactions between neuronal cells and other neural cell types,such as glia,compared to 2D cultures.They also enhance cell survival,allow the formation of longer neurites and additional network density,as well as more closely mimic in vitro human development and organ regeneration.Recently,organoids and spheroids have been used to explore the pathogenesis of TSC (as is shown in Table 2) [4,22,30].To identify howTSC1andTSC2mutations affect neuronal development in the 3D spheroids,cortical spheroids were generated from hESCs [4].InTSC1orTSC2heterozygous spheroids,they found that neurons and glia development show a normal profile.However,TSC1andTSC2homozygous spheroids presented elevated expression of glial markers,as well as decreased or delayed expression of neuronal markers.In addition,enlarged and dysmorphic neurons and glia were found in theseTSC1andTSC2homozygous spheroids.These results indicate that loss ofTSC1orTSC2significantly affects the development of neurons and glia in human cortical spheroids[4].Along these same lines,a TSC organoid model has been successfully established that exhibited a high number of GABAergic synapses and a lower number of glutamatergic synapses relative to control organoids [22].These findings may explain the synaptic balance changes and associated neurological deficits that are common in TSC patients.

Most recently,cerebral organoids were generated wherein the overproduction of mid-gestational human interneurons reportedly lead to the development of TSC[31].In these TSC cerebral organoids,a new population of caudal late interneuron progenitor (CLIP) cells was identif ied.InTSC2heterozygous organoids,the over-proliferation of CLIP cells causes the formation of cortical tubers and subependymal tumors as the result of mTOR hyperactivation [31].This provides a new insight to elaborate abnormalities of TSC and related neurodevelopmental disorders.In addition to the brain organoids reviewed above,kidney organoids were developed from TSC patient-specif ic hiPSCs to model angiomyolipoma (AML),which is a rare kidney tumor.TheseTSC2-/-AML organoids exhibited individual cysts [30],thus recapitulating a major feature of cystic kidney disease associated with TSC.To investigate tumor mechanisms of AML in vivo,they transplantedTSC2homozygous mutant renal organoids into the kidneys of immunodef icient rats,thus allowing the comparison of xenografts from diff erent genotypes within the same animal.TheTSC2-/-AML organoid xenografts showed signif icantly higher growth rates compared to heterozygousTSC2organoids and wild-type organoids [30].These studies indicate that three-dimensional cultures,such as spheroids or organoids,provide valuable means of exploring the pathogenesis of TSC and other developmental diseases.

With the rapid development of technology,iPSCs are proving to be an excellent tool for studying neurodevelopmental and neural diseases.In this review,we summarize the recent progress in modeling TSC using 2D and 3D cultures derived from hiPSCs.The above literature reveals that proliferative ability was increased in TSC patient-derived NPCs [17,19],astrocytes [22],oligodendrocytes [28],CLIP cells [31].Additionally,TSC patient-derived NPCs [17],neurons in 2D culture [19],as well as neurons and glia generated inTSC1-/-andTSC2-/-spheroids have all demonstrated enlarged somas[4].These results are consistent with the abnormal cellular proliferation and size that has been observed in pathological TSC lesions.Previous research has demonstrated that biallelic mutations ofTSC2are necessary to induce gene expression dysregulation in neurons [24] and that loss of both alleles ofTSC1orTSC2is required to alter diff erentiation and dysmorphic cell emergence in spheroids [4].However,several studies have produced neuronal abnormalities with heterozygousTSC2mutations [18,19,21,23,24,28].Therefore,additional experiments are needed to determine whether biallelic mutations are necessary to model diff erential expression gene expression in TSC.

Approximately 40% of TSC patients are aff ected with ASD.As previously mentioned,genotype-dependent changes in the expression of ASD,epilepsy,and intellectual disability associated-genes through transcriptome analysis inTSC1NPCs [17].ASD patient-derived neurons withTSC2gene mutations exhibit hyperactivity,leading to neuronal network dysfunction,as characterized by decreased synchronicity of neuronal bursting and lower spatial connectivity[23].However,to date,the etiology of ASD in TSC patients remains insuffi ciently illuminated.Therefore,it is urgent to establish iPSC lines from TSC patients with comorbid ASD to determine the impact ofTSC1/2mutations on neural development.

Furthermore,the 2D cultures have thus far included NPCs,neurons,Purkinje cells,and glial cells,such as astrocytes,and oligodendrocytes,which play a vital role in the pathology of TSC.However,human iPSC-derived microglia have yet to be included in 2D models of TSC.Examination of postmortem tissue has revealed activated microglia in the brains of TSC patients [34].Microglia are the immune cells of the central nervous system,and they are known to regulate brain development by modulating neurogenesis,neuron migration,synaptic maturity,connection,and pruning [46,47].Neuroinf lammation is a proven hallmark of TSC-related lesions [34] and microglial abnormalities have also been found in rodent models of TSC [48,49].These f indings suggest that microglia may play a role in the neurologic manifestations of TSC.Therefore,human iPSC-derived microglia are an important next step in the study of TSC pathogenesis.However,cortical organoids lack microglia,the immune cells in the brain,which originate from a non-ectodermal cell lineage [50].Therefore,co-cultures of microglia and brain organoids are required to study the role of microglia in the pathogenesis of TSC.Microglia and brain organoid co-culture models have been developed to study human microglial function in human brain development and disease [51-54],which could be used to explore the contribution of microglia to the pathogenesis of TSC.

Altered glutamatergic and GABAergic neurotransmission results in E/I imbalance,which is thought to underlie the pathogenesis of TSC.Brain assembloids can be used to simulate E/I balances that may involve network physiological alterations in neurodevelopmental disorders [55].Moreover,fused cerebral organoids can be used to model interactions between diff erent brain regions [55],such as the fusing dorsal and ventral forebrain organoids which contain cortical glutamatergic or GABAergic neurons [55-58] and thus to simulate E/I balances.These or other brain assembloids could be leveraged to investigate the pathogenesis of TSC.

Microf luidic 3D cell culture,based on tissue-on-a-chip devices,has also shown great potential for human disease research and drug screening [59-62].Most recently,an automated,high-throughput microf luidic 3D culture with the periodic f low has been developed to improve the uniformity and survival of organoids.This device has been used to signif icantly increase the neurogenesis of hiPSCs in developing brain organoids as well as in the extracellular matrix.Furthermore,dynamic culture in microf luidic chamber equipment has been used to improve the development of cerebral organoids,resulting in increased volume,and enhanced electrophysiological function [59].Additionally,a fully integrated,modular physical,biochemical,and optical sensing platform has also been established,which uses a f luid path breadboard to operate on organ-on-chip units in a continuous,dynamic,and automated manner [62].Through which dynamic drug therapy can be supplied by microf ulidic 3D cell culture and organoids can be monitored in real-time.These microf luidic 3D cell culture technologies could be used to screen drugs for patients with TSC,thus facilitating pre-clinical research and the development of precise therapies for TSC.

Modeling TSC using 2D and 3D cultures with hiPSC has limitations but remains the most promising tool for personalized disease modeling and the study of neurodevelopmental disorders.Moreover,the constant technological progress in this f ield continues to address the current limitations of these models.In the meantime,the integration of animal models,iPSC models,and clinical f indings provides a means of comprehensive data analysis.This combinatory approach promises to provide new insights into the pathogenesis and prevention of TSC,as well as the development of precision medicine for the treatment of TSC.

AcknowledgementsWe thank the support from the following funding sources: NIH grants R01AG065611,R01MH121102,R21MH123711,and Department of Defense grant W81XWH1910353 to ZW.

Author contributionsNW wrote the manuscript.SB helped with the revision.WZ contributed to manuscript revision,read,and approved the submitted version.

FundingThis work was supported by the following funding sources:NIH grants (Nos.R01AG065611,R01MH121102,R21MH123711),and Department of Defense grant (No.W81XWH1910353 to ZW).

Data availabilityAll data and articles supporting this review are available within the article in the reference section.

Ethical approvalNot needed.

Conflict of interestNo f inancial or non-f inancial benef its have been received or will be received from any party related directly or indirectly to the subject of this article.The authors have no conf lict of interest to declare.